HCM Vol 4

HIGHWAY CAPACITY
MANUAL 7th Edition

A Guide for Multimodal Mobility Analysis
Transportation Research Board publications are available by ordering individual publications
directly from the TRB Business Office, through the Internet at www.TRB.org or
nationalacademies.org/trb, or by annual subscription through organizational or individual
affiliation with TRB. Affiliates and library subscribers are eligible for substantial discounts. For
further information, contact the Transportation Research Board Business Office, 500 Fifth
Street, NW, Washington, DC 20001 (telephone 202-334-3213; fax 202-334-2519; or e-mail
TRBsales@nas.edu).
Copyright 2022 by the National Academy of Sciences. All rights reserved.
Printed in the United States of America
Hardcover International Standard Book Number-13: 978-0-309-08766-7

Volume 1 International Standard Book Number-13: 978-0-309-27566-8
eBook International Standard Book Number-13: 978-0-309-27562-0
Digital Object Identifier: https://doi.org/10.17226/26432
Library of Congress Control Number: 2022930290
Suggested citation: National Academies of Sciences, Engineering, and Medicine. 2022. Highway
Capacity Manual 7th Edition: A Guide for Multimodal Mobility Analysis. Washington, DC: The
National Academies Press. https://doi.org/10.17226/26432.
The National Academy of Sciences was established in 1863 by an Act of Congress,
signed by President Lincoln, as a private, nongovernmental institution to advise the
nation on issues related to science and technology. Members are elected by their peers
for outstanding contributions to research. Dr. Marcia McNutt is president.
The National Academy of Engineering was established in 1964 under the charter of the
National Academy of Sciences to bring the practices of engineering to advising the
nation. Members are elected by their peers for extraordinary contributions to
engineering. Dr. John L. Anderson is president.
The National Academy of Medicine (formerly the Institute of Medicine) was established
in 1970 under the charter of the National Academy of Sciences to advise the nation on
medical and health issues. Members are elected by their peers for distinguished
contributions to medicine and health. Dr. Victor J. Dzau is president.
The three Academies work together as the National Academies of Sciences,

Engineering, and Medicine to provide independent, objective analysis and advice to the
nation and conduct other activities to solve complex problems and inform public policy
decisions. The Academies also encourage education and research, recognize
outstanding contributions to knowledge, and increase public understanding in matters
of science, engineering, and medicine.
Learn more about the National Academies of Sciences, Engineering, and Medicine at
www.nationalacademies.org.
The Transportation Research Board is one of seven major programs of the National
Academies of Sciences, Engineering, and Medicine. The mission of the Transportation
Research Board is to provide leadership in transportation improvements and innovation
through trusted, timely, impartial, and evidence-based information exchange,
research, and advice regarding all modes of transportation. The Board’s varied
activities annually engage about 8,000 engineers, scientists, and other transportation
researchers and practitioners from the public and private sectors and academia, all of
whom contribute their expertise in the public interest. The program is supported by
state departments of transportation, federal agencies including the component
administrations of the U.S. Department of Transportation, and other organizations and
individuals interested in the development of transportation.
Learn more about the Transportation Research Board at www.TRB.org.

Highway Capacity Manual: A Guide for Multimodal Mobility Analysis
CHAPTER 25
FREEWAY FACILITIES: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 25-1
Chapter Scope ...................................................................................................... 25-1
Chapter Organization ......................................................................................... 25-1
Limitations of the Methodologies ..................................................................... 25-1
2. GLOSSARY OF VARIABLE DEFINITIONS .................................................... 25-3

Overview .............................................................................................................. 25-3
Global Variables ................................................................................................... 25-3
Segment Variables ............................................................................................... 25-4
Node Variables..................................................................................................... 25-5
On-Ramp Variables ............................................................................................. 25-6
Off-Ramp Variables ............................................................................................. 25-6
Facilitywide Variables ......................................................................................... 25-6
Travel Time Reliability Variables ...................................................................... 25-7
3. UNDERSATURATED SEGMENT EVALUATION ........................................ 25-9

Facility Speed Constraint .................................................................................... 25-9
Directional Facility Module.............................................................................. 25-10
4. OVERSATURATED SEGMENT EVALUATION .......................................... 25-11

Procedure Parameters ....................................................................................... 25-11
Flow Estimation ................................................................................................. 25-13
Segment and Ramp Performance Measures .................................................. 25-25
Oversaturation Analysis within Managed Lanes.......................................... 25-26
5. WORK ZONE ANALYSIS DETAILS ............................................................... 25-28

Special Work Zone Configurations ................................................................. 25-28
6. PLANNING-LEVEL METHODOLOGY FOR FREEWAY FACILITIES .... 25-34

Input Requirements ........................................................................................... 25-34
Step 1: Demand-Level Calculations ................................................................ 25-36
Step 2: Section Capacity Calculations and Adjustments .............................. 25-37
Step 3: Delay Rate Estimation .......................................................................... 25-38
Step 4: Average Travel Time, Speed, and Density Calculations ................. 25-39
Step 5: Level of Service ..................................................................................... 25-40
Chapter 25/Freeway Facilities: Supplemental Contents

Version 7.0 Page 25-i
7. MIXED-FLOW MODEL FOR COMPOSITE GRADES ................................25-41

Overview of the Methodology .........................................................................25-41
Step 1: Input Data ..............................................................................................25-41
Step 2: Capacity Assessment ............................................................................25-41
Step 3: Specify Initial Conditions .....................................................................25-44
Step 4: Compute Truck Spot and Space-based Travel Time Rates ................25-44
Step 5: Compute Automobile Spot and Space-Based Travel Time Rates ...25-50
Step 6: Compute Mixed-Flow Space-Based Travel Time Rate and
Speed ............................................................................................................25-51
Step 7: Overall Results ......................................................................................25-52
8. FREEWAY CALIBRATION METHODOLOGY .............................................25-53

Calibration at the Core Freeway Facility Level .............................................25-53
Calibration at the Travel Time Reliability Level ............................................25-60
Calibration at the Reliability Strategy Assessment Level .............................25-65
9. FREEWAY SCENARIO GENERATION ..........................................................25-68

Introduction ........................................................................................................25-68
Methodology ......................................................................................................25-71
10. COMPUTATIONAL ENGINE OVERVIEW ...............................................25-84
11. EXAMPLE PROBLEMS ...................................................................................25-85

Example Problem 1: Evaluation of an Undersaturated Facility ..................25-85
Example Problem 2: Evaluation of an Oversaturated Facility .....................25-92
Example Problem 3: Capacity Improvements to an Oversaturated
Facility ..........................................................................................................25-97
Example Problem 4: Evaluation of an Undersaturated Facility with a
Work Zone .................................................................................................25-102
Example Problem 5: Evaluation of an Oversaturated Facility with a
Managed Lane ...........................................................................................25-108
Example Problem 6: Planning-Level Analysis of a Freeway Facility .......25-113
Example Problem 7: Reliability Evaluation of an Existing Freeway
Facility ........................................................................................................25-118
Example Problem 8: Reliability Analysis with Geometric
Improvements ...........................................................................................25-122
Example Problem 9: Evaluation of Incident Management ........................25-123
Example Problem 10: Planning-Level Reliability Analysis ........................25-124
Example Problem 11: Estimating Freeway Composite Grade
Operations with the Mixed-Flow Model ...............................................25-125
12. REFERENCES..................................................................................................25-135
APPENDIX A: TRUCK PERFORMANCE CURVES .......................................25-137
Contents Chapter 25/Freeway Facilities: Supplemental

Page 25-ii Version 7.0
LIST OF EXHIBITS
Exhibit 25-1 Node–Segment Representation of a Directional Freeway

Facility ................................................................................................................. 25-11
Exhibit 25-2 Segment Flow–Density Function ...................................................... 25-13
Exhibit 25-3 Oversaturated Analysis Procedure .................................................. 25-14
Exhibit 25-4 Definitions of Mainline and Segment Flows ................................... 25-19
Exhibit 25-5 Flow–Density Function with a Shock Wave ................................... 25-21
Exhibit 25-6 Vertical Queuing from a Managed Lane Due to Queue
Presence on the General Purpose Lanes ......................................................... 25-27
Exhibit 25-7 On-Ramp Merge Diagram for 2-to-1 Freeway Work Zone
Configuration ..................................................................................................... 25-28
Exhibit 25-8 Proportion of Work Zone Queue Discharge Rate (Relative to
the Basic Work Zone Capacity) Available for Mainline Flow
Upstream of Merge Area .................................................................................. 25-29
Exhibit 25-9 Off-Ramp Diverge Diagram for a 2-to-1 Freeway Work Zone
Configuration ..................................................................................................... 25-30
Exhibit 25-10 Proportion of Work Zone Capacity Available for Mainline
Flow Downstream of Diverge Area ................................................................ 25-31
Exhibit 25-11 Proportion of Off-Ramp Demand Served in Work Zone ............ 25-31
Exhibit 25-12 Proportion of Available Work Zone Capacity for a
Directional Crossover in the Work Zone ........................................................ 25-31
Exhibit 25-13 Model Coefficients for Estimating the Proportion of Work
Zone Capacity in a Weaving Segment ............................................................ 25-33
Exhibit 25-14 Model Coefficients for Estimating the Proportion of Off-
Ramp Volume Served in the Weaving Area .................................................. 25-33
Exhibit 25-15 Schematics of Freeway Sections ...................................................... 25-34
Exhibit 25-16 Parameter Values for Undersaturated Model ............................... 25-39
Exhibit 25-17 LOS Criteria for Urban and Rural Freeway Facilities .................. 25-40
Exhibit 25-18 Schematic of a Composite Grade .................................................... 25-41
Exhibit 25-19 Mixed-Flow Methodology Overview............................................. 25-42
Exhibit 25-20 SUT Spot Rates Versus Distance with Initial Speeds of 75
and 30 mi/h ......................................................................................................... 25-45
Exhibit 25-21 TT Spot Rates Versus Distance with Initial Speeds of 75 and
20 mi/h................................................................................................................. 25-45
Exhibit 25-22 SUT Travel Time Versus Distance Curves for 70-mi/h Initial
Speed ................................................................................................................... 25-47
Exhibit 25-23 SUT Travel Time Versus Distance Curves for 30-mi/h Initial
Speed ................................................................................................................... 25-47
Exhibit 25-24 δ Values for SUTs .............................................................................. 25-48

Version 7.0 Page 25-iii
Exhibit 25-25 δ Values for TTs .................................................................................25-48

Exhibit 25-26 Calibration Steps for the Core Freeway Facility Level.................25-54
Exhibit 25-27 Effect of Calibrating Free-Flow Speed on Capacity......................25-55
Exhibit 25-28 Effects of Segment Capacity ............................................................25-56
Exhibit 25-29 Effects of Queue Discharge Rate Drop ...........................................25-56
Exhibit 25-30 Effects of Jam Density .......................................................................25-57
Exhibit 25-31 Effect of Demand Level ....................................................................25-58
Exhibit 25-32 Comprehensive Reliability Calibration Steps ...............................25-60
Exhibit 25-33 High Demand Level on the Seed Day ............................................25-62
Exhibit 25-34 Low Demand Level on the Seed Day .............................................25-62
Exhibit 25-35 Overestimating the Impacts of Nonrecurring Sources of
Congestion ..........................................................................................................25-63
Exhibit 25-36 Underestimating the Impacts of Nonrecurring Sources of
Congestion ..........................................................................................................25-64
Exhibit 25-37 Process Flow Overview for Freeway Scenario Generation .........25-69
Exhibit 25-38 Distribution of Number of Incidents in the Scenarios .................25-70
Exhibit 25-39 Detailed Freeway Scenario Generation Flowchart .......................25-72
Exhibit 25-40 Listing of Weather Stations with Available Weather Data ..........25-76
Exhibit 25-41 Incident Duration Distribution Parameters in Minutes ...............25-81
Exhibit 25-42 List of Example Problems ................................................................25-85
Exhibit 25-43 Example Problem 1: Freeway Facility ............................................25-85
Exhibit 25-44 Example Problem 1: Geometry of Directional Freeway
Facility .................................................................................................................25-85
Exhibit 25-45 Example Problem 1: Demand Inputs .............................................25-87
Exhibit 25-46 Example Problem 1: Segment Capacities .......................................25-88
Exhibit 25-47 Example Problem 1: Segment Demand-to-Capacity Ratios ........25-89
Exhibit 25-48 Example Problem 1: Volume-Served Matrix .................................25-89
Exhibit 25-49 Example Problem 1: Speed Matrix..................................................25-90
Exhibit 25-50 Example Problem 1: Density Matrix ...............................................25-90
Exhibit 25-51 Example Problem 1: LOS Matrix.....................................................25-90
Exhibit 25-52 Example Problem 1: Facility Performance Measure
Summary .............................................................................................................25-92
Exhibit 25-53 Example Problem 2: Demand Inputs .............................................25-93
Exhibit 25-54 Example Problem 2: Segment Capacities .......................................25-94
Exhibit 25-55 Example Problem 2: Segment Demand-to-Capacity Ratios ........25-95
Exhibit 25-56 Example Problem 2: Volume-Served Matrix .................................25-96
Exhibit 25-57 Example Problem 2: Speed Matrix..................................................25-96
Exhibit 25-58 Example Problem 2: Density Matrix ...............................................25-96

Page 25-iv Version 7.0
Exhibit 25-59 Example Problem 2: Expanded LOS Matrix.................................. 25-96

Summary............................................................................................................. 25-97
Exhibit 25-61 Example Problem 3: Freeway Facility ............................................ 25-98
Facility ................................................................................................................. 25-98
Exhibit 25-63 Example Problem 3: Segment Capacities ..................................... 25-100
Exhibit 25-64 Example Problem 3: Segment Demand-to-Capacity Ratios ...... 25-100
Exhibit 25-65 Example Problem 3: Speed Matrix ............................................... 25-101
Exhibit 25-66 Example Problem 3: Density Matrix ............................................ 25-101
Exhibit 25-67 Example Problem 3: LOS Matrix .................................................. 25-101
Summary........................................................................................................... 25-102
Exhibit 25-69 Example Problem 4: Freeway Facility .......................................... 25-102
Facility ............................................................................................................... 25-102
Exhibit 25-73 Example Problem 4: Volume-Served Matrix ............................... 25-106
Summary........................................................................................................... 25-107
Exhibit 25-78 Example Problem 5: Freeway Facility .......................................... 25-108
Facility ............................................................................................................... 25-108
Exhibit 25-80 Example Problem 5: Demand Inputs on the Mainline ............... 25-109
Summary for Lane Groups ............................................................................. 25-112
Summary........................................................................................................... 25-112
Exhibit 25-88 Example Problem 6: AADT Values for the Facility .................... 25-113
Exhibit 25-89 Example Problem 6: Section Definition for the Facility ............. 25-114

Version 7.0 Page 25-v
Exhibit 25-90 Example Problem 6: Demand Flow Rates (pc/h) on the

Subject Facility .................................................................................................25-114
Exhibit 25-91 Example Problem 6: Demand-to-Capacity Ratios by Section
and Analysis Period ........................................................................................25-115
Exhibit 25-92 Example Problem 6: Delay Rates by Section and Analysis
Period ................................................................................................................25-116
Exhibit 25-93 Example Problem 6: Travel Rates by Section and Analysis
Period ................................................................................................................25-117
Exhibit 25-94 Example Problem 6: Average Travel Times by Section and
Analysis Period ................................................................................................25-117
Exhibit 25-95 Example Problem 6: Density by Section and Analysis
Period ................................................................................................................25-117
Exhibit 25-96 Example Problem 6: Facility Performance Summary.................25-117
Exhibit 25-97 Example Problem 7: Freeway Facility ..........................................25-118
Facility ...............................................................................................................25-118
Exhibit 25-99 Example Problem 7: Demand Flow Rates (veh/h) by
Analysis Period in the Base Data Set.............................................................25-119
Exhibit 25-100 Example Problem 7: Demand Ratios Relative to AADT..........25-120
Exhibit 25-101 Example Problem 7: Weather Event Probabilities by
Season ................................................................................................................25-120
Exhibit 25-102 Example Problem 7: CAF, SAF, and Event Duration
Values Associated with Weather Events ......................................................25-120
Exhibit 25-103 Example Problem 7: Incident Frequencies by Month ..............25-121
Exhibit 25-104 Example Problem 7: Summary Reliability Performance
Measure Results ...............................................................................................25-121
Exhibit 25-105 Example Problem 7: VMT-Weighted TTI Probability and
Cumulative Distribution Functions ..............................................................25-122
Exhibit 25-106 Example Problem 8: Freeway Facility ........................................25-122
Measure Results ...............................................................................................25-123
Measure Results ...............................................................................................25-124
Exhibit 25-109 Example Problem 11: Spot Speeds of All Segments .................25-134
Exhibit 25-110 Example Problem 11: Space Mean Speeds of All Segments ....25-134
Exhibit 25-111 Example Problem 11: Overall Space Mean Speeds of All
Segments ...........................................................................................................25-134
Exhibit 25-A1 SUT Travel Time Versus Distance Curves for 35-mi/h
Initial Speed ......................................................................................................25-137
Initial Speed ......................................................................................................25-137

Page 25-vi Version 7.0
Initial Speed ...................................................................................................... 25-138
Initial Speed ...................................................................................................... 25-138
Initial Speed ...................................................................................................... 25-139
Initial Speed ...................................................................................................... 25-139
Initial Speed ...................................................................................................... 25-140
Initial Speed ...................................................................................................... 25-140
Exhibit 25-A9 TT Travel Time Versus Distance Curves for 20-mi/h Initial
Speed ................................................................................................................. 25-141
Exhibit 25-A10 TT Travel Time Versus Distance Curves for 25-mi/h
Initial Speed ...................................................................................................... 25-141
Initial Speed ...................................................................................................... 25-142
Initial Speed ...................................................................................................... 25-142
Initial Speed ...................................................................................................... 25-143
Initial Speed ...................................................................................................... 25-143
Initial Speed ...................................................................................................... 25-144
Initial Speed ...................................................................................................... 25-144
Initial Speed ...................................................................................................... 25-145
Initial Speed ...................................................................................................... 25-145
Initial Speed ...................................................................................................... 25-146
Initial Speed ...................................................................................................... 25-146

Version 7.0 Page 25-vii
1. INTRODUCTION
CHAPTER SCOPE VOLUME 4: APPLICATIONS

GUIDE
Chapter 25 is the supplemental chapter for Chapter 10, which describes the 25. Freeway Facilities:
Supplemental
core methodology for freeway facilities, and Chapter 11, which presents a 26. Freeway and Highway
methodology for evaluating freeway reliability and active traffic and demand Segments: Supplemental
27. Freeway Weaving:
management (ATDM) strategies. The computations used by these methodologies Supplemental
are detailed in this supplemental chapter. The documentation is closely tied to 28. Freeway Merges and
Diverges: Supplemental
FREEVAL-2015E, the computational engine for Chapter 10 and Chapter 11. 29. Urban Street Facilities:
Supplemental
The FREEVAL (FREeway EVALuation) tool was initially developed for the 30. Urban Street Segments:
2000 edition of the Highway Capacity Manual (HCM) (1, 2) and has been updated Supplemental
31. Signalized Intersections:
to reflect subsequent methodological changes in the HCM. All variable Supplemental
definitions and subroutine labels presented in this chapter are consistent with the 32. STOP-Controlled
Intersections:
computational code in FREEVAL-2015E. The Technical Reference Library in
Supplemental
Volume 4 contains a FREEVAL-2015E user guide, which provides more details 33. Roundabouts:
Supplemental
on how to use the computational engine. Other software implementations of this
34. Interchange Ramp
method are available and can be used instead of the computational engine. Terminals: Supplemental
35. Pedestrians and Bicycles:
Supplemental
CHAPTER ORGANIZATION 36. Concepts: Supplemental
37. ATDM: Supplemental
Section 2 presents a glossary of all relevant variables used in the procedures 38. Network Analysis
and the computational engine. Section 3 and Section 4, respectively, provide
details of the undersaturated and oversaturated flow procedures. Section 5
describes details for work zone analysis. Section 6 develops the planning-level
methodology for freeway facilities, and Section 7 discusses the mixed-flow
model for composite grades. Section 8 develops the freeway calibration
methodology at three levels. Section 9 discusses freeway scenario generation,
and Section 10 presents an overview of the computational engine structure.
Example problems are presented in Section 11, and Section 12 provides
references for the chapter.
LIMITATIONS OF THE METHODOLOGIES

The completeness of the analysis will be limited if freeway segment cells in
the first time interval, the final time interval, and the first freeway segment do
not have demand-to-capacity ratios of 1.00 or less. The methodology can handle
congestion in the first interval properly, although it will not quantify any
congestion that could have occurred before the first time interval. To ensure a
complete quantification of the effects of congestion, it is recommended that the
analysis contain an initial undersaturated time interval. If all freeway segments
in the final time interval do not exhibit demand-to-capacity ratios less than 1.00,
congestion will continue beyond the final time interval, and additional time
intervals should be added. This fact will be noted as a difference between the
vehicle miles of travel desired at the end of the analysis (demand flow) and the
corresponding vehicle miles of travel flow generated (volume served). If queues
extend upstream of the first segment, the analysis will not account for the
congestion outside the freeway facility but will store the vehicles vertically until
Chapter 25/Freeway Facilities: Supplemental Introduction

Version 7.0 Page 25-1
the congestion clears the first segment. The same process is followed for queues
on on-ramp segments.
The methodology for oversaturated conditions described in this chapter is
based on concepts of traffic flow theory and assumes a linear speed–flow
relationship for densities greater than 45 passenger cars per mile per lane
(pc/mi/ln). This relationship has not been extensively calibrated for field
observations on U.S. freeways, and analysts should therefore perform their own
validation from local data to obtain additional confidence in the results of this
procedure. For an example of a validation exercise for this methodology, the
reader is referred elsewhere (3).
The procedure described here becomes extremely complex when the queue
from a downstream bottleneck extends into an upstream bottleneck, causing a
queue interaction. When such cases arise, the reliability of the methodology is
questionable, and the user is cautioned about the validity of the results. For
heavily congested directional freeway facilities with interacting bottleneck
queues, a traffic simulation model might be more applicable. Noninteracting
bottlenecks are addressed by the methodology.
The procedure focuses on analyzing a directional series of freeway segments.
It describes the performance of a facility but falls short of addressing the broader
transportation network. The analyst is cautioned that severe congestion on a
freeway—especially freeway on-ramps—is likely to affect the adjacent surface
street network. Similarly, the procedure is limited in its ability to predict the
impacts of an oversaturated off-ramp and the associated queues that may spill
back onto the freeway. Alternative tools are suitable to evaluate these impacts.
Introduction Chapter 25/Freeway Facilities: Supplemental

Page 25-2 Version 7.0
2. GLOSSARY OF VARIABLE DEFINITIONS
OVERVIEW
This glossary defines internal variables used exclusively in the freeway
facilities methodology. The variables are consistent with those used in the
computational engine for the freeway facilities methodology.
If a managed lane facility is adjacent to the general purpose lanes, the
oversaturated freeway facilities methodology will analyze each facility
independently. As a result, the variables presented in this chapter will pertain to
general purpose and managed lane facilities separately.
The glossary of variables is presented in seven parts: global variables,
segment variables, node variables, on-ramp variables, off-ramp variables,
facilitywide variables, and travel time reliability variables. Global variables are
used across multiple aspects of the procedure. Segment variables represent
conditions on segments. Node variables denote flows across a node connecting
two segments. On- and off-ramp variables correspond to flow on ramps.
Facilitywide variables pertain to aggregate traffic performance over the entire
general purpose or managed lane facility. Reliability variables pertain to traffic
performance over a period of up to one year.
In addition to the spatial categories listed above, there are temporal divisions
that represent characteristics over a time step for oversaturated conditions or an
analysis period for undersaturated conditions. The first dimension associated
with each variable specifies whether the variable refers to segment or node
characteristics. The labeling scheme for nodes and segments is such that segment
i is immediately downstream of node i. The distinction of nodes and segments is
used primarily in the oversaturated flow regime as discussed in Section 4.
Thus, there is always one more node than the number of segments on a
facility. The second and third dimensions denote a time step t and a time interval
p. Facility variables are estimates of the average performance over the length of
the facility. The units of flow are in vehicles per time step. The selection of the
time step size is discussed later in this chapter.
The variable symbols used internally by the computational engine and
replicated in this chapter frequently differ from the symbols used elsewhere in
the HCM, particularly in Chapter 10, Freeway Facilities Core Methodology. For
example, the HCM uses n to represent the number of segments forming a facility,
whereas the computational engine and this chapter use NS.
GLOBAL VARIABLES
• i—index to segment or node number: i = 1, 2, . . . , NS (for segments) and
i = 1, 2, ..., NS + 1 (for nodes). In the computational engine, i is represented
as the index of the GPSegments/MLSegments Array List variable in the Seed
class.
• KC—ideal density at capacity (pc/mi/ln). The density at capacity is 45
pc/mi/ln.
Chapter 25/Freeway Facilities: Supplemental Glossary of Variable Definitions

• KJ—facilitywide jam density (pc/mi/ln).

• NS—number of segments on the facility. NS is represented as the size of
the GPSegments/MLSegments ArrayList variable in the Seed class.
• P—number of (15-min) analysis periods in the study period. Represented
as Period in the computational engine. For a 24-h analysis, the theoretical
maximum is 96 analysis periods.
• p—analysis period index: p = 1, 2, . . . , P.
• S—number of computational time steps in an analysis period (integer). S
is represented as Step in the computational engine. S is set as a constant of
60 in the computational engine, corresponding to a 15-s interval and
allowing a minimum segment length of 300 ft.
• t—time step index in a single analysis period: t = 1, 2, . . . , S.
• T—number of time steps in 1 h (integer). T is set as a constant of 240 in the
computational engine, or equal to four times the value of S.
• α—fraction of capacity drop in queue discharge conditions due to
congestion on the facility. This variable is represented as
inCapacityDropPercentage in the GPMLSegment class in the computational
engine.
SEGMENT VARIABLES
• ED(i, p)—expected demand (veh/h) that would arrive at segment i on the
basis of upstream conditions over time interval p. The upstream queuing
effects include the metering of traffic from an upstream queue but not the
spillback of vehicles from a downstream queue.
• K(i, p)—average traffic density (veh/mi/ln) of segment i over time interval
p as estimated by the oversaturated procedure. This variable is
represented as the scenAllDensity_veh variable in the GPMLSegment class
in the computational engine.
• KB(i, p)—background density: segment i density (veh/mi/ln) over time
interval p assuming there is no queuing on the segment. This density is
calculated by using the expected demand on the segment in the
corresponding undersaturated procedure in Chapters 12, 13, and 14.
• KQ(i, t, p)—queue density: vehicle density (veh/mi/ln) in the queue on
segment i during time step t in time interval p. Queue density is
calculated on the basis of a linear density–flow relationship in the
congested regime.
• L(i)—length of segment i (mi). This variable converts the inSegLength_ft
variable (in feet) to miles when necessary in equations.
• N(i, p)—number of lanes on segment i in time interval p. It could vary by
time interval if a temporary lane closure is in effect. N is represented as
the inMainlineNumLanes variable in the GPMLSegment class in the
computational engine.
• NV(i, p)—number of vehicles present on segment i at the end of time
interval p (veh).
Glossary of Variable Definitions Chapter 25/Freeway Facilities: Supplemental

• NV(i, t, p)—number of vehicles present on segment i at the end of time

step t during time interval p. The number of vehicles is initially based on
the calculations of Chapters 12, 13, and 14, but, as queues grow and
dissipate, input–output analysis updates these values in each time step.
• Q(i, t, p)—total queue length on segment i at the end of time step t in time
interval p (ft).
• SC(i, t, p)—segment capacity: maximum number of vehicles that can pass
through segment i at the end of time step t in time interval p based strictly
on traffic and geometric properties (veh).
• SD(i, p)—segment demand: desired flow rate (veh/h) through segment i
including on- and off-ramp demands in time interval p (veh). This
segment demand is calculated without any capacity constraints. It is
represented as the scenMainlineDemand_veh variable in the GPMLSegment
class in the computational engine.
• SF(i, p)—segment flow out of segment i in time interval p (veh/h).
• SF(i, t, p)—segment flow out of segment i during time step t in time
interval p (veh/time step).
• U(i, p)—average space mean speed over the length of segment i during
time interval p (mi/h). It is represented as the scenSpeed variable in the
GPMLSegment class in the computational engine.
• UV(i, t, p)—unserved vehicles: the additional number of vehicles stored
on segment i at the end of time step t in time interval p due to a
downstream bottleneck.
• WS(i, p)—wave speed: speed at which a front-clearing queue shock wave
travels through segment i during time interval p (mi/h).
• WTT(i, p)—wave travel time: time taken by the shock wave traveling at
wave speed WS to travel from the downstream end of segment i to the
upstream end of the segment during time interval p, in time steps.
NODE VARIABLES
• MF(i, t, p)—actual mainline flow rate that can cross node i during time
step t in time interval p.
• MI(i, t, p)—maximum mainline input: maximum flow desiring to enter
node i during time step t in time interval p, based on flows from all
upstream segments and taking into account all geometric and traffic
constraints upstream of the node, including queues accumulated from
previous time intervals.
• MO1(i, t, p)—maximum Mainline Output 1: maximum allowable mainline
flow rate across node i during time step t in time interval p, limited by the
flow from an on-ramp at node i.
flow rate across node i during time step t in time interval p, limited by
available storage on segment i due to a downstream queue.


flow rate across node i during time step t in time interval p, limited by the
presence of queued vehicles at the upstream end of segment i while the
queue clears from the downstream end of segment i.
ON-RAMP VARIABLES
• ONRC(i, p)—geometric carrying capacity of on-ramp at node i during
time interval p.
• ONRD(i, p)—demand flow rate for on-ramp at node i in time interval p.
• ONRF(i, t, p)—actual ramp flow rate that can cross on-ramp node i during
time step t in time interval p; it takes into account control constraints (e.g.,
ramp meters).
• ONRI(i, t, p)—input flow rate desiring to enter the merge point at on-
ramp i during time step t in time interval p, based on current ramp
demand and ramp queues accumulated from previous time intervals.
• ONRO(i, t, p)—maximum output flow rate that can enter the merge point
from on-ramp i during time step t in time interval p; it is constrained by
Lane 1 (shoulder lane) flow on segment i and the segment i capacity or by
a queue spillback filling the mainline segment from a bottleneck further
downstream, whichever governs.
• ONRQ(i, t, p)—unmet demand that is stored as a queue on the on-ramp
roadway at node i during time step t in time interval p (veh).
• RM(i, p)—maximum allowable rate of an on-ramp meter at the on-ramp at
node i during time interval p (veh/h).
OFF-RAMP VARIABLES
• DEF(i, t, p)—deficit: unmet demand from a previous time interval p that
flows past node i during time step t; it is used in off-ramp flow
calculations downstream of a bottleneck.
• OFRD(i, p)—desired off-ramp demand flow exiting at off-ramp i during
time interval p.
• OFRF(i, t, p)—actual flow that can exit at off-ramp i during time step t in
time interval p.
FACILITYWIDE VARIABLES
• K(NS, P)—average vehicle density over the entire facility during the entire
analysis period P.
• K(NS, p)—average vehicle density over the entire facility during time
interval p.
• SMS(NS, P)—average analysis period facility speed: average space mean
speed over the entire facility during the entire analysis period P.
• SMS(NS, p)—average time interval facility speed: average space mean
speed over the entire facility during time interval p.

TRAVEL TIME RELIABILITY VARIABLES

• CRj—crash rate per 100 million vehicle miles traveled (VMT) in month j.
• DSP—duration of study period SP (h).

• DAFs(i, p)—demand adjustment factor for scenario s, time interval p, and
segment i.
• DCs—demand combination associated with scenario s.
• DM(s)—demand multiplier associated with scenario s.
• DM(Seed)—demand multiplier associated with the seed file.
• 𝐷𝑀
̅̅̅̅̅𝑗 —weighted average demand multiplier for all days in month j
relative to seed value.
• E[nw, j]—expected frequency of weather event w in month j, rounded to
the nearest integer.
• E15min[Dw]—expected duration of weather event w, rounded to the nearest
15-min increment.
• 𝔾(i)—distribution function for incident with severity type i.
• ICR—incident-to-crash ratio.
• IncDur—incident duration (min).
• IncType—incident severity type (1–5).
• ninc—number of incidents.
• nj—expected frequency of all incidents in the study period for month j,
rounded to the nearest integer.
• nDay,k—number of days in the reliability reporting period associated with
demand combination 𝑘.
• NDC—number of demand-level combinations considered.
• NScen—number of scenarios in the analysis.
• NInc,i—number of incidents associated with severity type i.
• NScen,Inc—number of all incident events generated for all scenarios.
• NScen,j—number of scenarios associated with month j of the reliability
reporting period.
• N
‾ DC,WZ—adjusted number of replications of a demand combination for
which the work zone is active.
• P{s}— probability of scenario s.
• Pt{w, j}—time-wise probability of weather type w in month j.
• rDC—ratio of weekday types with an active work zone in a given month to
the total number of each weekday type occurring in a given month.
• VMTi,p—vehicle miles traveled on segment i during analysis period p in
the seed file.

• VMTSeed—vehicle miles of travel in the seed file.

• δx—adjustment parameters to satisfy equilibrium calibration equations.

3. UNDERSATURATED SEGMENT EVALUATION
FACILITY SPEED CONSTRAINT

This module begins with the first segment in the first time interval. For each
cell, the flow (or volume) is equal to demand, the volume-to-capacity ratio is
equal to the demand-to-capacity ratio, and undersaturated flow conditions
prevail. Performance measures for the first segment during the first time interval
are calculated by using the procedures for the corresponding segment type in
Chapters 12, 13, and 14.
The analysis continues to the next downstream freeway segment in the same
time interval, and the performance measures are calculated. The process is
continued until the final downstream freeway segment cell in this time interval
has been analyzed. For each cell, the volume-to-capacity ratio and performance
measures are calculated for each freeway segment in the first time interval. The
analysis continues in the second time interval beginning at the furthest upstream
freeway segment and moving downstream until all freeway segments in that
time interval have been analyzed. This pattern continues for the third time
interval, fourth time interval, and so on until the methodology encounters a time
interval that contains one or more segments with a demand-to-capacity ratio
greater than 1.00 or when the final segment in the final time interval is analyzed.
If no oversaturated segments are encountered, the segment performance
measures are taken directly from Chapters 12, 13, and 14, and the facility
performance measures are calculated as described next in the Directional Facility
Module subsection.
When the analysis moves from isolated segments to a facility, an additional On the facility level, a speed
constraint is introduced that
constraint is necessary that controls the relative speed between two segments. To limits the maximum achievable
limit the speeds downstream of a segment experiencing a low average speed, a speed downstream of a
segment experiencing a low
maximum achievable speed is imposed on the downstream segments. This average speed.
maximum speed is based on acceleration characteristics reported elsewhere (4)
and is shown in Equation 25-1.
𝑉𝑚𝑎𝑥 = 𝐹𝐹𝑆 − (𝐹𝐹𝑆 − 𝑉𝑝𝑟𝑒𝑣 ) × 𝑒 −0.00162×𝐿 Equation 25-1
where
Vmax = maximum achievable segment speed (mi/h),
FFS = segment free-flow speed (mi/h),

Vprev = average speed on immediate upstream segment (mi/h), and
L = distance from midpoints of the upstream segment and the subject

segment (ft).
Chapter 25/Freeway Facilities: Supplemental Undersaturated Segment Evaluation

DIRECTIONAL FACILITY MODULE

The traffic performance measures can be aggregated over the length of the
directional freeway facility, over the time duration of the study interval, or over
the entire time–space domain. Each measure is discussed in the following
paragraphs.
Aggregating the estimated traffic performance measures over the entire
length of the freeway facility provides facilitywide estimates for each time
interval. Facilitywide travel times, vehicle distance of travel, and vehicle hours of
travel and delay can be computed, and patterns of their variation over the
connected time intervals can be assessed. The computational engine is limited to
15-min time intervals and 1-min time steps.
Aggregating the estimated traffic performance measures over the time
duration of the study interval provides an assessment of the performance of each
segment along the freeway facility. Average and cumulative distributions of
speed and density for each segment can be determined, and patterns of the
variation over connected freeway segments can be compared. Average trip times,
vehicle distance of travel, and vehicle hours of travel are easily assessed for each
segment and compared.
Aggregating the estimated traffic performance measures over the entire
time–space domain provides an overall assessment over the study interval time
duration. Overall average speeds, average trip times, total vehicle distance
traveled, and total vehicle hours of travel and delay are the most obvious overall
traffic performance measures. Equation 25-2 through Equation 25-5 show how
the facilitywide performance measures are calculated.
Facility space mean speed in time interval p is calculated with Equation 25-2:
Equation 25-2
∑𝑁𝑆
𝑖=1 𝑆𝐹(𝑖, 𝑝) × 𝐿(𝑖)
𝑆𝑀𝑆(𝑁𝑆, 𝑝) =
𝐿(𝑖)
∑𝑁𝑆
𝑖=1 𝑆𝐹(𝑖, 𝑝) × 𝑈(𝑖, 𝑝)
Average facility density in time interval p is calculated with Equation 25-3:

∑𝑁𝑆
𝑖=1 𝐾(𝑖, 𝑝) × 𝐿(𝑖)
Equation 25-3 𝐾(𝑁𝑆, 𝑝) = 𝑁𝑆
∑𝑖=1 𝐿(𝑖) × 𝑁(𝑖, 𝑝)
Overall space mean speed across all intervals is calculated with Equation 25-4:
∑𝑃𝑝=1 ∑𝑁𝑆
𝑖=1 𝑆𝐹(𝑖, 𝑝) × 𝐿(𝑖)
Equation 25-4 𝑆𝑀𝑆(𝑁𝑆, 𝑃) =
𝐿(𝑖)
𝑖=1 𝑆𝐹(𝑖, 𝑝) × 𝑈(𝑖, 𝑝)
Overall average density across all intervals is calculated with Equation 25-5:
𝑖=1 𝐾(𝑖, 𝑝) × 𝐿(𝑖)
Equation 25-5 𝐾(𝑁𝑆, 𝑃) =
𝑖=1 𝐿(𝑖) × 𝑁(𝑖, 𝑝)
These performance measures can be compared for different alternatives to

assess the impacts of different volume scenarios or the effects of geometric
improvements to the facility.
Undersaturated Segment Evaluation Chapter 25/Freeway Facilities: Supplemental

4. OVERSATURATED SEGMENT EVALUATION
Oversaturated flow conditions occur when the demand on one or more

freeway segment cells exceeds its capacity. The oversaturated segment evaluation
procedure presented in this chapter is performed separately for general purpose
and managed lanes. To evaluate the effect of interactions between the general
purpose and managed lanes, additional delays are introduced and calculated in
the form of vertical queueing, which is discussed at the end of this section.
Once oversaturation is encountered, the methodology changes its temporal
and spatial units of analysis. The spatial units become nodes and segments, and
the temporal unit moves from a time interval to smaller time steps. A node is
defined as the junction of two segments. There is always one more node than
there are segments, with a node added at the beginning and end of each
segment. The numbering of nodes and segments begins at the upstream end and
moves to the downstream end, with the segment upstream of node i numbered
segment i – 1 and the downstream segment numbered i, as shown in Exhibit 25-1.
The intermediate segments and node numbers represent the division of the
section between Ramps 1 and 2 into three segments numbered 2 (ONR),
3 (BASIC), and 4 (OFR). The oversaturated analysis moves from the first node to
each downstream node in the same time step. After completion of a time step,
the same nodal analysis is performed for subsequent time steps.
Exhibit 25-1
Seg. 1 Seg. 2 Seg. 3 Seg. 4 Seg. 5 Seg. 6 Node–Segment
Representation of a
N1 N2 N3 N4 N5 N6 N7
Directional Freeway Facility
Ramp 1 Ramp 2
The oversaturated analysis focuses on the computation of segment average

flows and densities in each time interval. These parameters are later aggregated
to produce facilitywide estimates. Two key inputs into the flow estimation
procedures are the time step duration for flow updates and a flow–density
function. These two inputs are described in the next subsections.
PROCEDURE PARAMETERS
Time Step Duration
Segment flows are calculated in each time step and are used to calculate the
number of vehicles on each segment at the end of every time step. The number of
vehicles on each segment is used to track queue accumulation and discharge and
to calculate the average segment density.
To provide accurate estimates of flows in oversaturated conditions, the time
intervals are divided into smaller time steps. The conversion from time intervals
to time steps occurs during the first oversaturated time interval and remains
until the end of the analysis. The transition to time steps is essential because, at
Chapter 25/Freeway Facilities: Supplemental Oversaturated Segment Evaluation

certain points in the methodology, future performance estimates are made on the
basis of the past value of a variable.
The computational engine assumes a time step of 15 s for oversaturated flow
computations, which is adequate for most facilities with a minimum segment
length greater than 300 ft. This time step is based on the assumption that a
shockwave of (severe) congestion can travel at speeds up to 20 ft/s or 13.6 mi/h.
A minimum segment length of 300 ft ensures that the congestion shockwave
does not travel more than one segment length in one 15-s time step.
For shorter segments, two problem situations may arise. The first situation
occurs when segments are short and the rate of queue growth (shockwave speed)
is rapid. Under these conditions, a short segment may be completely
undersaturated in one time step and completely queued in another. The
methodology may store more vehicles in this segment during a time step than
space allows. Fortunately, the next time step compensates for this error, and the
procedure continues to track queues and store vehicles accurately after this
correction.
The oversaturated The second situation in which small time steps are important occurs when
methodology implemented in
the computational engine two queues interact. There is a temporary inaccuracy due to the maximum
assumes a time step of 15 s, output of a segment changing, thus causing the estimation of available storage to
which is adequate for segment
lengths greater than 300 ft. be slightly in error. This situation results in the storage of too many vehicles on a
particular segment. This “supersaturation” is temporary and is compensated for
in the next time step. Inadequate time step size will result in erroneous
estimation of queue lengths and may affect other performance measures as well.
Regardless, if queues interact, the results should be viewed with extreme caution.
Flow–Density Relationship
Analysis of freeway segments depends on the relationships between segment
speed, flow, and density. Chapter 12, Basic Freeway and Multilane Highway
Segments, defines a relationship between these variables and the calculation of
performance measures in the undersaturated regime. The freeway facilities
methodology presented here uses the same relationships for undersaturated
segments. In other words, when a segment is undersaturated the computations
of this methodology are identical to the results obtained from Chapters 12, 13,
and 14 for basic freeway segments, weaving segments, and ramp segments,
respectively.
The calculations for oversaturated segments assume a simplified linear flow–
density diagram in the congested region. Exhibit 25-2 shows this flow–density
diagram for a segment having a free-flow speed (FFS) of 75 mi/h. For other FFSs,
the corresponding capacities in Chapters 12, 13, and 14 should be used.
The oversaturated regime curve in Exhibit 25-2 is constructed from a user-
specified jam density (default is 190 pc/mi/ln) and the known value of capacity,
defined as the flow at a density of 45 pc/mi/ln. The flow–density relationship is
assumed to be linear between these two points. The slope of the resulting line
describes the speed of the shock wave at which queues grow and dissipate, as
discussed further below. The speed in a congested segment is obtained from the
prevailing density in the segment, read along the linear flow–density
Oversaturated Segment Evaluation Chapter 25/Freeway Facilities: Supplemental

relationship. Details on the theory of kinematic waves in highway traffic are

given elsewhere (5–7).
Exhibit 25-2
Segment Flow–Density
Function
Note: Assumed FFS = 75 mi/h.
FLOW ESTIMATION
The oversaturated portion of the methodology is detailed as a flowchart in
Exhibit 25-3. The flowchart is divided into several sections over several pages.
Processes that continue from one section of the flowchart to another are indicated
by capital letters within parallelograms. Computations are detailed and labeled
in the subsections that follow according to each step of the flowchart.
The procedure first calculates flow variables starting at the first node during
the first time step of oversaturation and followed by each downstream node and
segment in the same time step. After all computations in the first time step are
completed, calculations are performed at each node and segment during
subsequent time steps for all remaining time intervals until the analysis is
completed.

Exhibit 25-3
Oversaturated Analysis
Procedure

Exhibit 25-3 (cont’d.)

Procedure


Procedure


E D Oversaturated Analysis
Procedure
27
26
Last No Move to Next
Node? Downstream Node
F
Yes
28 Last
29
Time Step No Move to Next
in Current Time Time Step G
Segment and Facility Performance Measures
Interval?
Yes
30
Calculate Segment
Performance Measures
31 32
Last Move to First Time
No
Time Interval Step in Next Time
in Analysis? Interval
Yes
36 33
Calculate Calculate Background
Facilitywide Density for This
Performance Time Interval
Measures
37 34
Is There
END
a Front-Clearing
Queue in This
Time Interval?
Yes
35
No
Calculate
Wave Speed
H
Segment Initialization: Exhibit 25-3, Steps 1–4

Steps 1–4 of the oversaturated procedure prepare the flow calculations for
the first time step and specify return points for later time steps. To calculate the
number of vehicles on each segment at the various time steps, the segments must
contain the proper number of vehicles before the queuing analysis places
unserved vehicles on segments. The initialization of each segment is described
below. A simplified queuing analysis is initially performed to account for the
effects of upstream bottlenecks. These bottlenecks meter traffic downstream of
their location. The storage of unserved vehicles (those unable to enter the
bottleneck) on upstream segments is performed in a later module. To obtain the
proper number of vehicles on each segment, the expected demand ED is
calculated. Expected demand is based on demands for and capacities of the

segment and includes the effects of all upstream segments. The expected demand
is the flow of traffic expected to arrive at each segment if all queues were stacked
vertically (i.e., no upstream effects of queues). In other words, all segments
upstream of a bottleneck have expected demands equal to their actual demand.
The expected demand of the bottleneck segment and all further downstream
segments is calculated by assuming a capacity constraint at the bottleneck, which
meters traffic to downstream segments. The expected demand ED is calculated
for each segment with Equation 25-6:
Equation 25-6 𝐸𝐷(𝑖, 𝑝) = min[𝑆𝐶(𝑖, 𝑝), 𝐸𝐷(𝑖 − 1, 𝑝) + 𝑂𝑁𝑅𝐷(𝑖, 𝑝) − 𝑂𝐹𝑅𝐷(𝑖 − 1, 𝑝)]
The segment capacity SC applies to the length of the segment. With the
expected demand calculated, the background density KB can be obtained for
each segment by using the appropriate segment density estimation procedures in
Chapters 12, 13, and 14. The background density is used to calculate the number
of vehicles NV on each segment by using Equation 25-7. If there are unserved
vehicles at the end of the preceding time interval, the unserved vehicles UV are
transferred to the current time interval. Here, S refers to the final time step in the
preceding time interval. The (0) term in NV represents the start of the first time
step in time interval p. The corresponding term at the end of the time step is
NV(i, 1, p).
Equation 25-7 𝑁𝑉(𝑖, 0, 𝑝) = 𝐾𝐵(𝑖, 𝑝) × 𝑁(𝑖, 𝑝) × 𝐿(𝑖) + 𝑈𝑉(𝑖, 𝑆, 𝑝 − 1)
The number of vehicles calculated from the background density is the
minimum number of vehicles that can be on the segment at any time. This
constraint is a powerful check on the methodology because the existence of
queues downstream cannot reduce this minimum. Rather, the segment can only
store additional vehicles. The storage of unserved vehicles is determined in the
segment flow calculation module later in this chapter.
Mainline Flow Calculations: Exhibit 25-3, Steps 9 and 16–23

The description of ramp flows follows the description of mainline flows.
Thus, Steps 5–8 and 10–15 are skipped at this time to focus first on mainline flow
computations. Because of skipping steps in the descriptions, some computations
may include variables that have not been described but that have already been
calculated in the flowchart.
Flows analyzed in oversaturated conditions are calculated for every time
step and are expressed in terms of vehicles per time step. The procedure
separately analyzes the flow across a node on the basis of the origin and
destination of the flow across the node. The mainline flow is defined as the flow
passing from upstream segment i – 1 to downstream segment i. It does not
include the on-ramp flow. The flow to an off-ramp is the off-ramp flow. The flow
from an on-ramp is the on-ramp flow. Each of these flows is shown in Exhibit 25-
4 with the origin, destination, and relationship to segment i and node i.

Exhibit 25-4
Definitions of Mainline and
Segment Flows
The segment flow is the total output of a segment, as shown in Exhibit 25-4.
Segment flows are calculated by determining the mainline and ramp flows. The
mainline flow is calculated as the minimum of six constraints: mainline input
(MI), MO1, MO2, MO3, upstream segment i – 1 capacity, and downstream
segment i capacity, as explained next.
Mainline Input: Exhibit 25-3, Step 9

Mainline input MI is the number of vehicles that wish to travel through a
node during the time step. The calculation includes (a) the effects of bottlenecks
upstream of the analysis node, (b) the metering of traffic during queue
accumulation, and (c) the presence of additional traffic during upstream queue
discharge.
MI is calculated by taking the number of vehicles entering the node
upstream of the analysis node, adding on-ramp flows or subtracting off-ramp
flows, and adding the number of unserved vehicles on the upstream segment.
Thus, MI is the maximum number of vehicles that wish to enter a node during a
time step. MI is calculated by using Equation 25-8, where all values have units of
vehicles per time step.
𝑀𝐼(𝑖, 𝑡, 𝑝) = 𝑀𝐹(𝑖 − 1, 𝑡, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖 − 1, 𝑡, 𝑝) − 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝)
Equation 25-8
+𝑈𝑉(𝑖 − 1, 𝑡 − 1, 𝑝)
Mainline Output: Exhibit 25-3, Steps 16–21

The mainline output is the maximum number of vehicles that can exit a
node, constrained by downstream bottlenecks or by merging on-ramp traffic.
Different constraints on the output of a node result in three separate types of
mainline outputs (MO1, MO2, and MO3).
Mainline Output 1, Ramp Flows: Exhibit 25-3, Step 16

MO1 is the constraint caused by the flow of vehicles from an on-ramp. The
capacity of an on-ramp segment is shared by two competing flows. This on-ramp
flow limits the flow from the mainline through this node. The total flow that can
pass the node is estimated as the minimum of the segment i capacity and the
mainline outputs from the preceding time step. The sharing of Lane 1 (shoulder
lane) capacity is determined in the calculation of the on-ramp. MO1 is calculated
by using Equation 25-9.
𝑆𝐶(𝑖, 𝑡, 𝑝) − 𝑂𝑁𝑅𝐹(𝑖, 𝑡, 𝑝)
𝑀𝑂1 = min { 𝑀𝑂2(𝑖, 𝑡 − 1, 𝑝) Equation 25-9
𝑀𝑂3(𝑖, 𝑡 − 1, 𝑝)

Mainline Output 2, Segment Storage: Exhibit 25-3, Steps 20 and 21

The second constraint on the output of mainline flow through a node is
caused by the growth of queues on a downstream segment. As a queue grows on
a segment, it may eventually limit the flow into the current segment once the
boundary of the queue reaches the upstream end of the segment. The boundary
of the queue is treated as a shock wave. MO2 is a limit on the flow exiting a node
due to the presence of a queue on the downstream segment.
The MO2 limitation is determined first by calculating the maximum number
of vehicles allowed on a segment at a given queue density. The maximum flow
that can enter a queued segment is the number of vehicles that leave the segment
plus the difference between the maximum number of vehicles allowed on the
segment and the number of vehicles already on the segment. The density of the
queue is calculated by using Equation 25-10 for the linear density–flow
relationship shown in Exhibit 25-2 earlier.
Equation 25-10 𝐾𝑄(𝑖, 𝑡, 𝑝) = [𝐾𝐽 × 𝑓𝐻𝑉 (𝑖, 𝑝)]
− [(𝐾𝐽 − 𝐾𝐶) × 𝑓𝐻𝑉 (𝑖, 𝑝) × 𝑆𝐹(𝑖, 𝑡 − 1, 𝑝)]/𝑆𝐶(𝑖, 𝑡, 𝑝)
Once the queue density is computed, MO2 can be computed by using
Equation 25-11.
𝑀𝑂2 (𝑖, 𝑡, 𝑝) = 𝑆𝐹(𝑖, 𝑡 − 1, 𝑝) − 𝑂𝑁𝑅𝐹(𝑖, 𝑡, 𝑝) + [𝐾𝑄(𝑖, 𝑡, 𝑝) × 𝑁(𝑖, 𝑝) × 𝐿(𝑖)]
Equation 25-11 −𝑁𝑉(𝑖, 𝑡 − 1, 𝑝)
The performance of the downstream node is estimated by taking the
performance during the preceding time step. This estimation remains valid when
there are no interacting queues. When queues interact and the time steps are
small enough, the error in the estimations is corrected in the next time step.
Mainline Output 3, Front-Clearing Queues: Exhibit 25-3, Steps 17–19

The final constraint on exiting mainline flows at a node is caused by
downstream queues clearing from their downstream end. These front-clearing
queues are typically caused by incidents in which there is a temporary reduction
in capacity. A queue will clear from the front if two conditions are satisfied. First,
the segment capacity (minus the on-ramp demand if present) for this time
interval must be greater than the segment capacity (minus the ramp demand if
present) in the preceding time interval. The second condition is that the segment
capacity minus the ramp demand for this time interval must be greater than the
segment demand for this time interval. A queue will clear from the front if both
conditions in the following inequality (Equation 25-12) are met.
If [𝑆𝐶(𝑖, 𝑝) − 𝑂𝑁𝑅𝐷(𝑖, 𝑝)] > [𝑆𝐶(𝑖, 𝑝 − 1) − 𝑂𝑁𝑅𝐷(𝑖, 𝑝 − 1)]
Equation 25-12
and [𝑆𝐶(𝑖, 𝑝) − 𝑂𝑁𝑅𝐷(𝑖, 𝑝)] > 𝑆𝐷(𝑖, 𝑝)
A segment with a front-clearing queue will have the number of vehicles
stored decrease during recovery, while the back of the queue position is
unaffected. Thus, the clearing does not affect the segment throughput until the
recovery wave has reached the upstream end of the front-clearing queue. The
computational engine implementation is simplified by assuming the downstream
segment is fully queued when the MO3 constraint is applied. In the flow–density
graph shown in Exhibit 25-5, the wave speed is estimated by the slope of the
dashed line connecting the bottleneck throughput and the segment capacity points.

Exhibit 25-5
Flow–Density Function with a
Shock Wave
Note: Assumed FFS = 75 mi/h.
The assumption of a linear flow–density function greatly simplifies the

calculation of the wave speed. The bottleneck throughput value is not required to
estimate the speed of the shock wave that travels along a known line. All that is
required is the slope of the line, which is calculated with Equation 25-13.
𝑊𝑆(𝑖, 𝑝) = 𝑆𝐶(𝑖, 𝑝)/[𝑁(𝑖, 𝑝) × (𝐾𝐽 − 𝐾𝐶) × 𝑓𝐻𝑉 ] Equation 25-13
The wave speed is used to calculate the wave travel time WTT, which is the
time it takes the front queue-clearing shock wave to traverse this segment.
Dividing the wave speed WS by the segment length in miles gives WTT.
The recovery wave travel time is the time required for the conditions at the
downstream end of the current segment to reach the upstream end of the current
segment. To place a limit on the current node, the conditions at the downstream
node are observed at a time in the past. This time is the wave travel time. This
constraint on the current node is MO3. The calculation of MO3 uses Equation 25-
14 and Equation 25-15. If the wave travel time is not an integer number of time
steps, then the weighted average performance of each variable is taken for the
time steps nearest the wave travel time. This method is based on a process
described elsewhere (5–7).
𝑊𝑇𝑇 = 𝑇 × 𝐿(𝑖)/𝑊𝑆(𝑖, 𝑝) Equation 25-14
𝑀𝑂1 (𝑖 + 1, 𝑡 − 𝑊𝑇𝑇, 𝑝)
𝑀𝑂2 (𝑖 + 1, 𝑡 − 𝑊𝑇𝑇, 𝑝) + 𝑂𝐹𝑅𝐹(𝑖 + 1, 𝑡 − 𝑊𝑇𝑇, 𝑝)
𝑀𝑂3 (𝑖, 𝑡, 𝑝) = min 𝑀𝑂3 (𝑖 + 1, 𝑡 − 𝑊𝑇𝑇, 𝑝) + 𝑂𝐹𝑅𝐹(𝑖 + 1, 𝑡 − 𝑊𝑇𝑇, 𝑝) Equation 25-15
𝑆𝐶(𝑖, 𝑡 − 𝑊𝑇𝑇, 𝑝)
{ 𝑆𝐶(𝑖 + 1, 𝑡 − 𝑊𝑇𝑇, 𝑝) + 𝑂𝐹𝑅𝐹(𝑖 + 1, 𝑡 − 𝑊𝑇𝑇, 𝑝) }
− 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝)

Mainline Flow: Exhibit 25-3, Steps 22 and 23

The flow across a node is called the mainline flow MF and is the minimum of
the following variables: MI, MO1, MO2, MO3, upstream segment i – 1 capacity,
and downstream segment i capacity, as shown in Equation 25-16.
𝑀𝐼(𝑖, 𝑡, 𝑝)
𝑀𝑂1 (𝑖, 𝑡, 𝑝)
𝑀𝑂2 (𝑖, 𝑡, 𝑝)
Equation 25-16 𝑀𝐹(𝑖, 𝑡, 𝑝) = min
𝑀𝑂3 (𝑖, 𝑡, 𝑝)
𝑆𝐶(𝑖, 𝑡, 𝑝)
{𝑆𝐶(𝑖 − 1, 𝑡, 𝑝)}
In addition to mainline flows, ramp flows must be analyzed. The presence of
mainline queues also affects ramp flows.
On-Ramp Calculations: Exhibit 25-3, Steps 10–15

On-Ramp Input: Exhibit 25-3, Steps 10 and 11
The maximum on-ramp input ONRI is calculated by adding the on-ramp
demand and the number of vehicles queued on the ramp. The queued vehicles
are treated as unmet ramp demand that was not served in previous time steps.
The on-ramp input is calculated with Equation 25-17.
Equation 25-17 𝑂𝑁𝑅𝐼(𝑖, 𝑡, 𝑝) = 𝑂𝑁𝑅𝐷(𝑖, 𝑡, 𝑝) + 𝑂𝑁𝑅𝑄(𝑖, 𝑡 − 1, 𝑝)
On-Ramp Output: Exhibit 25-3, Step 12

The maximum on-ramp output ONRO is calculated on the basis of the
mainline traffic through the node where the on-ramp is located. The on-ramp
output is the minimum of two values. The first is segment i capacity minus MI, in
the absence of downstream queues. Otherwise, the segment capacity is replaced
by the throughput of the queue. This estimation implies that vehicles entering an
on-ramp segment will fill Lanes 2 to N (where N is the number of lanes on the
current segment) to capacity before entering Lane 1. This assumption is
consistent with the estimation of v12 from Chapter 14, Freeway Merge and
Diverge Segments.
The second case occurs when the Lane 1 flow on segment i is greater than
one-half of the Lane 1 capacity. At this point, the on-ramp maximum output is
set to one-half of Lane 1 capacity. This output limitation implies that when the
demands from the freeway and the on-ramp are very high, there will be forced
one-to-one merging on the freeway from the freeway mainline and the on-ramp
in Lane 1. An important characteristic of traffic behavior is that, in a forced
merging situation, ramp and right-lane freeway vehicles will generally merge
one on one, sharing the capacity of the rightmost freeway lane (8). In all cases,
the on-ramp maximum output is also limited to the physical ramp road capacity
and the ramp-metering rate, if present. The maximum on-ramp output is an
important limitation on the ramp flow. Queuing occurs when the combined
demand from the upstream segment and the on-ramp exceeds the throughput of
the ramp segment. The queue can be located on the upstream segment, on the
ramp, or on both and depends on the on-ramp maximum output. Equation 25-18
determines the value of the maximum on-ramp output.

𝑂𝑁𝑅𝑂(𝑖, 𝑡, 𝑝)
𝑅𝑀(𝑖, 𝑡, 𝑝)
𝑂𝑁𝑅𝐶(𝑖, 𝑡, 𝑝)
min {𝑀𝐹(𝑖 + 1, 𝑡 − 1, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)} − 𝑀𝐼(𝑖, 𝑡, 𝑝) Equation 25-18
= min
𝑀𝑂3 (𝑖, 𝑡 − 1, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)
max
min {𝑀𝐹(𝑖 + 1, 𝑡 − 1, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)} /2𝑁(𝑖, 𝑝)
{ { 𝑀𝑂3 (𝑖, 𝑡 − 1, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖, 𝑡 − 1, 𝑝) }}
This model incorporates the maximum mainline output constraints from
downstream queues, not just the segment capacity. This fact is significant
because as a queue spills over an on-ramp segment, the flow through Lane 1 is
constrained. This constraint, in turn, limits the flow that can enter Lane 1 from
the on-ramp. The values of MO2 and MO3 for this time step are not yet known,
so they are estimated from the preceding time step. This estimation is one
rationale for using small time steps. If there is forced merging during the time
step when the queue spills back over the current node, the on-ramp will
discharge more than its share of vehicles (i.e., more than 50% of the Lane 1 flow).
This situation will cause the mainline flow past node i to be underestimated. But
during the next time step, the on-ramp flow will be at its correct flow rate, and a
one-to-one sharing of Lane 1 will occur.
On-Ramp Flows, Queues, and Delays: Exhibit 25-3, Steps 13–15

Finally, the on-ramp flow is calculated on the basis of the on-ramp input and
output values computed above. If the on-ramp input is less than the on-ramp
output, then the on-ramp demand can be fully served in this time step and
Equation 25-19 is used.
𝑂𝑁𝑅𝐹(𝑖, 𝑡, 𝑝) = 𝑂𝑁𝑅𝐼(𝑖, 𝑡, 𝑝) Equation 25-19
Otherwise, the ramp flow is constrained by the maximum on-ramp output,

and Equation 25-20 is used.
𝑂𝑁𝑅𝐹(𝑖, 𝑡, 𝑝) = 𝑂𝑁𝑅𝑂(𝑖, 𝑡, 𝑝) Equation 25-20
In the latter case, the number of vehicles in the ramp queue is updated by
using Equation 25-21.
𝑂𝑁𝑅𝑄(𝑖, 𝑡, 𝑝) = 𝑂𝑁𝑅𝐼(𝑖, 𝑡, 𝑝) − 𝑂𝑁𝑅𝑂(𝑖, 𝑡, 𝑝) Equation 25-21
The total delay for on-ramp vehicles can be estimated by integrating the
value of on-ramp queues over time. The methodology uses the discrete queue
lengths estimated at the end of each interval ONRQ(i, S, p) to produce overall
ramp delays by time interval.
Off-Ramp Flow Calculation: Exhibit 25-3, Steps 5–8

The off-ramp flow is determined by calculating a diverge percentage on the
basis of the segment and off-ramp demands. The diverge percentage varies only
by time interval and remains constant for vehicles that are associated with a
particular time interval. If there is an upstream queue, traffic may be metered to
this off-ramp, which will cause a decrease in the off-ramp flow. When the
vehicles that were metered arrive in the next time interval, they use the diverge

percentage associated with the preceding time interval. A deficit in flow, caused
by traffic from an upstream queue meter, creates delays for vehicles destined to
this off-ramp and other downstream destinations. The upstream segment flow is
used because the procedure assumes a vehicle destined for an off-ramp is able to
exit at the off-ramp once it enters the off-ramp segment. This deficit is calculated
with Equation 25-22.
𝐷𝐸𝐹(𝑖, 𝑡, 𝑝)
0
𝑝−1 𝑝−1 𝑇
∑ 𝑆𝐷(𝑖 − 1, 𝑋) − ∑ ∑[𝑀𝐹(𝑖 − 1, 𝑡, 𝑋) + 𝑂𝑁𝑅𝐹(𝑖 − 1, 𝑡, 𝑋)]

Equation 25-22 = max 𝑋=1 𝑋=1 𝑡=1
𝑡−1
+ ∑[𝑀𝐹(𝑖 − 1, 𝑡, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖 − 1, 𝑡, 𝑝)]

{{ 𝑡=1 }}
If there is a deficit, then the off-ramp flow is calculated by using the deficit
method. The deficit method is used differently in two specific situations. If the
upstream mainline flow plus the flow from an on-ramp at the upstream node (if
present) is less than the deficit for this time step, then the off-ramp flow is equal
to the mainline and on-ramp flows times the off-ramp turning percentage in the
preceding time interval, as indicated in Equation 25-23.
𝑂𝐹𝑅𝐷(𝑖, 𝑝 − 1)
Equation 25-23 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝) = [𝑀𝐹(𝑖 − 1, 𝑡, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖 − 1, 𝑡, 𝑝)] × [ ]
𝑆𝐷(𝑖 − 1, 𝑝 − 1)
However, if the deficit is less than the upstream mainline flow plus the on-
ramp flow from an on-ramp at the upstream node (if present), then Equation 25-
24 is used. This equation separates the flow into the remaining deficit flow and
the balance of the arriving flow.
𝑂𝐹𝑅𝐷(𝑖, 𝑝 − 1)
𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝) = 𝐷𝐸𝐹(𝑖, 𝑡, 𝑝) × [ ] + [𝑀𝐹(𝑖 − 1, 𝑡, 𝑝)
Equation 25-24
𝑆𝐷(𝑖 − 1, 𝑝 − 1)
𝑂𝐹𝑅𝐷(𝑖, 𝑝)
+ 𝑂𝑁𝑅𝐹(𝑖 − 1, 𝑡, 𝑝) − 𝐷𝐸𝐹(𝑖, 𝑡, 𝑝)] × [ ]
𝑆𝐷(𝑖 − 1, 𝑝)
If there is no deficit, then the off-ramp flow is equal to the sum of the
upstream mainline flow plus the on-ramp flow from an on-ramp at the upstream
node (if present) multiplied by the off-ramp turning percentage for this time
interval according to Equation 25-25.
𝑂𝐹𝑅𝐷(𝑖, 𝑝)
Equation 25-25 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝) = [𝑀𝐹(𝑖 − 1, 𝑡, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖 − 1, 𝑡, 𝑝)] × [ ]
𝑆𝐷(𝑖 − 1, 𝑝)
The procedure does not incorporate any delay or queue length computations
for off-ramps.
Segment Flow Calculation: Exhibit 25-3, Steps 24 and 25
The segment flow is the number of vehicles that flow out of a segment
during the current time step. These vehicles enter the current segment either to
the mainline or to an off-ramp at the current node. The vehicles that entered the
upstream segment may or may not have become queued within the segment.
The segment flow SF is calculated with Equation 25-26.
Equation 25-26 𝑆𝐹(𝑖 − 1, 𝑡, 𝑝) = 𝑀𝐹(𝑖, 𝑡, 𝑝) + 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝)

The number of vehicles on each segment is calculated on the basis of the

number of vehicles that were on the segment in the preceding time step, the
number of vehicles that entered the segment in this time step, and the number of
vehicles that leave the segment in this time step. Because the number of vehicles
that leave a segment must be known, the number of vehicles on the current
segment cannot be determined until the upstream segment is analyzed. The
number of vehicles on each segment NV is calculated with Equation 25-27.
𝑁𝑉(𝑖 − 1, 𝑡, 𝑝) = 𝑁𝑉(𝑖 − 1, 𝑡 − 1, 𝑝) + 𝑀𝐹(𝑖 − 1, 𝑡, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖 − 1, 𝑡, 𝑝) Equation 25-27
− 𝑀𝐹(𝑖, 𝑡, 𝑝) − 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝)
The number of unserved vehicles stored on a segment is calculated as the
difference between the number of vehicles on the segment and the number of
vehicles that would be on the segment at the background density. The number of
unserved vehicles UV stored on a segment is calculated with Equation 25-28.
𝑈𝑉(𝑖 − 1, 𝑡, 𝑝) = 𝑁𝑉(𝑖 − 1, 𝑡, 𝑝) − [𝐾𝐵(𝑖 − 1, 𝑝) × 𝐿(𝑖 − 1)] Equation 25-28
If the number of unserved vehicles is greater than zero, then a queue is

present on the facility upstream of the node in question. The presence of a queue
and congestion indicates that the node capacity is in queue discharge mode, which
means the queue discharge capacity is reduced relative to the pre-breakdown
capacity by a factor α. To account for this queue discharge effect, Equation 25-29
is applied to any active bottleneck along the facility if UV(i – 1, t, p) > 0.001. This
tolerance over an absolute value of zero is necessary to account for potential
rounding errors in the procedure.
𝑆𝐶(𝑖, 𝑡, 𝑝) = (1 − 𝛼) × 𝑆𝐶(𝑖, 𝑡, 𝑝) Equation 25-29
SEGMENT AND RAMP PERFORMANCE MEASURES

In the final time step of a time interval, the segment flows are averaged over
the time interval, and the performance measures for each segment are calculated.
If there was no queue on a particular segment during the entire time interval,
then the performance measures are calculated from the corresponding Chapter
12, 13, or 14 method for that segment. Because there are T time steps in an hour,
the average segment flow rate in vehicles per hour in time interval p is calculated
𝑆
𝑇
𝑆𝐹(𝑖, 𝑝) = ∑ 𝑆𝐹(𝑖, 𝑡, 𝑝) Equation 25-30
𝑆
𝑡=1
If T = 240 (1-h time steps) and S = 60 (1–analysis period time steps), then T/S =
4. If there was a queue on the current segment in any time step during the time
interval, then the segment performance measures are calculated in three steps.
First, the average number of vehicles NV over a time interval is calculated for
each segment by using Equation 25-31.
𝑆
1
𝑁𝑉(𝑖, 𝑝) = ∑ 𝑁𝑉(𝑖, 𝑡, 𝑝) Equation 25-31
𝑆
𝑡=1
Second, the average segment density K is calculated by taking the average
number of vehicles NV for all time steps in the time interval and dividing it by
the segment length, as shown by Equation 25-32.

𝑁𝑉(𝑖, 𝑝)
Equation 25-32 𝐾(𝑖, 𝑝) =
𝐿(𝑖) × 𝑁(𝑖, 𝑝)
Third, the average speed U on the current segment i during the current time
interval p is calculated with Equation 25-33.
𝑆𝐹(𝑖, 𝑝)
Equation 25-33 𝑈(𝑖, 𝑝) =
𝐾(𝑖, 𝑝)
Additional segment performance measures can be derived from the basic
measures shown in Equation 25-30 through Equation 25-33. Most prominent is
segment delay, which can be computed as the difference in segment travel time
at speed U(i, p) and at the segment FFS.
The final segment performance measure is the length of the queue at the end
of the time interval (i.e., step S in time interval p). The length of a queue Q on the
segment, in feet, is calculated with Equation 25-34.
𝑈𝑉(𝑖, 𝑆, 𝑝)
Equation 25-34 𝑄(𝑖, 𝑝) = × 5,280
max[(𝐾𝑄(𝑖, 𝑆, 𝑝) − 𝐾𝐵(𝑖, 𝑝)), 1]
OVERSATURATION ANALYSIS WITHIN MANAGED LANES

Whenever oversaturated conditions occur (as defined in Chapter 10) on
freeway facilities that contain managed lanes, the freeway facilities methodology
invokes the oversaturated analysis described in this chapter for both the general
purpose and managed lane facilities. The analysis will be performed separately
for each facility, meaning that the queues in either the general purpose or
managed lanes do not interact with each other. For freeway facilities with
managed lanes that do not have any access segments connecting the two lane
groups, performing oversaturated analysis separately yields accurate
performance measures for both the general purpose and managed lanes.
However, when access segments connect the two lane groups, no method
currently exists to model the queue interaction between the two. In this situation,
the queue spillback between the general purpose and managed lanes is modeled
as a “vertical queue.” The vehicles that cannot enter the general purpose or
managed lane facilities due to the presence of a queue do not translate into actual
queuing on the origin lane group, as shown in Exhibit 25-6.
The freeway facilities methodology keeps track of vehicles that cannot enter
the downstream segment (past the access point) in the form of a vertical queue,
and it releases these vehicles as congestion dissipates. Note that there are two
vertical queues for each access segment, one for vehicles traveling from the
managed to the general purpose lanes, and the other for vehicles traveling from
the general purpose to the managed lanes. Exhibit 25-6 shows an example of a
vertical queue for the first situation. Note that the existence of a vertical queue
does not lead to actual queuing on the managed lane.

Exhibit 25-6
VERTICAL Vehicles that are not in the queue Vertical Queuing from a
QUEUE Access Vehicles in the vertical queue Managed Lane Due to Queue
Segment Presence on the General
Vehicles in the actual queue
Purpose Lanes
Queue on the GP Lanes
Despite this simplification of queue spillback modeling, the methodology

keeps track of the delays vehicles encounter in the vertical queues. The delay is
computed as the number of vehicles stored in the vertical queue, multiplied by
15 min of delay in each analysis period. The delay of the vehicles originating
from the managed lanes that are waiting in the vertical queue is estimated based
on Equation 25-35.
𝐷𝑀𝐿,𝑣𝑒𝑟𝑡 = 𝑁𝑀𝐿,𝑣𝑒𝑟𝑡 × 0.25 Equation 25-35
where
DML,vert = delay incurred by vehicles originating from the managed lanes waiting
in the vertical queue for one 15-min analysis period (h) and
NML,vert = average number of vehicles originating from the managed lanes that
are waiting in the vertical queue in one analysis period (veh).
Similar to the vehicle delay in the managed lanes, the delay of vehicles
originating from the general purpose lanes that are waiting in the vertical queue
is estimated based on Equation 25-36.
𝐷𝐺𝑃,𝑣𝑒𝑟𝑡 = 𝑁𝐺𝑃,𝑣𝑒𝑟𝑡 × 0.25 Equation 25-36
where
DGP,vert = delay incurred by vehicles originating from the general purpose lanes
waiting in the vertical queue for one 15-min analysis period (h) and
NGP,vert = average number of vehicles originating from the general purpose lanes
that are waiting in the vertical queue in one analysis period (veh).

5. WORK ZONE ANALYSIS DETAILS
This section provides additional computational details for work zone

analysis on freeway facilities. The analysis of work zones on basic segments on a
facility is described in Chapter 10, Freeway Facilities Core Methodology; this
section provides additional analysis details for work zones in merge, diverge,
and weaving segments, as well as the analysis of directional crossover work
zones. The information provided in this section is largely based on results from
National Cooperative Highway Research Program Project 03-107 (9).
SPECIAL WORK ZONE CONFIGURATIONS

The queue discharge rate model predictions explained in Chapter 10 apply to
basic freeway segments. These estimates should be adjusted for special freeway
work zone configurations, such as merge segments, diverge segments, weaving
segments, and work zones with directional crossovers. The relationships
presented in this section were derived from field-calibrated microsimulation
models for the special work zone configurations.
No data were available for the impacts of these work zone configurations on
FFS, and so FFS estimates for these configurations should be used only when
local data are not available. One exception is the FFS for a directional crossover,
which should be estimated from the geometric design of the configuration, and is
used as an input to the queue discharge rate estimation for that work zone
configuration.
Work Zone Capacity Adjustments for Merge Segments

The proportion of work zone capacity (in reference to the basic work zone
capacity calculated in Chapter 10) that is allocated to the mainline flow in a
merge segment is presented separately for locations upstream and downstream
of the special work zone activity segment. Exhibit 25-7 shows an example for a
merge area within a construction zone.
Exhibit 25-7
On-Ramp Merge Diagram for
2-to-1 Freeway Work Zone
Configuration
Note: WZ = work zone.
Exhibit 25-8 through Exhibit 25-12 give the proportion of work zone capacity
allocated to mainline flow in merge, diverge, and directional crossover segments.
For a weaving segment, a predictive model is presented following those exhibits.
In the exhibits, only a subset of potential work zone configurations is presented,
Work Zone Analysis Details Chapter 25/Freeway Facilities: Supplemental

as these are the only ones that were included in the simulation modeling effort in
the original research.
Exhibit 25-8 presents the proportion of available capacity upstream of a
merge area in a construction zone, as a function of work zone lane
configurations, different levels of on-ramp input volumes, and lengths of the
acceleration lane. Upstream of the work zone, the proportion of capacity
available to the mainline movement decreases considerably as the on-ramp
demand increases.
Work Zone On-Ramp Exhibit 25-8

Lane Input Demand Acceleration Lane Length (ft) Proportion of Work Zone
Configuration (pc/h) 100 300 500 700 900 1,100 1,300 1,500 Queue Discharge Rate
0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 (Relative to the Basic Work
250 1.00 0.86 0.86 0.86 0.86 0.86 0.86 0.86 Zone Capacity) Available for
2 to 1 500 1.00 0.70 0.70 0.70 0.70 0.70 0.70 0.70 Mainline Flow Upstream of
750 1.00 0.53 0.53 0.53 0.53 0.53 0.53 0.53 Merge Area
1,000 1.00 0.49 0.45 0.40 0.40 0.40 0.40 0.40
0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
250 1.00 0.92 0.92 0.92 0.92 0.92 0.92 0.92
2 to 2 500 1.00 0.84 0.84 0.84 0.84 0.84 0.84 0.84
750 1.00 0.75 0.75 0.75 0.75 0.75 0.75 0.75
1,000 1.00 0.67 0.67 0.67 0.67 0.67 0.67 0.67
0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
250 1.00 0.95 0.95 0.95 0.95 0.95 0.95 0.95
3 to 2 500 1.00 0.87 0.87 0.87 0.87 0.87 0.86 0.86
750 1.00 0.78 0.78 0.78 0.78 0.78 0.78 0.78
1,000 1.00 0.70 0.70 0.70 0.70 0.70 0.70 0.70
0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
250 1.00 0.97 0.97 0.98 0.98 0.98 0.98 0.98
4 to 3 500 1.00 0.91 0.91 0.91 0.92 0.92 0.92 0.92
750 1.00 0.85 0.85 0.85 0.86 0.86 0.86 0.86
1,000 1.00 0.79 0.79 0.79 0.79 0.80 0.80 0.80
The capacity of the merge segment is the same as a basic work zone segment,
with the caveat that the on-ramp flow consumes a portion of the mainline
capacity. As a result, the available capacity upstream of the merge area leading
into the work zone will be reduced once the queue spills back to the lane drop
point. The proportions presented in Exhibit 25-8 approximate the conditions of a
zipper merge configuration, with capacity divided approximately equally
between the on-ramp and the right-most freeway mainline lane. In other words,
the estimates correspond to a worst-case scenario for mainline flow in terms of
available capacity, and a best-case scenario for the on-ramp movement. Note that
the proportions for a 100-ft acceleration lane length are all 1.0 because on-ramp
vehicles will experience difficulty entering the mainline lanes with the extremely
short acceleration lane. These findings are based on results from microscopic
simulation models of this configuration.
Research (9) shows that the throughput downstream of a merge area is
approximately equal to the upstream queue discharge rate (before the merge) in
most cases, with some configurations actually showing a marginal increase in
flow. This slight increase occurs because additional demand from the on-ramp is
able to more efficiently utilize gaps in the work zone queue discharge flow
without the turbulence effects of the upstream lane drop. This effect was
primarily observed for long acceleration lanes. However, for a more conservative
Chapter 25/Freeway Facilities: Supplemental Work Zone Analysis Details

estimate of work zone operations, it is recommended not to consider this

increase in flow downstream of the merge area regardless of lane configuration,
on-ramp input volume, or acceleration lane length.
Work Zone Capacity Adjustments for Diverge Segments

Similar to merge segment analysis, the analysis of diverge segments
distinguishes between the diverge segment portions of the work zone that are
upstream and downstream of the diverge segment. Exhibit 25-9 shows an
example for a diverge area within a construction zone.
Exhibit 25-9
Off-Ramp Diverge Diagram for
a 2-to-1 Freeway Work Zone
Configuration
Note: WZ = work zone.
Exhibit 25-10 presents the proportion of available capacity downstream of a

diverge area for various freeway work zone lane configurations, different levels
of off-ramp volume percentage, and deceleration lane lengths. Upstream of the
diverge area, research (9) shows the available capacity is generally equivalent to
that of a basic work zone segment. Therefore, it is recommended to apply a fixed
adjustment of 1.00 upstream of the diverge area regardless of lane configuration,
off-ramp volume percentage, or deceleration lane length.
At the downstream end, however, the proportion of available capacity for
mainline volume decreases significantly as the off-ramp volume percentage
increases. Analysts should expect work zone operations to improve downstream
of a diverge segment (but still within the work zone) because some portion of
traffic will exit the freeway, thereby decreasing the processed volume below the
downstream capacity. However, if the deceleration lane lengths are shorter than
100 ft, exiting vehicles will need to slow down while still on the mainline to
complete the exit maneuver. This speed reduction may drop mainline capacity
by as much as 10% or more.
For a diverge area, the proportion of off-ramp demand that can be served in
the work zone under congested conditions can be predicted as presented in
Exhibit 25-11. This proportion is defined as the off-ramp observed volume
divided by the off-ramp demand volume.

Work Zone Exhibit 25-10

Lane Off-Ramp Volume Deceleration Lane Length (ft) Proportion of Work Zone
Configuration Percentage 100 300 500 700 900 1,100 1,300 1,500 Capacity Available for Mainline
0.0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Flow Downstream of Diverge
6.3 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.93 Area
2 to 1 12.5 0.87 0.88 0.88 0.88 0.88 0.88 0.87 0.87
18.8 0.79 0.82 0.82 0.82 0.82 0.81 0.81 0.81
25.0 0.72 0.76 0.76 0.75 0.75 0.75 0.75 0.75
0.0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
6.3 0.93 0.94 0.94 0.94 0.94 0.94 0.94 0.94
2 to 2 12.5 0.84 0.87 0.87 0.87 0.87 0.87 0.87 0.87
18.8 0.76 0.81 0.81 0.81 0.81 0.81 0.81 0.81
25.0 0.68 0.75 0.75 0.75 0.75 0.75 0.75 0.75
0.0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
6.3 0.93 0.94 0.94 0.94 0.94 0.94 0.94 0.94
3 to 2 12.5 0.86 0.87 0.87 0.87 0.87 0.87 0.87 0.87
18.8 0.78 0.81 0.81 0.81 0.81 0.81 0.81 0.81
25.0 0.69 0.74 0.74 0.74 0.74 0.74 0.74 0.74
0.0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
6.3 0.93 0.93 0.93 0.93 0.93 0.93 0.93 0.93
4 to 3 12.5 0.86 0.87 0.87 0.87 0.87 0.87 0.87 0.87
18.8 0.76 0.80 0.80 0.80 0.80 0.80 0.80 0.80
25.0 0.64 0.73 0.73 0.73 0.73 0.73 0.73 0.73
Proportion of Off-Ramp Demand Exhibit 25-11

Lane Configuration Served in Work Zone Proportion of Off-Ramp
2 to 1 0.39 Demand Served in Work Zone
2 to 2 0.82
3 to 2 0.53
4 to 3 0.60
Work Zone Capacity Adjustments for Crossover Segments

Exhibit 25-12 presents the proportion of work zone capacity available for a
directional crossover for various crossover vehicle speeds. As shown in the
exhibit, the crossover capacity is highly sensitive to average crossover speed. The
variation in capacity for different work zone lane configurations was found to be
negligible in crossovers. The estimates in Exhibit 25-12 should be applied as
multipliers of the basic segment work zone capacity described above.
Crossover Average Speed (mi/h) Exhibit 25-12

Lane Configuration 25 35 45 Proportion of Available Work
2 to 1 Zone Capacity for a
3 to 2 0.83 0.90 0.94 Directional Crossover in the
4 to 3 Work Zone
Work Zone Capacity Adjustments for Weaving Segments

In a weaving area, the proportion of work zone capacity available for
mainline flow can be predicted by using a two-step model. In Step 1, the analyst
estimates the maximum proportion of mainline flow that can be served through
the work zone based on the work zone lane configuration and the volume ratio.
This maximum becomes an upper bound on the actual estimated proportion,
which is estimated in Step 2. In Step 2, the actual proportion of work zone
capacity available for mainline flow is estimated based on the lane configuration,
volume ratio, and auxiliary lane length. The final proportion of mainline flow
that can be processed through the weaving segment is the lower of the two

estimated proportions from Steps 1 and 2. The model intercept and coefficient
values for Equation 25-37 and Equation 25-38 are presented in Exhibit 25-13.
Step 1: Estimate Maximum Mainline Allocation Proportion

Equation 25-37 𝑀𝑎𝑥𝑃𝑟𝑜𝑝𝑜𝑟𝑡𝑖𝑜𝑛 = Intercept + 𝛽1 (2-to-1) + 𝛽2 (2-to-2)
+𝛽3 (3-to-2)+𝛽4 (4-to-3) + 𝛽5 (𝑉𝑅)
where
MaxProportion = maximum proportion of work zone capacity available for
mainline flow at the weave area (decimal),
Intercept = model intercept,
β1 = model coefficient for 2-to-1 lane closures,
2-to-1 = indicator variable that is 1 when the work zone has a 2-to-1
configuration and 0 otherwise,
β5 = model coefficient for volume ratio, and
VR = volume ratio = weave volume/total volume.
Step 2: Predict Mainline Proportion

Equation 25-38 Proportion = Intercept + 𝛽1 (2-to-1) + 𝛽2 (2-to-2)
+𝛽3 (3-to-2)+𝛽4 (4-to-3) + 𝛽5 (𝑉𝑅) + 𝛽6 (𝐴𝑢𝑥𝐿𝑒𝑛𝑔𝑡ℎ)
where
Proportion = proportion of work zone capacity available for mainline flow
(decimal),
β6 = model coefficient for auxiliary lane length,
AuxLength = auxiliary lane length (ft), and
all other variables are as defined previously.
The off-ramp demand volume proportion Prop(off-ramp) in the weaving area
is estimated from Equation 25-39, with the intercept and model coefficients
given in Exhibit 25-14, and all other variables as defined previously.
𝑃𝑟𝑜𝑝(off-ramp) = Intercept + 𝛽1 (2-to-1) + 𝛽2 (2-to-2)
Equation 25-39
+𝛽3 (3-to-2)+𝛽4 (4-to-3) + 𝛽5 (𝑉𝑅)

Model Model Term Coefficient Exhibit 25-13

Intercept 1.0023 Model Coefficients for
β1 –0.1197 Estimating the Proportion of
β2 0.0105 Work Zone Capacity in a
Upstream Weaving Segment
β3 0.0085
β4 0.0000
Step 1: β5 –0.3048
Maximum
Proportion Intercept 1.0573
β1 0.1307
β2 –0.0623
Downstream
β3 0.0494
β4 0.0000
β5 -0.3332
Intercept 0.8491
β1 –0.0665
β2 0.0061
Upstream β3 0.0050
β4 0.0000
β5 –0.4687
Step 2: β6 9.0956 × 10-5
Predicted
Proportion Intercept 0.8962
β1 0.2702
β2 0.0535
Downstream β3 0.1073
β4 0.0000
β5 –0.9694
β6 30.5253 × 10-5
Model Model Term Coefficient Exhibit 25-14

Intercept 0.6162 Model Coefficients for
β1 –0.2201 Estimating the Proportion of
Off-Ramp Volume β2 0.2082 Off-Ramp Volume Served in
Proportion β3 –0.0551 the Weaving Area
β4 0.0000
β5 0.0850

6. PLANNING-LEVEL METHODOLOGY FOR

FREEWAY FACILITIES
This section presents a planning-level approach for freeway facility analysis

that is compatible with the operational method presented in Chapter 10, Freeway
Facilities Core Methodology. The planning level-approach is specifically
constructed to
1. Use default values for as many of the operational parameters as practical;
2. Omit the need to enter detailed data about segment attributes (e.g.,
acceleration lane length and detailed weaving section geometry);
3. Aggregate the analysis to a coarser spatial representation, reporting at the
freeway section level instead of the HCM segment level; and
4. Enable HCM users to manually carry out the analysis for a single peak
hour without an extensive computational burden.
The method covers both undersaturated and oversaturated conditions and
produces estimates of travel time, speed, density, and level of service (LOS). The
underlying methodology relies on developing a relationship between delay rate
per unit distance on a basic freeway segment, and the demand-to-capacity ratio.
For weaving segments, capacity adjustment factors (CAFs) are developed based
on the volume ratio and segment length. By using these factors, demand-to-
capacity ratios on weaving segments can be adjusted, and the segment is
subsequently treated similarly to a basic freeway segment. The capacities of
merge and diverge segments are determined from the demand level, FFS, and
space mean speed. CAFs are subsequently calculated for those segments, and
their demand-to-capacity ratios are adjusted accordingly.
INPUT REQUIREMENTS
Input variables are characterized into global and section inputs. Sections are
defined to occur between points where either demand or capacity changes, as
shown in Exhibit 25-15.
Exhibit 25-15
Schematics of Freeway
Sections
For instance, the first section in Exhibit 25-15 (starting from the left) is a basic
freeway section. This section is followed by an on-ramp, and the demand level
changes. Capacity and demand remain unchanged until the first off-ramp.
Consequently, the second freeway section in Exhibit 25-15 is defined as a ramp
section. The next section that follows is a basic freeway section. It is followed by
Planning-Level Methodology for Freeway Facilities Chapter 25/Freeway Facilities: Supplemental

a weaving section (this section is a weaving section due to the presence of an

auxiliary lane). The weaving section is followed by another ramp section (due to
an off-ramp), a basic section, and finally a ramp section (due to an on-ramp).
Introduction of freeway sections facilitates user input and is more compatible
with links in travel demand models as well as modern digital data sources.
In the operational freeway facilities method, the influence area of an on-ramp
or off-ramp is typically limited to a length of 1,500 ft. In the planning method,
ramp sections can be longer. For cases where a ramp section length exceeds 2 mi,
it is recommended to divide the section into multiple sections to avoid having
the lower ramp section capacity apply for a very long distance.
Global inputs include information about the facility of interest and are
applicable to all sections across all analysis periods. These inputs include
1. Free-flow speed (SFFS),
2. Peak hour factor (PHF),
3. Percentage heavy vehicles (%HV),
4. General terrain type for truck passenger-car equivalent (PCE) conversion,
5. K-factor [to convert directional annual average daily traffic (AADT) to
peak hour flows], and
6. Traffic growth factor (ftg).
The equation used to estimate section speeds in this planning method
(Equation 25-45) is fully consistent with the basic freeway segment speed–flow
models presented in Chapter 12, Basic Freeway and Multilane Highway Segments.
Section inputs cover information that is applicable to a given section across all
analysis periods and that may vary from one section to another as a function of
1. Section type (basic, weave, ramp),
2. Section length L (mi),
3. Section number of lanes, and
4. Section directional AADT.
This information, along with the global inputs, is used to calculate the free-
flow travel rate (the inverse of FFS), CAFs for weave and ramp sections, adjusted
lane capacity (the product of base capacity and CAF), and section capacity (the
product of adjusted lane capacity and number of lanes). The planning
methodology follows five basic steps:
1. Demand-level calculations;
2. Section capacity calculations and adjustments;
3. Delay rate estimation;
4. Average travel time, speed, and density calculations; and
5. Level of service.
All steps are described in detail below.
Chapter 25/Freeway Facilities: Supplemental Planning-Level Methodology for Freeway Facilities

STEP 1: DEMAND-LEVEL CALCULATIONS

The demand level for each section is determined from the entering demand,
exiting demand, and carryover demand from a previous analysis period (in the
case of oversaturated conditions).
The methodology uses the directional average annual daily traffic on section i
AADTi, K-factor, traffic growth factor ftg, and peak hour factor PHF during each
15-min analysis period t in the peak hour to compute the demand inflow and
outflow Vi,t as shown in Equation 25-40:
𝐴𝐴𝐷𝑇𝑖 × 𝑘 × 𝑓𝑡𝑔 𝑡 = 1, 3
1
Equation 25-40 𝑉𝑖,𝑡 = 𝐴𝐴𝐷𝑇𝑖 × 𝑘 × ( ) × 𝑓𝑡𝑔 𝑡=2
𝑃𝐻𝐹
1
{𝐴𝐴𝐷𝑇𝑖 × 𝑘 × (2 − ) × 𝑓𝑡𝑔 𝑡=4
𝑃𝐻𝐹
where all parameters were defined previously.
All demand inputs should be in units of passenger cars per hour per lane
(pc/h/ln). If demands are given in units of vehicles per hour per lane (veh/h/ln),
they need to be converted with Equation 25-41.
𝑉𝑖,𝑡
Equation 25-41 𝑞𝑖,𝑡 =
𝑓𝐻𝑉
where
qi,t = demand flow rate in PCEs (pc/h),
Vi,t = demand flow rate in vehicles per hour (veh/h), and
fHV = adjustment factor for presence of heavy vehicles in traffic stream.
Just as in the operational method, all heavy vehicles are classified as single-
unit trucks (SUTs) or tractor-trailers (TTs). Recreational vehicles and buses are
treated as SUTs. The heavy-vehicle adjustment factor fHV is computed from the
combination of the two heavy vehicle classes, which are added to get an overall
truck percentage PT, as shown by Equation 25-42.
1
Equation 25-42 𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1)
where
fHV = heavy-vehicle adjustment factor (decimal),
PT = proportion of SUT and TTs in traffic stream (decimal), and
ET = PCE of one heavy vehicle in the traffic stream (PCE).
The values for ET are 2.0 for level terrain and 3.0 for rolling terrain. For
specific grades, Chapter 12 provides other heavy-vehicle equivalency factors.
The converted demand flow rates qi,t can represent both inflow demand and
outflow demand. For the first facility section and all on-ramps, qi,t represents
inflow demand and is denoted by (qi,z)in. For all off-ramps, qi,t represents outflow
demand and is represented by (qi,z)out.

Demand level di,p (in passenger cars per hour) on section i in analysis period p
is computed as the demand level in section i – 1, plus the inflow at section i
during analysis period p, minus the outflow at the same section at analysis
period p, plus any carryover demand d′i,p–1 in section 𝑖 from the previous analysis
period p – 1. The relationship is as shown in Equation 25-43.
𝑑𝑖,𝑝 = 𝑑𝑖−1,𝑝 + (𝑞𝑖,𝑝 ) − (𝑞𝑖,𝑝 ) ′
+ 𝑑𝑖,𝑝−1 Equation 25-43
in out
where all variables are as defined previously.
The carryover demand d′i,p–1 on section i at analysis period p is the difference
between the section demand and capacity, as given by Equation 25-44.
′
𝑑𝑖,𝑝 = max(𝑑𝑖,𝑝 − 𝑐𝑖 , 0) Equation 25-44
The carryover demand is also used as an indication of a queue on the section.

Note that in this approach, queues are stacked vertically and do not spill back
into an upstream link. The section queue length is estimated by dividing the
difference in lane demand and capacity by the density. Essentially, it provides an
estimate for how long the queue would spill back at the given density, assuming
a fixed number of lanes upstream of the bottleneck.
STEP 2: SECTION CAPACITY CALCULATIONS AND ADJUSTMENTS

The capacity of basic freeway sections is found by using the FFS and the
percentage of heavy vehicles on the facility, as shown by Equation 25-45.
𝑐𝑖 = 2,200 + 10 × [min(70, 𝐹𝐹𝑆) − 50] Equation 25-45
where ci is the capacity of freeway section i (pc/h/ln) and FFS is the facility’s free-
flow speed (mi/h).
Equation 25-45 provides capacity values for basic freeway sections. This
capacity must be adjusted for weaving, merge, diverge, and ramp sections, as
described next.
Capacity Adjustments for Weaving Sections

As mentioned above, the planning method is derived from the basic freeway
segment speed–flow model to estimate a section’s delay rate and travel speed.
When applied to weaving sections, an adjustment to capacity is required to
account for the generally lower capacity in weaving segments. This capacity
adjustment factor CAFweave can be estimated with Equation 25-46.
𝐶𝐴𝐹weave = min(0.884 − 0.0752𝑉𝑟 + 0.0000243𝐿𝑠 , 1) Equation 25-46
where
CAFweave = capacity adjustment factor used for a weaving segment
(0 ≤ CAFweave ≤ 1.0) (decimal),
Vr = ratio of weaving demand flow rate to total demand flow rate in the
weaving segment (decimal), and
Ls = weaving segment length (ft).
Through this capacity adjustment, the basic section method can be extended
to weaving sections, as described elsewhere (10). The process for estimating

CAFweave is based on a representative weaving section with the following

characteristics (see Chapter 13 for additional details):
• Minimum number of lane changes that must be made by a single weaving
vehicle from the on-ramp to the freeway: LC(RF) = 1,
• Minimum number of lane changes that must be made by a single weaving
vehicle from the freeway to the off-ramp: LC(FR) = 1,
• Minimum number of lane changes that must be made by a ramp-to-ramp
vehicle to complete a weaving maneuver: LC(RR) = 0, and
• Number of lanes from which a weaving maneuver may be made with one
or no lane changes: N(WL) = 2.
Adjustments for Ramp Sections

Research shows an average CAF of 0.9 can be used for ramp sections with an
on-ramp or off-ramp (10, 11). It is recognized that known bottlenecks may have
significantly reduced capacities that require a lower CAF. Further calibration of
the CAF by the analyst is strongly encouraged when applying this method to on-
ramp sections with known capacity constraints and congestion impacts. Analyst
calibration of this factor is also possible for off-ramp sections.
STEP 3: DELAY RATE ESTIMATION

The planning-level approach estimates the delay rate per unit distance as a
function of a section’s demand-to-capacity ratio. The delay rate is the difference
between the actual and free-flow travel time per unit distance. For example, if a
facility’s space mean speed is 60 mi/h relative to an FFS of 75 mi/h for a 0.5-mi
segment, then the free-flow travel time is 0.4 min, and the actual travel time is 0.5
min. The delay rate per mile is the difference of those travel times divided by the
segment length, which gives a delay rate of 0.2 min/mi. The calculation of the
delay rate needs to be performed differently for undersaturated and
oversaturated conditions, as described next.
Undersaturated Conditions
For undersaturated conditions, the basic freeway segment speed–flow model
in Chapter 12 can be used to estimate delay rates. However, for a planning-level
analysis, it is desirable to further simplify the estimation of delay rate to be a
function of inputs readily available in a planning context. The delay rate ΔRUi,p (in
minutes per mile) for segment i in analysis period p as a function of the demand-
to-capacity ratio di,p/ci is given by Equation 25-47.
𝑑𝑖,𝑝
0 <𝐸
𝑐𝑖
Equation 25-47 ∆𝑅𝑈𝑖,𝑝 = 3 2
𝑑𝑖,𝑝 𝑑𝑖,𝑝 𝑑𝑖,𝑝 𝑑𝑖,𝑝
𝐴( ) +𝐵( ) +𝐶( )+𝐷 𝐸≤ ≤1
{ 𝑐𝑖 𝑐𝑖 𝑐𝑖 𝑐𝑖
where A, B, C, D, and E are parameters given in Exhibit 25-16 and all other
variables are as defined previously.

Free-Flow Speed (mi/h) A B C D E Exhibit 25-16

75 68.99 –77.97 34.04 –5.82 0.44 Parameter Values for
70 71.24 –85.48 35.58 –5.44 0.52 Undersaturated Model
65 92.45 –127.33 56.34 –8.00 0.62
60 121.35 –184.84 83.21 –9.33 0.72
55 156.43 –248.99 99.20 –0.12 0.82
Oversaturated Conditions
For oversaturated conditions, the additional delay rate is approximated
assuming uniform arrival and departures at the bottleneck location. With the
demand exceeding capacity, any demand that cannot be served through the
bottleneck must be stored upstream of the bottleneck in a queue. The additional
oversaturation delay rate ΔROi,p (in minutes per mile) for segment 𝑖 at analysis
period p, over a 15-min (900-s) analysis period, is obtained by Equation 25-48.
450 𝑑𝑖,𝑝
∆𝑅𝑂𝑖,𝑝 = [max ( − 1.0)] Equation 25-48
𝐿 𝑐𝑖
where all variables are as previously defined.
STEP 4: AVERAGE TRAVEL TIME, SPEED, AND DENSITY CALCULATIONS

After the delay rate is determined, the travel rate is computed by summing the
delay rate and travel rate under free-flow conditions, as shown by Equation 25-49.
𝑇𝑅𝑖,𝑝 = ∆𝑅𝑈𝑖,𝑝 + ∆𝑅𝑂𝑖𝑝 + 𝑇𝑅𝐹𝐹𝑆 Equation 25-49
where TRi,p is the travel rate on segment i in analysis period p (min/mi), TRFFS is
the travel rate under free-flow conditions (min/mi), and all other parameters are
as previously defined.
The section travel time is then computed by multiplying the travel rate and
segment length, as shown by Equation 25-50.
𝑇𝑖,𝑝 = 𝑇𝑅𝑖,𝑝 × 𝐿𝑖 Equation 25-50
where Ti,p is the travel time on segment i in analysis period p (min/mi), TRi,p is the
travel rate on segment i in analysis period p (min/mi), and Li is the length of
section i (mi).
The average speed Si,p (in miles per hour) on section i in analysis period p is
computed by using Equation 25-51.
𝐿𝑖
𝑆𝑖,𝑝 = Equation 25-51
𝑇𝑖,𝑝
Finally, the density is calculated as shown by Equation 25-52.
𝑑𝑖,𝑝
𝐷𝑖,𝑝 = Equation 25-52
𝑁𝑖 × 𝑆𝑖,𝑝
where Di,p is density on section i in analysis period p (pc/mi/ln), Ni is the number
of lanes in section i, di,p is section demand (pc/h), and Si,p is speed (mi/h).
Thus, the planning-level method provides a facility performance summary
that includes whether the facility is undersaturated or oversaturated, the total
facility travel time, the space mean speed, the average facility density, and the
total queue length.

STEP 5: LEVEL OF SERVICE

With the density obtained in Step 4, LOS can be estimated for urban or rural
facilities following the thresholds in Chapter 10.
The LOS criteria for urban and rural freeway facilities are repeated in Exhibit
25-17. Urban LOS thresholds are the same density-based criteria used for basic
freeway segments. Studies on LOS perception by rural travelers indicate lower-
density thresholds than those of their urban freeway counterparts. The average
LOS applies to each 15-min analysis period.
Exhibit 25-17 LOS Urban Freeway Facility Density Rural Freeway Facility Density
LOS Criteria for Urban and (pc/mi/ln) (pc/mi/ln)
Rural Freeway Facilities A ≤11 ≤6
B >11–18 >6–14
C >18–26 >14–22
D >26–35 >22–29
E >35–45 >29–39
F >45 or >39 or
any component section vd/c ratio > 1.00 any component section vd/c ratio >1.00

7. MIXED-FLOW MODEL FOR COMPOSITE GRADES
This section presents the application of the mixed-flow model in the case of
composite grades. The procedure builds on the single-grade methodology
described in Chapter 26, Freeway and Highway Segments: Supplemental, and
uses the same basic set of equations. The procedure computes LOS, capacity,
speed, and density for each segment and for the composite grade as a whole.
Many of the equations in this section are identical to those presented in Chapter
26, although they have different equation numbers. The major difference with
composite grades is that the analyst must compute the spot travel rates or spot
speeds at the start and end of each segment on the composite grade as an input
to the analysis of the next grade segment.
OVERVIEW OF THE METHODOLOGY

The methodology assumes the composite grade both begins and ends with a
long, level segment. The example shown in Exhibit 25-18 has five segments.
Exhibit 25-18
Schematic of a Composite
Grade
Exhibit 25-19 presents the methodology flowchart. The remainder of this

section provides the computational details for each step in the process.
STEP 1: INPUT DATA

The user must supply the length dj (mi) and the grade gj (decimal) for each
segment j, including the tangent segment immediately preceding the composite
grade. In addition, the auto-only free-flow speed FFS (mi/h), peak hour factor
PHF (decimal), the flow rate of mixed traffic vmix (veh/h/ln), and the fraction of
SUTs and TTs in the traffic stream must be specified for the facility as a whole.
STEP 2: CAPACITY ASSESSMENT

Before the composite grade is examined in detail, the capacity of the
individual segments j is determined. A mixed-flow capacity adjustment factor
CAFmix,j converts auto-only capacities into mixed-traffic-stream capacities. It is
computed with Equation 25-53. The third term in this equation changes for each
segment.
Chapter 25/Freeway Facilities: Supplemental Mixed-Flow Model for Composite Grades

Exhibit 25-19
Mixed-Flow Methodology
Overview
𝐶𝐴𝐹mix,𝑗 = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix,𝑗

Equation 25-53
where
CAFmix,j = mixed-flow capacity adjustment factor for segment j (decimal),
CAFao = capacity adjustment factor for the auto-only case (e.g., due to
weather or incidents) (decimal),
CAFT,mix = capacity adjustment factor for the percentage of trucks in mixed-flow
conditions (decimal), and
CAFg,mix,j = capacity adjustment factor for grade for segment j in mixed-flow
conditions (decimal).
CAF for the Auto-Only Case

Because CAFao is used to convert auto-only capacities into mixed-traffic
capacities, it defaults to a value of 1.0 unless other capacity adjustments are in
effect (e.g., weather, incidents, driver population factor).
Mixed-Flow Model for Composite Grades Chapter 25/Freeway Facilities: Supplemental

CAF for Truck Percentage

The CAF for truck percentage CAFT,mix is computed with Equation 25-54.
𝐶𝐴𝐹𝑇,mix = 0.53 × 𝑃𝑇 0.72 Equation 25-54
where PT is the total percentage of SUTs and TTs in the traffic stream (decimal).
CAF for Grade Effect

The CAF for grade effect CAFg,mix accounts for the grade severity, grade length,
and truck presence. It is computed by using Equation 25-55 with Equation 25-56.
𝐶𝐴𝐹𝑔,mix = 𝜌𝑔,mix × max[0, 0.69 × (𝑒 12.9𝑔𝑗 − 1)]
Equation 25-55
× max[0, 1.72 × (1 − 1.71𝑒 −3.16𝑑𝑗 )]
with
8 × 𝑃𝑇 𝑃𝑇 < 0.01
𝜌𝑔,mix = { Equation 25-56
0.126 − 0.03𝑃𝑇 otherwise
where
ρg,mix = coefficient for grade term in the mixed-flow CAF equation (decimal),
PT = total truck percentage (decimal),
gj = grade of segment j (decimal), and
dj = length of segment j (mi).
Once CAFmix,j is computed, the mixed-flow capacity for each segment j is
calculated with Equation 25-57.
𝐶mix,𝑗 = 𝐶𝑎𝑜 × 𝐶𝐴𝐹mix,𝑗
Equation 25-57
where
Cmix,j = mixed-flow capacity for segment j (veh/h/ln);
Cao = auto-only capacity for the given FFS, from Exhibit 12-6 (pc/h/ln); and
CAFmix,j = mixed-flow capacity adjustment factor for segment j (decimal).
The procedure identifies the smallest of these capacities and designates it as
Cmix. It also notes the segment that produces this capacity as jc. The capacity Cmix
is checked against the mixed-flow rate vmix to check if vmix ≥ Cmix. If this condition
occurs, the system is deemed to be oversaturated, LOS F is reported, and no
further analysis is carried out. However, if vmix < Cmix, the procedure continues.

STEP 3: SPECIFY INITIAL CONDITIONS

Starting with Step 3, the methodology analyzes each segment in sequence.
Steps 3 through 6 are repeated for each segment until the final segment on the
composite grade is reached. The main focus is on computing travel times and
speeds for SUTs, TTs, and autos.
Step 3 specifies the initial kinematics-based spot speeds for SUTs and TTs.
The effects of the traffic interaction terms are omitted for the time being. The
focus is on the kinematic behavior of the trucks as they ascend and descend the
individual grades. For the first segment, the initial kinematic spot speed is the
speed for SUTs and TTs on the long, level segment that precedes the composite
grade. For all subsequent segments, it is the kinematic spot speed at the end of
the previous segment. The kinematic spot speeds are speeds without traffic
interaction, which will be added to the final kinematic spot speeds to obtain final
spot speeds of each segment.
STEP 4: COMPUTE TRUCK SPOT AND SPACE-BASED TRAVEL TIME RATES

This step computes the SUT and TT space-based travel time rates for each of
the segments and the spot rates at the end of each segment. The procedure
follows a process similar to Step 5 of the mixed-flow model procedure described
in Chapter 26.
The first substep involves analyzing the kinematic behavior of the trucks on
the grade. The final spot rates are needed, as well as a determination of whether
the trucks accelerated or decelerated on the grade.
Exhibit 25-20 and Exhibit 25-21 can be used for these purposes. These graphs
are based on kinematic relationships given elsewhere (12). Alternative models of
propulsive and resistive forces, such as more complex ones that account for gear
shifting (e.g., 13, 14), can produce longer travel times. Such considerations can be
incorporated into the mixed-flow model by adjusting the parameter values that
affect the tractive effort to account for the additional losses. The travel time rates
presented here are based on a model that assumes constant peak-engine power.
Other models (e.g., 13, 14) account for the power losses that occur for the time
intervals prior to and after gear shifting when the engine speed is outside the
range that produces peak power.
Exhibit 25-20 shows the trends in SUT spot rates for various grades starting
from travel rates of 48 s/mi (75 mi/h) and 120 s/mi (30 mi/h). Exhibit 25-21 shows
the same trends for a TT. Clearly, trucks decelerate as upgrades become steeper.
For milder grades, trucks can often accelerate.

Exhibit 25-20
SUT Spot Rates Versus
Distance with Initial Speeds of
75 and 30 mi/h
Notes: Curves in this graph assume a weight-to-horsepower ratio of 100. Solid curves are for an initial speed of
75 mi/h (48 s/mi) and dashed curves are for an initial speed of 30 mi/h (120 s/mi).
Exhibit 25-21
TT Spot Rates Versus
Distance with Initial Speeds of
75 and 20 mi/h
Notes: Curves in this graph assume a weight-to-horsepower ratio of 100. Solid curves are for an initial speed of
75 mi/h (48 s/mi) and dashed curves are for an initial speed of 20 mi/h (180 s/mi).

In both Exhibit 25-20 and Exhibit 25-21, the x-axis gives the distance d
traveled by the truck, and the y-axis gives the spot travel rate τkin,j at the end of
that distance. The different curves are for various upgrades and downgrades.
To ascertain whether trucks accelerate or decelerate on segment j, consider
the travel time rate trends shown in Exhibit 25-20 and Exhibit 25-21. If an SUT’s
final spot rate for segment j τSUT,kin,f,j is greater than the SUT’s initial spot rate for
segment j τSUT,kin,i,j and the TT’s spot rate at the end of segment j τTT,kin,f,j is greater
than the TT’s spot rate at the beginning of segment j τTT,kin,i,j, then both truck
classes decelerate. If τSUT,kin,f,j < τSUT,kin,i,j and τTT,kin,f,j < τTT,kin,i,j, then both truck classes
accelerate.
To determine the end-of-grade spot travel time rates, start by finding the
point on the applicable grade that corresponds to the initial kinematic rate. Treat
that point as the zero distance location. Next, proceed along the grade length
(x-axis) for a distance equal to the length d of the segment and read the spot rate
at that distance. This reading is the final spot rate. For example, an SUT travels
2,000 ft starting from 60 mi/h (60 s/mi) on a 5% grade. Point 1 in Exhibit 25-20 is
the 60-mi/h speed (60-s/mi rate) from which the SUT starts to travel on the 5%
grade. Point 2 is the distance that is treated as the zero distance of the SUT. Point
3 represents the distance the SUT has traveled after 2,000 ft. The final spot rate
can be read at Point 4. The initial kinematic SUT and TT spot rates for segment j
τSUT,kin,i,j and τTT,kin,i,j are the kinematic spot rates at the end of the preceding
segment. For remaining segments, τSUT,kin,i,j and τTT,kin,i,j are the kinematic spot
rates at the end of the preceding segment j – 1, which are τSUT,kin,f,j–1 and τTT,kin,f,j–1.
The second substep involves determining the space-based travel time rates
for SUTs and TTs. Exhibit 25-22 and Exhibit 25-23 provide examples. Exhibit 25-
22 shows the time versus distance relationships for SUTs starting at 70 mi/h with
a desired speed of 75 mi/h as they accelerate or decelerate on various grades.
Exhibit 25-23 shows time versus distance relationships for SUTs starting at 30
mi/h as they ascend or descend grades. Relationships for a range of initial rates
for both SUTs and TTs are provided in Appendix A.
In all exhibits, the x-axis is the distance d traveled by the truck, while the
y-axis is the travel time T to cover the grade length d. The various curves in each
exhibit represent different upgrades. All the truck profiles have a desired speed
of 75 mi/h. For example, the 2% curve in Exhibit 25-23 shows travel time versus
distance for SUTs starting from 30 mi/h with a desired speed of 75 mi/h.
When necessary, symbols are placed on the curves to indicate where a truck
reaches 55, 60, 65, and 70 mi/h, for use when the speed limit is less than 75 mi/h,
as indicated in the notes for Exhibit 25-23. For example, if the speed limit is 55
mi/h, it is assumed trucks will maintain a constant speed of 55 mi/h after
reaching that speed. The analyst would use the graph to determine the travel
time to accelerate to 55 mi/h and then perform the remainder of the travel time
calculation using 55 mi/h as the truck speed. Not all curves have these symbols,
as (a) the truck’s crawl speed would be less than 55 mi/h for the particular grade,
(b) the truck would take more than 10,000 ft to reach that speed, or (c) the graph
being used starts from a relatively high speed (e.g., Exhibit 25-22).

Exhibit 25-22
SUT Travel Time Versus
Distance Curves for 70-mi/h
Initial Speed
Note: Curves in this graph assume a weight-to-horsepower ratio of 100.
Exhibit 25-23
Initial Speed
Notes: Curves in this graph assume a weight-to-horsepower ratio of 100.

Triangles indicate where a truck reaches 55 mi/h, circles indicate 60 mi/h, diamonds indicate 65 mi/h, and
squares indicate 70 mi/h.
The analyst should use the Appendix A graph that has a starting spot speed
closest to the value computed in the first substep. Because the graphs are
provided in 5-mi/h increments, this choice means using the graph that is within
2.5 mi/h of the speed corresponding to the segment’s initial spot rate.
The kinematic space-based travel time rate τkin (in seconds per mile) can then
be computed with Equation 25-58.
𝜏𝑘𝑖𝑛 = 𝑇/𝑑 Equation 25-58

where T is the segment travel time (s) and d is the grade length (mi).
The maximum grade length shown in the graphs is 10,000 ft. When the grade
length exceeds 10,000 ft, the travel rate can be computed using Equation 25-59.
𝑇10000 10,000
Equation 25-59 𝜏𝑘𝑖𝑛 = + 𝛿 (1 − ) × 5,280
𝑑 5,280𝑑
where
τkin = kinematic travel rate (s/mi),
T10000 = travel time at 10,000 ft (s),
δ = slope of the travel time versus distance curve (s/ft),
d = grade length (mi), and
5,280 = number of feet in 1 mi.
The 𝛿 values for SUTs and TTs are shown in Exhibit 25-24 and Exhibit 25-25,
respectively.
Exhibit 25-24 Free-Flow Speed (mi/h)

δ Values for SUTs Grade 50 55 60 65 70 75
–5% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
0% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
2% 0.0136 0.0124 0.0114 0.0105 0.0100 0.0099
3% 0.0136 0.0124 0.0114 0.0113 0.0112 0.0112
4% 0.0136 0.0129 0.0128 0.0128 0.0128 0.0127
5% 0.0146 0.0146 0.0146 0.0146 0.0145 0.0145
6% 0.0165 0.0165 0.0165 0.0165 0.0165 0.0165
7% 0.0186 0.0186 0.0186 0.0186 0.0186 0.0186
8% 0.0208 0.0208 0.0208 0.0208 0.0208 0.0208

δ Values for TTs Grade 50 55 60 65 70 75
–5% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
0% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
2% 0.0136 0.0124 0.0119 0.0118 0.0116 0.0115
3% 0.0143 0.0143 0.0142 0.0141 0.0140 0.0138
4% 0.0171 0.0171 0.0171 0.0170 0.0169 0.0168
5% 0.0202 0.0202 0.0202 0.0202 0.0202 0.0202
6% 0.0236 0.0236 0.0236 0.0236 0.0236 0.0236
7% 0.0272 0.0272 0.0272 0.0272 0.0272 0.0272
8% 0.0310 0.0310 0.0310 0.0310 0.0310 0.0310
Once the end-of-grade spot travel time rates and the space-based rates are
obtained for the current segment, Equation 25-60 and Equation 25-61 are used to
account for the traffic interaction term to obtain the actual truck spot and space-
based travel time rates.
Equation 25-60 𝜏∗,𝑆𝑈𝑇,𝑗 = 𝜏∗,𝑆𝑈𝑇,𝑘𝑖𝑛,𝑗 + 𝛥𝜏 𝑇𝐼
Equation 25-61 𝜏∗,𝑇𝑇,𝑗 = 𝜏∗,𝑇𝑇,𝑘𝑖𝑛,𝑗 + 𝛥𝜏 𝑇𝐼
where
* = placeholder that can either be f to designate the spot travel time rate
at the end of the segment or S to indicate the space-based rate across
the segment,

τ ,SUT,j = spot travel time rate for SUTs at the end of segment j or the space-
*
based rate (s/mi),
τ ,SUT,kin,j = kinematic final spot travel time rate or space-based rate for SUTs
*
(s/mi),
ΔτTI = traffic interaction term (s/mi) from Equation 25-62,
τ ,TT,j = spot travel time rate for TTs at the end of segment j or the space-
*
based rate (s/mi), and
τ ,TT,kin,j = kinematic final spot travel time rate or space-based rate for TTs
*
(s/mi).
The traffic interaction term represents the contribution of other traffic to
truck speeds or travel time rates in mixed flow. It is computed by Equation 25-62.
3,600 3,600 1 Equation 25-62
𝛥𝜏𝑇𝐼 = ( − ) × [1 + 3 ( − 1)]
𝑆𝑎𝑜 𝐹𝐹𝑆 𝐶𝐴𝐹mix
where
ΔτTI = traffic interaction term (s/mi),
Sao = auto-only speed for the given flow rate (mi/h) from Equation 25-63,
FFS = base free-flow speed of the basic freeway segment (mi/h), and
CAFmix = mixed-flow capacity adjustment factor for the segment (decimal)
from Equation 25-53.
The auto-only travel time rate for the given flow rate can be computed with
Equation 25-63.
𝑣mix
𝐹𝐹𝑆 ≤ 𝐵𝑃𝑎𝑜
𝐶𝐴𝐹mix
𝑆𝑎𝑜 = 𝐶 𝑣mix 2 Equation 25-63
(𝐹𝐹𝑆 − 𝐷𝑎𝑜 ) (𝐶𝐴𝐹 − 𝐵𝑃𝑎𝑜 ) 𝑣mix
𝑐 mix
𝐹𝐹𝑆 − > 𝐵𝑃𝑎𝑜
{ (𝐶𝑎𝑜 − 𝐵𝑃𝑎𝑜 )2 𝐶𝐴𝐹mix }
where
Sao = auto-only speed for the given flow rate (mi/h),
FFS = base free-flow speed of the basic freeway segment (mi/h),
Cao = base segment capacity (pc/h/ln) from Exhibit 12-6,
BPao = breakpoint in the auto-only flow condition (pc/h/ln) from Exhibit
12-6,
Dc = density at capacity = 45 pc/mi/ln,
vmix = flow rate of mixed traffic (veh/h/ln), and
CAFmix = mixed-flow capacity adjustment factor for the basic freeway segment
(decimal).

STEP 5: COMPUTE AUTOMOBILE SPOT AND SPACE-BASED TRAVEL

TIME RATES
Whether trucks accelerate or decelerate, the automobile spot travel time rates
at the end of the segment are computed with Equation 25-64. The analyst should
check that the automobile spot rates are always less than or equal to the truck
spot rates (i.e., automobile speeds are greater than or equal to truck speeds).
3,600
Equation 25-64 𝜏𝑓,𝑎,𝑗 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆
𝑣mix 0.77
64.50 × ( ) × (𝑃𝑆𝑈𝑇 )0.34
1,000
+ 1.53
𝜏𝑓,𝑆𝑈𝑇,𝑘𝑖𝑛,𝑗 3,600
[× max (0, 100 − 𝐹𝐹𝑆 × 100) ]
𝑣mix 0.81
79.50 × ( ) × (𝑃𝑇𝑇 )0.56
1,000
+ 1.32
𝜏𝑓,𝑇𝑇,𝑘𝑖𝑛,𝑗 3,600
[ × max (0, − ) ]
100 𝐹𝐹𝑆 × 100
where
τf,a,j = end-of-grade spot travel time rate for automobiles (s/mi),
τf,SUT,kin,j = spot kinematic travel time rate of SUTs at the end of segment j (s/mi),
τf,TT,kin,j = spot kinematic travel time rate of TTs at the end of segment j (s/mi),
vmix = flow rate of mixed traffic (veh/h/ln),
PSUT = proportion of SUTs in the traffic stream (decimal), and
PTT = proportion of TTs in the traffic stream (decimal).
In Step 4, it was determined whether trucks accelerate or decelerate across a
segment. If they decelerate, Equation 25-65 is used to compute the auto space-
based travel time rate. If trucks accelerate, Equation 25-66 is employed. The auto
space mean rates are always less than or equal to the truck space mean rates.
Equation 25-65
3,600
𝜏𝑆,𝑎,𝑗 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆
𝑣mix 0.46
100.42 × ( ) × (𝑃𝑆𝑈𝑇 )0.68
1,000
+ 2.76
𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,𝑗 3,600
[× max (0, − ) ]
100 𝐹𝐹𝑆 × 100
𝑣mix 1.36
110.64 × ( ) × (𝑃𝑇𝑇 )0.62
1,000
+ 1.81
𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,𝑗 3,600
[× max (0, − ) ]
100 𝐹𝐹𝑆 × 100

3,600
𝜏𝑆,𝑎,𝑗 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆 Equation 25-66
𝑣mix 1.16
54.72 × ( ) × (𝑃𝑆𝑈𝑇 )0.28
1,000
+ 1.73
𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,𝑗 3,600
[× max (0, 100
−
𝐹𝐹𝑆 × 100
) ]
𝑣mix 1.32
69.72 × ( ) × (𝑃𝑇𝑇 )0.61
1,000
+ 1.33
𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,𝑗 3,600
[× max (0, 100 − 𝐹𝐹𝑆 × 100) ]
where
τS,a,,j = auto space-based travel time rate (s/mi),
τS,SUT,kin,j = kinematic space-based travel time rate of SUTs (s/mi),
τS,TT,kin,j = kinematic space-based travel time rate of TTs (s/mi),
The traffic interaction term is the same for all the travel time rate equations
and can be computed with Equation 25-62.
STEP 6: COMPUTE MIXED-FLOW SPACE-BASED TRAVEL TIME RATE

AND SPEED
The mixed-flow space-based travel time rate τmix,j and the space-based speed
Smix,j are computed with Equation 25-67 and Equation 25-68, respectively.
𝜏mix,𝑗 = 𝑃𝑎 𝜏𝑆,𝑎,𝑗 + 𝑃𝑆𝑈𝑇 𝜏𝑆,𝑆𝑈𝑇,𝑗 + 𝑃𝑇𝑇 𝜏𝑆,𝑇𝑇,𝑗 Equation 25-67
3,600
𝑆mix,𝑗 = Equation 25-68
𝜏mix,𝑗
where
τmix,j = mixed-flow space-based travel time rate for segment j (s/mi),
τS,a,j = automobile space-based travel time rate for segment j (s/mi),
τS,SUT,j = space-based travel time rate of SUTs (s/mi),
τS,TT,j = space-based travel time rate of TTs (s/mi),
As indicated above, Steps 3 through 6 are repeated for each segment until the
end of the composite grade is reached.

STEP 7: OVERALL RESULTS

Once spot and space mean speeds and travel time rates have been developed
for all vehicle types on all segments, the overall performance of the composite
grade can now be estimated. The mixed-flow travel time for each segment can be
computed with Equation 25-69.
Equation 25-69 3,600𝑑𝑗
𝑡mix,𝑗 =
𝑆mix,𝑗
where
tmix,j = mixed-flow travel time segment j (s),
dj = grade length of segment j (mi), and
Smix,j = mixed-flow speed for segment j (mi/h).
The overall mixed-flow travel time tmix,oa is the summation of mixed-flow
travel times on all segments. The overall space-based travel speed can then be
3,600𝑑𝑜𝑎
Equation 25-70 𝑆mix,𝑜𝑎 =
𝑡mix,𝑜𝑎
where
Smix,oa = overall mixed-flow speed (mi/h);
doa = overall distance, the summation of all the segment grade lengths on
the composite grade (mi); and
tmix,oa = overall mixed-flow travel time (s).

8. FREEWAY CALIBRATION METHODOLOGY
This section presents a calibration methodology for the procedures described

in Chapter 10, Freeway Facilities Core Methodology, and Chapter 11, Freeway
Reliability Analysis. The freeway calibration methodology is carried out at three
main levels:
1. Calibration at the core freeway facility level,
2. Calibration at the reliability level, and
3. Calibration at the Active Traffic and Demand Management (ATDM)
strategy assessment level.
The procedure uses sequential calibration to calibrate these three distinct
methodological parts, meaning that the calibration is carried out sequentially for
each level. After a level is fully calibrated, no further change is allowed from a
different level. As a result, this approach requires that the calibration parameters
of different levels be mutually exclusive.
The approach first calibrates the base scenario, then focuses on reliability-
level calibration, and concludes with ATDM-level calibration. It is logical both
that the base scenario (i.e., core freeway facility) should be fully calibrated before
evaluating reliability or ATDM strategies and that the base scenario calibration
should not be affected by any subsequent changes from the reliability or ATDM
calibration levels. Consequently, it is critical to select a suitable base scenario
with oversaturated flow conditions to ensure that the bottlenecks are calibrated
appropriately. More information about the development of the methodology is
provided in a paper (15) located in the Technical Reference Library section of
online HCM Volume 4.
Calibration relies on field measurements of key input variables, including the
segment capacity. Chapter 12, Basic Freeway and Multilane Highway Segments,
provides definitions for prebreakdown and queue discharge capacity. Chapter
26, Freeway and Highway Segments: Supplemental, provides guidance for field
measuring and estimating capacity from sensor data.
CALIBRATION AT THE CORE FREEWAY FACILITY LEVEL

The core freeway facility analysis is calibrated for a specific day, called the
seed day. Exhibit 25-26 depicts five steps of the calibration process for a core
facility analysis. After gathering input data, the actual calibration consists of
three steps (Steps 2, 3, and 4), the order of which is somewhat flexible. Multiple
iterations may be needed to achieve satisfactory performance. A detailed
explanation of each step follows.
Step 1: Gather Input Data

In this step, all input data required for a single freeway facility analysis
(computational engine seed file) need to be gathered. These data include
1. Geometric information such as segment type, segment length, and
number of lanes;
Chapter 25/Freeway Facilities: Supplemental Freeway Calibration Methodology

2. Facility free-flow speed (FFS);

3. Capacity estimate for bottleneck segment(s); and
4. Demand-level data for all segments in all time intervals.
Geometric data are model input parameters and will not be changed in the
calibration process. The other three inputs (FFS, capacity, and demand) are used
as calibration parameters.
Exhibit 25-26
Calibration Steps for the Core
Freeway Facility Level
Step 2: Calibrate Free-Flow Speed

FFS can be field measured or estimated by using the procedure given in
Chapter 12. The FFS calibration procedure may be applied in either case;
however, if accurate field measurements of FFS are available, great care should
be taken before changing a field-measured input.
To start, the analyst should select a time interval with a low demand level
and no active bottleneck. The analyst should then compare the estimated free-
flow travel time of this interval with the field measurements. Because a later step
requires the analyst to look at congested periods, the study period should be
sufficiently long to include free-flow conditions before or after the onset of
congestion.
The calibration process involves making a computational engine run for the
seed day, recording the average travel time for a low-demand time interval, and
comparing it to the observed travel time. The user needs to repeatedly perform
one of the following actions until the predicted facility travel time is within a
predefined threshold (e.g., 10% error tolerance) of the observed facility travel time:
• Reduce the FFS in 1- to 5-mi/h increments if the predicted travel time is less
than the observed travel time, or
• Increase the FFS in 1- to 5-mi/h increments if the predicted travel time is
more than the observed travel time.
Freeway Calibration Methodology Chapter 25/Freeway Facilities: Supplemental

This process should only be used for analysis periods with demand levels far
less than oversaturation (i.e., free-flow conditions). The speed–flow diagram in
Exhibit 25-27 illustrates the effect of different FFSs on the overall facility speed–
flow–density relationship. A higher free-flow speed FFS1 and a lower free-flow
speed FFS2 are shown. A 5-mi/h drop in FFS is associated with a drop in capacity
equal to 50 pc/h/ln, except at very high FFSs.
Exhibit 25-27
Effect of Calibrating Free-Flow
Speed on Capacity
Step 3: Calibrate Bottleneck Capacity

In this step, the location and extent of bottlenecks are calibrated, which
requires a freeway facility to feature at least some periods of oversaturated flow
conditions. Guidance for selecting capacity measurement locations and for
reducing the collected data is provided in Chapter 26.
It is very important to calibrate for capacity, as research (11) shows the
controlling capacity at the bottleneck is often significantly less than the HCM’s
base capacity. Three parameters are used to calibrate for the location and extent
of bottlenecks:
1. Prebreakdown capacity at the bottleneck, implemented through a capacity
adjustment factor (CAF) relative to the base capacity for a freeway
segment. In the HCM, the prebreakdown flow rate is defined as the 15-
min average flow rate immediately prior to the breakdown event. For the
purposes of this chapter, the prebreakdown flow rate is equivalent to the
segment capacity;
2. Queue discharge rate at the bottleneck following breakdown, as
implemented through a percentage capacity drop α. In the HCM, the
queue discharge rate is defined as the average flow rate during
oversaturated conditions (i.e., during the time interval after breakdown
and prior to recovery); and
3. Jam density of the queue forming upstream of the bottleneck, which
describes the maximum density (minimum intervehicle spacing) in a
queued condition.

The prebreakdown capacity and the queue-discharge capacity loss influence

the actual throughput of the bottleneck, as well as the speed of shock waves
describing the rate of change of the back of the queue. Jam density does not affect
throughput; it only influences the formation and dissipation of queues at a
bottleneck. The following exhibits illustrate the effects of these three calibration
parameters in a shock wave diagram format.
In Exhibit 25-28, the number 1 denotes the base condition (dashed gray line)
and the number 2 denotes the alternative condition (solid gray line). Two
demand levels D are shown. Demand rates that are greater than the bottleneck
capacity are noted with an asterisk.
Exhibit 25-28
Effects of Segment Capacity
Reducing the prebreakdown capacity increases the speed of the forming

shock wave, but the speed of the recovery wave is decreased. As a result, a
reduction in the segment’s prebreakdown capacity is expected to increase
congestion throughout the segment. Note that it is assumed a reduction in the
segment capacity has no impact on the queue discharge rate at the bottleneck in
the example above. The effects of a drop in queue discharge rate are shown in
Exhibit 25-29.
Exhibit 25-29
Effects of Queue Discharge
Rate Drop
Exhibit 25-29 shows that including a queue discharge rate drop in the
freeway model results in a reduction in bottleneck throughput after breakdown.
The factor α describes the percentage reduction from prebreakdown capacity to

queue discharge rate. A larger α corresponds to a larger drop and lower

throughput. Implementing this factor results in a drop in throughput, an increase
in the speed of the forming shockwave, and a decrease in the speed of the
recovery wave. The result is a threefold effect that leads to a higher level of
congestion, which has been demonstrated in the literature (16). It is therefore
expected that the capacity drop has a nonlinear effect on the overall facility
performance.
Exhibit 25-30 shows the effect of an increase in the jam density on wave
speeds. Interestingly, an increase in the jam density value reduces both the
forming and recovery wave speeds, thus canceling each other’s effects to some
degree. The opposite situation occurs if jam density is decreased, in which case
both the forming and recovery speeds will increase. Although jam density is
likely to affect the queue size (a higher jam density results in a smaller queue
size), it may not influence travel time values as much as the prebreakdown
capacity and queue discharge rate do.
Exhibit 25-30
Effects of Jam Density
To calibrate for bottlenecks, the analyst needs to change the capacity and
capacity drop values for different segments of the freeway facility to recreate the
bottlenecks that are observed in the field. Therefore, the analyst must first
identify recurring bottlenecks in the field.
Next, the calibration process begins with setting the segment capacity to the
HCM value for the facility’s FFS (e.g., 2,400 pc/h/ln for a 70-mi/h FFS). A value of
7% for capacity drop is recommended.
If these initial values predict the bottleneck location correctly, the analysis
proceeds to the validation step. If the model fails to identify a bottleneck, the
analyst should reduce capacity in increments of 50 pc/h/ln until a bottleneck
occurs. However, if the HCM model identifies a bottleneck that does not exist in
the field, the analyst should increase capacity in increments of 50 pc/h/ln until
the bottleneck disappears.
It is recommended that analysts wait to adjust the capacity drop value until
after the bottleneck locations have been fixed. This procedure is performed as
part of validating the queue length and travel time, as explained in Step 5.

Step 4: Calibrate Facility Demand Level

The demand level is a model input that can serve as a calibration parameter
as a last resort. Presumably, demand has been measured based on field data, and
therefore can be considered to be a fixed input. However, given the variability of
demand (i.e., day-to-day fluctuation), as well as potential errors in volume and
demand measurements, demand can become a calibration parameter after the
FFS and capacity adjustment possibilities have been exhausted.
Two potential problems may be encountered with demand levels. First, in
oversaturated conditions, it is not possible to measure the demand level
downstream of a bottleneck or within a queued segment. The volume served is
measured, rather than the demand level. Second, demand data vary from day to
day, and the selected demand levels may not represent a “typical” day. This
second problem is also true if AADT demand values are used to estimate peak
period demands. As a result, although demand level is one of the inputs to the
core freeway facility analysis, it may be subject to calibration.
To provide an example of the effect of the demand level on segment and
facility travel time, a shockwave representation of the oversaturation model used
in the core HCM freeway facilities methodology is presented. Although the
HCM uses an adaptation of the cell-transmission model to estimate queue
propagation and dissipation patterns at a bottleneck, the shockwave approach is
useful to illustrate the calibration concepts here.
Exhibit 25-31 shows the flow–density relationship under high- and low-
volume conditions for a segment that is just upstream of a bottleneck with a
reduced capacity. As before, the number 1 denotes the base condition (dashed
gray line), the number 2 denotes the alternative condition (solid gray line), and
demand rates greater than the bottleneck capacity are denoted with an asterisk.
Exhibit 25-31
Effect of Demand Level
In Exhibit 25-31 it is evident that an overall increase in demand level (from

D1* to D2* and from D1 to D2) would result in both an increase in the forming shock
wave speed and a reduction in the recovery wave speed, assuming a fixed
bottleneck capacity. In other words, an overall increase in demand level results
in a higher level of congestion throughout. The greater the difference between
upstream demand and downstream bottleneck capacity, the faster the resulting

shock wave either grows the queue (demand-to-capacity ratio > 1.0) or dissipates
the queue (demand-to-capacity ratio ≤ 1.0).
The analyst should increase the demand level in increments of 50 pc/h/ln
until all bottlenecks that are observed in the field are activated in the freeway
facility core analysis. However, if the model predicts bottlenecks that do not exist
in the field, the user should decrease the demand level in increments of 50
pc/h/ln until those bottlenecks are deactivated. This activity should be performed
in conjunction with Step 3: Calibrate Bottleneck Capacity.
Step 5: Validate Travel Time and Queue Length

The validation step has two major components:
1. Validate facility travel time, and
2. Validate queue length at active bottlenecks.
Travel Time Validation

After fixing the FFS and the bottleneck locations, the analyst should adjust
the calibration parameters further to match predicted and observed facility travel
times within a defined range (a 10% or less difference is recommended). Note
that FFS has already been fixed in Step 3 and will not be adjusted further in this
step. This process can be done by adjusting
1. Demand level,
2. Prebreakdown capacity,
3. Capacity drop, and
4. Jam density.
The analyst is trying to match reasonably well the estimated and observed
facility and segment travel times. If the model underestimates the travel time, the
analyst should consider one of the following actions:
1. Increase the demand level (in increments of 100 pc/h/ln),
2. Reduce prebreakdown capacity (in increments of 100 pc/h/ln), or
3. Increase the capacity drop (in increments of 1%).
If the model overestimates travel time, the analyst should consider one of the
following actions:
1. Reduce the demand level (in increments of 50 pc/h/ln),
2. Increase prebreakdown capacity (in increments of 50 pc/h/ln), or
3. Reduce the capacity drop (in increments of 1%).
Note that jam density is unlikely to have a significant impact on facility
travel time and is therefore not included in the steps above.
Queue Length Validation

After the facility travel time is fixed, the queue lengths at the facility’s active
bottlenecks should be matched reasonably well (i.e., within 10%) through further
adjustments to the capacity drop and jam density.

If the predicted queue length at an active bottleneck is shorter than observed

in the field, the capacity drop should be increased and the jam density should be
decreased.
However, if the predicted queue length is longer than that observed in the
field, the capacity drop should be decreased and the jam density should be
increased. It is recommended that the capacity drop be changed in increments of
1% and that the jam density be changed in increments of 10 pc/mi/ln.
CALIBRATION AT THE TRAVEL TIME RELIABILITY LEVEL

After calibrating the core freeway facility methodology and fixing the value
of its parameters, a comprehensive travel time reliability calibration is
performed. Note that the process does not allow any change in the parameters
that were calibrated in the previous step. The process requires a host of different
input variables and calibration parameters. Comprehensive reliability-level
calibration, as shown in Exhibit 25-32, starts with gathering the necessary input
data. Some of these parameters, including facility geometry and FFS, are already
known and fixed.
The process includes three major steps: whole-year demand calibration,
incident calibration, and weather calibration. In the rest of this section, each step
is presented in more detail.
To calibrate the methodology for a particular site, it is recommended that the
analyst perform an initial comprehensive reliability run using default values for
all input parameters and subsequently compare the predicted travel time index
(TTI) cumulative distribution to the observed distribution. This section provides
suggestions on how to change calibration parameters on the basis of the
difference between the two TTI distributions.
Exhibit 25-32
Comprehensive Reliability
Calibration Steps


In this step, all the input data required for a reliability analysis are gathered.
These data include
1. Demand distribution over the reliability reporting period, converted to
monthly and day-of-week demand multipliers;
2. Incident or crash rates and event durations, with the corresponding speed
and capacity adjustment factors;
3. Weather probabilities, with the corresponding speed and capacity
adjustment factors; and
4. Work zone and special event data, with the corresponding speed and
capacity adjustment factors.
Specific details about these input data are provided in Chapter 11, Freeway
Reliability Analysis.
Step 2: Determine Demand Multipliers

As mentioned above, the demand level for the seed day is either known or
calibrated at the core freeway facility analysis level. However, in addition to the
seed day, the reliability analysis requires the demand level for the other days
included in the reliability reporting period. Because it is not feasible to measure
demand level for all days, the methodology uses demand multipliers to convert
the seed day demand to demand level for different days.
Although the demand level of the seed day may be accurately measured, the
seed day may have experienced unusually low or high demand levels. In that
event, the seed day demand either inflates or deflates the demand level for the
other days of the reliability reporting period. In the example shown in Exhibit 25-
33, a high demand level on the seed day causes the resulting TTI distribution to
be consistently shifted to the right compared to the distribution observed in the
field, across the full range of the distribution. Key reliability performance
measures, such as TTImean or TTI95, are also overestimated by the procedure in the
case shown. To fix this problem (i.e., an inflated demand level for the seed day),
the analyst needs to reduce the demand level in the seed file and make additional
runs to determine whether the problem is resolved.

Exhibit 25-33
High Demand Level on the
Seed Day
Note also in Exhibit 25-33 that the intercept with the x-axis is the same for
both distributions, suggesting that the free-flow travel time at very low demands
is the same in both cases. If the two distributions do not match at very low flow
rates, this may be an indication that the free-flow speed calibration step for the
core method was not performed correctly.
In contrast, in the example shown in Exhibit 25-34, the predicted TTI values
are consistently lower than the observed values, suggesting that the seed day has
an unusually low demand level. To resolve the problem, the demand level on the
seed day should be increased and additional reliability runs performed.
Exhibit 25-34
Low Demand Level on the
Seed Day
Another calibration lever is to change the distribution of the demand

multipliers over the days of the reliability reporting period. This effort can
improve the calibration of the methodology; however, its outcome is harder to
predict. Users should change the distribution only when they have additional
field information about seasonal and daily changes in the demand level that can
bring it closer to reality.

When adjusting the demand level, users should try to bring the estimated
50th percentile TTI value to within 10% of the field-observed value. This is an
iterative process that requires adjusting either the seed day demand level or the
distribution of the demand multipliers, performing an additional comprehensive
reliability run, and comparing the modeled and field-measured 50th percentile
TTI values.
Step 3: Calibrate Incident Probabilities

When the demand level is calibrated, the predicted and observed TTI
distributions are expected to closely follow each other up to the 50th to 60th TTI
percentiles. However, nonrecurring sources of congestion usually influence the
higher percentiles of the TTI distribution. They may cause a drift in distributions
for higher percentiles, as shown in Exhibit 25-35. The figure shows a match
between the predicted (red) and observed (blue) TTI distributions, but then
suggests an overestimation of TTIs for higher percentiles with the red curve
shifted to the right. As a result, to more accurately calibrate the comprehensive
reliability analysis, the focus should be on incident and weather events. Incidents
are known to have a more considerable impact on congestion level, and therefore
the model is calibrated for incidents first, followed by weather events.
Exhibit 25-35
Overestimating the Impacts of
Nonrecurring Sources of
Congestion
Incidents can be calibrated by using a number of parameters as listed below:

1. Probability of incident severity for each month, or crash rate per 100
million vehicle-miles traveled for each month and crash-to-incident rate
and incident severity distribution, depending on the approach used for
scenario generation;
2. Incident duration attributes by severity type (mean, standard deviation,
and distribution);
3. Capacity and speed adjustment factors by severity type; and
4. Demand adjustment factors by severity type.
Incident attributes can be used to address overestimation in the tail of the
predicted TTI distribution and to bring it closer to the observed distribution. For

the example shown in Exhibit 25-35, the predicted and observed TTI
distributions almost match each other up to the 60th TTI percentile, indicating
that the demand level and base congestion level (i.e., recurring congestion) are
calibrated well. After the 60th percentile, the reliability methodology
overestimated TTI values in this case.
To reduce TTI values, the analyst should start by reducing the crash rate or
incident probability. The same effect is expected by reducing the demand
adjustment factor (for incidents). Note that in the case of severe incidents, a
significant reduction in the demand level is expected, as drivers start to reroute
to avoid the congestion. Finally, increasing the capacity and speed adjustment
factors are expected to reduce the impacts of incidents as well.
On the other hand, if the method underestimates TTI values at the tail of the
distribution (see Exhibit 25-36), the user can increase the crash rate, incident
probability, or demand adjustment factor. (Note that the maximum allowable
value for the demand adjustment factor is 1.) In addition, reducing capacity and
speed adjustment factors for incidents is expected to magnify the impacts of
incidents on travel time and consequently increase TTI values.
Exhibit 25-36
Underestimating the Impacts
of Nonrecurring Sources of
Congestion
Step 4: Calibrate Weather Probabilities

Similar to incidents, weather events influence the tail of the TTI distribution,
but to a lesser extent. The following calibration parameters are available:
1. Probability of different weather events by month,
2. Duration of each weather event,
3. Capacity and speed adjustment factors, and
4. Demand adjustment factor.
These calibration parameters are expected to impact the TTI distribution
similarly to those parameters mentioned in Step 3 for incident calibration. Note
that weather information is more likely to be accurate as it is based on 10 years of
data, while incident data are more difficult to gather. In addition, incidents have
a more considerable impact on the TTI distribution. Therefore, as mentioned

previously, it is recommended that the methodology be calibrated first through

the demand and incident data, with the analyst turning to the weather-related
parameters only if additional calibration is required.
For the example shown previously in Exhibit 25-35, the model overestimated
TTI values in the tail of the distribution. The analyst can bring the two
distributions closer to each other by reducing the probability of different weather
events or by reducing their duration. The same effect is possible by increasing
the capacity and speed adjustment factors or by reducing the demand
adjustment factor. Note that in the case of extreme weather events, a significant
reduction in the demand level is expected as travelers might decide to cancel
their trips. However, data on such trends are very scarce and hard to collect. It is
recommended that analysts adjust the demand adjustment factors only when
there is evidence or knowledge of the trends on the study facility.
On the other hand, when the methodology underestimates TTI values in the
tail of the distribution, as in Exhibit 25-36, the analyst can increase the probability
of weather events or increase their durations. In addition, a reduction in capacity
and speed adjustment factors is expected to move the distribution to the right.
Step 5: Validation
Changing all of the calibration parameters at the same time might lead to
unexpected results. Therefore, the user is encouraged to change only one
parameter at a time, run the comprehensive reliability methodology, plot and
evaluate the new TTI distribution, and only then decide whether and how to
change other parameters. The use of a computational engine makes running
repeated reliability analyses with changing inputs a straightforward process.
The analyst should try to bring at least the predicted 80th and 95th percentile
TTI values within 10% of the field-observed values. Preferably, additional
percentiles should match the field data, although a perfect match may not be
achievable. The collected field data should span the same reliability reporting
period that was selected for the analysis, to ensure that results are comparable.
CALIBRATION AT THE RELIABILITY STRATEGY ASSESSMENT LEVEL

Calibration at the reliability strategy assessment level is only possible for
strategies that have already been implemented in the field. For other strategies,
calibration is not possible, other than based on expert judgment or comparison to
an alternative tools analysis. However, the user can run a set of sensitivity
analyses for each strategy to identify the trends and make sure that they match
expectations. For example, a ramp-metering strategy is expected to shift the TTI
distribution to the left, toward lower TTI values. The lower the metering rate, the
larger the expected shift. If such a trend is observed, and if its extent is in a
reasonable range, one can conclude that methodology works reasonably.
Similar to the calibration procedure at the comprehensive reliability level, the
analyst must first gather all input data on facility geometry, free-flow speed, and
demand level. Note that an important assumption is that the demand, incident,
and weather calibration parameters are already fixed in the comprehensive

reliability calibration step. As a result, the analyst is left with the remaining
calibration parameters that are specific to each scenario.
In general, different scenarios may change a facility’s free-flow speed,
capacity, demand, incident probability, and average incident duration.
Therefore, “scenario-specific” calibration parameters are
1. Speed adjustment factor,
2. Capacity adjustment factor,
3. Metering rate,
4. Demand adjustment factor,
5. Incident probability, and
6. Average incident duration.
It is recommended that the analyst make a reliability strategy assessment run
based on a combination of field measurements and default values, plot the
predicted TTI distribution, and then compare the result to the field observation.
Similar to the comprehensive reliability calibration procedure, the analyst can
then make changes in the calibration parameters to bring the predicted
distribution closer to the observed one.
Based on the modifications that each strategy makes in the freeway
methodology, the user can adjust the corresponding calibration parameters.
Similar to calibrating the comprehensive reliability methodology, increasing the
speed adjustment factor is expected to reduce travel time across the facility,
while reducing it has an opposite effect. Increasing the value of the capacity
adjustment factor is expected to reduce the facility travel time. Increasing the
metering rate will allow more vehicles to enter the mainline and is expected to
increase the facility travel time and perhaps activate bottlenecks in merge areas.
On the other hand, reducing the metering rate is likely to reduce travel time
across the facility and eliminate bottlenecks at merge areas. Increasing the
demand adjustment factor is expected to increase travel time throughout the
facility and shift the TTI distribution toward larger TTI values, while reducing it
has the opposite effect. Increasing the incident probability is expected to shift the
tail of the TTI distribution toward higher TTI values, while reducing it shifts the
tail toward lower values. Finally, changing the average incident duration is
expected to influence the TTI distribution similarly to incident probability.
The analyst should avoid making several changes in calibration parameters
at the same time, as this may result in changes in TTI distribution that are hard to
explain and may make the calibration procedure more difficult. Instead, analysts
should select one calibration parameter at a time, make changes, rerun the
strategy assessment procedure, plot the TTI distribution, compare it to the field
distribution, and make other changes as necessary.
The user needs to first identify the main source of difference between the
predicted and field TTI distributions. If a difference between the two
distributions is observed throughout all ranges of TTIs (similar to Exhibit 25-33
and Exhibit 25-34), changing parameters such as the speed adjustment factor,
capacity adjustment factor, demand adjustment factor, and metering rate is

expected to bring the two distributions closer. The analyst should aim for a
maximum of 10% difference between the 50th percentile of the predicted and
observed TTI distributions at this stage.
On the other hand, if the difference between TTI distributions is observed
mostly in the tail of the distribution (similar to Exhibit 25-35 and Exhibit 25-36),
changing the incident probability and duration is expected to move the predicted
distribution to the right. The analyst should aim for a maximum 10% difference
between the 80th and 95th percentiles of the predicted and observed TTI
distributions at this stage as well.

9. FREEWAY SCENARIO GENERATION
INTRODUCTION
This section provides details of the freeway scenario generation process. An
overview of this process is provided in Chapter 11, Freeway Reliability Analysis,
and elsewhere (17).
Freeway scenario generation utilizes a hybrid process, which includes
deterministic and stochastic methods for modeling traffic demand, weather
events, work zones, and incidents. The freeway reliability methodology uses a
deterministic, calendar-based approach to model traffic demand levels and
scheduled, significant work zone events. It uses a stochastic (Monte Carlo)
approach to assign the occurrence of incident and weather events to scenarios.
The method enumerates the different operational conditions on a freeway facility
on the basis of varying combinations of factors affecting the facility travel time.
Each unique set of operational conditions constitutes a scenario. A single
replication of a scenario represents a unique combination of a day of week and
month of year. The following seven principal stages, depicted in Exhibit 25-37,
are involved in the scenario generation process:
• Stage 1, based on the user inputs, computes the number of different
demand combinations and the resulting number of scenarios, along with
their probabilities. These values also depend on the duration of the
reliability reporting period.
• Stage 2 uses local traffic demand data to characterize the demand levels in
the generated scenarios in a deterministic, calendar-based manner.
• Stage 3 incorporates scheduled work zones deterministically based on the
calendar.
• Stage 4 incorporates published local weather event information, and
generates the number and type of weather events, consistent with local
data.
• Stage 5 randomly assigns the generated weather events in Stage 4 to the
scenarios generated in Stage 1.
• Stage 6 utilizes the local crash or incident database to generate the
number and severity of incident events, consistent with local data.
• Stage 7 randomly assigns incidents and their characteristics to each
generated scenario in Stage 1.
The time frame within a given day when the reliability analysis is performed
is called a study period. It consists of several contiguous 15-min analysis periods,
which is the smallest temporal unit of analysis used in reliability procedures. The
smallest spatial unit on the facility is an HCM analysis segment (see Chapters 12–
14). The reliability reporting period is the time period over which the travel time
distribution is generated (typically, but not necessarily, one year).
Each scenario representing a study period is characterized by a unique set of
segment capacities, demands, free flow speeds, and number of lanes, for both
general purpose and managed lane segments on the freeway facility. Various
Freeway Scenario Generation Chapter 25/Freeway Facilities: Supplemental

scenarios are created by adjusting one or more of the above parameters. A

probability value is associated with each scenario that represents its likelihood of
occurrence. This probability is computed on the basis of the number of scenarios
and replications.
Stage 1 Exhibit 25-37

Specify Demand Combinations, Process Flow Overview for
Number of Replications, and Freeway Scenario Generation
Scenario Probabilities
Stage 2
Traffic Demand
Incorporate Demand Variation in the
Database
Scenarios
Stage 3
Incorporate Scheduled Work Zone
Events in Scenarios Deterministically
Stage 4
Weather
Generate Weather Events for
Database
Scenarios
Stage 5
Randomly Assign Weather Events to
the Scenarios
Stage 6
Incident
Generate Incident Events for
Database
Scenarios
Stage 7
Randomly Assign Incident Events to
the Scenarios
Scenarios are generated in such a manner that the characteristics of the

factors affecting travel time within scenarios best match the input, field-observed
conditions. For example, the distribution of the number of incidents generated in
various scenarios should yield a distribution similar to that observed in the field.
Exhibit 25-38 depicts such an example, in which the number of incidents
modeled in all scenarios (histogram) is designed to match field-observed values
(curve).
Chapter 25/Freeway Facilities: Supplemental Freeway Scenario Generation

Exhibit 25-38
Distribution of Number of
Incidents in the Scenarios
Therefore, the process of generating scenarios effectively turns into an

optimization problem. The objective is to maximize the match (or minimize the
difference) between the predicted and field-observed distributions by assigning
appropriate traffic demand levels, weather events, work zones, and incidents
within the different scenarios. Eight distributions are considered in the scenario
generation procedure:
1. Temporal distribution of traffic demand level (typically expressed as a
ratio of scenario demand to AADT),
2. Temporal distribution of weather event frequency (by calendar month,
randomly assigned to scenarios),
3. Distribution of average weather event duration by weather event type (by
calendar month),
4. Temporal distribution of incident event frequency (by calendar month,
weighted in the facility by segment VMT),
5. Distribution of incident severity (user specified),
6. Distribution of incident duration by severity (user specified),
7. Distribution of incident event start time (random), and
8. Spatial distribution of incident events (random).
The scenario generation method attempts to generate scenarios such that all
eight specified distributions match field observations, with consideration for the
need to round to integer values and to the 15-min duration of the analysis
period. Such rounding is not likely to generate any significant systematic bias in
the analysis.

METHODOLOGY
The freeway reliability scenario generation methodology consists of 34 steps.
Exhibit 25-39 shows the methodology’s process flow. Note that when managed
lanes are present on the facility, the reliability scenarios should also consider
their varying operational characteristics. The methodology assumes traffic
demand levels and weather events affect both general purpose and managed
lane operations simultaneously. However, the methodology does not account for
scheduled work zone events on the managed lanes. Analysts should repeat Steps
19–34 should they desire to model incident events on the managed lanes
separately. An explanation of each step in the process flow follows. All variables
used in this section are defined in Section 2.
Step 1: Prepare Necessary Data for the Reliability Analysis

In this step, the analyst provides all necessary data for executing the scenario
generation method. The starting point is preparing a complete seed file describing
the facility’s demand and geometry for a single study period. Developing the
seed file is akin to developing a data set for the core methodology, as described
in Chapter 10. In addition, for scenario generation purposes, additional data
must include (a) the start and end clock times of the study period, (b) the
duration of the reliability reporting period, (c) the seed file date, (d) the series of
demand multipliers (see Step 4) for each demand combination, (e) the nearest
metropolitan area to the facility (for weather station data), (f) the crash or
incident rates by month of year on the facility, and (g) other local inputs.
Step 2: Determine the Number of Demand Combinations

The freeway scenario generation method defines a demand combination as
the combination of a specific weekday and month of year. Although demand
levels in different demand combinations might be very similar (e.g., Tuesday and
Wednesday afternoon volumes), the methodology handles them separately to
keep the process simple. For a 1-year, weekday-only analysis, there are 60 such
combinations (5 × 12). The number of demand combinations is defined by the
variable NDC.
Step 3: Create Scenario Sets and Associate Them with Demand

Combinations
As a default, the methodology creates four scenario replications for each
demand combination. The rationale behind four replications is that each demand
combination usually consists of four or five calendar days. However, if a short-
duration reliability reporting period is considered, the number of replications
must be increased to capture sufficient variability in the travel time distribution.
Typically, however, the default number of scenarios for a 1-year, weekday-only
analysis would be 4 × 60 = 240 scenarios. The method allows the analyst to
specify the number of replications per reliability analysis.

Exhibit 25-39 Legend:

Detailed Freeway Scenario
START User interaction
Generation Flowchart
Monte Carlo modeling (computational engine)
STEP 1
Prepare necessary data for the Deterministic modeling (computational engine)
reliability analysis
STEP 2 STEP 3 STEP 4

Determine number of Create scenario sets and associate Assign traffic demand level to each
demand combinations (DCs) [60] them with DCs [4] scenario set [240]
Modeling Traffic Demand
STEP 5
Calculate scenario probabilities [240]
STEP 7
Select an unassigned WZ and calculate the
active WZ ratios for each month in which
the WZ is active No
STEP 8
Calculate the adjusted number of DC
STEP 6
replications for which the WZ is active
All work zones
(WZs) assigned?
STEP 9
For each DC, assign the WZ to the
calculated number of WZ replications in
Yes
Step 8
Modeling Work Zones
STEP 10
Group scenarios by month [20]
STEP 11
Compute expected frequency of weather
events by month
STEP 12
Select a month in which weather events
have not been assigned [12]
STEP 13
Update the list of weather events (LWE)
STEP 14
From the LWE for current month, randomly
select a weather event and assign it to a
scenario with a start time
STEP 15 No
STEP 17
Is there any
STEP 16 No Are all weather
Yes overlap with
Undo the last weather events in current
another weather
event assignment month assigned
event in a
to scenarios?
scenario?
Yes
STEP 18 No
Are all weather
PROCEED TO STEP 19 events for all
months assigned
to scenarios?
Modeling Weather Events

Note: Numbers in brackets are default values.


Detailed Freeway Scenario
Generation Flowchart
FROM STEP 18 STEP 27
Generate a set of incident durations
for each incident severity type
STEP 19
Select a month to which incident
STEP 28
events are not yet assigned [12]
Randomly assign incident durations
with respect to incident severity types
STEP 20
Compute expected frequency of
incidents in current month STEP 29
Generate the distribution of incident
start times and locations
STEP 21
Generate a set of incident frequencies
for all scenarios in the current month
STEP 30
Generate a set of incident start times
and locations for all incidents in LIE
STEP 22
Randomly assign each generated
incident frequency to a scenario in
current month
STEP 31
From LIE, select an incident whose
start time and location are not
STEP 23 assigned, and randomly assign a start
Update the list of incident events time and location from previous step
(LIE)
STEP 32
STEP 33
Is there any Yes
STEP 24 Undo the last start
conflict with
No Are incident time and location
another incident
frequencies assignment
event in LIE?
assigned for all
months?
No
Yes
STEP 34
No Do all incidents
STEP 25 have a start time
Generate incident severity types for and location
each incident event assigned?
Yes
STEP 26
For each incident event, randomly
assign a generated incident severity FINISH
type
Modeling Incident Events

Note: Numbers in brackets are default values.
For each scenario, a set of adjustment factors is created for capacity, speed,
demand, and number of lanes (CAF, SAF, DAF, and NLAF, respectively). At this
point, each scenario contains default values for CAF, SAF, and DAF (all equal to
1) and NLAF (equal to 0), but the scenarios do not yet contain any demand,
weather, or incident data. NScen represents the total number of scenarios and is
computed as:
𝑁𝑆𝑐𝑒𝑛 = 4 × 𝑁𝐷𝐶 Equation 25-71
Step 4: Assign a Traffic Demand Level to Each Scenario Set

In this step, a traffic demand level is assigned to each scenario set (i.e., the
number of replications used per scenario). For this purpose, demand multipliers,
representing the ratio of the traffic demand level in each demand combination to
the AADT are used to generate each scenario demand level. Because each
scenario is associated with a unique demand combination, the ratio of the

scenario demand multiplier to the seed file demand multiplier is used to

determine the scenario demand, as shown in Equation 25-72.
Equation 25-72 𝐷𝑀(𝑠)

𝐷𝐴𝐹𝑠 (𝑡𝑝, 𝑠𝑒𝑔) = ∀𝑡𝑝 ∈ 𝑆𝑃 and 𝑠𝑒𝑔 ∈ 𝑆𝑒𝑔𝑚𝑒𝑛𝑡𝑠
𝐷𝑀(𝑆𝑒𝑒𝑑𝑡𝑝 )
where
DAFs(tp, seg) = demand adjustment factor for scenario s, period tp, and
segment seg;
DM(Seedtp) = demand multiplier associated with the seed file; and
DM(s) = demand multiplier associated with scenario s.
The process to calculate any demand value of any cell in a scenario is to
multiply the cell demand value in the corresponding seed file (for the same HCM
segment and analysis period) with the appropriate DAF, as shown in Equation
25-72. Note that if the facility contains managed lanes, the traffic demand level
generated in this step will be effective for both the general purpose and managed
lanes.
Step 5: Calculate Scenario Probabilities

The probability of a scenario occurrence is strictly a function of the number
of days in the associated demand combination. Note that the probability of a
scenario is fixed at this step and will not be altered in any subsequent steps.
Simply stated, the probability of each scenario does not change by incorporating
weather and incident events. The probability of each scenario is computed based
on Equation 25-73.
𝑛Day,𝐷𝐶𝑠
Equation 25-73 𝑃{𝑠} = 𝑁
4 × ∑𝑘=1
𝐷𝐶
𝑛Day,𝑘
where
P{s} = probability of scenario s,
DCs = demand combination associated with scenario s,
nDay,k = number of days in the reliability reporting period associated with
demand combination k (typically four for a 1-year weekday analysis),
and
NDC = number of demand combinations.
After computing each scenario’s probability, the probabilities are assigned to
the scenarios created in Step 3. The probability of a scenario is a function of the
number of days in the associated demand combination, which is typically four or
five for a whole-year analysis. For a typical 1-year, weekday-only analysis, the
probability of each scenario is approximately 1/240 or 4.33%.
Step 6: Determine Whether All Work Zones Have Been Assigned

If there are no scheduled work zones during the reliability reporting period,
or if all scheduled work zones have been assigned to scenarios, the process flow
proceeds to Step 10. Otherwise, the process moves to Step 7 and assigns the next

work zone. If there are no work zones considered in the reliability analysis, the
process flow proceeds to Step 10.
Step 7: Calculate Active Work Zone Ratios

In this step, the parameter rDC is calculated. This parameter is the ratio of
each weekday type in which the work zone is active in a given month to the total
number of each weekday type occurring in a given month. An unassigned work
zone event is selected, and rDC is calculated for each month in which the work
zone is active.
Step 8: Calculate the Adjusted Number of Replications

For each affected demand combination in which a work zone is present,
Equation 25-74 is used to calculate N
‾ DC,WZ, the adjusted number of replications of
a demand combination for which the work zone is active.
̅𝐷𝐶,𝑊𝑍 = round(𝑟𝐷𝐶 × 𝑁𝑟 , 0)
𝑁 Equation 25-74
Step 9: Assign the Work Zone to the Work Zone Replications

For each demand combination of each month in which the work zone is
active, assign the work zone to the adjusted number of replications of each
demand combination (equivalently scenarios) calculated in Step 8.
Step 10: Group Scenarios by Month

The attributes of inclement weather events are assumed to vary only by the
month of the year. As such, in Step 10, all scenarios associated with a given
month of year are grouped. Typically, this step involves grouping 20 scenarios
(four replications of five weekdays each per month.)
Step 11: Compute the Expected Frequency of Weather Events by Month

The method uses the expected frequencies of weather events to create and
characterize weather events. Historical data are used to estimate the probability,
average duration, and standard deviation of duration of different weather
conditions. Weather event likelihoods are reported in timewise probabilities that
were computed for 103 metropolitan areas in the United States on the basis of 10
years of data. The resulting probability tables are provided as resource material
in the Technical Reference Library in online HCM Volume 4. A listing of the 97
locations used to create the weather data is provided in Exhibit 25-40.
Only weather events that reduce capacity by more than 5% are included in
the probability calculations. The average event duration and the standard
deviation for each weather category are calculated by using the 10-year weather
data set for each weather station. The probability of weather event type i in
month j is found from Equation 25-75.
𝑃𝑊 {𝑖, 𝑗}
Sum of all SP durations in minutes in month 𝑗 that weather type 𝑖 is present
= Equation 25-75
Sum of all SP durations in minutes in month 𝑗

where SP indicates study period, and PW{i, j} is the probability of encountering

weather type i in month j.
Exhibit 25-40 Airport Airport

Listing of Weather Stations # City, State # City, State
Code Code
with Available Weather Data 1 KBHM Birmingham, AL 50 KGSO Greensboro, NC
2 KLIT Little Rock, AR 51 KRIC Raleigh, NC
3 KPHX Phoenix, AZ 52 KOMA Omaha, NE
4 KTUS Tucson, AZ 53 KABQ Albuquerque, NM
5 KBFL Bakersfield, CA 54 KLAS Las Vegas, NV
6 KFAT Fresno, CA 55 KALB Albany, NY
7 KLAX Los Angeles, CA 56 KBUF Buffalo, NY
8 KMOD Modesto, CA 57 KLGA New York, NY
9 KCMA Oxnard, CA 58 KPOU Poughkeepsie, NY
10 KROC Riverside, CA 59 KSAC Rochester, NY
11 KSAN Sacramento, CA 60 KSYR Syracuse, NY
12 KSAT San Diego, CA 61 KCAK Akron, OH
13 KSJC San Francisco, CA 62 KCVG Cincinnati, OH
14 KSLC San Jose, CA 63 KCLE Cleveland, OH
15 KSDF Stockton, CA 64 KCMH Columbus, OH
16 KCOS Colorado Springs, CO 65 KDAY Dayton, OH
17 KDEN Denver, CO 66 KTOL Toledo, OH
18 KBDL Hartford, CT 67 KYNG Youngstown, OH
19 KDCA Washington, DC 68 KOKC Oklahoma City, OK
20 KFMY Cape Coral, FL 69 KTUL Tulsa, OK
21 KJAX Jacksonville, FL 70 KPDX Portland, OR
22 KTPA Lakeland, FL 71 KABE Allentown, PA
23 KMIA Miami, FL 72 KMDT Harrisburg, PA
24 KSRQ North Port, FL 73 KLNS Lancaster, PA
25 KMCO Orlando, FL 74 KPHL Philadelphia, PA
26 KMLB Palm Bay, FL 75 KPIT Pittsburgh, PA
27 KATL Atlanta, GA 76 KAVP Scranton, PA
28 KAGS Augusta, GA 77 KPVD Providence, RI
29 PHNL Honolulu, HI 78 KCHS Charleston, SC
30 KDSM Des Moines, IA 79 KCAE Columbia, SC
31 KBOI Boise City, ID 80 KGSP Greenville, SC
32 KORD Chicago, IL 81 KCHA Chattanooga, TN
33 KIND Indianapolis, IN 82 KTYS Knoxville, TN
34 KICT Wichita, KS 83 KMEM Memphis, TN
35 KSEA Louisville, KY 84 KBNA Nashville, TN
36 KBTR Baton Rouge, LA 85 KAUS Austin, TX
37 KMSY New Orleans, LA 86 KDFW Dallas, TX
38 KBOS Boston, MA 87 KELP El Paso, TX
39 KCEF Springfield, MA 88 KIAH Houston, TX
40 KORH Worcester, MA 89 KMFE McAllen, TX
41 KBWI Baltimore, MD 90 KSCK San Antonio, TX
42 KPWM Portland, ME 91 KOGD Ogden, UT
43 KDTW Detroit, MI 92 KPVU Provo, UT
44 KGRR Grand Rapids, MI 93 KRIV Richmond, VA
45 KMSP Minneapolis, MN 94 KORF Virginia Beach, VA
46 KMCI Kansas City, MO 95 KSFO Seattle, WA
47 KSTL St. Louis, MO 96 KMSN Madison, WI
48 KJAN Jackson, MS 97 KMKE Milwaukee, WI
49 KCLT Charlotte, NC
Source: Zegeer et al. (18).
Equation 25-76 is used to convert those reported probabilities into rounded

expected monthly weather event frequencies.
𝑃𝑡 {𝑤, 𝑗} × 𝐷𝑆𝑃 × 𝑁𝑆𝑐𝑒𝑛,𝑗
Equation 25-76 𝐸[𝑛𝑤 , 𝑗] = round ( )
𝐸15𝑚𝑖𝑛 [𝐷𝑤 ]

where
E[nw, j] = expected frequency of weather event w in month j, rounded to
the nearest integer;
Pt{w, j} = timewise probability of weather type w in month j;
DSP = duration of study period SP (h);

NScen,j = number of scenarios associated with month j of the reliability
reporting period; and
E15min[Dw] = expected duration of weather event w rounded to the nearest 15-
min increment.
In this step, the E[nw, j] values for each weather type w are computed in each
month j of the reliability reporting period. Note that the unit of the expected
frequency is events per total scenario hours in each month. Also note that the
minimum value for E15min[Dw] is 0.25 h.
For example, if the study period is 5 h, if the probability of light rain during
that month and time period (typically associated with about 20 scenarios) is 0.10,
and if the average light rain event lasts 1 h, then the expected number of light
rain events in that month is (0.1 × 5 × 20)/1, which rounds to 10 light rain weather
events in that month, or 10 h of light rain in the month.
Step 12: Select a Month with Unassigned Weather Events

The process of assigning weather events in a month is independent of other
months in the reliability reporting period. The process is carried out on a
monthly basis. For this purpose, one month from the reliability reporting period
without an assigned weather event is selected in the next steps.
Step 13: Update the List of Weather Events

In this step, the list of weather events is updated. That is, the weather events
associated with the current month will have their characteristics (durations,
CAFs, and SAFs) assigned.
Step 14: Assign Weather Events and Start Times to Scenarios

In this step, a weather event that was updated in the list of weather events in
Step 13 is selected and randomly assigned to a scenario in the current month. The
assignment of weather events to scenarios is carried out consistent with the
relative scenario probabilities. In addition, a start time is randomly assigned to
the selected weather event from the list of weather events. Because actual data on
the start time of weather events are lacking, those are assigned randomly based
on a uniform distribution.
Step 15: Identify Overlaps Between Weather Events in a Single Scenario

This step ensures there will be no temporal overlap between two weather
events within a single scenario. Possible overlaps between weather events are
checked, and if they exist, then Step 16 is executed. Otherwise, the process moves
to Step 17.

Step 16: Undo the Most Recent Weather Event Assignment

If there is an overlap between weather events, the most recent weather
assignment is undone. The process then goes back to Step 14 to reassign a
scenario and a start time for the weather event.
Step 17: Check for Unassigned Weather Events in the Current Month
This step checks that all weather events present in the list of weather events
have been assigned. If one or more unassigned weather events exist for the
current month, the process returns to Step 14 to select another unassigned
weather event.
Step 18: Check for Unassigned Weather Events in All Months

Once all weather events have been assigned to scenarios across all months in
the reliability reporting period, the methodology proceeds to the incident
modeling stage. Otherwise, the process returns to Step 12 to select another month
from the reliability reporting period to have its weather events modeled in the
associated scenarios.
Step 19: Select a Month with Unassigned Incidents

The methodology allows the user to directly enter monthly incident
occurrences on a given facility during the study period into the procedure,
should these values be available. Optimally, the distribution of incident
durations, the start times, and the distribution of incidents by severity (e.g.,
number of lanes closed) could also be entered directly from a local incident
database.
However, in most cases (including predictive reliability applications), these
data will not be available, and incident events will need to be estimated from
incident or crash rates (which vary by month and traffic demand levels). The
methodology accounts for the correlation between incident and crash-only rates.
Because the method attempts to generate the number of incident events based on
their distributions, a high number of incidents could be assigned to a scenario
that is associated with a low traffic demand level. The average traffic demand
level for each month is therefore computed and used to characterize the incident
events within scenarios in each month. Incident events are assigned to different
months of the reliability reporting period independently. Therefore, a month
from the reliability reporting period without any assigned incidents is first
selected in the next steps.
Step 20: Compute the Expected Incident Frequency

The expected frequency of all incidents on the facility per study period in a
given month j is computed with Equation 25-77.
Equation 25-77 𝑛𝑗 = 𝐼𝑅𝑗 × 𝑉𝑀𝑇𝑗
where
nj = expected frequency of all incidents in the study period for month j,
rounded to the nearest integer;
IRj = incident rate per 100 million VMT in month j; and

VMTj = average vehicle miles traveled for scenarios in month j, after adjusting
the demand in the base scenario with the appropriate demand
multipliers and multiplying by the facility length in miles.
If IRj is not locally available, Equation 25-78 can be used to estimate it.
𝐼𝑅𝑗 = 𝐶𝑅𝑗 × 𝐼𝐶𝑅 Equation 25-78
where CRj is the local facilitywide crash rate per 100 million VMT in month j and
ICR is the local incident-to-crash ratio. In the absence of other data, a national
default value for ICR is 4.9.
When the crash rate is not available locally, the Highway Economic
Requirements System (HERS) model can be used to estimate it (19). Agencies
may also use other predictive models such as the Highway Safety Manual (20). The
crash or incident rate is estimated per 100 million VMT. The HERS model uses
Equation 25-79 to estimate the crash rate.
𝐶𝑅 = (154.0 − 1.203 × 𝐴𝐶𝑅 + 0.258 × 𝐴𝐶𝑅2 − 0.00000524 × 𝐴𝐶𝑅5 ) Equation 25-79
× 𝑒 0.0082×(12−𝐿𝑊)
where CR is the crash rate per 100 million VMT, ACR is the facility AADT
divided by its two-way hourly capacity, and LW is the lane width in feet.
Step 21: Generate a Set of Incident Frequencies

The distribution of the number of incidents in a study period can be
characterized by a Poisson distribution. Assume there are NScen,j scenarios
(typically 20) associated with the current month j. Then, on average, nj × NScen,j
incidents (rounded to the nearest integer) need be to generated and assigned to
scenarios. Therefore, a set of NScen,j numbers should be generated that best
matches a Poisson distribution with a mean value of nj, per Equation 25-80.
For this purpose, an adjustment parameter δ1 is defined. By solving Equation
25-80, the frequency of incidents for a set of NScen,j scenarios can be computed,
following the Poisson distribution. The values of the adjustment parameter
usually hover around 1 and are estimated from the equality.
+∞
∑(round[𝛿1 × 𝑁𝑆𝑐𝑒𝑛,𝑗 × Prob{𝑛𝑖𝑛𝑐 = 𝑘}]) = 𝑁𝑆𝑐𝑒𝑛,𝑗 Equation 25-80

𝑘=0
where ninc is the number of incidents and other variables are as defined
previously. Subsequently, the number of scenarios that are assigned k incidents
(k = 0 → ) is determined by Equation 25-81.
Number of scenarios with 𝑘 incident events = round[𝛿1 × 𝑁𝑆𝑐𝑒𝑛,𝑗 × Prob{𝑛𝑖𝑛𝑐 =
Equation 25-81
𝑘}]
where all variables are as defined previously. By setting different k-values in the
above equation, a set of monthly incident frequencies will be generated in this
step.

Step 22: Assign Incidents to Scenarios

The incidents generated in Step 21 are randomly assigned to the scenarios
associated with the current month. A random number is drawn with respect to
scenario probabilities to determine the assigned scenario number.
Step 23: Update the List of Incident Events

The list of incident events is updated after the incident frequencies are
generated. This list holds information for each incident event in the entire
reliability analysis. The associated incident event information includes the
assigned scenario number, calendar month, incident duration, incident impact
factors (e.g., CAF, SAF), incident segment location, and incident start time.
Step 24: Check for Unassigned Incidents

This step ensures that incident event frequencies are generated and assigned
to scenarios for all months in the reliability reporting period. Once incidents in
all months have been processed in Steps 20–23, the scenario generation process
continues to Step 25.
Step 25: Generate Incident Severities for Each Incident Event

A set of incident severities is generated for the entire set of incidents
developed in Step 21. Note that this step is not carried out on a monthly basis.
The distribution of incident severities must be known a priori for incorporation
in the methodology. This distribution is defined by 𝔾(i), which is assumed to be
homogeneous across the facility and different demand levels.
Agencies can estimate this distribution by analyzing their incident logs or
they can use national default values. Equation 25-82 gives the definition of 𝔾(i) as
a discrete distribution, where i denotes the incident severity type (e.g., 𝑖 = 1 is a
shoulder closure, and 𝑖 = 5 is a four-lane closure).
ℊ1 𝑖=1
ℊ2 𝑖=2
ℊ3 𝑖=3
Equation 25-82 𝔾(𝑖) =
ℊ4 𝑖=4
ℊ5 𝑖=5
{
Suppose a total of NScen,Inc incidents was generated in Steps 19–24. To generate
incident severities, an adjustment parameter δ2 is defined. By solving Equation
25-83, incident severities for all incidents in the list of incident events will be
estimated that will follow the prespecified 𝔾(i) distribution.
Equation 25-83 ∑(round[𝛿2 × 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐 × 𝔾(𝑖)]) = 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐

𝑖
where all variables are as previously defined. The adjustment parameter is
determined with Equation 25-83, and the number of scenarios that are assigned
incident severity type 𝑖 is determined by Equation 25-84.
Equation 25-84 Number of incidents with severity 𝑖 = (round[𝛿2 × 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐 × 𝔾(𝑖)])

The distribution of incident severity 𝔾(i) is shown in Equation 25-85. These

values are based on national default values (18).
0.754 i = 1 (shoulder closed)
0.196 i = 2 (one lane closed)
𝔾(𝑖) = 0.031 i = 3 (two lanes closed) Equation 25-85
0.019 i = 4 (three lanes closed)
{ 0 i = 5 (four or more lanes closed)
Step 26: Assign Incident Severity Type

The incident severities generated in Step 25 are randomly assigned to the
incidents in the list of incident events.
Step 27: Generate Incident Durations by Incident Severity Type

The duration of each incident severity type is assumed to follow a lognormal
distribution (15). Exhibit 25-41 shows default parameters for the incident
duration distribution developed through research (18).
No. of Lanes Closed Exhibit 25-41

Statistics Shoulder 1 2 3 or more Incident Duration Distribution
Range 8.7‒58 16‒58.2 30.5‒66.9 36‒93.3 Parameters in Minutes
Average 34.0 34.6 53.6 69.6
Median 36.5 32.6 60.1 67.9
Standard deviation 15.1 13.8 13.9 21.9
Because NInc,i incidents are associated with severity i, a set of NInc,i numbers
can be generated that best matches a lognormal distribution of incident
durations. For this purpose, an adjustment parameter δ3 is defined, as shown in
Equation 25-86.
∑(round[𝛿3 × 𝑁𝐼𝑛𝑐,𝑖 × Prob{𝐼𝑛𝑐𝐷𝑢𝑟 = 𝑡, 𝐼𝑛𝑐𝑇𝑦𝑝𝑒 = 𝑖}]) = 𝑁𝐼𝑛𝑐,𝑖 Equation 25-86

𝑡
where IncDur is the incident duration in minutes, IncType is the incident severity
type (1–5, as listed in Equation 25-85), and other variables are as defined
previously.
By solving Equation 25-86, the adjustment parameter is determined. The
number of scenarios that are assigned an incident duration t are then determined
by Equation 25-87.
Number of scenarios assigned incident severity 𝑖 = round[𝛿3 × 𝑁𝐼𝑛𝑐,𝑖 × Equation 25-87
Prob{𝐼𝑛𝑐𝐷𝑢𝑟 = 𝑡, 𝐼𝑛𝑐𝑇𝑦𝑝𝑒 = 𝑖}]

By inserting different t-values in Equation 25-87, a set of incident durations
for each incident severity type will be generated.
Step 28: Randomly Assign Incident Durations by Severity

The incident durations generated in Step 27 are randomly assigned to the
incidents in the list of incident events on the basis of the incident severity.

Step 29: Generate the Distribution of Incident Start Times and Locations
In this step, the distribution of each incident start time and location is
assigned based on Step 20, with the likelihood of having an incident on a
segment in a given analysis period being correlated to the segment VMT. The
distribution of incident start times will coincide with the distribution of facility
VMT across all analysis periods. Further, the distribution of the location of an
incident will be similarly tied to the distribution of VMT for each segment across
the study period. Since VMTseg,p represents the VMT on segment seg during
analysis period p in the seed file, the distribution of the incident locations will be
determined by Equation 25-88.
∑𝑝 𝑉𝑀𝑇𝑥,𝑝
Equation 25-88 Prob{Location = segment 𝑥} =
∑𝑠𝑒𝑔,𝑝 𝑉𝑀𝑇𝑖,𝑝
where Location is the segment in which the incident occurs.
In a similar manner, the distribution of the incident start time will be
determined by Equation 25-89.
∑𝑖 𝑉𝑀𝑇𝑖,𝑦
Equation 25-89 Prob{𝑆𝑡𝑎𝑟𝑡𝑇𝑖𝑚𝑒 = analysis period 𝑦} =
∑𝑖,𝑝 𝑉𝑀𝑇𝑖,𝑝
where StartTime is the analysis period in which the incident starts.
Step 30: Generate Incident Start Times and Locations for All Incidents
Assuming there are NScen,Inc incidents in the list of incident events, two sets of
NScen,Inc numbers should be generated that best match the incident start time and
location distributions. For this purpose, two adjustment variables, δ4 and δ5, are
defined by Equation 25-90 and Equation 25-91, respectively.
Equation 25-90 ∑(round[𝛿4 × 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐 × Prob{𝐿𝑜𝑐𝑎𝑡𝑖𝑜𝑛 = 𝑥}]) = 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐

𝑥
Equation 25-91 ∑(round[𝛿5 × 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐 × Prob{𝑆𝑡𝑎𝑟𝑡𝑇𝑖𝑚𝑒 = 𝑦}]) = 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐

𝑦
By solving Equation 25-90 and Equation 25-91, the adjustment parameters

are determined. The number of incidents that are assigned to any segment seg are
then determined from Equation 25-92.
Equation 25-92 Number of incidents assigned to segment 𝑠𝑒𝑔 = round[𝛿4 × 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐 ×

Prob{𝐿𝑜𝑐𝑎𝑡𝑖𝑜𝑛 = 𝑠𝑒𝑔}]
Finally, the number of incidents that are assigned a starting time (analysis
period p) is determined from Equation 25-93.
Number of incidents assigned a starting time in analysis period 𝑝 =
Equation 25-93
round[𝛿5 × 𝑁𝑆𝑐𝑒𝑛,𝐼𝑛𝑐 × Prob{𝑆𝑡𝑎𝑟𝑡𝑇𝑖𝑚𝑒 = 𝑝}]
By inserting different seg and p values in the above equations, a set of
incident locations and start times will be generated.

Step 31: Assign Start Times and Locations to Incidents

In this step, an incident from the list of incident events is selected whose start
time and location have not been assigned. A start time and location already
generated in Step 30 are randomly assigned to the selected incident.
Step 32: Check for Overlap with Previously Assigned Incidents

This step checks if there is any overlap between other incident events for
which the start time and location have been assigned in the list of incident
events. If there is an overlap, the process proceeds to Step 33. Otherwise, it
proceeds to Step 34.
Step 33: Undo the Previous Start Time and Location Assignment
This step undoes the previous start time and location assignment from Step
31 that led to the identification of a conflict in the list of incident events in Step 32.
Step 34: Check Whether All Incident Start Times and Locations Have
Been Assigned
If there are incidents in the list of incident events that have not been assigned
a start time and location, the process returns to Step 31 for further assignment.
Otherwise, all the incidents in the list of incident events have been fully
described and are ready to be modeled in the scenarios.

10. COMPUTATIONAL ENGINE OVERVIEW
The FREEVAL-2015E computational engine is written in the Java

programming language. Java is a free, open source, object-oriented programming
language that is highly portable and will run on almost all platforms. Unlike
procedural languages, which largely consist of code broken up into subroutines,
object-oriented languages require that the code be expressed in terms of objects.
These objects have functions that either operate on the data associated with them
or on other objects. In Java, groups of objects are called classes. Classes are then
grouped into packages, which seek to provide organization based on some shared
purpose or similarity.
The computational engine consists of nine packages, each of which contains
a group of classes specific to a certain aspect of the HCM analysis. The main
package contains the two most important classes for the methodology. First, the
Seed class contains all input data for the freeway facility (e.g., freeway geometry,
demand) and is the backbone of the engine. Once the analysis has been run, the
Seed class will also contain all output performance measures. Further, any
reliability or ATDM analysis performed will use Seed as the basis for its analysis.
The second class in the main package is the GPMLSegment class. This class is
used to represent the segments of the freeway facility (general purpose or
managed lane), and contains the code for both the undersaturated and
oversaturated computational modules. Much of this code is an exact translation
of the HCM methodology, with differences only occurring when it was necessary
to either improve the performance of the code, or to match Java programming
conventions. An example of a difference is that some variable values may not be
explicitly stored but rather are calculated only as needed.
The other eight packages build on these two main classes. Four of the packages
consist of “helper” functions that are used throughout the code. These helper
classes provide functionality ranging from general input-output actions, such as
opening and saving files, to more specific purposes, such as creating facility
output summaries and specifying parameters for ramp-metering methodologies.
The final four packages relate to reliability and ATDM analysis. These packages
contain the reliability scenario generator, as well as many additional data
structures to facilitate data input for both reliability and ATDM analysis.
The Java programming language provides the integrated ability to generate
its own documentation. Developers simply provide descriptions of classes,
functions, and variables throughout the code, and Java compiles them into a set
of documentation referred to as a “Javadoc.” This Javadoc follows the format of
the official documentation of the language, thus allowing it to be easily
understood and used by anyone familiar with the language. This documentation
has been generated and is packaged with the computational engine. A user guide
for the graphical user interface version of the engine is available to provide
guidance on its use. These items can be found in the Technical Reference Library
in online HCM Volume 4.
Computational Engine Overview Chapter 25/Freeway Facilities: Supplemental

11. EXAMPLE PROBLEMS
This section presents eleven example problems illustrating the evaluation of

freeway facilities using the core methodology, the reliability methodology, and
the ATDM methodology. Exhibit 25-42 presents a list of these problems.
Example Exhibit 25-42

Problem Description Application List of Example Problems
1 Evaluation of an undersaturated facility Operational analysis
2 Evaluation of an oversaturated facility Operational analysis
3 Capacity improvements to an oversaturated facility Operational analysis
4 Evaluation of an undersaturated facility with a work zone Operational analysis
5 Evaluation of an oversaturated facility with a managed lane Operational analysis
6 Planning-level analysis of a freeway facility Planning analysis
7 Reliability evaluation of an existing freeway facility Reliability analysis
8 Reliability analysis with geometric improvements Reliability analysis
9 Evaluation of incident management ATDM analysis
10 Planning-level reliability analysis Planning analysis
11 Estimating freeway composite grade operations with the Specialized truck
mixed-flow model analysis
EXAMPLE PROBLEM 1: EVALUATION OF AN UNDERSATURATED

FACILITY
The Facility
The subject of this operational analysis is a 6-mi-long urban freeway facility
that is composed of 11 individual analysis segments, as shown in Exhibit 25-43.
Exhibit 25-43
Example Problem 1:
Freeway Facility
The facility has three on-ramps and three off-ramps. Geometric details are
given in Exhibit 25-44.
Segment No. 1 2 3 4 5 6 7 8 9 10 11 Exhibit 25-44

Segment type B ONR B OFR B B or W B ONR R OFR B Example Problem 1: Geometry
Segment of Directional Freeway Facility
length (ft) 5,280 1,500 2,280 1,500 5,280 2,640 5,280 1,140 360 1,140 5,280
No. of lanes 3 3 3 3 3 4 3 3 3 3 3
Notes: B = basic freeway segment; W = weaving segment; ONR = on-ramp (merge) segment; OFR = off-ramp
(diverge) segment; R = overlapping ramp segment.
The on- and off-ramps in Segment 6 are connected by an auxiliary lane, and
the segment may therefore operate as a weaving segment, depending on traffic
patterns. The separation of the on-ramp in Segment 8 and the off-ramp in
Segment 10 is less than 3,000 ft. Because the ramp influence area of on-ramps and
off-ramps is 1,500 ft, according to Chapter 14, the segment affected by both
ramps is analyzed as a separate overlapping ramp segment (Segment 9), labeled
“R” in Exhibit 25-44.
Chapter 25/Freeway Facilities: Supplemental Example Problems

The analysis question at hand is the following: What is the operational

performance and LOS of the directional freeway facility shown in Exhibit 25-43?
The Facts
In addition to the information contained in Exhibit 25-43 and Exhibit 25-44,
the following characteristics of the freeway facility are known:
SUTs and buses = 1.25% (all movements);
TTs = 1.00% (all movements);
Driver population = regular commuters;
FFS = 60 mi/h (all mainline segments);
Ramp FFS = 40 mi/h (all ramps);
Acceleration lane length = 500 ft (all ramps);
Deceleration lane length = 500 ft (all ramps);
Djam = 190 pc/mi/ln;
cIFL = 2,300 pc/h/ln (for FFS = 60 mi/h);
Ls = 1,640 ft (for Weaving Segment 6);
Total ramp density TRD = 1.0 ramp/mi;
Terrain = level; and
Analysis duration = 75 min (divided into five 15-min intervals).
A queue discharge capacity drop of 7% is assumed.
Comments
The facility was divided into analysis segments on the basis of the guidance
given in Chapter 10, Freeway Facilities Core Methodology. The facility shown in
Exhibit 25-43 depicts seven freeway sections (measured between ramps) that are
divided into 11 analysis segments. The facility contains each of the possible
segment types for illustrative purposes, including basic segment (B), weaving
segment (W), merge segment (ONR), diverge segment (OFR), and overlapping
ramp segment (R). The input data contain the required information needed for
each of the segment methodologies.
The classification of the weave in Segment 6 is preliminary until it is
determined whether the segment operates as a weave. For this purpose, the short
length must be compared with the maximum length for weaving analysis to
determine whether the Chapter 13, Freeway Weaving Segments, or the Chapter
12, Basic Freeway and Multilane Highway Segments, methodology is applicable.
The short length of the weaving segment used for calculation is shorter than the
weaving influence area over which the calculated speed and density measures
are applied.
Chapter 12 must be consulted to find appropriate values for the heavy-
vehicle adjustment factor fHV. The computational engine automatically
determines these adjustment factors for general terrain conditions, but user input
is needed for specific upgrades and composite grades.
Example Problems Chapter 25/Freeway Facilities: Supplemental

All input parameters have been specified, so default values are not needed.
Fifteen-minute demand flow rates are given in vehicles per hour under
prevailing conditions. These demands must be converted to passenger cars per
hour under equivalent ideal conditions for use in the parts of the methodology
related to segment LOS estimation. Details of the steps of the methodology
follow.
Step A-1: Define Study Scope

In this initial step, the analyst defines the spatial extent of the facility (start
and end points, total length) and the temporal extent of the analysis (number of
15-min analysis periods). The analyst should further decide which study
extensions (if any) apply to the analysis (i.e., managed lanes, reliability, ATDM).
According to the inputs provided in the example, the number of analysis
periods is five and the facility has 11 segments. The analysis does not involve a
methodological extension.
Step A-2: Divide Facility into Sections and Segments

In this step, the analyst first defines the number of sections from gore point
to gore point along the selected facility. These gore-to-gore sections are more
consistent with modern freeway performance databases than HCM segments,
and this consistency is critical for calibrating and validating the freeway facility.
The analyst later divides sections into HCM segments (basic, merge, diverge,
weave, overlapping ramp, or managed lane segment) as described in Chapter 10.
The subject facility has already been segmented as shown in Exhibit 25-43.
Step A-3: Input Data

Data concerning demand, geometry, and other data are specified in this step.
As the methodology builds on segment analysis, all data for each segment and
each analysis period must be provided. Traffic demand inputs for all 11
segments and five analysis periods are given in Exhibit 25-45.
Exhibit 25-45
Analysis Entering Exiting
Ramp Flow Rates by Analysis Period (veh/h) Example Problem 1:
Period Flow Rate Flow Rate
Demand Inputs
(15 min) (veh/h) ONR1 ONR2a ONR3 OFR1 OFR2 OFR3 (veh/h)
1 4,505 450 540 (50) 450 270 360 270 5,045
2 4,955 540 720 (100) 540 360 360 270 5,765
3 5,225 630 810 (150) 630 270 360 450 6,215
4 4,685 360 360 (80) 450 270 360 270 4,955
5 3,785 180 270 (50) 270 270 180 180 3,875
a
Note: Numbers in parentheses indicate ONR-2 to OFR-2 demand flow rates in Weaving Segment 6.
The volumes in Exhibit 25-45 represent the 15-min demand flow rates on the
facility as determined from field observations or other sources. The actual
volume served in each segment will be determined by the methodology. The
demand flows are given for the extended time–space domain, consistent with the
recommendations in Chapter 10. Peaking occurs in the third 15-min period.
Because inputs are in the form of 15-min flow rates, no peak hour factor
adjustment is necessary. Additional geometric and traffic-related inputs are as
specified in Exhibit 25-44 and the Facts section of the problem statement.

Step A-4: Balance Demands

The traffic flows in Exhibit 25-45 are already given in the form of actual
demands. Therefore, balancing demand is not necessary.
Step A-5: Identify Global Parameters

Global inputs are jam density and queue discharge capacity drop. Values for
both parameters are given in the example problem’s Facts section.
Step A-6: Code Base Facility

Step 6 is the first step requiring the use of a computational engine or
software. Data input needs for the computational engine include all items
collected or estimated in the previous steps. These data generally need to be
entered for each segment and each analysis period, making this one of the most
time-consuming steps in the analysis.
Step A-7: Compute Segment Capacities

Segment capacities are determined by using the methodologies of Chapter 12
for basic freeway segments, Chapter 13 for weaving segments, and Chapter 14
for merge and diverge segments. The resulting capacities are shown in Exhibit
25-46. Because the capacity of a weaving segment depends on traffic patterns,
including the weaving ratio, it varies by analysis period. The remaining segment
capacities are constant in all five time intervals. The capacities for Segments 1–5
and 7–11 are the same because the segments have the same basic cross section.
The units shown are in vehicles per hour.
Exhibit 25-46 Analysis Capacities (veh/h) by Segment

Example Problem 1: Period 1 2 3 4 5 6 7 8 9 10 11
Segment Capacities
1 8,273
2 8,281
3 6,748 6,748 6,748 6,748 6,748 8,323 6,748 6,748 6,748 6,748 6,748
4 8,403
5 8,463
Step A-8: Calibrate with Adjustment Factors

This step allows the analyst to adjust demands, capacities, and FFSs for the
purpose of calibration. The demand adjustment factor (DAF), capacity
adjustment factor (CAF), and speed adjustment factor (SAF) can be modified for
each segment and each analysis period. There is no adjustment needed for the
subject facility according to the problem statement.
Step A-9: Adjust Managed Lane Cross Weave

This step is only required for facilities with managed lanes. The subject
facility does not have a managed lane; therefore, this step is not required.
Step A-10: Compute Demand-to-Capacity Ratios

The demand-to-capacity ratios in Exhibit 25-47 are calculated from the
demand flows in Exhibit 25-45 and the segment capacities in Exhibit 25-46.

Demand-to-Capacity Ratios by Segment Exhibit 25-47

Analysis
Example Problem 1: Segment
Period 1 2 3 4 5 6 7 8 9 10 11
Demand-to-Capacity Ratios
1 0.67 0.73 0.73 0.73 0.69 0.63 0.72 0.79 0.79 0.79 0.75
2 0.73 0.81 0.81 0.81 0.76 0.71 0.81 0.89 0.89 0.89 0.85
3 0.77 0.87 0.87 0.87 0.83 0.77 0.89 0.99 0.99 0.99 0.92
4 0.69 0.75 0.75 0.75 0.71 0.61 0.71 0.77 0.77 0.77 0.73
5 0.56 0.59 0.59 0.59 0.55 0.47 0.56 0.60 0.60 0.60 0.57
The computed demand-to-capacity ratio matrix in Exhibit 25-47 shows no

segments with a vd/c ratio greater than 1.0 in any time interval. Consequently, the
facility is categorized as globally undersaturated, and the analysis proceeds with
computing the undersaturated service measures in Step A-11. Further, it is
expected that no queuing will occur on the facility and that the volume served in
each segment is identical to the input demand flows. Consequently, the matrix of
volume-to-capacity ratios would be identical to the demand-to-capacity ratios in
Exhibit 25-47. The resulting matrix of volumes served by segment and time
interval is shown in Exhibit 25-48.
Volumes Served (veh/h) by Segment Exhibit 25-48

Analysis
Example Problem 1:
Period 1 2 3 4 5 6 7 8 9 10 11
Volume-Served Matrix
1 4,505 4,955 4,955 4,955 4,685 5,225 4,865 5,315 5,315 5,315 5,045
2 4,955 5,495 5,495 5,495 5,135 5,855 5,495 6,035 6,035 6,035 5,765
3 5,225 5,855 5,855 5,855 5,585 6,395 6,035 6,665 6,665 6,665 6,215
4 4,685 5,045 5,045 5,045 4,775 5,135 4,775 5,225 5,225 5,225 4,955
5 3,785 3,965 3,965 3,965 3,695 3,965 3,785 4,055 4,055 4,055 3,875
Step A-11: Compute Undersaturated Segment Service Measures

Because the facility is globally undersaturated, the methodology proceeds to
calculate service measures for each segment and each analysis period, starting
with the first segment in Analysis Period 1. The computational details for each
segment type are exactly as described in Chapters 12, 13, and 14. The weaving
methodology in Chapter 13 checks whether the weaving short length LS is less
than or equal to the maximum weaving length Lmax. It is assumed, for any time
interval where LS is longer than or equal to Lmax, that the weaving segment will
operate as a basic freeway segment.
The basic performance measures computed for each segment and each
analysis period are the segment speed (Exhibit 25-49), density (Exhibit 25-50),
and LOS (Exhibit 25-51).

Exhibit 25-49 Speed (mi/h) by Segment

Analysis
Example Problem 1:
Period 1 2 3 4 5 6 7 8 9 10 11
Speed Matrix
1 60.0 53.9 59.7 56.1 60.0 48.0 59.9 53.4 53.4 56.0 59.7
2 59.9 53.2 58.6 55.8 59.6 46.8 58.6 52.3 52.3 55.7 57.6
3 59.4 52.6 57.2 55.7 58.3 46.2 56.2 50.6 50.6 51.8 55.1
4 60.0 53.8 59.7 56.1 60.0 49.7 60.0 53.6 53.6 56.0 59.9
5 60.0 54.9 59.8 56.3 60.0 52.5 60.0 54.8 54.8 56.5 60.0
Exhibit 25-50 Density (veh/mi/ln) by Segment

Analysis
Example Problem 1:
Period 1 2 3 4 5 6 7 8 9 10 11
Density Matrix
1 25.0 30.6 27.6 29.4 26.0 27.2 27.1 33.2 33.2 31.6 28.1
2 27.6 34.5 31.2 32.8 28.7 31.3 31.2 38.5 38.5 36.1 33.4
3 29.3 37.1 34.1 35.0 31.9 34.6 35.8 43.9 43.9 42.9 37.6
4 26.0 31.3 28.1 30.0 26.5 25.8 26.5 32.5 32.5 31.1 27.6
5 21.0 24.1 22.0 23.5 20.5 18.9 21.0 24.7 24.7 23.9 21.5
Exhibit 25-51 LOS by Segment

Analysis
Example Problem 1:
Period 1 2 3 4 5 6 7 8 9 10 11
LOS Matrix
1 C C D C D C D D D D D
2 D D D D D D D D E D D
3 D D D D D D E E E D E
4 D C D C D C D C D D D
5 C C C C C B C C C C C
Step A-13: Apply Managed Lane Adjacent Friction Factor

This step is only required for facilities with managed lanes.
Step A-14: Compute Lane Group Performance

Step A-15: Compute Freeway Facility Service Performance Measures by

Time Interval
In this analysis step, facilitywide performance measures are calculated for
each analysis period. Example calculations are provided for the first analysis
period only; summary results are shown for all five analysis periods.
First, the facility space mean speed SMS is calculated for analysis period p = 1
from the 11 individual segment flows SF(i, p), segment lengths L(i), and space
mean speeds in each segment and analysis period U(i, p).
∑11
𝑖=1 𝑆𝐹(𝑖, 1) × 𝐿(𝑖)
𝑆𝑀𝑆(𝑁𝑆, 𝑝 = 1) =
𝐿(𝑖)
∑11
𝑖=1 𝑆𝐹(𝑖, 1) × 𝑈(𝑖, 1)
11
∑ 𝑆𝐹(𝑖, 1) × 𝐿(𝑖) = (4,505 × 5,280) + (4,955 × 1,500) + (4,955 × 2,280) +

𝑖=1 (4,955 × 1,500) + (4,685 × 5,280) + (5,225 × 2,640) +
(4,865 × 5,280) + (5,315 × 1,140) + (5,315 × 360) +
(5,315 × 1,140) + (5,045 × 5,280)
= 154,836,000 veh-ft

11
𝐿(𝑖)
∑ 𝑆𝐹(𝑖, 1) × = (4,505 × 5,280 / 60.00) + (4,955 × 1,500 / 53.90)
𝑈(𝑖, 1)
𝑖=1 + (4,955 × 2,280 / 59.70) + (4,955 × 1,500 / 56.10)
+ (4,685 × 5,280 / 60.00) + (5,225 × 2,640 / 48.00)
+ (4,865 × 5,280/ 59.90) + (5,315 × 1,140/ 53.40)
+ (5,315 × 360 / 53.40) + (5,315 × 1,140 / 56.00)
+ (5,045 × 5,280 / 59.70)
= 2,688,234 veh-ft/mi/h
154,836,000
𝑆(𝑁𝑆, 𝑝 = 1) = = 57.6 mi/h
2,688,234
Second, the average facility density is calculated for Analysis Period 1 from
the individual segment densities K, segment lengths L, and number of vehicles in
each segment N.
∑11
𝑖=1 𝐷(𝑖, 1) × 𝐿(𝑖) × 𝑁(𝑖, 1)
𝐾(𝑁𝑆, 𝑝 = 1) =
∑11
𝑖=1 𝑆𝐿(𝑖)𝑁(𝑖, 1)
11
∑ 𝐷(𝑖, 1) × 𝐿(𝑖) × 𝑁(𝑖, 1) = (25.0 × 5,280 × 3) + (30.6 × 1,500 × 3) + (27.6 × 2,280 × 3)

𝑖=1 + (29.4 × 1,500 × 3) + (26.0 × 5,280 × 3) + (27.2 × 2,640 × 4)
+ (27.1 × 5,280 × 3) + (33.2 × 1,140 × 3) + (33.2 × 360 × 3)
+ (31.6 × 1,140 × 3) + (28.1 × 5,280 × 3)
= 2,685,696 (veh/mi/ln)(ln-ft)
11
∑ 𝑆𝐿(𝑖)𝑁(𝑖, 1) = (5,280 × 3) + (1,500 × 3) + (2,280 × 3) + (1,500 × 3)

𝑖=1 + (5,280 × 3) + (2,640 × 4) + (5,280 × 3) + (1,140 × 3)
+ (360 × 3) + (1,140 × 3) + (5,280 × 3)
= 97,680 ln-ft
2,685,696
𝐾(𝑁𝑆, 𝑝 = 1) = = 27.5 veh/mi/ln
97,680
These calculations are repeated for all five analysis periods. The overall space
mean speed across all analysis periods is calculated as follows:
∑5𝑝=1 ∑11
𝑖=1 𝑆𝐹(𝑖, 𝑝) × 𝐿(𝑖)
𝑆𝑀𝑆(𝑃 = 5) =
𝐿(𝑖)
∑5𝑝=1 ∑11
𝑖=1 𝑆𝐹(𝑖, 𝑝) × 𝑈(𝑖, 𝑝)
The overall average density across all analysis periods is calculated as follows:
∑5𝑝=1 ∑11
𝑖=1 𝐾(𝑖, 𝑝) × 𝐿(𝑖) × 𝑁(𝑖, 𝑝)
𝐾(𝑃 = 5) =
∑5𝑝=1 ∑11
𝑖=1 𝑆𝐿(𝑖)𝑁(𝑖, 𝑝)
The resulting performance and service measures for Analysis Periods 1–5
and the facility totals are shown in Exhibit 25-52.

Exhibit 25-52 Performance Measure

Example Problem 1: Facility Space Mean Average
Performance Measure Analysis Speed Density
Summary Period (mi/h) (veh/mi/ln) LOS
1 57.6 27.5 D
2 56.6 31.3 D
3 55.0 34.8 E
4 57.9 27.5 D
5 58.4 21.4 C
Total 56.9 28.4 —
Step A-16: Aggregate to Section Level and Validate Against Field Data
This step is used to validate the analysis and is performed only when field
data are available.
Step A-17: Estimate LOS and Report Performance Measures for Lane
Groups and Facility
The LOS for each time interval is determined directly from the average
density for each time interval by using Exhibit 10-7. No LOS is defined for the
average across all time intervals.
Discussion
This facility turned out to be globally undersaturated. Consequently, the
facility-aggregated performance measures could be calculated directly from the
individual segment performance measures. An assessment of the segment
service measures across the time–space domain can begin to highlight areas of
potential congestion. Visually, this process can be facilitated by plotting the vd/c,
va/c, speed, or density matrices in contour plots.
EXAMPLE PROBLEM 2: EVALUATION OF AN OVERSATURATED FACILITY

The Facility
The facility used in Example Problem 2 is identical to the one in Example
Problem 1, which is shown in Exhibit 25-43 and Exhibit 25-44.
The Facts
In addition to the information in Exhibit 25-43 and Exhibit 25-44, the
following characteristics of the freeway facility are known:
TTs = 1.00% (all movements);


TRD = 1.0 ramp/mi;
Terrain = level;
Analysis duration = 75 min (divided into five 15-min analysis periods);
and
Demand adjustment = +11% increase in demand volumes across all
segments and analysis periods relative to Example
Problem 1.
As before, a queue discharge capacity drop of 7% is assumed.
Comments
The facility and all geometric inputs are identical to Example Problem 1. The
same general comments apply. The results of Example Problem 1 suggested a
globally undersaturated facility, but some segments were close to their capacity
(vd/c ratios approaching 1.0). In the second example, a facilitywide demand
increase of 11% is applied to all segments and all analysis periods. Consequently,
it is expected parts of the facility may become oversaturated and queues may
form on the facility.

Similar to Example Problem 1, there are five analysis periods and the facility
has 11 segments. The analysis does not include any extensions such as managed
lanes, reliability, ATDM, or work zones.

The subject facility segmentation is given in Exhibit 25-43. Therefore, there is
no need to go through the segmentation process.

The revised traffic demand inputs for all 11 segments and five analysis
periods are shown in Exhibit 25-53.
Entering Exiting Exhibit 25-53

Analysis
Flow Rate Ramp Flow Rates by Analysis Period (veh/h) Flow Rate Example Problem 2:
Period
(veh/h) (veh/h) Demand Inputs
(15 min) ONR1 ONR2a ONR3 OFR1 OFR2 OFR3
1 5,001 500 599 (56) 500 300 400 300 5,600
2 5,500 599 799 (111) 599 400 400 300 6,399
3 5,800 699 899 (167) 699 300 400 500 6,899
4 5,200 400 400 (89) 500 300 400 300 5,500
5 4,201 200 300 (56) 300 300 200 200 4,301
a
Note: Numbers in parentheses indicate ONR-2 to OFR-2 demand flow rates in Weaving Segment 6.
The values in Exhibit 25-53 represent the adjusted demand flows on the
facility as determined from field observations or demand projections. The actual
volume served in each segment will be determined during the application of the
methodology and is expected to be less downstream of a congested segment. The
demand flows are given for the extended time–space domain, consistent with the
methodology presented in Chapter 10. Peaking occurs in the third 15-min period.

Because inputs are in the form of 15-min observations, no peak hour factor
adjustment is necessary. Additional geometric and traffic-related inputs are as

The traffic flows in Exhibit 25-53 have already been given in the form of
actual demands and no balancing is necessary.

both parameters are given in the Facts section of the problem statement.

In this step, all input data for the subject are coded in the computational
engine. Note that this facility can be coded by increasing entry demand across
the facility by 11% relative to the Example Problem 1 demands.

Because no changes to segment geometry were made, the segment capacities
for basic and ramp segments are consistent with Example Problem 1. Capacities
for weaving segments are a function of weaving flow patterns, and the increased
demand flows resulted in slight changes as shown in Exhibit 25-54.

Segment Capacities
1 8,273
2 8,281
3 6,748 6,748 6,748 6,748 6,748 8,323 6,748 6,748 6,748 6,748 6,748
4 8,403
5 8,463

purpose of calibration. There is no adjustment needed for the subject capacity
according to the problem statement.

facility does not have a managed lane.

demand flows in Exhibit 25-53 and the segment capacities in Exhibit 25-54.

Demand-to-Capacity Ratios by Segment Exhibit 25-55

Analysis
Period 1 2 3 4 5 6 7 8 9 10 11
1 0.74 0.82 0.82 0.82 0.77 0.70 0.80 0.87 0.87 0.87 0.83
2 0.82 0.90 0.90 0.90 0.84 0.78 0.90 0.99 0.99 0.99 0.95
3 0.86 0.96 0.96 0.96 0.92 0.85 0.99 1.10 1.10 1.10 1.02
4 0.77 0.83 0.83 0.83 0.79 0.68 0.79 0.86 0.86 0.86 0.82
5 0.62 0.65 0.65 0.65 0.61 0.52 0.62 0.67 0.67 0.67 0.64
The computed vd/c matrix in Exhibit 25-55 shows Segments 8–11 have vd/c
ratios greater than 1.0 (bold values). Consequently, the facility is categorized as
oversaturated, and the analysis proceeds with computing the oversaturated
service measures in Step A-12. It is expected that queuing will occur on the
facility upstream of the congested segments and that the volume served in each
segment downstream of the congested segments will be less than its demand.
This residual demand will be served in later time intervals, provided the
upstream demand drops and queues are allowed to clear.
Step A-12: Compute Oversaturated Segment Service Measures

Computations for oversaturation apply to any segment with a vd/c ratio
greater than 1.0 as well as any segments upstream of those segments that
experience queuing as a result of the bottleneck. All remaining segments are
analyzed by using the individual segment methodologies of Chapters 12, 13, and
14, as applicable, with the caveat that volumes served may differ from demand
flows.
Similar to Example Problem 1, in Example Problem 2 the methodology
calculates performance measures for each segment and each analysis period,
starting with the first segment in Analysis Period 1. The computations are
repeated for all segments for Analysis Periods 1 and 2 without encountering a
segment with vd/c > 1.0. Once the methodology enters Analysis Period 3 and
Segment 8, the oversaturated computational module is invoked.
At the first active bottleneck, the va/c ratio for Segment 8 will be exactly 1.0
and the segment will process traffic at its capacity. Consequently, demand for all
downstream segments will be metered by that bottleneck. The unsatisfied
demand is stored in upstream segments, which causes queuing in Segment 7 and
perhaps segments further upstream depending on the level of excess demand.
The rate of growth of the vehicle queue (wave speed) is estimated from shock
wave theory. The performance measures (speed and density) of any segment
with queuing are recomputed, and the newly calculated values override the
results from the segment-specific procedures.
Any unsatisfied demand is served in later analysis periods. As a result,
volumes served in later analysis periods may be higher than the period demand
flows. The resulting matrix of volumes served for Example Problem 2 is shown
in Exhibit 25-56.

Exhibit 25-56 Volumes Served (veh/h) by Segment

Analysis
Example Problem 2:
Period 1 2 3 4 5 6 7 8 9 10 11
1 5,001 5,500 5,500 5,500 5,200 5,800 5,400 5,900 5,900 5,900 5,600
2 5,500 6,099 6,099 6,099 5,700 6,499 6,099 6,699 6,699 6,699 6,399
3 5,800 6,499 6,499 6,499 5,831 6,281 5,584 6,284 6,284 6,284 5,859
4 5,200 5,600 5,600 5,600 5,668 6,311 5,776 6,276 6,276 6,276 5,934
5 4,201 4,401 4,401 4,401 4,102 4,608 4,840 5,140 5,140 5,140 4,912
As a result of the bottleneck activation in Segment 8 in Analysis Period 3,

queues form in upstream Segments 7, 6, and 5. The queuing is associated with
reduced speeds and increased densities in those segments. The results in this
chapter were obtained from the computational engine. The resulting
performance measures computed for each segment and time interval are speed
(Exhibit 25-57), density (Exhibit 25-58), and LOS (Exhibit 25-59).

Analysis
Example Problem 2:
Period 1 2 3 4 5 6 7 8 9 10 11
Speed Matrix
1 59.8 53.2 58.6 55.9 59.5 46.8 59.0 52.5 52.5 55.7 58.3
2 58.6 52.1 55.8 55.5 57.9 45.4 55.8 50.6 50.6 51.5 53.9
3 57.4 51.1 53.1 53.1 45.3 24.2 28.1 51.6 51.6 54.7 57.1
4 47.2 47.5 51.5 48.3 56.5 24.7 29.6 51.7 51.7 54.7 56.8
5 60.0 54.5 59.7 56.2 60.0 51.4 50.9 53.7 53.7 56.1 59.9
Exhibit 25-58 Density (veh/mi/ln) by Segment

Analysis
Example Problem 2:
Period 1 2 3 4 5 6 7 8 9 10 11
Density Matrix
1 27.9 34.5 31.3 32.8 29.2 31.0 30.5 37.4 37.4 35.3 32.0
2 31.3 39.0 36.4 36.7 32.8 35.8 36.4 44.2 44.2 43.3 39.6
3 33.7 42.4 40.8 40.8 42.9 64.8 66.4 40.6 40.6 38.3 34.2
4 36.7 39.3 36.3 38.6 33.4 63.9 65.1 40.4 40.4 38.2 34.8
5 23.3 26.9 24.5 26.1 22.8 22.4 31.7 31.9 31.9 30.5 27.3
Exhibit 25-59 Density-Based LOS by Segment

Analysis
Example Problem 2:
Period 1 2 3 4 5 6 7 8 9 10 11
Expanded LOS Matrix
1 D D D D D D D D E D D
2 D D E D D E E E E D E
3 D D E D E F F D E D D
4 E E E E D F F D E D E
5 C C C C C C D C D C D
Analysis Demand-Based LOS by Segment
Period 1 2 3 4 5 6 7 8 9 10 11
1
2
3 F F F F
4
5
The LOS table for oversaturated facilities (Exhibit 25-59) distinguishes

between the conventional density-based LOS and a segment demand-based LOS.
The density-based stratification strictly depends on the prevailing average
density on each segment. Segments downstream of the bottleneck, whose
capacities are greater than or equal to the bottleneck capacity, operate at LOS E
(or better), even though their vd/c ratios are greater than 1.0. The demand-based
LOS identifies those segments with demand-to-capacity ratios exceeding 1.0 as if
they had been evaluated in isolation (i.e., using the methodologies of Chapters

12, 13, and 14). By contrasting the two parts of the LOS table, the analyst can
develop an understanding of the metering effect of the bottleneck.



Time Interval
In the final analysis step, facilitywide performance measures are calculated
for each time interval (Exhibit 25-60), consistent with Example Problem 1.
Because the computations have already been shown, only summary results are
shown here.
Performance Measure Exhibit 25-60

Example Problem 2: Facility
Space Mean Average
Performance Measure
Analysis Speed Density
Summary
Period (mi/h) (veh/mi/ln) LOS
1 56.8 31.0 D
2 54.4 36.2 E
3 42.5 45.6 F
4 42.5 43.8 E
5 56.4 26.2 D
Total 50.5 35.6 —
This step validates the analysis and is performed only when field data are
available.
Groups and Facility
density for each time interval. The facility operates at LOS F in Analysis Period 3
because one or more individual segments have demand-to-capacity ratios ≥ 1.0,
even though the average facility density is below the LOS F threshold.
EXAMPLE PROBLEM 3: CAPACITY IMPROVEMENTS TO AN

OVERSATURATED FACILITY
The Facility
In this example, portions of the congested facility in Example Problem 2 are
being improved in an attempt to alleviate the congestion resulting from the
Segment 8 bottleneck. Exhibit 25-61 shows the upgraded facility geometry.

Exhibit 25-61
Example Problem 3:
Freeway Facility
The modified geometry of the 6-mi directional freeway facility is reflected in

Exhibit 25-62.
Exhibit 25-62 Segment No. 1 2 3 4 5 6 7 8 9 10 11

Example Problem 3: Geometry Segment type B ONR B OFR B B or W B ONR R OFR B
of Directional Freeway Facility Segment length
5,280 1,500 2,280 1,500 5,280 2,640 5,280 1,140 360 1,140 5,280
(ft)
No. of lanes 3 3 3 3 3 4 4 4 4 4 4
Bold type indicates geometry changes from Example Problems 1 and 2.
The facility improvements consisted of adding a lane to Segments 7–11 to

give the facility a continuous four-lane cross section starting in Segment 6. The
active bottleneck in Example Problem 2 was in Segment 8, but prior analysis
showed that other segments (Segments 9–11) showed similar demand-to-capacity
ratios greater than 1.0. Consequently, any capacity improvements that are limited
to Segment 8 would have merely moved the spatial location of the bottleneck
farther downstream rather than improving the overall facility. Segments 9–11
may also be referred to as “hidden” or “inactive” bottlenecks, because their
predicted congestion is mitigated by the upstream metering of traffic.
The Facts
Mainline TTs = 1.00% (all movements);
TRD = 1.0 ramp/mi;
Terrain = level;
Analysis duration = 75 min (divided into five 15-min intervals); and

Demand adjustment = +11% (all segments and all time intervals).

Comments
The traffic demand flow inputs are identical to those in Example Problem 2,
which reflected an 11% increase in traffic applied to all segments and all analysis
periods relative to Example Problem 1. In an attempt to solve the congestion
effect found in the earlier example, the facility was widened in Segments 7
through 11. This change directly affects the capacities of those segments.
In a more subtle way, the proposed modifications also change some of the
defining parameters of Weaving Segment 6. With the added continuous lane
downstream of the segment, the required number of lane changes from the ramp
to the freeway is reduced from one to zero, following the guidelines in Chapter
13. These changes need to be considered when the undersaturated performance
of that segment is evaluated. The weaving segment’s capacity is unchanged
relative to Example Problem 2 because, even with the proposed improvements,
the number of weaving lanes remains two.

Similar to the previous example, the number of analysis periods is five and
the facility has 11 segments. The analysis does not include any methodological
extensions (i.e., managed lanes, reliability, ATDM, work zones).

The segmentation of the subject facility is the same as in Example Problems 1
and 2 and is given in Exhibit 25-61. Therefore, the segmentation process is not
repeated.

Traffic demand inputs for all 11 segments and five analysis periods are
identical to those in Example Problem 2, as shown in Exhibit 25-53. The values
represent the adjusted demand flows on the facility as determined from field
observations or other sources. The actual volume served in each segment will be
determined by using the methodologies and is expected to be less downstream of
a congested segment. Additional geometric and traffic-related inputs are as




In this step, all input data for the subject are coded in the computational
engine.

for basic freeway segments, Chapter 13 for weaving segments, and Chapter 14
for merge and diverge segments. The resulting capacities are shown in Exhibit
25-63. Because the capacity of a weaving segment depends on traffic patterns, it
varies by analysis period. The remaining capacities are constant for all five
analysis periods. The capacities for Segments 1–5 and Segments 7–11 are the
same because the segments have the same basic cross section.

Segment Capacities
1 8,273
2 8,281
3 6,748 6,748 6,748 6,748 6,748 8,323 8,998 8,998 8,998 8,998 8,998
4 8,403
5 8,463

This step allows the user to adjust demands, capacities, and FFSs for the
purpose of calibration. There is no adjustment needed for the subject capacity
according to the problem statement.


demand flows in Exhibit 25-53 and segment capacities in Exhibit 25-63.
Exhibit 25-64 Demand-to-Capacity Ratio by Segment

Analysis
Example Problem 3:
Period 1 2 3 4 5 6 7 8 9 10 11
Segment Demand-to-Capacity
1 0.74 0.82 0.82 0.82 0.77 0.70 0.60 0.66 0.66 0.66 0.62
Ratios
2 0.82 0.90 0.90 0.90 0.84 0.78 0.68 0.74 0.74 0.74 0.71
3 0.86 0.96 0.96 0.96 0.92 0.85 0.74 0.82 0.82 0.82 0.77
4 0.77 0.83 0.83 0.83 0.79 0.68 0.59 0.64 0.64 0.64 0.61
5 0.62 0.65 0.65 0.65 0.61 0.52 0.47 0.50 0.50 0.50 0.48
The demand-to-capacity ratio matrix for Example Problem 3 (Exhibit 25-64)

shows the capacity improvements successfully reduced all the previously
congested segments to vd/c < 1.0. Therefore, it is expected that the facility will
operate as globally undersaturated and that all segment performance measures can
be directly computed by using the methodologies in Chapters 12, 13, and 14.


Because the facility is globally undersaturated, the methodology proceeds to
calculate service measures for each segment and each analysis period, starting
with the first segment in Analysis Period 1. The computational details for each
segment type are exactly as described in Chapters 12, 13, and 14. The basic
performance service measures computed for each segment and each time
interval include segment speed (Exhibit 25-65), density (Exhibit 25-66), and LOS
(Exhibit 25-67).
Speed (mi/h) by Segment Exhibit 25-65

Analysis
Example Problem 3:
Period 1 2 3 4 5 6 7 8 9 10 11
Speed Matrix
1 59.8 53.2 58.6 55.9 59.5 50.5 60.0 54.9 54.9 58.1 60.0
2 58.6 52.1 55.8 55.5 57.9 50.1 60.0 54.3 54.3 57.7 60.0
3 57.4 51.1 53.1 53.1 55.2 49.7 59.8 53.6 53.6 57.2 59.5
4 59.5 53.0 58.3 55.8 59.2 50.8 60.0 55.0 55.0 58.1 60.0
5 60.0 54.5 59.7 56.2 60.0 53.4 60.0 55.9 55.9 58.8 60.0
Density (veh/mi/ln) by Segment Exhibit 25-66

Analysis
Example Problem 3:
Period 1 2 3 4 5 6 7 8 9 10 11
Density Matrix
1 27.9 34.5 31.3 32.8 29.2 28.7 22.5 26.8 26.8 25.4 23.3
2 31.3 39.0 36.4 36.7 32.8 32.5 25.4 30.9 30.9 29.0 26.7
3 33.7 42.4 40.8 40.8 37.4 35.7 28.0 34.5 34.5 32.4 29.0
4 29.2 35.2 32.0 33.4 29.8 28.1 22.1 26.4 26.4 24.9 22.9
5 23.3 26.9 24.5 26.1 22.8 20.6 17.5 20.1 20.1 19.1 17.9
LOS by Segment Exhibit 25-67

Analysis
Example Problem 3:
Period 1 2 3 4 5 6 7 8 9 10 11
LOS Matrix
1 D D D D D D C C D C C
2 D D E D D D C C D C D
3 D D E D E E D D D D D
4 D D D D D D C C D C C
5 C C C C C C B B C B B



Time Interval
In this analysis step, facilitywide performance measures are calculated for
each analysis period (Exhibit 25-68), consistent with Example Problem 2. Because
the computations have already been shown, only summary results are shown
here. The improvement restored the facility LOS to the values experienced in the
original pregrowth scenario, as shown in Exhibit 25-68.


Space Mean Average
Performance Measure
Summary
1 57.9 26.8 D
2 57.1 30.3 D
3 55.9 33.5 D
4 57.8 26.9 D
5 58.6 20.8 C
Total 57.5 27.7 —
available.
Groups and Facility
density for each time interval. The improvement restored the facility LOS to the
values experienced in the original pregrowth (undersaturated) scenario shown in
Exhibit 25-51.
EXAMPLE PROBLEM 4: EVALUATION OF AN UNDERSATURATED

FACILITY WITH A WORK ZONE
The Facility
In this example, a long-term work zone is placed on the final segment of
Example Problem 1. Exhibit 25-69 shows the change to the facility.
Exhibit 25-69
Example Problem 4:
Freeway Facility
The modified geometry of the 6-mi directional freeway facility is reflected in

Exhibit 25-70.

5,280 1,500 2,280 1,500 5,280 2,640 5,280 1,140 360 1,140 5,280
(ft)
No. of lanes 3 3 3 3 3 4 3 3 3 3 2

The Facts
TRD = 1.0 ramp/mi;
Terrain = level; and
Analysis duration = 75 min (divided into five 15-min intervals).
A queue discharge capacity drop of 7% is assumed for non–work zone
conditions.
Comments
The traffic demand flow inputs are identical to those in Example Problem 1.
The work zone has a single lane closure (in Segment 11), plastic drum barriers,
and a lateral distance of 0 ft in an urban area. Daytime performance is of interest
throughout the analysis.

Similar to the previous examples, there are five analysis periods and the
facility has 11 segments. The work zone extension to the methodology will be
included as part of the analysis.

The segmentation of the subject facility is given in Exhibit 25-69. Therefore,
there is no need to go through the segmentation process.

Traffic demand inputs for all 11 segments and five analysis periods are
identical to those in Example Problem 1, as shown in Exhibit 25-45. The values
represent the adjusted demand flows on the facility as determined from field
observations or other sources. Additional geometric and traffic-related inputs are
as specified in Exhibit 25-70 and the Facts section of the problem statement.




In this step, all input data for the subject facility are coded in the

The resulting capacities are shown in Exhibit 25-71. Because the capacity of a
weaving segment depends on traffic patterns, it varies by analysis period. The
remaining capacities are constant for all five analysis periods. The capacities for
Segments 1–5 and for Segments 7–10 are the same because the segments have the
same basic cross section. The lane closure on Segment 11 reduces its base
capacity by 33%. The impacts of work zone presence on further capacity
reduction are assessed in the next step.

Segment Capacities
1 8,273
2 8,281
3 6,748 6,748 6,748 6,748 6,748 8,323 6,748 6,748 6,748 6,748 4,499
4 8,403
5 8,463

To calculate the CAF for the work zone (Segment 11), the queue discharge
and prebreakdown capacities are required. As a result of the work zone, Segment
11 has two open lanes and one closed lane. Therefore, from Exhibit 10-15, its lane
closure severity index LCSI value is equal to 0.75. Equation 10-8 gives the
segment’s queue discharge capacity as follows:
𝑄𝐷𝑅𝑤𝑧 = 2,093 − 154 × 𝐿𝐶𝑆𝐼 − 194 × 𝑓𝐵𝑟 − 179 × 𝑓𝐴𝑇 + 9 × 𝑓𝐿𝐴𝑇 − 59 × 𝑓𝐷𝑁
𝑄𝐷𝑅𝑤𝑧 = 2,093 − 154 × 𝐿𝐶𝑆𝐼 − 194 × 𝑓𝐵𝑟 − 179 × 𝑓𝐴𝑇 + 9 × 𝑓𝐿𝐴𝑇 − 59 × 𝑓𝐷𝑁
= 2,093 − 154 × 0.75 − 194 × 1 − 179 × 0 − 59 × 0 + 9 × 0
= 1,783.5 pc/h/ln
Using Equation 10-9 and assuming a 13.1% queue discharge capacity drop in
work zone conditions, prebreakdown capacity is calculated as follows:
𝑄𝐷𝑅𝑤𝑧
𝑐𝑊𝑍 = × 100
100 − 𝛼𝑤𝑧
1,783.5
𝑐𝑊𝑍 = × 100
100 − 13.1
𝑐𝑊𝑍 = 2,052.3 pc/h/ln

Then, from Equation 10-11, the work zone CAF is equal to

𝑐𝑤𝑧 2,052.3
𝐶𝐴𝐹𝑤𝑧 = = = 0.892
𝑐 2,300
Using a similar approach, the work zone SAF can be found as follows from
Equation 10-10 and Equation 10-12.
𝐹𝐹𝑆𝑤𝑧 = 9.95 + 33.49 × 𝑓𝑆𝑟 + 0.53 × 𝑆𝐿𝑤𝑧 − 5.60 × 𝐿𝐶𝑆𝐼 − 3.84 × 𝑓𝐵𝑟
−1.71 × 𝑓𝐷𝑁 − 8.7 × 𝑇𝑅𝐷
60
𝐹𝐹𝑆𝑤𝑧 = 9.95 + 33.49 × ( ) + 0.53 × 55 − 5.60 × 0.75 − 3.84 × 1
55
−1.71 × 0 − 8.7 × 1
𝐹𝐹𝑆𝑤𝑧 = 58.9 mi/h
𝐹𝐹𝑆𝑤𝑧 58.9
𝑆𝐴𝐹𝑤𝑧 = = = 0.982
𝐹𝐹𝑆 60
These values will be used to update the capacity and FFS of Segment 11 in all
analysis periods. In addition, the number of lanes in the segment will be reduced
to two.


The demand-to-capacity ratios shown in Exhibit 25-72 are calculated from
the demand flows in Exhibit 25-45 and segment capacities in Exhibit 25-71.
Demand-to-Capacity Ratio by Segment Exhibit 25-72

Analysis
Period 1 2 3 4 5 6 7 8 9 10 11
1 0.67 0.73 0.73 0.73 0.69 0.63 0.72 0.79 0.79 0.79 1.26
2 0.73 0.81 0.81 0.81 0.76 0.71 0.81 0.89 0.89 0.89 1.44
3 0.77 0.87 0.87 0.87 0.83 0.77 0.89 0.99 0.99 0.99 1.56
4 0.69 0.75 0.75 0.75 0.71 0.61 0.71 0.77 0.77 0.77 1.24
5 0.56 0.59 0.59 0.59 0.55 0.47 0.56 0.60 0.60 0.60 0.97

shows the presence of the work zone significantly increases the demand-to-
capacity ratio on Segment 11. Queues are very likely to start to grow and spill
back to upstream segments, and the facility is expected to operate in
oversaturated conditions.
Step A-12: Compute Oversaturated Segment Service Measures

The computations for oversaturation apply to any segment with a vd/c ratio
greater than 1.0, as well as any segments upstream of those segments that
14, as applicable, with the caveat that the volumes served may differ from the
demand flows.

Similar to Example Problem 1, in Example Problem 4, the methodology

calculates performance measures for each segment and each analysis period,
starting with the first segment in Analysis Period 1. The computations are
repeated for the first 10 segments for Analysis Period 1 without encountering a
segment with vd/c > 1.0. Once the methodology enters Segment 11 in Analysis
Period 1, the oversaturated computational module is invoked.
The va/c ratio for Segment 11, which has the first active bottleneck, will be
more than 1.0 and the segment will process traffic at its capacity. Consequently,
demand for all downstream segments will be metered by that bottleneck. The
unsatisfied demand is stored in upstream segments, which causes queuing in
Segment 10 and perhaps additional upstream segments, depending on the level
of excess demand. The rate of growth of the vehicle queue (wave speed) is
estimated from shock wave theory. The performance measures (speed and
density) of any segment with queuing are recomputed, and the newly calculated
values override the results from the segment-specific procedures.
Any unsatisfied demand is served in later analysis periods. As a result,
volumes served in later analysis periods may be higher than the period demand
flows. The resulting matrix of volumes served for Example Problem 4 is shown
in Exhibit 25-73.
Exhibit 25-73 Volumes Served (veh/h) by Segment

Analysis
Example Problem 4:
Period 1 2 3 4 5 6 7 8 9 10 11
1 4,505 4,955 4,955 4,955 4,685 5,225 3,924 4,185 4,126 3,929 3,719
2 4,955 5,495 5,495 5,446 3,947 3,701 3,325 3,878 3,882 3,895 3,714
3 3,275 3,476 3,094 3,031 2,912 3,391 3,250 3,899 3,905 3,929 3,714
4 2,831 3,398 3,474 3,416 3,424 3,914 3,597 4,014 4,004 3,965 3,714
5 3,589 3,991 4,096 3,957 3,452 3,912 3,675 3,923 3,916 3,897 3,714
As a result of the bottleneck activation (due to the work zone’s presence) in

Segment 11 in Analysis Period 1, queues form in upstream Segments 10, 9, 8, 7,
and 6. The queuing is associated with reduced speeds and increased densities in
those segments. These and subsequent results were obtained from the
computational engine. The resulting performance measures computed for each
segment and time interval are speed (Exhibit 25-74), density (Exhibit 25-75), and
LOS (Exhibit 25-76). Similar trends are observed in the following time intervals,
with queueing reaching the beginning of the facility.

Analysis
Example Problem 4:
Period 1 2 3 4 5 6 7 8 9 10 11
Speed Matrix
1 60.0 53.9 59.7 56.1 60.0 48.0 24.2 15.9 13.0 13.0 50.4
2 59.9 53.2 54.5 52.3 22.2 8.9 9.4 12.3 12.2 12.2 50.5
3 12.9 12.8 13.1 9.7 8.0 6.5 9.1 12.4 12.4 12.4 50.5
4 5.9 11.0 12.9 12.8 11.5 8.3 11.0 13.1 12.7 12.7 50.5
5 11.0 16.4 18.6 16.4 12.3 8.3 11.2 12.5 12.3 12.3 50.5

Density (veh/mi/ln) by Segment Exhibit 25-75

Analysis
Example Problem 4:
Period 1 2 3 4 5 6 7 8 9 10 11
Density Matrix
1 25.0 30.6 27.6 29.4 26.0 27.2 54.1 87.5 100.6 100.6 36.9
2 27.6 34.5 33.6 34.7 59.1 104.2 117.8 105.5 106.2 106.2 36.8
3 84.6 90.6 78.7 104.6 121.4 130.1 119.1 104.4 105.4 105.4 36.8
4 159.3 103.4 89.8 88.7 99.4 117.3 109.0 102.5 104.2 104.2 36.8
5 108.6 81.0 73.5 80.4 93.5 118.2 109.2 105.0 106.0 106.0 36.8
LOS by Segment Exhibit 25-76

Analysis
Example Problem 4:
Period 1 2 3 4 5 6 7 8 9 10 11
LOS Matrix
1 C C D C D C F F F F E
2 D D D D F F F F F F E
3 F F F F F F F F F F E



Time Interval
for each analysis period (Exhibit 25-77). Because the computations have already
been demonstrated in previous example problems, only summary results are
shown. The work zone presence created significant congestion on the subject
facility.
Performance Measure Exhibit 25-77

Example Problem 4:
Space Mean Average
Facility Performance Measure
Summary
1 39.2 38.4 F
2 21.8 66.1 F
3 11.5 99.1 F
4 11.3 105.5 F
5 13.7 93.4 F
Total 19.5 80.5 —
available.
Groups and Facility
density for each time interval. Work zone presence eroded the facility LOS to F in
all time intervals.

EXAMPLE PROBLEM 5: EVALUATION OF AN OVERSATURATED FACILITY

WITH A MANAGED LANE
The Facility
In this example, a managed lane will be added to the freeway facility
described in Example Problem 2. Exhibit 25-78 shows the new facility geometry.
Exhibit 25-78
Example Problem 5:
Freeway Facility
Details of the modified geometry of the 6-mi directional freeway facility are
provided in Exhibit 25-79.

5,280 1,500 2,280 1,500 5,280 2,640 5,280 1,140 360 1,140 5,280
(ft)
No. of GP lanes 3 3 3 3 3 4 3 3 3 3 3
No. of ML 1 1 1 1 1 1 1 1 1 1 1
(diverge) segment; R = overlapping ramp segment; GP = general purpose; ML = managed lanes.
The Facts
TRD = 1.0 ramp/mi;
Terrain = level;
Analysis duration = 75 min (divided into five 15-min intervals); and
Demand adjustment = +11% (all segments and all time intervals).

Comments
The traffic demand flow inputs are identical to those in Example Problem 2.
The facility includes a single managed lane separated with marking with FFS
equal to 60 mi/h. The lane is a basic managed lane with no intermediate access
points. It is assumed 20% of entry traffic demand on the mainline will use the
managed lane.

Similar to the previous examples, there are five analysis periods and the
facility has 11 segments. The managed lane extension to the methodology will be
used for this analysis.

The segmentation of the subject facility is given in Exhibit 25-78. Therefore,
the segmentation process is not repeated.

On- and off-ramp demand flow rates are identical to those of Example
Problem 2, shown in Exhibit 25-53. It is assumed total entry volume is identical to
that of Example Problem 2; however, 20% of total demand is allocated to the
managed lane, and the remaining 80% to the general purpose lanes, as shown in
Exhibit 25-80.
Entering Flow Rate on Entering Flow Rate on Sum of Entering Flow Exhibit 25-80
Analysis General Purpose Lanes Managed Lane Rate to the Facility Example Problem 5: Demand
Period (veh/h) (veh/h) (veh/h) Inputs on the Mainline
1 4,001 1,000 5,001
2 4,400 1,100 5,500
3 4,640 1,160 5,800
4 4,160 1,040 5,200
5 3,361 840 4,201

The traffic flows in Exhibit 25-53 and Exhibit 25-80 have already been given
in the form of actual demands and no balancing is necessary.

both parameters are given in the problem statement.

In this step, all input data for the subject facility are coded in the

for basic freeway segments (general purpose and managed lanes), Chapter 13 for
weaving segments, and Chapter 14 for merge and diverge segments. The
resulting capacities are shown in Exhibit 25-81.

Exhibit 25-81 Analysis Capacities (veh/h) by Segment for General Purpose Lanes
Segment Capacities
1 8,177
2 8,189
3 6,748 6,748 6,748 6,748 6,748 8,244 6,748 6,748 6,748 6,748 6,748
4 8,331
5 8,403
Analysis Capacities (veh/h) by Segment for Managed Lane
Period 1 2 3 4 5 6 7 8 9 10 11
1
2
3 1,614 1,614 1,614 1,614 1,614 1,614 1,614 1,614 1,614 1,614 1,614
4
5

purpose of calibration. According to the problem statement, there is no
adjustment needed for the subject facility’s capacity.

This facility does not have a cross weave. Therefore, this step is skipped.

The demand-to-capacity ratios shown in Exhibit 25-82 are calculated from
the demand flows in Exhibit 25-53 and Exhibit 25-80 and segment capacities in
Exhibit 25-81.
Exhibit 25-82 Demand-to-Capacity Ratio by Segment (General Purpose Lanes)

Analysis
Period 1 2 3 4 5 6 7 8 9 10 11
1 0.59 0.67 0.67 0.67 0.62 0.59 0.65 0.73 0.73 0.73 0.68
2 0.65 0.74 0.74 0.74 0.68 0.66 0.74 0.83 0.83 0.83 0.79
3 0.69 0.79 0.79 0.79 0.75 0.72 0.82 0.92 0.92 0.92 0.85
4 0.62 0.68 0.68 0.68 0.63 0.56 0.63 0.71 0.71 0.71 0.66
5 0.50 0.53 0.53 0.53 0.48 0.42 0.50 0.54 0.54 0.54 0.51
Analysis Demand-to-Capacity Ratio by Segment (Managed Lane)
Period 1 2 3 4 5 6 7 8 9 10 11
1 0.62 0.62 0.62 0.62 0.62 0.62 0.62 0.62 0.62 0.62 0.62
2 0.68 0.68 0.68 0.68 0.68 0.68 0.68 0.68 0.68 0.68 0.68
3 0.72 0.72 0.72 0.72 0.72 0.72 0.72 0.72 0.72 0.72 0.72
4 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64
5 0.52 0.52 0.52 0.52 0.52 0.52 0.52 0.52 0.52 0.52 0.52

shows the addition of the managed lane improves traffic operations on the
general purpose lanes. As such, it is expected the facility will operate in
undersaturated conditions.

The computations for oversaturation apply to any segment with a vd/c ratio
greater than 1.0 as well as any segments upstream of those segments that

14, as applicable, with the caveat that volumes served may differ from demand
flows.
The basic performance service measures computed for each segment and
each time interval include segment speed (Exhibit 25-83), density (Exhibit 25-84),
and LOS (Exhibit 25-85).
Speed (mi/h) by Segment (General Purpose Lanes) Exhibit 25-83

Analysis
Example Problem 5:
Period 1 2 3 4 5 6 7 8 9 10 11
Speed Matrix
1 60.0 54.4 59.7 56.2 60.0 48.0 60.0 54.0 54.0 56.1 60.0
2 60.0 53.8 59.7 55.9 60.0 46.8 59.8 53.0 53.0 55.8 59.2
3 60.0 53.3 59.1 55.9 59.7 46.2 58.5 51.7 51.7 55.0 57.7
4 60.0 54.3 59.7 56.2 60.0 49.9 60.0 54.1 54.1 56.1 60.0
5 60.0 55.2 59.8 56.3 60.0 52.7 60.0 55.1 55.1 56.5 60.0
Analysis Speed (mi/h) by Segment (Managed Lane)
Period 1 2 3 4 5 6 7 8 9 10 11
1 59.3 59.3 59.3 59.3 59.3 59.3 59.3 59.3 59.3 59.3 59.3
2 58.9 58.9 58.9 58.9 58.9 58.9 58.9 53.5 53.5 58.1 58.9
3 58.6 58.6 58.6 58.6 58.6 58.6 58.6 52.1 52.1 52.1 58.6
4 59.2 59.2 59.2 59.2 59.2 59.2 59.2 59.2 59.2 59.2 59.2
5 59.7 59.7 59.7 59.7 59.7 59.7 59.7 59.7 59.7 59.7 59.7
Density (veh/mi/ln) by Segment (General Purpose Lanes) Exhibit 25-84

Analysis
Example Problem 5:
Period 1 2 3 4 5 6 7 8 9 10 11
Density Matrix
1 22.2 27.6 25.0 26.7 23.3 25.0 24.4 30.3 30.3 29.1 25.6
2 24.4 31.0 27.9 29.8 25.6 28.9 27.9 35.2 35.2 33.4 29.8
3 25.8 33.4 30.1 31.8 28.1 32.2 31.6 40.2 40.2 37.8 33.2
4 23.1 28.0 25.3 27.1 23.7 23.4 23.7 29.3 29.3 28.3 24.8
5 18.7 21.5 19.8 21.1 18.1 16.9 18.7 22.1 22.1 21.6 19.2
Analysis Density (veh/mi/ln) by Segment (Managed Lane)
Period 1 2 3 4 5 6 7 8 9 10 11
1 16.9 16.9 16.9 16.9 16.9 16.9 16.9 16.9 16.9 16.9 16.9
2 18.7 18.7 18.7 18.7 18.7 18.7 18.7 20.6 20.6 18.7 18.7
3 19.8 19.8 19.8 19.8 19.8 19.8 19.8 22.3 22.3 22.3 19.8
4 17.6 17.6 17.6 17.6 17.6 17.6 17.6 17.6 17.6 17.6 17.6
5 14.1 14.1 14.1 14.1 14.1 14.1 14.1 14.1 14.1 14.1 14.1
LOS by Segment (General Purpose Lanes) Exhibit 25-85

Analysis
Example Problem 5:
Period 1 2 3 4 5 6 7 8 9 10 11
LOS Matrix
1 C C C C C C C C D C C
2 C C D C C D D D E D D
3 C D D D D D D D E D D
4 C C C C C C C C D C C
5 C B C C C B C B C C C
Analysis LOS by Segment (Managed Lane)
Period 1 2 3 4 5 6 7 8 9 10 11
1 B B B B B B B B B B B
2 C C C C C C C C C C C
3 C C C C C C C C C C C

The subject facility has densities in excess of 35 pc/mi/ln. As a result, friction
effects are applied according to the process described in Chapter 12. The
indicator variable 𝐼𝑐 in Equation 12-12 will have a nonzero value for the segments

and analysis periods during which the general purpose lane density is greater
than 35 pc/mi/ln. Consequently, the 𝑆3 term in Equation 12-12 will reduce the
estimated general purpose lane speed as a result of the friction.

In this step, performance measures for all the facility’s lane groups are
computed. The subject facility has two lane groups, one for general purpose
lanes and one for the managed lane, as shown in Exhibit 25-86.
Exhibit 25-86 General Purpose Lane Group Managed Lane Group

Example Problem 5: Facility Performance Measure Performance Measure
Performance Measure Analysis Space Mean Average Density Space Mean Average Density
Summary for Lane Groups Period Speed (mi/h) (veh/mi/ln) Speed (mi/h) (veh/mi/ln)
1 57.7 24.9 59.3 16.9
2 57.3 28.1 58.6 18.8
3 56.5 31.0 58.0 20.0
4 58.0 24.6 59.2 17.6
5 58.5 19.1 59.7 14.1

Time Interval
for each analysis period (Exhibit 25-87). Because the computations have been
demonstrated previously, only summary results are shown here. The addition of
the managed lane reduced traffic congestion on the subject facility.

Analysis Space Mean Average Density
Performance Measure
Period Speed (mi/h) (veh/mi/ln) LOS
Summary
1 58.0 23.4 C
2 57.5 26.4 D
3 56.7 29.1 D
4 58.2 23.3 C
5 58.7 18.1 C
Total 57.8 24.0 —
available.
Groups and Facility
density for each time interval. The addition of the managed lane improved traffic
conditions over the entire facility.

EXAMPLE PROBLEM 6: PLANNING-LEVEL ANALYSIS OF A FREEWAY

FACILITY
The Facility
In this example, the planning-level methodology is used to analyze a
freeway facility with geometric characteristics identical to the facility used in
Example Problem 1. Exhibit 25-43 shows the facility geometry. Note that the
planning methodology uses annual average daily traffic (AADT) values to
calculate demand levels at the facility’s entry and exit points based on the hourly
(K) and annual growth factors (fg). As a result, although the AADTs have been
manipulated in this example to create demand levels close to those of Example
Problem 1, the results will not match precisely. Furthermore, because the
planning-level methodology uses freeway sections rather than segments and is
limited to four analysis periods, a direct comparison is not possible.
The Facts
In addition to the information given in Exhibit 25-43 and Exhibit 25-44, the
following characteristics of the freeway facility are known:
Heavy-vehicle percentage = 0%,
Driver population = regular commuters on an urban facility,
FFS = 60 mi/h (all mainline segments),
Ramp FFS = 40 mi/h (all ramps),
Djam = 190 pc/mi/ln,
K-factor = 0.09,
Growth factor = 1,
PHF = 0.9,
Terrain = level, and
Analysis duration = 60 min (divided into four 15-min analysis
periods).
Average Annual Daily Traffic

The planning-level approach uses directional AADT values to approximate
demand levels on different freeway sections. Exhibit 25-88 depicts AADT values
on all entry points (i.e., the first basic freeway section and all on-ramps) and all
exit points (all off-ramps).
Entering AADT Ramp AADT (veh/day) Exhibit 25-88

(veh/day) Example Problem 6:
ONR1 ONR2 ONR3 OFR1 OFR2 OFR3
AADT Values for the Facility
55,000 4,500 5,400 4,500 2,700 3,600 2,700
Sections
The facility and all geometric inputs are identical to Example Problem 1.
Exhibit 25-89 presents the different freeway sections for the facility of interest.

Exhibit 25-89
Example Problem 6: Section
Definition for the Facility
Section 1 is a basic section, identical to the HCM segmentation definition. An

on-ramp roadway is located just downstream of Section 1 that results in changes
in the demand level. As a result, a new section needs to be defined. The demand
level on the new section remains fixed up to the first off-ramp roadway, at which
point both the capacity and the demand change. As a result, Section 2 is defined
as a ramp section. After the off-ramp roadway, the facility demand drops and
remains fixed until the next on-ramp roadway. As a result, Section 3 is defined as
a basic freeway section. Sections on the rest of the freeway facility are defined
following a similar process. The result is that seven distinct sections are defined.
Step 1: Demand Level Calculations

The demand level on each section in each analysis period is determined by
using the given AADT values, PHF, K-factor, heavy-vehicle factor, and growth
factor.
𝑞1,1 = 𝐴𝐴𝐷𝑇1 × 𝐾 × 𝑓𝑡𝑔 × 𝑓𝐻𝑉 = 55,000 × 0.09 × 1 × 1 = 4,950 pc/h
1 1
𝑞1,2 = 𝐴𝐴𝐷𝑇1 × 𝐾 × ( ) × 𝑓𝑡𝑔 × 𝑓𝐻𝑉 = 55,000 × 0.09 × ( ) × 1
𝑃𝐻𝐹 0.9
= 5,500 pc/h
𝑞1,3 = 𝐴𝐴𝐷𝑇1 × 𝐾 × 𝑓𝑡𝑔 = 55,000 × 0.09 × 1 × 𝑓𝐻𝑉 = 4,950 pc/h
1 1
𝑞1,4 = 𝐴𝐴𝐷𝑇1 × 𝐾 × (2 − ) × 𝑓𝑡𝑔 × 𝑓𝐻𝑉 = 55,000 × 0.09 × (2 − )×1
𝑃𝐻𝐹 0.9
= 4,400 pc/h
By following the same approach, the demand levels for all facility entry and
exit points are found. The results are summarized in Exhibit 25-90.
Exhibit 25-90 Analysis On-Ramp Off-Ramp On-Ramp Off-Ramp On-Ramp Off-Ramp

Example Problem 6: Period Entry 1 1 2 2 3 3
Demand Flow Rates (pc/h) on 1 4,950 405 243 486 324 405 243
the Subject Facility 2 5,500 450 270 540 360 450 270
3 4,950 405 243 486 324 405 243
4 4,400 360 216 432 288 360 216
After calculation of the entry and exit demand flow rates from the AADT
values, the demand level in each section in each analysis period is found.

Step 2: Section Capacity Calculations and Adjustments

Equation 25-45 is used to determine the base capacity of each section. The
base capacity of each section is then adjusted by using the appropriate
adjustment factor for a weaving, ramp, merge, or diverge section. For instance,
the capacity of Section 1 (a basic section) is determined as follows:
𝑐1 = (2,200 + 10 × (min(70, 𝑆𝐹𝐹𝑆 ) − 50)) = (2,200 + 10 × (min(70, 60) − 50))
𝑐1 = 2,300 pc/h/ln
Because FFS and percentage heavy vehicles are global inputs, the capacity of
each of the facility’s basic freeway sections is equal to 2,300 pc/h/ln. However, for
all other sections, this base capacity needs to be adjusted.
Section 2 is a ramp section. The CAF for a ramp section is 0.9. Therefore, the
capacity of Section 2 is computed as follows:
𝑐2 = 2300 × 0.90 = 2,070 pc/h/ln
Section 3 is a basic freeway section; therefore, its capacity remains at 2,300
pc/h/ln. However, Section 4 is a weaving section and its capacity will need to be
adjusted. The CAF for a weaving section is determined by the volume ratio and
section length.
The volume ratio (the ratio of weaving demand to total demand) is
approximated by summing the weaving section’s ramp AADT values and
dividing the result by the total AADT on the weaving section, as follows:
(5,400 + 3,600) 9,000
𝑉𝑟 = = = 0.158
55,000 + 4,500 − 2,700 56,800
The length of the weaving section is 0.5 mi. As a result, the CAF is calculated
as follows:
𝐶𝐴𝐹weave = min(0.884 − 0.0752𝑉𝑟 + 0.0000243𝐿𝑠 , 1)
𝐶𝐴𝐹weave = 0.884 − 0.0752 × 0.164 + 0.0000243 × 0.5 × 5,280 = 0.94
Therefore, the capacity of Section 4 is
𝑐4 = 2,300 × 0.94 = 2,162 pc/h/ln
The capacities of Section 5 (basic), Section 6 (ramp), and Section 7 (basic) are
2,300, 2,070, and 2,300 pc/h/ln, respectively. At this stage, demand-to-capacity
ratios for all sections in all analysis periods can be determined, as presented in
Exhibit 25-91.
Demand-to-Capacity Ratios by Section Exhibit 25-91

Analysis
Example Problem 6:
Period 1 2 3 4 5 6 7
Demand-to-Capacity Ratios by
1 0.72 0.86 0.74 0.65 0.76 0.91 0.79
Section and Analysis Period
2 0.80 0.96 0.82 0.72 0.85 1.02 0.88
3 0.72 0.86 0.74 0.65 0.76 0.93 0.80
4 0.64 0.77 0.66 0.58 0.68 0.81 0.70
As shown in Exhibit 25-91, the demand-to-capacity ratio in the sixth section

in the second analysis period is greater than one. As a result, queue formation
and low space mean speeds are expected on this section. The demand-to-capacity
ratios on the remaining segments are below one across all analysis periods.

Step 3: Delay Rate Estimation

In this step, demand-to-capacity ratios are used to determine delay rates for
all sections of the facility across all analysis periods. FFS on the facility is 60 mi/h,
and all demand-to-capacity ratios are below one. As a result, the delay rates for
each section are found by using Equation 25-47.
𝑑𝑖,𝑡
0 < 0.72
𝑐𝑖
∆𝑅𝑈𝑖,𝑡 =
𝑑𝑖,𝑡 3 𝑑𝑖,𝑡 2 𝑑𝑖,𝑡 𝑑𝑖,𝑡
121.35 ( ) + (−184.84) ( ) + 83.21 ( ) + (−9.33) 0.72 ≤ ≤1
{ 𝑐𝑖 𝑐𝑖 𝑐𝑖 𝑐𝑖
For instance, the delay rate for Section 1 in the first analysis period is 0 s/mi,
because its demand-to-capacity ratio of 0.717 is less than the 0.72 threshold used
in Equation 25-47. Section 2’s demand-to-capacity ratio is 0.86, which is greater
than the threshold. Therefore, its delay rate is calculated as follows:
∆𝑅𝑈2,1 = 121.35(0.86)3 + (−184.84)(0.86)2 + 83.21(0.86) + (0.86) = 2.8 s/mi
Delay rates for other sections of the facility are determined in the same way
and are summarized in Exhibit 25-92.
Exhibit 25-92 Delay Rate by Section (s/mi)

Analysis
Example Problem 6:
Period 1 2 3 4 5 6 7
Delay Rates by Section and
1 0.0 2.8 0.2 0.0 0.5 5.0 0.8
Analysis Period
2 1.0 7.4 1.6 0.1 2.3 11.7 3.3
3 0.0 2.8 0.2 0.0 0.5 5.8 1.1
4 0.0 0.5 0.0 0.0 0.0 1.3 0.0
Step 4: Average Travel Time, Speed, and Density Calculations

Delay rates are used to compute travel times and, consequently, speeds. To
determine a section’s travel time, its travel rate is calculated by summing the
section’s travel rate under free-flow conditions and its delay rates for
undersaturated and oversaturated conditions. This calculation is repeated for
each section across all analysis periods. The following equations demonstrate the
calculation for the first two sections during the first analysis period:
3,600 3,600
𝑇𝑅1,1 = ∆𝑅𝑈1,1 + ∆𝑅𝑂1,1 + 𝑇𝑅𝐹𝐹𝑆 = 0.00 + 0.00 + = = 60 s/mi
𝑆𝐹𝐹𝑆 60
3,600 3,600
𝑇𝑅2,1 = ∆𝑅𝑈2,1 + ∆𝑅𝑂2,1 + 𝑇𝑅𝐹𝐹𝑆 = 0.00 + 0.00 + = 2.8 +
𝑆𝐹𝐹𝑆 60
= 62.8 s/mi
Travel rates for all sections across all analysis periods are shown in Exhibit
25-93.

Travel Rate by Section (s/mi) Exhibit 25-93

Analysis
Example Problem 6:
Period 1 2 3 4 5 6 7
Travel Rates by Section and
1 60.0 62.8 60.2 60.0 60.5 65.0 60.8
Analysis Period
2 61.0 67.4 61.6 60.1 62.3 71.7 63.3
3 60.0 62.8 60.2 60.0 60.5 65.8 61.1
4 60.0 60.5 60.0 60.0 60.0 61.3 60.0
Each section’s travel time is calculated by multiplying its travel rate by its
length. The results are presented in Exhibit 25-94.
Travel Time by Section (s) Exhibit 25-94

Analysis
Example Problem 6:
Period 1 2 3 4 5 6 7
Average Travel Times by
1 60.0 62.8 60.2 30.0 60.5 32.5 60.8
Section and Analysis Period
2 61.0 67.4 61.6 30.0 62.3 35.8 63.3
3 60.0 62.8 60.2 30.0 60.5 32.9 61.1
4 60.0 60.5 60.0 30.0 60.0 30.7 60.0
Density is determined for each section across all analysis periods by dividing
the section’s demand by its speed (section length divided by travel time). The
results are shown in Exhibit 25-95.
Density by Section (pc/mi/ln) Exhibit 25-95

Analysis
Example Problem 6:
Period 1 2 3 4 5 6 7
Density by Section and
1 27.5 31.1 28.5 23.3 29.5 34.2 30.6
Analysis Period
2 31.1 37.2 32.4 25.9 33.8 41.2 35.4
3 27.5 31.1 28.5 23.3 29.5 35.2 31.3
4 24.4 26.7 25.2 20.7 26.0 28.7 26.8
Finally, the approach provides a high-level summary that includes a capacity

assessment, the aggregated travel time, the space mean speed, the average
facility density, the total queue length, and the facility LOS by analysis period, as
Exhibit 25-96
Space Average Total
Example Problem 6:
High-Level Travel Mean Facility Queue
Facility Performance Summary
Analysis Capacity Time Speed Density Length
Period Assessment (min) (mi/h) (pc/mi/ln) (mi) LOS
1 Undersaturated 6.1 58.9 29.2 0.0 D
2 Oversaturated 6.4 56.6 33.7 0.8 F
3 Undersaturated 6.1 58.8 29.4 0.0 D
4 Undersaturated 6.0 59.8 25.5 0.0 C
The average facility travel time in each analysis period is calculated by

summing each section’s travel time and dividing the result by 60 to convert the
units to minutes. Space mean speed in each analysis period is then calculated by
dividing the total facility length by the facility travel time in each analysis period.
The facility density is a length-weighted average of each section’s density, and
the total queue length is the sum of each section’s queue length. Finally, LOS is
calculated based on the urban freeway density thresholds if the demand-to-
capacity ratio is less than 1; otherwise, LOS is set to F if any section operates at a
demand-to-capacity ratio greater than 1.0.
The facility is oversaturated during the second analysis period, with one of
the sections experiencing a demand-to-capacity ratio greater than 1.0. The

method estimates that a 0.8-mi queue will result from an active bottleneck. With
at least one time interval operating at LOS F, it is recommended that a more
detailed operational analysis of this facility be conducted to obtain a more
accurate estimate of congestion patterns.
EXAMPLE PROBLEM 7: RELIABILITY EVALUATION OF AN EXISTING

FREEWAY FACILITY
The Facility
This example problem uses the same 6-mi facility used in Example Problem
1. The facility consists of 11 segments with the properties indicated in Exhibit 25-
97. Other facility characteristics are identical to those given in Example Problem
1, except that the study period in this example has been extended from 75 to 180
min. Exhibit 25-98 shows the facility geometry.
Exhibit 25-97
Example Problem 7:
Freeway Facility

of Directional Freeway Facility Segment
5,280 1,500 2,280 1,500 5,280 2,640 5,280 1,140 360 1,140 5,280
length (ft)
No. of lanes 3 3 3 3 3 4 3 3 3 3 3
Input Data
This example illustrates the use of defaults and lookup tables to substitute
for desirable but difficult to obtain data. Minimum facility inputs for the example
problem include the following.
Facility Geometry
All the geometric information about the facility normally required for an
HCM freeway facility analysis (Chapters 10–14) is also required for a reliability
analysis. These data are supplied as part of the base scenario.
Study Parameters
These parameters specify the study period, the reliability reporting period,
and the date represented by the traffic demand data used in the base scenario.
The study period in this example is from 4 to 7 p.m., which covers the
afternoon and early evening peak hour and shoulder periods. Recurring
congestion is typically present in the study direction of this facility during that
period, which is why it has been selected for reliability analysis. The reliability
reporting period is set as all weekdays in the calendar year. (For simplicity of
presentation in this example, holidays have not been removed from the

reliability reporting period.) The demand data are reflective of AADT variations
across the weekdays and months in a calendar year for the subject facility.
Base Demand
Demand flow rates in vehicles per hour are supplied for each 15-min analysis
period in the base scenario. Care should be taken that demand data are measured
upstream of any queued traffic. If necessary, demand can be estimated as the
sum of departing volume and the change in the queue size at a recurring
bottleneck.
Exhibit 25-99 provides the twelve 15-min demand flow rates required for the
entire 3-h study period.
Demand Exhibit 25-99

Analysis Entry Flow Example Problem 7: Demand
Period Rate ONR1 ONR2 ONR3 OFR1 OFR2 OFR3 Flow Rates (veh/h) by
1 3,095 270 270 270 180 270 180 Analysis Period in the Base
2 3,595 360 360 360 270 360 270 Data Set
3 4,175 360 450 450 270 360 270
4 4,505 450 540 450 270 360 270
5 4,955 540 720 540 360 360 270
6 5,225 630 810 630 270 360 450
7 4,685 360 360 450 270 360 270
8 3,785 180 270 270 270 180 180
9 3,305 180 270 270 270 180 180
10 2,805 180 270 270 270 180 180
11 2,455 180 180 180 270 180 180
12 2,405 180 180 180 180 180 180
Note: ONR = on-ramp; OFR = off-ramp.
Incident Data
Detailed incident logs are not available for this facility, but local data are
available about the facility’s crash rate: 150 crashes per 100 million VMT. An
earlier study conducted by the state in which the facility is located found that an
average of seven incidents occur for every crash.
Computational Steps
Base Data Set Analysis
The Chapter 10 freeway facilities core methodology is applied to the base
data set to ensure the specified facility boundaries and study period are sufficient
to cover any bottlenecks and queues. In addition, because incident data are
supplied in the form of a facility crash rate, the VMT associated with the base
data set are calculated so that incident probabilities can be calculated in a
subsequent step. In this case, 71,501 vehicle miles of travel occur on the facility
over the 3-h base study period. The performance measures normally output by
the Chapter 10 methodology are compiled for each combination of segment and
analysis period during the study period and stored for later use. Of particular
note, the facility operates just under capacity, with a maximum demand-to-
capacity ratio of 0.99 in Segments 7–10.

Incorporating Demand Variability

Exhibit 25-100 provides demand ratios relative to AADT by month and day
derived from a permanent traffic recorder on the facility. The demand values for
the seed file were collected on a Tuesday in November.
Exhibit 25-100 Month Monday Tuesday Wednesday Thursday Friday

Example Problem 7: Demand January 0.822 0.822 0.839 0.864 0.965
Ratios Relative to AADT February 0.849 0.849 0.866 0.892 0.996
March 0.921 0.921 0.939 0.967 1.080
April 0.976 0.976 0.995 1.025 1.145
May 0.974 0.974 0.993 1.023 1.142
June 1.022 1.022 1.043 1.074 1.199
July 1.133 1.133 1.156 1.191 1.329
August 1.033 1.033 1.054 1.085 1.212
September 1.063 1.063 1.085 1.117 1.248
October 0.995 0.995 1.016 1.046 1.168
November 0.995 0.995 1.016 1.046 1.168
December 0.979 0.979 0.998 1.028 1.148
Incorporating Weather Variability

In the absence of facility-specific weather data, the default weather data for
the metropolitan area closest to the facility are used.
In the absence of local data, the default CAF and SAF for an FFS of 60 mi/h
are used for each weather event. These values are applied in a later step to each
scenario involving a weather event. Exhibit 25-101 summarizes the probabilities
of each weather event by season, and Exhibit 25-102 summarizes the CAF, SAF,
and event duration values associated with each weather event.
Exhibit 25-101 Weather Event Probability by Season (%)

Example Problem 7: Weather Weather Event Winter Spring Summer Fall
Event Probabilities by Season Medium rain 0.80 1.01 0.71 0.86
Heavy rain 0.47 0.81 1.33 0.68
Light snow 0.91 0.00 0.00 0.00
Light–medium snow 0.29 0.00 0.00 0.00
Medium–heavy snow 0.04 0.00 0.00 0.00
Heavy snow 0.00 0.00 0.00 0.00
Severe cold 0.00 0.00 0.00 0.00
Low visibility 0.97 0.12 0.16 0.34
Very low visibility 0.00 0.00 0.00 0.00
Minimal visibility 0.44 0.10 0.00 0.03
Nonsevere weather 96.09 97.95 97.80 98.08
Note: Winter = December, January, and February; spring = March, April, and May; summer = June, July, and
August; fall = September, October, and November.
Exhibit 25-102 Weather Event CAF SAF Average Duration (min)

Example Problem 7: Medium rain 0.93 0.95 40.2
CAF, SAF, and Event Duration Heavy rain 0.86 0.93 33.7
Values Associated with Light snow 0.96 0.92 93.1
Weather Events Light–medium snow 0.94 0.90 33.4
Medium–heavy snow 0.91 0.88 21.7
Heavy snow 0.78 0.86 7.3
Severe cold 0.92 0.95 0.0
Low visibility 0.90 0.95 76.2
Very low visibility 0.88 0.94 0.0
Minimal visibility 0.90 0.94 145
Nonsevere weather 1.00 1.00 N/A
Note: N/A = not applicable.

Incorporating Incident Variability

For an existing freeway facility such as this one, detailed incident logs would
be desirable so that facility-specific monthly or seasonal probabilities of various
incident severities could be determined. However, in this case, incident logs of
sufficient detail are not available.
Therefore, incident probabilities and severities are estimated by the
alternative method of using local crash rates and ratios of incidents to crashes, in
combination with default values, by using Equation 25-77 through Equation 25-79.
The expected number of incidents during a study period under a specified
demand pattern is the product of the crash rate, the local incident-to-crash ratio,
the demand volume during the study period, and the facility length. The crash
rate is 150 crashes per 100 million VMT; the ratio of incidents to crashes is given
as 7. The resulting incident frequencies for different months of the reliability
reporting period are determined as shown in Exhibit 25-103.
Incident Exhibit 25-103

Month Frequency Example Problem 7: Incident
January 0.65 Frequencies by Month
February 0.67
March 0.72
April 0.77
May 0.77
June 0.80
July 0.89
August 0.82
September 0.83
October 0.83
November 0.79
December 0.77
Results and Discussion

Exhibit 25-104 provides key reliability performance measure results for this
example problem. The number of replications for each scenario was four,
resulting in 240 scenarios. Exhibit 25-105 shows the generated probability and
cumulative distributions of travel time index (TTI) for this example problem. A
seed number of 1 was chosen to generate random numbers in the computational
engine.
Value from All Exhibit 25-104

Reliability Performance Measure Scenarios Example Problem 7: Summary
TTI50 1.03 Reliability Performance
TTImean 1.30 Measure Results
PTI (TTI95) 1.67
Maximum observed facility TTI (TTImax) 33.57
Misery index 5.76
Reliability rating 90.8%
Semi-standard deviation 2.05
Percentage VMT at TTI >2 2.95%
Note: PTI = planning time index; TTI = travel time index.

Exhibit 25-105
Example Problem 7:
VMT-Weighted TTI Probability
and Cumulative Distribution
Functions
(a) Probability Distribution Function
(b) Cumulative Distribution Function
EXAMPLE PROBLEM 8: RELIABILITY ANALYSIS WITH GEOMETRIC

IMPROVEMENTS
The Facility
In this example, the freeway facility from Example Problem 6 is widened by
a lane in Segments 7–11. These segments operated close to capacity in the base
scenario and were definitely over capacity in scenarios with severe weather or
incident conditions. The revised geometry also improves the operation of
weaving Segment 6, because no lane changes are required of traffic entering at
On-Ramp 2. Exhibit 25-106 provides a schematic of the freeway facility.
Exhibit 25-106
Example Problem 8:
Freeway Facility

Data Inputs
All the input data used in Example Problem 6 remain unchanged, except for
the number of lanes on the facility. The extra lane creates the possibility of having a
three-lane-closure incident scenario in the four-lane portion of the facility.

Exhibit 25-107 provides key reliability performance measure results for this
example problem. The mean TTI across the reliability reporting period decreases
from 1.54 to 1.18, corresponding to a speed improvement from 38.96 to 50.8
mi/h—more than a 10% increase and perhaps enough to justify the improvement,
once non-reliability-related factors are taken into account. Similar results occur
for most other performance measures.
Value from All Exhibit 25-107

Reliability Performance Measure Scenarios Example Problem 8: Summary
TTI50 1.02 Reliability Performance
TTImean 1.18 Measure Results
PTI (TTI95) 1.17
Misery index 4.07
EXAMPLE PROBLEM 9: EVALUATION OF INCIDENT MANAGEMENT

This example problem illustrates the analysis of a nonconstruction
alternative that focuses on improved incident management strategies. In this
example, the size of the motorist response fleet is increased and communication
is improved between the various stakeholders (e.g., traffic management center,
emergency responders, and motorist response fleet), allowing faster clearance of
incidents than before.
Data Inputs
All the input data used in Example Problem 6 remain unchanged, except for
the assumed incident durations and standard deviations. The default incident
mean durations and standard deviations are reduced by 30% each for all incident
severity types. Note that these values have been created for the purposes of this
example problem and do not necessarily reflect results that would be obtained in
an actual situation.

The key congestion and reliability statistics for this example problem are
summarized in Exhibit 25-108. The mean TTI across the reliability reporting
period decreases from 1.35 to 1.20, corresponding to a speed improvement from
44.4 to 50.0 mi/h—more than a 10% increase and perhaps enough to justify the
improvement, once non-reliability-related factors are taken into account. Similar
results occur for most other performance measures.

Exhibit 25-108 Value from All

Example Problem 9: Reliability Performance Measure Scenarios
Summary Reliability TTI50 1.03
Performance Measure Results TTImean 1.25
PTI (TTI95) 1.59
Misery index 4.88
EXAMPLE PROBLEM 10: PLANNING-LEVEL RELIABILITY ANALYSIS

This example illustrates the planning-level reliability analysis methodology
described in Chapter 11. The method estimates the mean and 95th percentile TTI,
as well as the percentage of trips occurring below a speed of 45 mi/h.
The Facts
The segment under study has three lanes in the analysis direction, an FFS of
75 mi/h, and a peak hour speed of 62 mi/h. The volume-to-capacity ratio during
the peak hour is 0.95.
Solution
The value of TTImean is calculated from Equation 11-1, and is a function of the
recurring delay rate RDR and the incident delay rate IDR. These rates are
calculated from Equation 11-2 and Equation 11-3, respectively.
1 1
𝑅𝐷𝑅 =
−
𝑆 𝐹𝐹𝑆
1 1
𝑅𝐷𝑅 = − = 0.00280
62 75
𝐼𝐷𝑅 = [0.020 − (𝑁 − 2) × 0.003] × 𝑋12
𝐼𝐷𝑅 = [0.020 − (3 − 2) × 0.003] × (0.95)12 = 0.00919
TTImean can now be calculated as
𝑇𝑇𝐼mean = 1 + 𝐹𝐹𝑆 × (𝑅𝐷𝑅 + 𝐼𝐷𝑅)
𝑇𝑇𝐼mean = 1 + 75 × (0.00280 + 0.00919)
𝑇𝑇𝐼mean = 1.899
TTI95 is calculated from Equation 11-4 as follows:
𝑇𝑇𝐼95 = 1 + 3.67 × ln (𝑇𝑇𝐼mean )
𝑇𝑇𝐼95 = 1 + 3.67 × ln (1.899)
𝑇𝑇𝐼95 = 3.353
Finally, the percentage of trips made at a speed below 45 mi/h is calculated
with Equation 11-5.
𝑃𝑇45 = 1 − exp (−1.5115 × (𝑇𝑇𝐼mean − 1))
𝑃𝑇45 = 1 − exp(−1.5115 × (1.899 − 1))
𝑃𝑇45 = 74.3%

EXAMPLE PROBLEM 11: ESTIMATING FREEWAY COMPOSITE GRADE

OPERATIONS WITH THE MIXED-FLOW MODEL
This example problem addresses a composite grade section on a six-lane
freeway. It illustrates how the mixed-flow model procedures can be applied to
the case of composite grades.
The Facts
• Three segments with the following grades and lengths:
o First segment: 1.5-mi basic segment on a 3% upgrade
o Second segment: 2-mi basic segment on a 2% upgrade
o Third segment: 1-mi basic segment on a 5% upgrade
• 5% SUTs and 10% TTs
• FFS of 65 mi/h
• 15-min mixed-traffic flow rate is 1,500 veh/h/ln (PHF = 1.0)
Comments
Chapter 26, Basic Freeway and Highway Segments: Supplemental, presents
the procedure for estimating the speed on a single-grade basic freeway segment
using the mixed-flow model. The task here is to estimate the speed by mode for
each segment, along with the overall mixed-flow speed and travel time for the
composite grade.
Step 1: Input Data

All input data are specified above.
Step 2: Capacity Assessment

The CAF for mixed flow allows for the conversion of auto-only capacities
into mixed-traffic-stream capacities. It can be computed with Equation 25-53.
For the first segment,
𝐶𝐴𝐹mix,1 = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix,1
There are four terms in the equation. The CAF for auto-only conditions CAFao
is assumed to be 1, because no auto adjustments are necessary.

The truck effect term is computed from Equation 25-54.
𝐶𝐴𝐹𝑇,mix = 0.53 × 𝑃𝑇 0.72 = 0.53 × 0.150.72 = 0.135

The grade effect term is computed from Equation 25-55 and Equation 25-56.
Given that the total truck percentage is 15%, the coefficient ρg,mix is calculated as
𝜌𝑔,mix = 0.126 − 0.03𝑃𝑇 = 0.126 − 0.03 × 0.15 = 0.1215
and the CAF for grade effect for Segment 1 is calculated as

𝐶𝐴𝐹𝑔,mix,1 = 𝜌𝑔,mix × max[0, 0.69 × (𝑒12.9𝑔𝑗 − 1)]

× max[0, 1.72 × (1 − 1.71𝑒 −3.16𝑑𝑗 )]
𝐶𝐴𝐹𝑔,mix,1 = 0.1215 × max[0, 0.69 × (𝑒12.9×0.03 − 1)]

× max[0, 1.72 × (1 − 1.71𝑒 −3.16×1.5 )]
𝐶𝐴𝐹𝑔,mix,1 = 0.067
Mixed-Flow CAF
The mixed-flow CAF for Segment 1 can now be calculated from Equation
25-53.
𝐶𝐴𝐹mix,𝑗 = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix,1 = 1.000 − 0.135 − 0.067 = 0.798
Segment Capacity
The mixed-flow capacity of segment 1 is computed from the segment’s auto-
only capacity and mixed-flow CAF. The auto-only capacity is determined from
an equation in Exhibit 12-6.
𝐶𝑎𝑜 = 2,200 + 10(𝐹𝐹𝑆 − 50) = 2,200 + 10 × (65 − 50) = 2,350 pc/h/ln
Segment 1’s mixed-flow capacity is then determined with Equation 25-57.
𝐶mix,1 = 𝐶𝑎𝑜 × 𝐶𝐴𝐹mix,1 = 2,350 × 0.798 = 1,875 veh/h/ln
Because the mixed-flow CAFs and capacities for Segments 2 and 3 can be
computed by following the same procedure, the results are presented directly
without showing the computational details.
𝐶𝐴𝐹mix,2 = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix,2 = 1 − 0.135 − 0.042 = 0.823
𝐶𝐴𝐹mix,3 = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix,3 = 1 − 0.135 − 0.122 = 0.743
As the mixed-flow demand of 1,500 veh/h/ln is less than the smallest of the
three segment capacities, 1,746 veh/h/ln, the analysis can proceed.
Steps 3 to 6
Steps 3 through 6 are repeated for each segment, as shown below.
Segment 1
Step 3: Specify Initial Conditions
Because this is the first segment, an FFS of 65 mi/h is used as the initial truck
kinematic spot travel time rate. The effect of traffic interactions on truck speed is
accounted for in Step 4.
Step 4: Compute Truck Space-Based and Spot Travel Time Rates

Kinematic Spot Rates. The initial truck kinematic spot travel time rates for
both SUTs and TTs are 65 mi/h. These rates are located on the curves
representing a 3% upgrade starting from 75 mi/h (48 s/mi) in Exhibit 25-20
(SUTs) and Exhibit 25-21 (TTs).

The SUT and TT spot rates versus distance curves starting from 65 mi/h will
be applied to obtain τf,SUT,kin,1 and τf,TT,kin,1. In Exhibit 25-20, 65 mi/h (55.4 s/mi)
occurs about 4,100 ft into the 3% grade. After an SUT travels for 1.5 mi (7,920 ft)
starting at an initial speed of 65 mi/h, its spot rate can be read at 12,020 ft. That
distance is outside the plot range, but Exhibit 25-20 shows SUTs reach a crawl
speed of 59 s/mi (61 mi/h) at around 10,000 ft. Therefore, the kinematic spot rate
for SUTs at the end of the first segment τf,SUT,kin,1 is 59 s/mi.
In Exhibit 25-21, 65 mi/h (55.4 s/mi) is found at about 2,100 ft. After a TT
travels for 1.5 mi (7,920 ft) from an initial speed of 65 mi/h, its spot rate can be
read at 12,020 ft, which is outside the plot range in Exhibit 25-21. However,
similar to SUTs, TTs approach their crawl speed at 10,000 ft, namely 73 s/mi (49.3
mi/h).
Because this is the first segment, the initial truck kinematic rates τi,SUT,kin,1 and
τi,TT,kin,1 are equivalent to the free-flow rate of 55.4 s/mi. Because τi,SUT,kin,1 is less
than τf,SUT,kin,1 and τi,TT,kin,1 is less than τf,TT,kin,1, both types of trucks decelerate on
Segment 1, from 65 to 61 mi/h for SUTs and from 65 to 49.3 mi/h for TTs.
Kinematic Space-Based Rates. Because this is the first segment, the space-
based speed at 0 ft is the FFS of 65 mi/h. Therefore, the 65-mi/h curve is applied
to obtain τS,SUT,kin,1 and τS,TT,kin,1.
The time for an SUT to travel 7,920 feet starting from 65 mi/h on a 3% grade
can be read from Exhibit 25-A7 and is 87 s. The corresponding travel time for a
TT can be read from Exhibit 25-A18 and is 99 s. The space mean rate at 7,920 ft
for an SUT τS,SUT,kin,65,7920 and a TT τS,TT,kin,65,7920 starting from a FFS of 65 mi/h on a
3% grade can then be computed by Equation 25-58:
𝑇𝑆𝑈𝑇,65,7920 87
𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,65,7920 = = = 58 s/mi
𝑑1 7,920/5,280
𝑇𝑇𝑇,65,7920 99
𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,65,7920 = = = 66 s/mi
𝑑1 7,920/5,280
Auto-Only Speed for the Given Flow Rate. The auto-only space mean speed
for the given flow rate is computed with Equation 25-63.
𝑣mix
𝐶𝐴𝐹mix
𝑆𝑎𝑜 = 𝐶 𝑣mix 2
(𝐹𝐹𝑆 − 𝐷𝑎𝑜 ) (𝐶𝐴𝐹 − 𝐵𝑃𝑎𝑜 ) 𝑣mix
𝑐 mix
{ (𝐶𝑎𝑜 − 𝐵𝑃𝑎𝑜 )2 𝐶𝐴𝐹mix }
The choice of equation depends on whether demand volumes are greater
than or less than the breakpoint. An equation in Exhibit 12-6 is used to compute
the breakpoint. For an auto-only condition, the CAF defaults to 1.0.
𝐵𝑃𝑎𝑜 = [1000 + 40 × (75 − 𝐹𝐹𝑆)] × 𝐶𝐴𝐹 2
𝐵𝑃𝑎𝑜 = [1000 + 40 × (75 − 65)] × 12 = 1,400 veh/h/ln
As the demand volume of 1,500 veh/h/ln is greater than the breakpoint, the
second of the two auto-only speed equations will be used. This equation requires
knowing the auto-only capacity, which can be computed from Exhibit 12-6.

𝐶𝑎𝑜 = 2,200 + 10 × (65 − 50) = 2,350 pc/h/ln

Then
2
2,350 1,500
(65 − 45 ) (0.798 − 1,400)
𝑆𝑎𝑜 = 65 − = 61.74 mi/h
(2,350 − 1,400)2
Traffic Interaction Term. The incremental traffic interaction term is
3,600 3,600 1
𝛥𝜏𝑇𝐼 = ( − ) × (1 + 3 ( − 1)) = 5.15 s/mi
61.74 65 0.798
Actual Spot Rates. The actual spot travel time rates of SUTs and TTs at the
end of Segment 1 are computed from Equation 25-60 and Equation 25-61,
respectively.
𝜏𝑓,𝑆𝑈𝑇,1 = 𝜏𝑓,𝑆𝑈𝑇,𝑘𝑖𝑛,1 + 𝛥𝜏𝑇𝐼 = 59 + 5.15 = 64.15 s/mi
𝜏𝑓,𝑇𝑇,1 = 𝜏𝑓,𝑇𝑇,𝑘𝑖𝑛,1 + 𝛥𝜏𝑇𝐼 = 73 + 5.15 = 78.15 s/mi
The initial spot rates of SUTs and TTs in Segment 1 can also be computed
from Equation 25-60 and Equation 25-61.
𝜏𝑖,𝑆𝑈𝑇,1 = 𝜏𝑖,𝑆𝑈𝑇,𝑘𝑖𝑛,1 + 𝛥𝜏𝑇𝐼 = (3,600/65) + 5.15 = 60.5 s/mi
𝜏𝑖,𝑇𝑇,1 = 𝜏𝑖,𝑇𝑇,𝑘𝑖𝑛,1 + 𝛥𝜏𝑇𝐼 = (3,600/65) + 5.15 = 60.5 s/mi
Actual Space-Based Rates. Equation 25-60 and Equation 25-61 are also used
to calculate the actual space-based travel time rates for SUTs and TTs. The traffic
interaction term is the same as the term used for the spot rate calculations.
𝜏𝑆,𝑆𝑈𝑇,1 = 𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,1 + 𝛥𝜏𝑇𝐼 = 58 + 5.15 = 63.15 s/mi
𝜏𝑆,𝑇𝑇,1 = 𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,1 + 𝛥𝜏𝑇𝐼 = 66 + 5.15 = 71.15 s/mi
Step 5: Compute Spot and Space-Based Travel Time Rates for Autos
Equation 25-64 is used to compute the spot-based travel time rate for
automobiles on the basis of the kinematic truck spot rate at the end of the
segment.
3,600
𝜏𝑓,𝑎,1 = + 5.15
65
1,500 0.77 0.34
59 3,600 1.53
+ [64.50 × ( ) × 0.05 × max (0, − ) ]
1,000 100 65 × 100
1,500 0.81 73 3,600 1.32
+ [79.5 × ( ) × 0.100.56 × max (0, − ) ]
1,000 100 65 × 100
𝜏𝑓,𝑎,1 = 63.8 s/mi
When the initial auto spot travel time rate is computed, the trucks’ kinematic
spot rates are the same as the FFS, so the last two terms are 0. Therefore,
Equation 25-64 can also be used to compute the initial auto spot rate, with the
last two terms equal to 0.
3,600
𝜏𝑖,𝑎,1 = + 5.15 + 0 + 0
65
𝜏𝑖,𝑎,1 = 60.5 s/mi

It was determined in Step 4 that trucks decelerate in the first segment, so

Equation 25-65 is used to compute the auto space-based rate on the basis of the
kinematic truck space-based rates.
3,600
𝜏𝑆,𝑎,1 = + 5.15
65
1,500 0.46 58 3,600 2.76
+ [100.42 × ( ) × 0.050.68 × max (0, − ) ]
1,000 100 65 × 100
1,500 1.36 66 3,600 1.81
+ [110.64 × ( ) × 0.100.62 × max (0, − ) ]
1,000 100 65 × 100
𝜏𝑆,𝑎,1 = 61.3 s/mi
Step 6: Compute Mixed-Flow Space-Based Travel Time Rate and Speed

The mixed-flow travel rate τmix,1 and the mixed speed Smix,1 are computed
with Equation 25-67 and Equation 25-68, respectively.
𝜏mix,1 = 0.85 × 61.3 + 0.05 × 63.15 + 0.10 × 71.15 = 62.4 s/mi
3,600
𝑆mix,1 = = 57.7 mi/h
62.4
Segment 2
For the second segment, the initial truck kinematic spot travel time rates are
the final truck kinematic spot rates from the preceding segment. These are 59
s/mi (61.0 mi/h) for SUTs and 73 s/mi (49.3 mi/h) for TTs.

both SUTs and TTs were determined in Step 3.
In Exhibit 25-20, the initial SUT kinematic spot rate of 59 s/mi (61.0 mi/h)
occurs on the curve for a 2% upgrade, starting from 30 mi/h (120 s/mi) at
approximately 4,000 ft along the curve. After an SUT travels for 2 mi (10,560 ft),
its spot rate can be read at 14,560 ft, which is outside the plot range. However,
Exhibit 25-20 shows SUTs approach their crawl speed of 67.9 mi/h (53 s/mi) on a
2% grade. Because the specified FFS is 65 mi/h, SUTs will maintain a speed of 65
mi/h (55.4 s/mi) when the kinematic spot speeds exceed 65 mi/h. Therefore, the
SUT spot rate at the end of Segment 2, τf,SUT,kin,2, is 55.4 s/mi.
In Exhibit 25-21, the initial TT kinematic spot rate of 73 s/mi (49.3 mi/h)
approximately 3,360 ft. After a TT travels for 2 mi (10,560 ft), its spot rate can be
read at 13,920 ft, which is outside the plot range. However, Exhibit 25-21 shows
TTs reach their crawl speed of 57.1 mi/h (63 s/mi) on a 2% grade. Thus, the TT
spot rate at the end of Segment 2, τf,TT,kin,2, is 63 s/mi.
On this segment, the final SUT and TT kinematic rates are greater than the
initial rates, so both truck types accelerate on the second grade. The nomographs
for the time versus distance relationships are applicable to both cases where

trucks are decelerating, and where they are accelerating. Acceleration is evident
if the time required to cover a given distance is reducing as the distance increases.
Kinematic Space-Based Rates. The kinematic space-based speeds at 0 ft into
Segment 2 equal the final kinematic spot speeds of Segment 1.
For SUTs, the final kinematic spot speed of Segment 1 was 61.0 mi/h
(59 s/mi). As this speed is within 2.5 mi/h of 60 mi/h, Exhibit 25-A6 is used to
obtain the SUT kinematic space-based travel time rate τS,SUT,kin,2. The time for an
SUT to travel 10,000 ft starting from an FFS of 60 mi/h on a 2% grade can be read
from Exhibit 25-A6 and is 105 s.
For TTs, the final kinematic spot speed of Segment 1 was 49.3 mi/h (73 s/mi).
As this speed is within 2.5 mi/h of 50 mi/h, Exhibit 25-A15 is applied to obtain the
TT kinematic space-based rate τS,TT,kin,2. The time for a TT to travel 10,000 ft
starting from an FFS of 50 mi/h on a 2% grade can be read from Exhibit 25-A15
and is 125 s.
The space mean travel time rates for SUTs and TTs can now be computed by
Equation 25-58.
𝑇𝑆𝑈𝑇,60,10000 105
𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,60,10000 = = = 55.4 s/mi
𝑑2 10,000/5,280
𝑇𝑇𝑇,50,10000 125
𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,50,10000 = = = 66.0 s/mi
𝑑2 10,000/5,280
The SUT and TT kinematic rates at a distance of 2 mi (10,560 ft) can be
computed from Equation 25-59. The δ values for SUTs (0.0104) and TTs (0.0136)
can be read from Exhibit 25-24 and Exhibit 25-25, respectively. The rates are
computed as follows:
105 10,000
𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,60,10560 = + 0.0105 × (1 − ) × 5,280 = 55.4 s/mi
2 2 × 5,280
125 10,000
𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,60,10560 = + 0.0118 × (1 − ) × 5,280 = 65.8 s/mi
2 2 × 5,280
for the given flow rate is computed with Equation 25-63. The breakpoint of the
speed–flow curve was already determined to be 1,400 veh/h/ln, as part of the
computations for the first segment. Thus,
2
2,350 1,500
(65 − ) (0.823 − 1,400)
𝑆𝑎𝑜 = 65 − 45 = 62.46 mi/h
(2,350 − 1,400)2
computed by Equation 25-62.
3,600 3,600 1
𝛥𝜏𝑇𝐼 = ( − ) × (1 + 3 ( − 1)) = 3.71 s/mi
62.46 65 0.823
Actual Spot Rates. The actual spot rates of SUTs and TTs at the end of
Segment 2 are computed from Equation 25-60 and Equation 25-61, respectively.
𝜏𝑓,𝑆𝑈𝑇,2 = 𝜏𝑓,𝑆𝑈𝑇,𝑘𝑖𝑛,2 + 𝛥𝜏𝑇𝐼 = 55.4 + 3.71 = 59.11 s/mi

Similarly, the space-based rates are

𝜏𝑆,𝑆𝑈𝑇,2 = 𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,2 + 𝛥𝜏𝑇𝐼 = 55.4 + 3.71 = 59.11 s/mi
𝜏𝑆,𝑇𝑇,2 = 𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,2 + 𝛥𝜏𝑇𝐼 = 65.8 + 3.71 = 69.51 s/mi
automobiles.
3,600
𝜏𝑓,𝑎,2 = + 3.71
65
1,500 0.77 0.34
55.4 3,600 1.53
+ [64.50 × ( ) × 0.05 × max (0, − ) ]
1,000 100 65 × 100
1,500 0.81 66.0 3,600 1.32
+ [79.5 × ( ) × 0.100.56 × max (0, − ) ]
1,000 100 65 × 100
𝜏𝑓,𝑎,2 = 60.1 s/mi
In this case, the auto spot rate of 60.1 s/mi is higher than the SUT spot rate of
59.1 s/mi. As the auto spot rate should always be less than or equal to the truck
spot rate, the auto spot rate is set equal to 59.11 s/mi.
In Step 4, it was determined that trucks accelerate in Segment 2, so Equation
25-66 is used to compute the auto space-based rate.
3,600
𝜏𝑆,𝑎,2 = + 3.71
65
1,500 1.16 55.4 3,600 1.73
+ [54.72 × ( ) × 0.050.28 × max (0, - ) ]
1,000 100 65×100
1,500 1.32 0.61
65.8 3,600 1.33
+ [69.72 × ( ) × 0.10 × max (0, − ) ]
1,000 100 65 × 100
𝜏𝑆,𝑎,2 = 60.5 s/mi

with Equation 25-67 and Equation 25-68, respectively.
𝜏mix,2 = 0.85 × 61.4 + 0.05 × 62.01 + 0.10 × 73.51 = 62.6 s/mi
3,600
𝑆mix,2 = = 58.7 mi/h
61.3
Segment 3
The initial truck kinematic spot travel time rates for Segment 3 are the final
truck kinematic spot rates for Segment 2. These are 55.4 s/mi (65 mi/h) for SUTs
and 63.0 s/mi (57.1 mi/h) for TTs.

both SUTs and TTs were determined in Step 3.

In Exhibit 25-20, the initial SUT kinematic spot rate of 55.4 s/mi (65 mi/h)
approximately 1,500 ft along the curve. After an SUT travels 1 mi (5,280 ft), its
spot rate can be read at 6,780 ft and is approximately 75 s/mi (48 mi/h). Thus, the
SUT spot rate at the end of Segment 3 is 75 s/mi.
In Exhibit 25-21, the initial TT kinematic spot rate of 63 s/mi (57.1 mi/h)
approximately 2,050 ft along the curve. After a TT travels 1 mi (5,280 ft), its spot
rate can be read at 7,330 ft and is approximately 103 s/mi (35.0 mi/h). Thus, the
TT spot rate at the end of Segment 3 is 103 s/mi.
In Segment 3, the initial kinematic rates for both truck types are less than the
final kinematic rates. Therefore, both truck types decelerate in Segment 3.
Kinematic Space-Based Rates. The kinematic space-based speeds at 0 ft into
Segment 3 equal the final kinematic spot speeds of Segment 2.
The final kinematic spot speed of SUTs in Segment 2 was 65 mi/h (55.4 s/mi).
Exhibit 25-A7 is therefore used to obtain the SUT kinematic space-based rate
τS,SUT,kin,3. The travel time for SUTs at 5,280 ft, starting from 65 mi/h on a 5% grade,
can be read from Exhibit 25-A7 and equals 67 s.
The final kinematic spot speed of TTs in Segment 2 was 57.2 mi/h (63.0 s/mi).
As this value is within 2.5 mi/h of 55 mi/h, Exhibit 25-A16 is applied to obtain the
TT kinematic space-based rate τS,TT,kin,3. The travel time for TTs at 5,280 ft, starting
from an FFS of 55 mi/h on a 5% grade, can be read from Exhibit 25-A16 and
equals 89 s.
The space mean rate at 5,280 ft for SUTs and TTs can be computed by
Equation 25-58.
𝑇𝑆𝑈𝑇,65,5280 67
𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,65,5280 = = = 67.0 s/mi
𝑑3 5,280/5,280
𝑇𝑇𝑇,55,5280 89
𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,55,5280 = = = 89.0 s/mi
𝑑3 5,280/5,280
for the given flow rate is computed with Equation 25-63. The breakpoint of the
speed–flow curve was already determined to be 1,400 veh/h/ln as part of the
computations for the first segment. Thus
2
2,350 1,500
(65 − 45 ) (0.743 − 1,400)
𝑆𝑎𝑜 = 65 − = 59.58 mi/h
(2,350 − 1,400)2
computed by Equation 25-62.
3,600 3,600 1
𝛥𝜏𝑇𝐼 = ( − ) × (1 + 3 ( − 1)) = 10.27 s/mi
59.58 65 0.743
Actual Spot Rates. The actual spot rates of SUTs and TTs at the end of
Segment 2 are computed from Equation 25-60 and Equation 25-61, respectively.
𝜏𝑓,𝑆𝑈𝑇,3 = 𝜏𝑓,𝑆𝑈𝑇,𝑘𝑖𝑛,3 + 𝛥𝜏𝑇𝐼 = 75 + 10.27 = 85.27 s/mi


Similarly the space-based rates are:
𝜏𝑆,𝑆𝑈𝑇,3 = 𝜏𝑆,𝑆𝑈𝑇,𝑘𝑖𝑛,3 + 𝛥𝜏𝑇𝐼 = 67.0 + 10.27 = 77.27 s/mi
𝜏𝑆,𝑇𝑇,3 = 𝜏𝑆,𝑇𝑇,𝑘𝑖𝑛,3 + 𝛥𝜏𝑇𝐼 = 89.0 + 10.27 = 99.27 s/mi
automobiles.
3,600
𝜏𝑓,𝑎,3 = + 10.27
65
1,500 0.77 75 3,600 1.53
+ [64.50 × ( ) × 0.050.34 × max (0, − ) ]
1,000 100 65 × 100
1,500 0.81 103 3,600 1.32
+ [79.5 × ( ) × 0.100.56 × max (0, − ) ]
1,000 100 65 × 100
𝜏𝑓,𝑎,3 = 79.7 s/mi
In Step 4, it was determined that trucks decelerate in Segment 3, so Equation
25-65 is used to compute the auto space-based rate.
3,600
𝜏𝑆,𝑎,3 = + 10.27
65
1,500 0.46 67.0 3,600 2.76
+ [100.42 × ( ) × 0.050.68 × max (0, − ) ]
1,000 100 65 × 100
1,500 1.36 89.0 3,600 1.81
+ [110.64 × ( ) × 0.100.62 × max (0, − ) ]
1,000 100 65 × 100
𝜏𝑆,𝑎,3 = 72.1 s/mi

using Equation 25-67 and Equation 25-68, respectively.
𝜏mix,3 = 0.85 × 72.1 + 0.05 × 77.27 + 0.10 × 99.27 = 75.1 s/mi
3,600
𝑆mix,3 = = 47.9 mi/h
75.1
Step 7: Overall Results

Now that results have been developed for all three segments, the overall
performance of the composite grade can be computed. The mixed-flow travel
time for each segment is computed with Equation 25-69.
3,600𝑑1 3,600 × 1.5
𝑡mix,1 = = = 93.6 s
𝑆mix,1 57.7
3,600𝑑2 3,600 × 2
𝑡mix,2 = = = 122.7 s
𝑆mix,2 58.7
3,600𝑑3 3,600 × 1
𝑡mix,3 = = = 75.2 s
𝑆mix,3 47.9

The overall mixed-flow travel time tmix,oa is the sum of the mixed-flow travel
times for all three segments and equals 294 s. Equation 25-70 can be used to
compute the mixed-flow speed.
3,600𝑑𝑜𝑎 3600 × 4.5
𝑆mix,𝑜𝑎 = = = 55.6 mi/h
𝑡mix,𝑜𝑎 291.5
Exhibit 25-109 shows the spot speeds of all the segments in the example.
Exhibit 25-109
Example Problem 11:
Spot Speeds of All Segments
Exhibit 25-110 shows the space mean speeds of all the segments in the
example.
Exhibit 25-110
Example Problem 11: Space
Mean Speeds of All Segments
Exhibit 25-111 shows the overall space mean speeds of all the segments in the
example.
Exhibit 25-111
Example Problem 11: Overall
Space Mean Speeds of All
Segments

12. REFERENCES
1. Highway Capacity Manual. Transportation Research Board, National Research Some of these references can
be found in the Technical
Council, Washington, D.C., 2000. Reference Library in Volume 4.
2. Eads, B. S., N. M. Rouphail, A. D. May, and F. Hall. Freeway Facilities
Methodology in Highway Capacity Manual 2000. In Transportation Research
Record: Journal of the Transportation Research Board, No. 1710, Transportation
Research Board, National Research Council, Washington, D.C., 2000, pp. 171–
180.
3. Hall, F. L., L. Bloomberg, N. M. Rouphail, B. Eads, and A. D. May. Validation
Results for Four Models of Oversaturated Freeway Facilities. In Transportation
Research Record: Journal of the Transportation Research Board, No. 1710,
Transportation Research Board, National Research Council, Washington,
D.C., 2000, pp. 161–170.
4. A Policy on Geometric Design of Highways and Streets, 5th ed. American
Association of State Highway and Transportation Officials, Washington,
D.C., 2004.
5. Newell, G. F. A Simplified Theory of Kinematic Waves in Highway Traffic.
Part I: General Theory. Transportation Research, Vol. 27B, No. 4, 1993, pp. 281–
287.
Part II: Queuing at Freeway Bottlenecks. Transportation Research, Vol. 27B,
No. 4, 1993, pp. 289–303.
Part III: Multidestination Flows. Transportation Research, Vol. 27B, No. 4, 1993,
pp. 305–313.
8. Newman, L. Freeway Operations Analysis. Course Notes. University of
California Institute of Transportation Studies University Extension, Berkeley,
1986.
9. Schoen, J. M., J. A. Bonneson, C. Safi, B. Schroeder, A. Hajbabaie, C. H. Yeom,
N. Rouphail, Y. Wang, W. Zhu, and Y. Zou. Work Zone Capacity Methods for
the Highway Capacity Manual. National Cooperative Highway Research
Program Project 3-107 final report, preliminary draft. Kittelson & Associates,
Inc., Tucson, Ariz., April 2015.
10. Hajbabaie, A., N. M. Rouphail, B. J. Schroeder, and R Dowling. Planning-
Level Methodology for Freeway Facilities. In Transportation Research Record:
Journal of the Transportation Research Board, No. 2483, Transportation Research
Board of the National Academies, Washington, D.C., 2015, pp. 47–56.
11. Elefteriadou, L., A. Kondyli, and B. St. George. Estimation of Capacities on
Florida Freeways. Final Report. Gainesville, Fla., Sept. 2014.
Chapter 25/Freeway Facilities: Supplemental References

12. Dowling, R., G. F. List, B. Yang, E. Witzke, and A. Flannery. NCFRP Report
31: Incorporating Truck Analysis into the Highway Capacity Manual.
Transportation Research Board of the National Academies, Washington,
D.C., 2014.
13. Washburn, S. S., and S. Ozkul. Heavy Vehicle Effects on Florida Freeways and
Multilane Highways. Report TRC-FDOT-93817-2013. Florida Department of
Transportation, Tallahassee, Oct. 2013.
14. Ozkul, S., and S. S. Washburn. Updated Commercial Truck Speed Versus
Distance-Grade Curves for the Highway Capacity Manual. In Transportation
D.C., 2015, pp. 91–101.
15. Hajbabaie, A., B. J. Schroeder, N. M. Rouphail, and S. Aghdashi. Freeway
Facility Calibration Procedure. NCHRP 03-115 Working Paper U-8b. ITRE at
North Carolina State University, Raleigh, Aug. 2014. (Available in the
Technical Reference Library section of HCM Volume 4,
http://hcmvolume4.org)
16. Hu, J., B. Schroeder, and N. Rouphail. Rationale for Incorporating Queue
Discharge Flow into Highway Capacity Manual Procedure for Analysis of
Freeway Facilities. In Transportation Research Record: Journal of the
Transportation Research Board, No. 2286, Transportation Research Board of the
National Academies, Washington, D.C., 2012, pp. 76–83.
17. Aghdashi, S., A. Hajbabaie, B. J. Schroeder, J. L. Trask, and N. M. Rouphail.
Generating Scenarios of Freeway Reliability Analysis: Hybrid Approach. In
Transportation Research Record: Journal of the Transportation Research Board, No.
2483, Transportation Research Board of the National Academies,
Washington, D.C., 2015, pp. 148–159.
18. Zegeer, J., J. Bonneson, R. Dowling, P. Ryus, M. Vandehey, W. Kittelson, N.
Rouphail, B. Schroeder, A. Hajbabaie, B. Aghdashi, T. Chase, S. Sajjadi, R.
Margiotta, and L. Elefteriadou. Incorporating Travel Time Reliability in the
Highway Capacity Manual. SHRP 2 Report S2-L08-RW-1. Transportation
Research Board of the National Academies, Washington, D.C., 2014.
19. Federal Highway Administration. Highway Economic Requirements System—
State Version (HERS-ST). Technical Report. U.S. Department of
Transportation, Washington, D.C., 2005.
20. Highway Safety Manual, 2014 Supplement to the Highway Safety Manual, 1st
ed. American Association of State Highway and Transportation Officials,
Washington, D.C., 2014.
References Chapter 25/Freeway Facilities: Supplemental

APPENDIX A: TRUCK PERFORMANCE CURVES
This appendix provides travel time versus distance curves for SUTs for
initial speeds between 35 and 75 mi/h in 5-mi/h increments. Curves for SUTs for
30- and 70-mi/h initial speeds are presented in Section 7 as Exhibit 25-23 and
Exhibit 25-22, respectively. The appendix also provides travel time versus distance
curves for TTs for initial speeds between 20 and 75 mi/h in 5-mi/h increments.
Exhibit 25-A1
Initial Speed

Exhibit 25-A2
Initial Speed

Chapter 25/Freeway Facilities: Supplemental Appendix A: Truck Performance Curves

Exhibit 25-A3
Initial Speed

Exhibit 25-A4
Initial Speed

Appendix A: Truck Performance Curves Chapter 25/Freeway Facilities: Supplemental

Exhibit 25-A5
Initial Speed

Circles indicate where a truck reaches 60 mi/h, diamonds indicate 65 mi/h, and squares indicate 70 mi/h.
Exhibit 25-A6
Initial Speed

Diamonds indicate where a truck reaches 65 mi/h and squares indicate 70 mi/h.

Exhibit 25-A7
Initial Speed

Squares indicate where a truck reaches 70 mi/h.
Exhibit 25-A8
Initial Speed

Exhibit 25-A9
TT Travel Time Versus
Initial Speed

Exhibit 25-A10
Initial Speed


Exhibit 25-A11
Initial Speed

Exhibit 25-A12
Initial Speed


Exhibit 25-A13
Initial Speed

Exhibit 25-A14
Initial Speed


Exhibit 25-A15
Initial Speed

Exhibit 25-A16
Initial Speed


Exhibit 25-A17
Initial Speed

Exhibit 25-A18
Initial Speed


Exhibit 25-A19
Initial Speed
Exhibit 25-A20
Initial Speed

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 26
FREEWAY AND HIGHWAY SEGMENTS: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 26-1
2. STATE-SPECIFIC HEAVY-VEHICLE DEFAULT VALUES .......................... 26-2
3. TRUCK ANALYSIS USING THE MIXED-FLOW MODEL .......................... 26-4

Introduction .......................................................................................................... 26-4
Overview of the Methodology ........................................................................... 26-4
4. ADJUSTMENTS FOR DRIVER POPULATION EFFECTS ......................... 26-14
5. GUIDANCE FOR FREEWAY CAPACITY ESTIMATION .......................... 26-15

Freeway Capacity Definitions .......................................................................... 26-15
Capacity Measurement Locations ................................................................... 26-16
Capacity Estimation from Field Data .............................................................. 26-18
6. CONNECTED AND AUTOMATED VEHICLES .......................................... 26-22

Introduction ........................................................................................................ 26-22
Concepts ............................................................................................................. 26-22
Capacity Adjustment Factors ........................................................................... 26-27
Service Volume Tables ...................................................................................... 26-28
7. FREEWAY AND MULTILANE HIGHWAY EXAMPLE PROBLEMS ....... 26-30

Example Problem 1: Four-Lane Freeway LOS ............................................... 26-30
Example Problem 2: Number of Lanes Required for Target LOS ............... 26-33
Example Problem 3: Six-Lane Freeway LOS and Capacity ......................... 26-35
Example Problem 4: LOS on a Five-Lane Highway with a Two-Way
Left-Turn Lane ............................................................................................ 26-38
Example Problem 5: Mixed-Flow Freeway Operations ............................... 26-40
Example Problem 6: Severe Weather Effects on a Basic Freeway
Segment ....................................................................................................... 26-47
Example Problem 7: Basic Managed Lane Segment ..................................... 26-49
Chapter 26/Freeway and Highway Segments: Supplemental Contents

8. TWO-LANE HIGHWAY EXAMPLE PROBLEMS .........................................26-54

Example Problem 1: Level, Straight, Passing Constrained Segment ..........26-54
Example Problem 2: Passing Constrained Segment with Horizontal
Curves ..........................................................................................................26-60
Example Problem 3: Facility Analysis—Level Terrain .................................26-63
Example Problem 4: Facility Analysis – Mountain Road .............................26-85
Example Problem 5: Two-Lane Highway Bicycle LOS ................................26-98
9. REFERENCES .....................................................................................................26-100
APPENDIX A: TRUCK PERFORMANCE CURVES .......................................26-102
APPENDIX B: WORK ZONES ON TWO-LANE HIGHWAYS ....................26-107

Concepts ............................................................................................................26-107
Work Zone Capacity .......................................................................................26-107
Queuing and Delay Analysis .........................................................................26-114
Example Calculation .......................................................................................26-116
Reference ...........................................................................................................26-120
Contents Chapter 26/Freeway and Highway Segments: Supplemental

LIST OF EXHIBITS
Exhibit 26-1 State-Specific Default Values for Percentage of Heavy

Vehicles on Freeways .......................................................................................... 26-2
Exhibit 26-2 State-Specific Default Values for Percentage of Heavy
Vehicles on Multilane and Two-Lane Highways ............................................ 26-3
Exhibit 26-3 Overview of Operational Analysis Methodology for Mixed-
Flow Model........................................................................................................... 26-5
Exhibit 26-4 Speed–Flow Models for 70-mi/h Auto-Only Flow and a
Representative Mixed Flow................................................................................ 26-5
Exhibit 26-5 SUT Travel Time Versus Distance Curves for 70-mi/h FFS ............ 26-9
Exhibit 26-6 TT Travel Time Versus Distance Curves for 70-mi/h FFS ............... 26-9
Exhibit 26-7 δ Values for SUTs ................................................................................ 26-10
Exhibit 26-8 δ Values for TTs................................................................................... 26-10
Exhibit 26-9 Recommended CAF and SAF Adjustments for Driver
Population Impacts ........................................................................................... 26-14
Exhibit 26-10 Recommended Capacity Measurement Location for Merge
Bottlenecks .......................................................................................................... 26-17
Exhibit 26-11 Recommended Capacity Measurement Location for
Diverge Bottlenecks ........................................................................................... 26-17
Exhibit 26-12 Recommended Capacity Measurement Location for
Weaving Bottlenecks ......................................................................................... 26-17
Exhibit 26-13 Illustrative Example of the Capacity Estimation Procedure ....... 26-20
Exhibit 26-14 Capacity Estimation Using the 15% Acceptable Breakdown
Rate Method ....................................................................................................... 26-21
Exhibit 26-15 Capacity Adjustment Factors for CAVs for Basic Freeway
and Freeway Diverge Segments ...................................................................... 26-27
Exhibit 26-16 Capacity Adjustment Factors for CAVs for Freeway Merge
Segments ............................................................................................................. 26-28
Exhibit 26-17 Capacity Adjustment Factors for CAVs for Freeway
Weaving Segments ............................................................................................ 26-28
Exhibit 26-18 Daily Maximum Service Volumes for Basic Freeway
Segments with CAV Presence (2-way veh/day/ln) ....................................... 26-29
Exhibit 26-19 Hourly Maximum Service Volumes for Basic Freeway
Segments with CAV Presence (veh/h/ln) ....................................................... 26-29
Exhibit 26-20 List of Freeway and Multilane Highway Example Problems ..... 26-30
Exhibit 26-21 Example Problem 1: Graphical Solution ......................................... 26-32
Exhibit 26-22 List of Two-Lane Highway Example Problems ............................ 26-54
Exhibit 26-23 Example Problem 2: Horizontal Curve Inputs ............................... 26-60
Exhibit 26-24 Example Problem 2: Horizontal Curve Average Speed Results ... 26-62
Exhibit 26-25 Example Problem 2: Average Speeds by Subsegment ................... 26-62
Exhibit 26-26 Example Problem 3: Input Data ....................................................... 26-63
Chapter 26/Freeway and Highway Segments: Supplemental Contents

Exhibit 26-27 Example Problem 3: LOS Results ....................................................26-85

Exhibit 26-28 Example Problem 4: Facility Diagram ............................................26-86
Exhibit 26-29 Example Problem 4: Facility Volume and Speed Data ................26-86
Exhibit 26-30 Example Problem 4: Facility Grade and Horizontal Curve
Data ......................................................................................................................26-86
Exhibit 26-31 Example Problem 4: Vertical Alignment Classifications by
Segment ...............................................................................................................26-88
Exhibit 26-32 Example Problem 4: Free-Flow Speed Results ..............................26-89
Exhibit 26-33 Example Problem 4: Unadjusted Average Speed Results ...........26-91
Exhibit 26-34 Example Problem 4: Adjusted Average Speed Results ................26-92
Exhibit 26-35 Example Problem 4: Percent Follower and Unadjusted
Follower Density Results ..................................................................................26-95
Exhibit 26-36 Example Problem 4: LOS Results ....................................................26-97
Exhibit 26-A1 SUT Travel Time Versus Distance Curves for 50-mi/h FFS .....26-102
Exhibit 26-A6 TT Travel Time Versus Distance Curves for 50-mi/h FFS ........26-104
Exhibit 26-A10 TT Travel Time Versus Distance Curves for 75-mi/h FFS ......26-106
Exhibit 26-B1 Traffic Control for a Two-Lane Highway Work Zone
Involving a Lane Closure ................................................................................26-108
Exhibit 26-B2 Two-Lane Highway Work Zone Grade Adjustment Factor
(fg) for Level Terrain, Rolling Terrain, and Specific Downgrades .............26-109
Exhibit 26-B3 Two-Lane Highway Work Zone Grade Adjustment Factor
(fg) for Specific Upgrades ................................................................................26-110
Exhibit 26-B4 Two-Lane Highway Work Zone Passenger Car Equivalents
for Trucks (ET) and RVs (ER) for Level Terrain, Rolling Terrain, and
Specific Downgrades .......................................................................................26-111
for Trucks (ET) on Specific Upgrades ............................................................26-112
for RVs (ER) on Specific Upgrades .................................................................26-112
Exhibit 26-B7 Directional Queueing Diagram for a Two-Lane Highway
Lane-Closure Work Zone ...............................................................................26-115
Exhibit 26-B8 Example Calculation: Work Zone Roadway Parameters ..........26-117
Exhibit 26-B9 Example Calculation: Work Zone Traffic Parameters ...............26-117
Contents Chapter 26/Freeway and Highway Segments: Supplemental

1. INTRODUCTION
Chapter 26 is the supplemental chapter for Chapter 12, Basic Freeway and VOLUME 4: APPLICATIONS
GUIDE
Multilane Highway Segments, and Chapter 15, Two-Lane Highways, which are 25. Freeway Facilities:
found in Volume 2 of the Highway Capacity Manual (HCM). Supplemental
26. Freeway and Highway
Section 2 provides state-specific heavy-vehicle default values that can be Segments:
Supplemental
applied to freeway, multilane highway, and two-lane highway analysis. 27. Freeway Weaving:
Supplemental
Section 3 presents a supplemental procedure for basic freeway segments that 28. Freeway Merges and
can be used to assess their operating performance under mixed-flow conditions Diverges: Supplemental
29. Urban Street Facilities:
when significant truck presence, a prolonged single upgrade, or both exist. Supplemental
Appendix A provides travel time versus distance curves for single-unit trucks 30. Urban Street Segments:
Supplemental
(SUTs) and tractor-trailers (TTs) for a range of free-flow speeds (FFS) for use with 31. Signalized Intersections:
this procedure. Chapter 25, Freeway Facilities: Supplemental, presents an Supplemental
32. STOP-Controlled
extension of this method for composite grades on freeway facilities. Intersections:
Supplemental
Section 4 provides suggested capacity and FFS adjustments to account for the 33. Roundabouts:
effects of different proportions of motorists on a freeway or multilane highway Supplemental
who are not regular users of the facility. Terminals: Supplemental
Section 5 presents freeway capacity definitions, guidance on locating sensors
Supplemental
for use in measuring freeway capacity, and guidance on estimating capacity from 36. Concepts: Supplemental
the collected sensor data.
38. Network Analysis
Section 6 provides guidance on incorporating the potential freeway capacity
benefits of connected and automated vehicles into an HCM analysis.
Section 7 provides seven example problems demonstrating the basic freeway
and multilane highway segment procedure presented in Chapter 12.
Section 8 provides five example problems demonstrating the motorized
vehicle and bicycle methodologies for two-lane highways presented in Chapter 15.
Appendix B describes a methodology for calculating capacity and related
performance measures for work zones along two-lane highways that involve the
closure of a single lane.
Chapter 26/Freeway and Highway Segments: Supplemental Introduction

2. STATE-SPECIFIC HEAVY-VEHICLE DEFAULT VALUES
Research into the percentage of heavy vehicles on uninterrupted-flow

facilities (1) found such a wide range of average values from state to state that
not even regional default values could be developed. Exhibit 26-1 presents
default values for the percentage of heavy vehicles on freeways by state and area
population based on data from the 2004 Highway Performance Monitoring
System. Exhibit 26-2 presents similar default values for multilane and two-lane
highways. In cases in which states or local jurisdictions have developed their
own default values, those values should be used in lieu of the values presented
here. Analysts may also wish to develop their own default values based on local
or more recent data.
Exhibit 26-1 Small Medium Large Small Medium Large

State-Specific Default Values State Rural Urban Urban Urban State Rural Urban Urban Urban
for Percentage of Heavy AL* 14a 7 7 7a MT 22c 16c 12c NA
Vehicles on Freeways AK 4 5b 5 3b NC* 19b 12b 12 10a
AR 30 24 13 14 ND 21c 22c 10c NA
AZ 21 19 18 11 NE 36 37 11 8
CA 16 10 7 6 NH 15b 12b 6b 7b
CO 12 10 8 7 NJ 8 6 6 9
CT 13 6 6 5 NM 26 12 21 12
DC NA NA NA 4b NV 34b 26 18b 11b
DE ⎯ ⎯ 9b 8b NY 18 11 11 7
FL* 11 7 12 6 OH 24 13 10 8
GA* 19b 7b 12 8b OK 28 27 12 10
HI 5 19b 2 3 OR 26 19 10 7
IA 20c 24c 11c 10c PA 16 13 9 8
ID 29c 28b 12b 7b PR* 6 7b 7 4b
IL 21 23 16 9 RI 3 ⎯ NA 4
IN 26 25 23 14 SC* 19b 7b 7 8b
KS 21c 17c 8c 9b SD 20c 14c 9c NA
KY* 20a 16 12 10a TN* 19 12 12 8
LA* 12c 7b 12 10c TX 16 28c 8 5
MA 7a 5 4a 4 UT 34c ⎯ 18 13
MD 18 14 17 8 VA* 9 7 7 4
ME 5 5 5 NA VT 15 12 6 NA
MI 18 12 13 8 WA 11 10 7 6
MN 11 10 6 4 WI 6 6 6 6
MO 29b 23b 13b 10b WV 16b 13b 9b NA
MS* 9b 7b 7 6b WY 33c 36a 28c,d NA
Source: Zegeer et al. (1 ).
Notes: Populations are as follows: rural: <5,000; small urban: 5,000–50,000; medium urban: 50,000–250,000;
large urban: >250,000.
Values shown represent mean values for the state for each population type except as otherwise noted.
NA = population group does not exist within the state; ⎯ = data not available.
* Because of limited data, small urban values were combined for two groups of states: AL, MS, PR, SC, and
VA and FL, GA, KY, LA, NC, and TN. Medium urban values were combined for AL, FL, and VA.
a
Reported values appeared to be a mix of field observations and statewide values. The latter were
discounted, such that the averages shown are based primarily on values deemed to be field observations,
with some consideration given to nearby states and the value state personnel thought was statewide.
b
The default value was estimated from field observations from nearby states because of insufficient field
data, a lack of data for this road type, or too-heavy reliance on statewide values.
c
The peak period percentage is identical to the daily average percentage for nearly all observations in the
2004 Highway Performance Monitoring System data set. Default values were estimated primarily from the
daily average value but took into account the results from nearby states, particularly the difference
between peak and daily values in those states.
d
This distribution was bimodal, with one group centered on 19% and the other on 44%.
State-Specific Heavy-Vehicle Default Values Chapter 26/Freeway and Highway Segments: Supplemental
Two-Lane Multilane Two-Lane Multilane Exhibit 26-2

Highways Highways Highways Highways State-Specific Default Values
Small Small Small Small for Percentage of Heavy
State Rural Urban Rural Urban State Rural Urban Rural Urban Vehicles on Multilane and
AL 6a 6a 4a 6a MT 10c 4c 6c 3c Two-Lane Highways
AK 10 2 6 3 NC 8b 4b 6b 6b
AR 14 7 11 12 ND 14c 3c 12c 7c
AZ 9 11 9 9 NE 10 3 12 5
CA 9 5 9 6 NH 6b 6a 6b 6b
CO 11 4 5 5 NJ 8 7 8 6b
CT 3 3 2 6b NM 17 7 23 12
DC NA NA NA NA NV 17b 5c 10c 6c
DE 7 6 9 8 NY 8 5 8 5
FL 8 4 7 7 OH 11 4 14 9
GA 8b 5b 6b 6b OK 14a 5 17 11
HI 3 3 2 2 OR 12 5 6 9
IA 4c 5c 5c 4c PA 6 3 5 4
ID 12c 7c 16c 9c PR 5 5b 5 6
IL 8 5 8 6 RI 2 1 2 6b
IN 10 6a 12 10 SC 8b 5b 6b 6b
KS 15a 3 12c 6c SD 13c 4c 12 c
7c
KY 16a 6a 9a 6a TN 5 4a 6 4
LA 16c 10c 6b 16 TX 13 9 12 9
MA 3a 3a 7b 6b UT 20c 9c 22c 14c
MD 10 6 12 8 VA 4 2 5 2
ME 5 3 4 3 VT 8 5a 7 6b
MI 9 7a 8 4 WA 15 8a 10 7
MN 9 8a 8 6 WI 4 5a 4 5a
MO 9c 6c 12b 10c WV 6b 6b 5b 6b
MS 14a 5a 6b 6a WY 15c 6c 10c 9c
Notes: Populations are as follows: rural: <5,000; small urban: 5,000–50,000.
Values shown represent mean values for the state for each population type except as otherwise noted.
NA = population group does not exist within the state.
a
Reported values appeared to be a mix of field observations and statewide values. The latter were
discounted, such that the averages shown are based primarily on values deemed to be field observations,
with some consideration given to nearby states and the value state personnel thought was statewide.
b
Either there are insufficient field data, such that regional averages were used, or there are no usable field
data, either because there are no data in the state for this road type or because there is a too-heavy
reliance on statewide values for both the peak period and the daily average. In these cases, the default
value was estimated from field observations for nearby states.
c
The peak period percentage is identical to the daily average percentage for all or almost all observations
in the 2004 Highway Performance Monitoring System data set for this cell. Default values were estimated
primarily from the daily average value for this cell, taking into account the results for other similar states in
the same region, and in particular the difference between peak and daily average values in those states.
Chapter 26/Freeway and Highway Segments: Supplemental State-Specific Heavy-Vehicle Default Values
3. TRUCK ANALYSIS USING THE MIXED-FLOW MODEL
INTRODUCTION
This section presents a supplemental procedure that can be used to assess the
operating performance of freeway segments under mixed-flow conditions when
significant truck presence, a prolonged single upgrade, or both exist. This
procedure must be used if the analyst is interested in estimating space mean
speeds and densities for cars and trucks separately or for the mixed-traffic stream.
Chapter 12, Basic Freeway and Multilane Highway Segments, describes a
methodology drawn from this procedure that can be used to assess a segment’s
level of service (LOS) by converting heavy vehicles into passenger cars by using
passenger car equivalent (PCE) values. However, users are cautioned that the
auto-only speeds and densities estimated by the PCE-based procedure are likely
to be an approximation of reality at high truck percentages and on steep
upgrades. For these situations, the mixed-flow model described here is
recommended.
Analysts can also use the mixed-flow model for analyzing downgrades and
both types of general terrain (level and rolling). When the truck percentage is
low or the upgrade is not steep, both the mixed-flow model and the Chapter 12
PCE-based method provide similar results. Chapter 25, Freeway Facilities:
Supplemental, extends the mixed-flow model to freeway facilities with multiple,
composite grades. National research (2) shows that when the truck presence is
low or the upgrade is not steep, both the mixed-flow model and the procedure
applying PCE values provide similar results.
OVERVIEW OF THE METHODOLOGY

The process flow for applying the mixed-flow model is depicted in Exhibit
26-3. Selected parameters referenced in the methodology are indicated in Exhibit
26-4 for a 70-mi/h auto-only traffic stream and a representative mixed-traffic
stream.
Truck Analysis Using the Mixed-Flow Model Chapter 26/Freeway and Highway Segments: Supplemental
Exhibit 26-3
Overview of Operational
Analysis Methodology for
Mixed-Flow Model
Notes: SUT = single-unit truck; TT = tractor-trailer; FFS = free-flow speed; MFM = mixed-flow model.
Exhibit 26-4
Speed–Flow Models for 70-
mi/h Auto-Only Flow and a
Representative Mixed Flow
Notes: BP = breakpoint; FFS = free-flow speed; c = capacity.
Chapter 26/Freeway and Highway Segments: Supplemental Truck Analysis Using the Mixed-Flow Model
Step 1: Input Data

For a typical operational analysis, the analyst must specify the flow rate of
the mixed-traffic stream vmix, grade g, grade length d, SUT percentage PSUT, and
TT percentage PTT for the traffic stream.
Step 2: Compute Mixed-Flow Capacity Adjustment Factor and Capacity

The capacity adjustment factor (CAF) for mixed-flow CAFmix converts auto-
only capacities into mixed-traffic capacities. It is computed with Equation 26-1.
Equation 26-1 𝐶𝐴𝐹mix = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix
where
(decimal);
CAFao = capacity adjustment factor for the auto-only case (decimal);
CAFT,mix = capacity adjustment factor for percentage of trucks for the mixed-
flow case (decimal); and
CAFg,mix = capacity adjustment factor for grade for the mixed-flow case
(decimal).
CAF for the Auto-Only Case

Because CAFao is used to convert auto-only capacities into mixed-traffic
capacities, it defaults to a value of 1.0 unless other capacity adjustments are in
effect (e.g., weather, incidents, driver population factor).

The CAF for truck percentage CAFT,mix is computed with Equation 26-2.
Equation 26-2 𝐶𝐴𝐹𝑇,mix = 0.53 × 𝑃𝑇 0.72
where PT is the total percentage of SUTs and TTs in the traffic stream (decimal).

The CAF for grade effect CAFg,mix accounts for the grade severity, grade length,
and truck presence. It is computed by using Equation 26-3 with Equation 26-3.
Equation 26-3 𝐶𝐴𝐹𝑔,𝑚𝑖𝑥 = 𝜌𝑔,mix × max[0, 0.69 × (𝑒 12.9𝑔 − 1)]
× max[0, 1.72 × (1 − 1.71𝑒 −3.16𝑑 )]
with
8 × 𝑃𝑇 𝑃𝑇 < 0.01
Equation 26-4 𝜌𝑔,mix = {
0.126 − 0.03𝑃𝑇 otherwise
where
ρg,mix = coefficient for grade term in the mixed-flow CAF equation (decimal),
PT = total truck percentage (decimal),
g = grade (decimal), and
d = grade length (mi).
Once CAFmix is computed, the mixed-flow capacity can be computed with

Equation 26-5.
𝐶mix = 𝐶𝑎𝑜 × 𝐶𝐴𝐹mix Equation 26-5
where
Cmix = mixed-flow capacity (veh/h/ln);
Cao = auto-only capacity for the given FFS, from Exhibit 12-6 (pc/h/ln); and
(decimal).
If the input flow rate of the mixed-traffic stream vmix exceeds the mixed-flow
capacity computed in Equation 26-5, then LOS F prevails, and the segment
procedure stops. A facility analysis is recommended under these conditions.
Step 3: Compute Mixed-Flow FFS and FFS Adjustment Factor

Equation 26-6 through Equation 26-8 compute the free-flow travel rates (in
seconds per mile) for SUTs, TTs, and automobiles, respectively, for a specific
segment with a steep grade, high truck percentage, or both. For the purposes of
calculating the automobile free-flow travel rate, the flow rate of the mixed-traffic
stream vmix is assumed to be 1 veh/h/ln when Equation 26-8 is used.
𝜏𝑆𝑈𝑇 = 𝜏𝑆𝑈𝑇,𝑘𝑖𝑛 + 𝛥𝜏 𝑇𝐼 Equation 26-6
𝜏𝑇𝑇 = 𝜏𝑇𝑇,𝑘𝑖𝑛 + 𝛥𝜏𝑇𝐼 Equation 26-7
3,600 Equation 26-8

𝜏𝑎 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆
2.76
𝑣mix 0.46 𝜏𝑆𝑈𝑇,𝑘𝑖𝑛 3,600
+100.42 × ( ) × 𝑃𝑆𝑈𝑇 0.68 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
1.81
𝑣mix 1.36 𝜏𝑇𝑇,𝑘𝑖𝑛 3,600
+110.64 × ( ) × 𝑃𝑇𝑇 0.62 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
where
τa = automobile free-flow travel rate (s/mi),
τSUT = SUT free-flow travel rate (s/mi),
τTT = TT free-flow travel rate (s/mi),
τSUT,kin = kinematic travel rate of SUTs (s/mi),
τTT,kin = kinematic travel rate of TTs (s/mi),
PSUT = SUT percentage (decimal),
PTT = TT percentage (decimal), and
3,600 = number of seconds in 1 h.
Traffic Interaction Term

The traffic interaction term computed by Equation 26-9 is the contribution of
traffic interactions to mixed-flow FFS. For the purposes of calculating the
automobile free-flow travel rate, the traffic interaction term ΔτTI is set to 0 when
3,600 3,600 1
Equation 26-9 𝛥𝜏𝑇𝐼 = ( − ) × (1 + 3 [ − 1])
𝑆𝑎𝑜 𝐹𝐹𝑆 𝐶𝐴𝐹mix
where
ΔτTI = traffic interaction term (s/mi);
Sao = auto-only speed for the given flow rate, from Equation 26-10 (mi/h);
FFS = base free-flow speed of the basic freeway segment (mi/h); and
from Equation 26-1 (decimal).
Auto-Only Speed for the Given Flow Rate

The auto-only travel rate for the given flow rate is computed with Equation
26-10.
𝑣mix
𝐶𝐴𝐹mix
2
Equation 26-10 𝑆𝑎𝑜 = 𝑐 𝑣
(𝐹𝐹𝑆 − ) ( mix − 𝐵𝑃𝑎𝑜) 𝑣mix
𝐷𝑐 𝐶𝐴𝐹mix
{ (𝑐 − 𝐵𝑃𝑎𝑜)2 𝐶𝐴𝐹mix
where
FFS = base free-flow speed of the basic freeway segment (mi/h);
c = base segment capacity, from Exhibit 12-6 (pc/h/ln);
BPao = breakpoint for the auto-only flow condition, from Exhibit 12-6
(pc/h/ln);
Dc = density at capacity = 45 pc/mi/ln; and
CAFmix = mixed-flow capacity adjustment factor for the basic freeway
segment, from Equation 26-1 (decimal).
Kinematic Travel Rates for SUTs and TTs

The kinematic travel rates for SUTs and TTs are obtained from truck travel
time versus distance performance curves on the basis of the truck weight-to-
horsepower ratio, grade, and grade length. Exhibit 26-5 shows truck travel time
versus distance curves for a representative SUT starting from a speed of 70 mi/h.
Alternate representations of how the propulsive and resistive forces vary with
speed can produce slightly different results (e.g., 3, 4).
Exhibit 26-6 shows the corresponding curves for TTs for a base FFS of 70
mi/h. These curves can be used when the base FFS is within 2.5 mi/h of 70 mi/h.
Appendix A provides additional curves for SUTs and TTs for FFS values of 50,
55, 60, 65, and 75 mi/h.
On downgrades, trucks are able to maintain their FFS, and their kinematic
performance is the same as passenger cars. The analyst could use the Chapter 12
PCE-based method instead of the mixed-flow model in those cases.
Exhibit 26-5
FFS
Exhibit 26-6
FFS
The x-axis in Exhibit 26-5 and Exhibit 26-6 represents the distance d traveled
by the truck, and the y-axis represents the travel time T to cover the grade length
d. Different curves provide the travel times for different upgrades. The kinematic
space mean travel rate can be computed with Equation 26-11.
𝜏𝑘𝑖𝑛 = 𝑇/𝑑 Equation 26-11
where
T = travel time (s), and
The maximum grade length shown in Exhibit 26-5 and Exhibit 26-6 is 10,000
ft. When the grade is longer than 10,000 ft, the kinematic travel rate can be
𝑇10000 10,000
Equation 26-12 𝜏𝑘𝑖𝑛 = + 𝛿 (1 − ) × 5,280
𝑑 5,280𝑑
where
T10000 = travel time at 10,000 ft (s),
δ = slope of the travel time versus distance curve (s/ft),
d = grade length (mi), and
5,280 = number of feet in 1 mi.
The δ value for SUTs and TTs is shown in Exhibit 26-7 and Exhibit 26-8,
respectively, for different combinations of grade and FFS.

δ Values for SUTs Grade 50 55 60 65 70 75
–5% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
0% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
2% 0.0136 0.0124 0.0114 0.0105 0.0100 0.0099
3% 0.0136 0.0124 0.0114 0.0113 0.0112 0.0112
4% 0.0136 0.0129 0.0128 0.0128 0.0128 0.0127
5% 0.0146 0.0146 0.0146 0.0146 0.0145 0.0145
6% 0.0165 0.0165 0.0165 0.0165 0.0165 0.0165
7% 0.0186 0.0186 0.0186 0.0186 0.0186 0.0186
8% 0.0208 0.0208 0.0208 0.0208 0.0208 0.0208

δ Values for TTs Grade 50 55 60 65 70 75
–5% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
0% 0.0136 0.0124 0.0114 0.0105 0.0097 0.0091
2% 0.0136 0.0124 0.0119 0.0118 0.0116 0.0115
3% 0.0143 0.0143 0.0142 0.0141 0.0140 0.0138
4% 0.0171 0.0171 0.0171 0.0170 0.0169 0.0168
5% 0.0202 0.0202 0.0202 0.0202 0.0202 0.0202
6% 0.0236 0.0236 0.0236 0.0236 0.0236 0.0236
7% 0.0272 0.0272 0.0272 0.0272 0.0272 0.0272
8% 0.0310 0.0310 0.0310 0.0310 0.0310 0.0310
Once τSUT,kin and τTT,kin are obtained, Equation 26-6 and Equation 26-7 can be
used to add the traffic interaction term to obtain the truck free-flow travel rates
τSUT and τTT. Equation 26-8 can then be used to compute the automobile free-flow
travel rate τa. Again, the mixed-flow rate vmix is assumed to be 1 veh/h/ln when
Equation 26-8 is used to estimate the automobile free-flow travel rate.
Mixed-Flow FFS
Equation 26-13 converts individual free-flow travel rates by mode into a
mixed-flow free-flow travel rate, and Equation 26-14 then converts the mixed-
flow free-flow travel rate into a mixed-flow FFS.
𝜏 = 𝑃𝑎 𝜏𝑎 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇 + 𝑃𝑇𝑇 𝜏𝑇𝑇 Equation 26-13
3,600 3,600
𝐹𝐹𝑆mix = = Equation 26-14
𝜏 𝑃𝑎 𝜏𝑎 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇 + 𝑃𝑇𝑇 𝜏𝑇𝑇
where
τ = mixed-flow free-flow travel rate (s/mi),
τa = automobile free-flow travel rate (s/mi),
τSUT = SUT free-flow travel rate (s/mi),
τTT = TT free-flow travel rate (s/mi),
Pa = automobile percentage (decimal),
PSUT = SUT percentage (decimal),
PTT = TT percentage (decimal), and
FFSmix = mixed-flow free-flow speed (mi/h).
FFS Adjustment Factor

The segment’s speed adjustment factor (SAF) is estimated with Equation
26-15.
𝑆𝐴𝐹mix = 𝐹𝐹𝑆mix /𝐹𝐹𝑆 Equation 26-15
where
SAFmix = mixed-flow speed adjustment factor for the basic freeway segment
(decimal),
FFSmix = mixed-flow free-flow speed (mi/h), and
FFS = base free-flow speed of the basic freeway segment (mi/h).
Step 4: Compute the Speed–Flow Relationship Breakpoint for the

Mixed-Flow Model
The breakpoint is the maximum flow rate up to which speed is maintained at
the adjusted FFS level. It is computed by Equation 26-16 and is depicted in
Exhibit 26-4.
𝐵𝑃mix = max[0, 𝐵𝑃𝑎𝑜 (1 − 0.4𝑃𝑇 0.1 × max[0, 𝑒 30𝑔 + 1] × 𝑑 0.01 )] Equation 26-16
where
BPmix = breakpoint for mixed flow (veh/h/ln);
BPao = breakpoint for the auto-only flow condition, from Exhibit 12-6 (pc/h/ln);
PT = total truck percentage (decimal);
g = grade (decimal); and
Step 5: Compute Mixed-Flow Speeds at Capacity and 90% of Capacity

To determine the mixed-flow speeds for the given mixed-flow rate, mixed-
flow speeds at capacity and 90% of capacity are computed for calibration
purposes. This computation, in turn, requires applying Equation 26-6 through
Equation 26-8 to calculate individual speeds for SUTs, TTs, and automobiles,
respectively. The equations are applied twice, first applying the value of Cmix as
vmix to calculate speed at capacity, and then applying the value of 0.9Cmix as vmix to
calculate speed at 90% of capacity.
The resulting modal travel time rates are converted to modal speeds Sm by
3,600
Equation 26-17 𝑆𝑚 =
𝜏𝑚
where Sm is the speed (mi/h) for mode m (SUT, TT, or automobile), and τm is the
travel time rate (s/mi) for mode m.
Next, densities for individual modes are computed with Equation 26-18.
Equation 26-18 𝐷𝑚 = 𝑣𝑚 /𝑆𝑚
where Dm is the density (SUT/mi, TT/mi, or pc/mi, depending on the mode) for
mode m, vm is the flow rate (SUT/h, TT/h, or pc/h) for mode m, and Sm is the
speed (mi/h) for mode m.
Finally, the mixed-flow speed used for calibration Scalib is calculated with
Equation 26-19.
3,600
Equation 26-19 𝑆𝑐𝑎𝑙𝑖𝑏 =
𝑃𝑎 𝜏𝑎 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇 + 𝑃𝑇𝑇 𝜏𝑇𝑇
Equation 26-19 is applied twice (i.e., two calibration points are needed), once
using τ values at capacity and again using τ values for 90% of capacity.
Mixed-flow travel rates and mixed-flow speeds are calculated with
Equations 26-13 and 26-14 twice (i.e., two calibration points are needed), once at
capacity and once at 90% capacity.
Step 6: Compute the Exponent for the Mixed-Flow Model Speed–Flow

Curve
The exponent for the speed–flow curve, which describes the rate at which
speed drops as the flow rate increases in the nonlinear portion of the mixed-flow
speed–flow curve (see Exhibit 26-4), is computed with Equation 26-20.
𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,90𝑐𝑎𝑝
ln ( )
𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝
Equation 26-20 𝜙mix = 1.195 ×
0.9𝐶 − 𝐵𝑃
ln ( 𝐶 mix− 𝐵𝑃 mix )
mix mix
where
φmix = exponent for the speed–flow curve (decimal),
FFSmix = mixed-flow free-flow speed (mi/h),
Scalib,90cap = mixed-flow speed at 90% of capacity (mi/h),
Scalib,cap = mixed-flow speed at capacity (mi/h),

Cmix = mixed-flow capacity (veh/h/ln), and
BPmix = breakpoint for mixed flow (veh/h/ln).
Step 7: Compute the Mixed-Flow Speed Under Mixed-Flow Conditions

The mixed-flow speed for mixed-flow conditions is computed by using the
generic form of the basic freeway segment speed–flow model, as shown in
Equation 26-21.
𝐹𝐹𝑆mix 𝑣mix ≤ 𝐵𝑃mix
𝑆mix = Equation 26-21
𝑣mix − 𝐵𝑃mix 𝜙mix
𝐹𝐹𝑆mix − (𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝 ) ( ) 𝑣mix > 𝐵𝑃mix
{ 𝐶mix − 𝐵𝑃mix
where
Smix = mixed-flow speed (mi/h),
FFSmix = mixed-flow free-flow speed (mi/h),
Scalib,cap = mixed-flow speed at capacity (mi/h),
BPmix = breakpoint for mixed flow (veh/h/ln),
Cmix = mixed-flow capacity (veh/h/ln), and
φmix = exponent for the speed–flow curve (decimal).
Step 8: Compute the Mixed-Flow Density Under Mixed-Flow Conditions

The mixed-flow density is computed by Equation 26-22.
𝐷mix = 𝑣mix /𝑆mix Equation 26-22
where
Dmix = mixed-flow density (veh/mi/ln),
vmix = flow rate of mixed traffic (veh/h/ln), and
Smix = mixed-flow speed (mi/h).
4. ADJUSTMENTS FOR DRIVER POPULATION EFFECTS
The base traffic stream characteristics for basic freeway and multilane
highway segments are representative of traffic streams composed primarily of
commuters or drivers who are familiar with the facility. It is generally accepted
that traffic streams with different characteristics (e.g., recreational trips) use
freeways less efficiently. Although data are sparse and reported results vary
substantially, significantly lower capacities have been reported on weekends,
particularly in recreational areas. Thus, it may generally be assumed the
reduction in capacity extends to service flow rates and service volumes for other
levels of service as well. In addition, it is expected that a reduction in FFS would
be observed when large numbers of unfamiliar drivers are present in a freeway
or multilane highway traffic stream.
The driver population adjustment factor fp has previously been used in the
HCM to reflect the effects of unfamiliar drivers in the traffic stream; it was
applied as an increase in demand volume. The values of fp ranged from 0.85 to
1.00 in most cases, although lower values have been observed in isolated cases.
The HCM recommended the analyst use a value of 1.00 for this factor (reflecting
a traffic stream composed of commuters or other regular drivers), unless there
was sufficient evidence that a lower value should be used. When greater
accuracy was needed, comparative field studies of commuter and noncommuter
traffic flow and speeds were recommended.
With the addition of a unified speed–flow equation in Chapter 12, Basic
Freeway and Multilane Highway Segments, and the ability to adjust both the
base FFS and capacity in all freeway segment chapters (Chapters 12, 13, and 14)
to account for incidents and weather events, the driver population factor is no
longer used. Instead, FFS and capacity adjustment factors SAFpop and CAFpop are
applied in combination with other applicable SAFs and CAFs.
In the absence of new research on driver population effects, recommended
values of SAFpop and CAFpop have been developed that produce similar density
results as those predicted using the former driver population factor approach.
This conversion was performed by using the unified equation of Chapter 12 and
therefore represents a slight approximation in the cases of weaving, merge, and
diverge segments.
Judgment is still required when the analyst applies these adjustments and, in
the absence of information to the contrary, the default value for SAFpop and CAFpop
is always 1.0. Should the analyst expect a significant presence of unfamiliar
drivers, the values shown in Exhibit 26-9 can serve as a guide for the analysis.
Exhibit 26-9 Level of Driver Familiarity CAFpop SAFpop

Recommended CAF and SAF All familiar drivers, regular commuters 1.000 1.000
Adjustments for Driver Mostly familiar drivers 0.968 0.975
Population Impacts Balanced mix of familiar and unfamiliar drivers 0.939 0.950
Mostly unfamiliar drivers 0.898 0.913
All or overwhelmingly unfamiliar drivers 0.852 0.863
Adjustments for Driver Population Effects Chapter 26/Freeway and Highway Segments: Supplemental
5. GUIDANCE FOR FREEWAY CAPACITY ESTIMATION
This section presents guidance for field measuring and estimating freeway
capacity. The section is organized as follows: overall definitions of freeway
capacity, guidance for field data collection using sensors, and guidance for
estimating capacity from the collected data.
FREEWAY CAPACITY DEFINITIONS

Freeway segment capacity is commonly understood to be a maximum flow
rate that is associated with the occurrence of some type of breakdown that in
turn results in lower speeds and higher densities after the breakdown event.
When oversaturation begins, queues develop and vehicles discharge from the
bottleneck at a queue discharge rate that is usually lower than the throughput
rate before the breakdown. This lower discharge rate after a breakdown is also
known as the capacity drop phenomenon. Several key terms related to freeway
capacity are defined below as they apply to this chapter.
Freeway Breakdown
A flow breakdown on a freeway represents the transition from uncongested
to congested conditions, as evidenced by the formation of queues upstream of
the bottleneck and reduced prevailing speeds.
In the HCM freeway methodology, the breakdown event on a freeway
bottleneck is defined as a sudden drop in speed at least 25% below the FFS for a
sustained period of at least 15 min that results in queuing upstream of the
bottleneck.
Recovery
A freeway segment is considered to have recovered from the breakdown
event and the resulting oversaturated conditions when the average speed (or
density) reaches prebreakdown conditions for a minimum duration of 15 min.
The definition of recovery is therefore the inverse of the definition of breakdown,
requiring a recovery to be near prebreakdown conditions (operations above the
speed threshold) for at least 15 min.
The HCM defines the breakdown recovery on a freeway bottleneck as a
return of the prevailing speed to within 10% of the FFS for a sustained period of
at least 15 min, without the presence of queuing upstream of the bottleneck.
Prebreakdown Flow Rate

The prebreakdown flow rate is the flow rate that immediately precedes the
occurrence of a breakdown event. The literature suggests this flow rate does not
have a fixed value, as evidence shows breakdowns are stochastic in nature and
can occur following a range of flow rates. The prebreakdown flow rate is
typically expressed in units of passenger cars per hour per lane. To achieve a
uniform expression, trucks and other heavy vehicles are converted into an
equivalent passenger car traffic stream.
Chapter 26/Freeway and Highway Segments: Supplemental Guidance for Freeway Capacity Estimation
In the HCM, the prebreakdown flow rate is defined as the 15-min average
flow rate that occurs immediately prior to the breakdown event. For the
purposes of this chapter, the prebreakdown flow rate is equivalent to the
segment capacity.
Postbreakdown Flow Rate or Queue Discharge Flow Rate

The postbreakdown flow rate is also referred to as the queue discharge flow
rate or the average discharge flow rate. This flow rate is usually lower than the
prebreakdown flow rate, resulting in a significant loss of freeway throughput
during congestion. Cases in which the postbreakdown flow rate exceeds the
prebreakdown flow rate have been observed, mostly when the prebreakdown
flow rate is low. Studies (5) have indicated the average difference between
postbreakdown and prebreakdown flow rates varies widely, from as little as 2%
to as much as 20%. In the absence of local information, a default value of 7% is
recommended.
In the HCM, the queue discharge flow rate is defined as the average flow
rate during oversaturated conditions (i.e., during the time interval after
breakdown and prior to recovery).
CAPACITY MEASUREMENT LOCATIONS

Research at freeway merging segments (6) has found a breakdown may first
be observed either upstream or downstream of the actual bottleneck. Some
research has indicated a breakdown may first be observed upstream of the
bottleneck, slowly spreading downstream as vehicles accelerate past the start of
the bottleneck. Other research has found the breakdown initially occurs
downstream of the merge point and then moves upstream as a shock wave.
To identify the breakdown event from field data, the following process
should be followed:
• Data are obtained at three sensors: (a) a bottleneck location (e.g., just
downstream of the end of the acceleration lane), (b) at a nearby sensor
location downstream of the bottleneck, and (c) at a nearby sensor location
upstream of the bottleneck.
• Upstream and downstream sensors should be within 0.5 mi of the
bottleneck, and the freeway ideally should have no entry or exit points
between the three sensors (other than, for example, a bottleneck on-ramp).
• The bottleneck detector should be upstream of the beginning of the
deceleration lane or downstream of the end of the acceleration lane to
avoid missing flow in those lanes.
• The analyst evaluates data from the bottleneck sensor to identify a
breakdown by using the definitions provided above.
• The analyst evaluates data from the downstream sensor for the same time
period to ensure no breakdown exists, which indicates congestion at the
bottleneck sensor is unlikely due to spillback from downstream
congestion.
Guidance for Freeway Capacity Estimation Chapter 26/Freeway and Highway Segments: Supplemental
• The analyst evaluates data from the upstream sensor to verify queues are
forming as a result of breakdown at the bottleneck. This check ensures
observed drops in speeds and increases in density at the bottleneck sensor
are indeed due to breakdown.
It is important that the measurements of flows, speeds, and densities used to
estimate capacity are carried out at the correct locations, especially if the data
will be generated from existing fixed freeway sensors, which may or may not be
at the optimal locations to detect breakdown events. Capacity should always be
measured at the bottleneck location. At merge bottlenecks or lane drops, this
location is downstream of the merge point (Exhibit 26-10). At diverge
bottlenecks, this location is upstream of the diverge point (Exhibit 26-11). At
weaving bottlenecks, this location is within the weaving area (Exhibit 26-12).
Exhibit 26-10
Recommended Capacity
Measurement Location for
Merge Bottlenecks
Source: Elefteriadou, Kondyli, and St. George (6).
Upstream Bottleneck Downstream Exhibit 26-11

traffic Recommended Capacity
detector detector detector
Measurement Location for
Diverge Bottlenecks
Location of
capacity
measurement Diverge point
Upstream Bottleneck Downstream Exhibit 26-12

traffic Recommended Capacity
detector detector detector Measurement Location for
Weaving Bottlenecks
Location of
capacity
measurement
Regardless of the bottleneck type, the analyst will be able to identify and
measure capacity only if a breakdown occurs. As discussed below, the
breakdown event is associated with the development of queues that form
upstream of the bottleneck location (i.e., merge point, diverge point, weaving
section) and propagate further upstream, but queues also propagate downstream
as vehicles accelerate past the start of the bottleneck. Once breakdown events are
identified, the analyst will be able to identify the prebreakdown and
postbreakdown flow rates and estimate segment capacity based on the method
discussed in the next section.
CAPACITY ESTIMATION FROM FIELD DATA

To estimate the capacity of the various freeway segments it is important to
analyze data obtained under recurring congestion and under similar operational
and weather conditions. Observations in which adverse weather, incidents, work
zones, or special events were present must be analyzed separately to obtain
capacities under various prevailing conditions. To obtain a reasonable capacity
estimate, it is important to analyze a considerable amount of data over a period
of several months to an entire year.
The recommended method for capacity estimation from sensor data takes
into account that capacity is stochastic. That is, the same flow rate may or may
not be followed by a breakdown. Therefore, during an observation period, both
prebreakdown flow rates and flow rates that are not followed by breakdown
events (uncongested flow rates) are considered. From these flow rates, the
method develops a capacity distribution and then selects a capacity value based
on an acceptable rate of breakdown. Two plausible (and equivalent) freeway
segment capacity definitions are offered:
1. Definition A: Freeway segment capacity is the maximum 15-min flow rate
(in passenger cars per hour per lane) that produces an acceptable (λ%)
rate of breakdown.
2. Definition B: Freeway segment capacity is the maximum 15-min flow rate
(in passenger cars per hour per lane) that ensures stable flow (100 – λ%)
of the time.
The rate of breakdown λ is the ratio of the total number of periods observed
under prebreakdown conditions, divided by the total number of 15-min
uncongested observations under the same flow rate. A default acceptable rate of
breakdown λ of 15% is recommended.
The capacity estimation process follows a series of eight steps and assumes
sensors are placed at the appropriate locations (as discussed above) and are
available to measure prebreakdown flows and ensure the absence of
downstream congestion, which may bias the results described below.
1. Record the distribution of 15-min flow rates (in passenger cars per hour
per lane) during the observation period (preferably a long period). For
example, sampling from the sensor 24 h per day on weekdays over a year
gives approximately 24 × 4 × 250 = 24,000 flow rate observations.
2. Exclude the 15-min analysis periods when the freeway is in breakdown

mode, as defined earlier, which will result in a distribution of
uncongested 15-min flow rates. It is recommended to filter breakdowns
due to nonrecurring sources of congestion, such as severe weather events
or incidents, as the focus is on estimating the bottleneck’s capacity under
recurring congestion conditions.
3. Bin the uncongested flow rates into 100- or 200-pc/h/ln bins.
4. Compute the average flow rate in each bin.
5. For each bin, count the number of times the flow rates in the bin were
immediately followed by the occurrence of a breakdown. In other words,
bin the prebreakdown 15-min flow rates.
6. Calculate the actual probability of breakdown P(BF) in each bin, defined
as the number of times a flow rate bin was in a prebreakdown condition
n(B), divided by the number of times that bin was observed to have
occurred, or n(F). The probability of breakdown P(BF) in each bin is
simply P(BF) = n(B)/n(F).
7. Fit a Weibull distribution (7) to the empirical probability of breakdown
computed in Step 6.
8. Based on the selected threshold breakdown (or stable flow) rate λ or (1 – λ),
determine the resulting capacity value from the Weibull distribution
developed in Step 6 by using Equation 26-23. A value of λ of 15% is
recommended.
𝛽
Capacity = 𝛾 × √−ln (1 − 𝜆) Equation 26-23
where β and γ, respectively, are the shape and scale parameters of the
fitted Weibull distribution, and λ is as defined previously. When λ = 0.15,
the equation simplifies to c = γ (0.163)1/β.
The following example is based on actual data and involves estimating the
capacity of a bottleneck on southbound I-440 in Raleigh, North Carolina. In this
example, sensor data in the vicinity of an on-ramp bottleneck were collected for
260 weekdays from June 2014 to May 2015. The average percentage of trucks
observed in the traffic stream was less than 1%; therefore, the conversion of
trucks into PCEs is ignored for the purposes of this example.
The theoretical number of 15-min observations is 260 days × 96 observations
per day = 24,960 observations. After outliers were removed (observations from
incident and weather events and congested-flow periods), there remained 22,984
periods when flow was deemed uncongested and that represented similar
operational and weather conditions. Within these periods, 192 breakdowns were
identified that met the criteria described above.
Exhibit 26-13 summarizes the computations for this example, using the eight
steps given above. The example illustrates how the process yields a capacity
value based on the recommended 15% breakdown rate.
Exhibit 26-13 [1] [2] [3] [4] [5] [6] [7]

Illustrative Example of the
No. of No. of Observed
Capacity Estimation Procedure
Average Observed 15-min Periods
Flow Rate in Flow Rate 15-min at a Probability of Cumulative
Bins (pc/h/ln) in Bin Uncongested Prebreakdown Breakdown Probability of
From To (pc/h/ln) Periods Flow Rate in Bin Breakdown
0 99 50 4,570 0 0.0% 0.0%
100 199 150 1,657 1 0.1% 0.5%
200 299 250 1,009 3 0.3% 2.1%
300 399 350 765 2 0.3% 3.1%
400 499 450 889 2 0.2% 4.2%
500 599 550 913 0 0.0% 4.2%
600 699 650 746 0 0.0% 4.2%
700 799 750 657 0 0.0% 4.2%
800 899 850 534 0 0.0% 4.2%
900 999 950 458 0 0.0% 4.2%
1,000 1,099 1,050 798 0 0.0% 4.2%
1,100 1,199 1,150 1,801 1 0.1% 4.7%
1,200 1,299 1,250 2,171 2 0.1% 5.7%
1,300 1,399 1,350 1,662 5 0.3% 8.3%
1,400 1,499 1,450 1,185 8 0.7% 12.5%
1,500 1,599 1,550 866 10 1.2% 17.7%
1,600 1,699 1,650 618 13 2.1% 24.5%
1,700 1,799 1,750 495 22 4.4% 35.9%
1,800 1,899 1,850 322 6 1.9% 39.1%
1,900 1,999 1,950 258 16 6.2% 47.4%
2,000 2,099 2,050 301 45 15.0% 70.8%
2,100 2,199 2,150 227 37 16.3% 90.1%
2,200 2,299 2,250 79 18 22.8% 99.5%
2,300 2,399 2,350 3 1 33.3% 100.0%
2,400 2,499 2,450 0 0 NA 100.0%
Sum 22,984 192
Notes: Numbers in brackets indicate column numbers. NA = not applicable.
The exhibit shows 22,984 15-min flow rate observations in Column 4,

equivalent to 5,746 h of observations. Column 5 shows 192 breakdown events.
The probability of breakdown in a bin is computed in Column 6, which is used to
estimate capacity based on the defined λ threshold. Finally, Column 7 shows the
cumulative distribution of prebreakdown flow rates, based on the data in
Column 5.
The information in Exhibit 26-13 is shown graphically in Exhibit 26-14. The
solid black curve to the right shows the Weibull distribution fitted to the data in
Column 6; the actual data are also plotted. The distribution parameters were β =
9.13 and γ = 2,569. Substituting these values into Equation 26-23 and using λ =
0.15 yields a capacity value of 2,105 pc/h/ln. The gray dashed curve to the left in
the exhibit represents the cumulative distribution of prebreakdown flow rates
(i.e., Column 7). In this case, the calculated capacity value corresponded to
approximately the 85th percentile of the prebreakdown flow rate distribution, as
represented by the dotted lines.
Exhibit 26-14
Capacity Estimation Using the
15% Acceptable Breakdown
Rate Method
In summary, the capacity estimation method considers the fact that flow
rates preceding breakdown can also occur at other times without being followed
by a breakdown. The definition of capacity is clear and unambiguous and can be
explained to the HCM user or practitioner without much difficulty. However, the
analyst needs to ensure there are a sufficient number of breakdown observations to be
confident in the calculated capacity value.
6. CONNECTED AND AUTOMATED VEHICLES
INTRODUCTION
This section provides capacity adjustment factors (CAFs) for freeway system
elements to account for the presence of connected and automated vehicles
(CAVs) in the traffic stream. It also provides daily and hourly maximum service
volumes for basic freeway segments for different proportions of CAVs in the
traffic stream. Although CAVs are still a developing technology, transportation
agencies have an immediate need as part of their long-range planning efforts to
account for CAVs’ potential ability to increase existing roadways’ throughput.
At the time of writing, CAVs capable of fully controlling the vehicle for an
entire trip without the possible need for human intervention, either under
specified operated conditions or under any operating condition [i.e., Society of
Automotive Engineers automation levels 4 and 5 (8)], were not yet in production
for consumer use. Although other HCM methodologies are based on empirical
observations of actual vehicles using actual roadway facilities, calibrated simulation,
or both, these approaches are currently infeasible given the absence of level 4 and
5 CAVs in the traffic stream. Instead, uncalibrated simulation modeling was
conducted using CAV logic developed for the Federal Highway Administration.
Details about this modeling are available in a paper (9) available online in HCM
Volume 4 (hcmvolume4.org) in the Technical Reference Library section for
Chapter 26.
CONCEPTS
CAV Technology
CAVs integrate two separate types of technology, communications and
automation. The combination of these technologies is required to achieve
roadway capacity increases, as described below:
• Connected vehicles transmit data about their status to their surroundings
(e.g., roadside infrastructure, other road users). They also receive
information about their surroundings (e.g., traffic conditions, weather
conditions, presence of potential conflicting vehicles, traffic signal timing)
that motorists can use to adjust their driving behavior in response to
conditions present at a given time and location. This exchange of
information offers potential safety, fuel economy, and environmental
benefits. However, it is not clear how connectivity affects car following
and driver behavior and subsequently freeway capacity.
• Automated vehicles take over all or a portion of the driving task. Depending
on the level of automation, a human may still need to take over under
certain conditions. In the absence of connectivity, the information
available to automated vehicles is limited to that which can be gathered
by on-board sensors, which is typically constrained by a sensor’s line of
sight and the rate at which the sensor takes measurements (e.g., 10 times
per second). As a result, for both safety and passenger comfort reasons,
current adaptive cruise control systems offer minimum time gaps that are
Connected and Automated Vehicles Chapter 26/Freeway and Highway Segments: Supplemental
similar to, or longer than, the gaps used by human drivers, and thus may
decrease roadway capacity when in widespread use (10).
• Connected and automated vehicles communicate with each other and with
roadside infrastructure. The connectivity element provides automated
driving systems with more complete information about a vehicle’s
surroundings and enables cooperative vehicle maneuvers that improve
roadway operations. The vehicle’s enhanced detection capabilities, as well
as redundancy in detection, enable an automated driving system to
operate more efficiently and more safely than with only an on-board
system (11). In particular, the cooperative adaptive cruise control (CACC)
feature enabled by vehicle-to-vehicle communication allows CAVs to
safely operate in platoons at shorter headways than possible by either
human-driven vehicles or automated vehicles using adaptive cruise
control only.
Factors Influencing Freeway Capacity with CAVs in the Traffic Stream

Basic Freeway Segments
The HCM’s base capacity for basic freeway segments of 2,400 pc/h/ln
represents an average headway (i.e., back bumper to back bumper) of 1.5 s
between successive vehicles in a lane. Depending on vehicle length and speed, it
represents an intervehicle gap (i.e., back bumper to front bumper) of around 1.1 s.
To the extent that CAVs can (a) safely reduce this average intervehicle gap and
(b) reduce oscillations in the size of the gap between successive vehicles, shorter
headways and thus greater capacities are feasible on basic freeway segments.
Basic freeway segment capacity is also affected by less-than-ideal roadway
geometric conditions (e.g., lane widths less than 12 ft, right-side lateral clearance
less than 6 ft), because human motorists drive more cautiously under these
conditions. However, lane-guidance technology is expected to allow CAVs to
travel along more-constrained freeway sections without changing their speed or
separation from another vehicle.
Drivers’ familiarity with a freeway can also affect capacity, with those less
familiar with a freeway’s geometry and typical traffic patterns likely to drive
more cautiously. CAVs, on the other hand, can be expected to have current
information about roadway geometry and conditions available to them and
therefore would be considered “familiar drivers” wherever they operate.
The proportion of the traffic stream that is composed of CAVs will also
influence the achievable capacity increase. The greater the proportion of CAVs in
the traffic stream, the more frequently the benefits of connectivity can be
realized, because it becomes more likely that one or more CAVs will trail another
CAV and can form a short-headway platoon. In contrast, a CAV following a
human-driven vehicle or an automated vehicle without connectivity will rely on
adaptive cruise control and therefore travel at headways similar to or longer than
those used at present by human drivers. Thus, under base conditions, a much
greater increase in capacity is seen when the proportion of CAVs on a freeway
increases from 80% to 100% than when it increases from 0% to 20% (9). However,
Chapter 26/Freeway and Highway Segments: Supplemental Connected and Automated Vehicles
when a freeway has a lower initial capacity (e.g., due to a lower design speed),
capacity tends to increase more linearly with increasing proportions of CAVs (9).
Freeway Merge, Diverge, and Weaving Segments

The HCM’s capacity-estimation methods for freeway merge, diverge, and
weaving segments use the basic freeway segment capacity as a starting point;
therefore, all of the CAV-related factors that affect basic segment capacity also
affect the capacity of merge, diverge, and weaving segments. CACC is used to
insert a merging CAV into a platoon of CAVs, assuming the platoon’s maximum
length has not yet been reached, or as the new leader of a trailing platoon
otherwise.
Freeway Managed Lane Segments

No specific research on CAV effects on managed lane segments is available.
However, the results for a market penetration of 100% may be used to approximate
the capacity effects of a CAV-only managed lane.
Oversaturated Conditions
No specific research on CAV effects on oversaturated freeways is available.
As such, this section’s CAFs are primarily intended to be used for planning-level
estimates of freeway capacities, and not for detailed operational analyses.
Assumptions Affecting CAV Ability to Provide Higher Capacities

Intervehicle Gap
Given that CAV technology and regulation is still in development,
assumptions necessarily have to be made when estimating CAVs’ potential
capacity benefit. A key assumption used in developing this section’s CAFs was
the minimum achievable intervehicle gap. Factors that could affect the eventual
intervehicle gap include:
• Legal or regulatory requirements dictating a minimum gap.
• Liability concerns on the part of vehicle manufacturers that cause them to
use a more conservative gap length than strictly needed for safety.
• Passenger comfort concerns on the part of vehicle manufacturers to
minimize the amount and magnitude of acceleration and deceleration
needed to maintain intervehicle gaps and facilitate cooperative merging
and lane changing.
• Passenger lack of trust concerns on the part of vehicle owners related to
traveling at high speed close behind another vehicle.
• Need for sufficient gaps to accommodate lane-changing and merging.
• Mechanical differences between vehicles that affect their operational
characteristics, such as braking and acceleration.
The simulation modeling that developed this section’s CAFs incorporated

the following assumptions related to intervehicle gaps (9):
• CAV capability. The modeled CAVs had vehicle-to-vehicle communication
abilities and a working CACC system. CAVs acting as platoon leaders
reverted to adaptive cruise control (i.e., relying on on-board sensors only).
• Human-driven vehicle capability. The operation of human-driven vehicles
was calibrated for three scenarios: 2,400 pc/h/ln (matching the HCM’s
base capacity for basic freeway segments with a 70 mi/h free-flow speed),
2,100 pc/h/ln, and 1,800 pc/h/ln. The latter two scenarios represent
freeway segments with some combination of lower base free-flow speeds,
narrow lanes, limited or no lateral clearance, high ramp density, and
unfamiliar drivers.
• Platooning behavior. CAVs formed platoons. A CAV became the leader of a
platoon when the vehicle in front of it was either a non-CAV or a CAV
that was the last vehicle in a platoon that had reached its maximum
length. Otherwise, a CAV that followed another CAV joined the
preceding CAV’s platoon. One-vehicle platoons were possible, and
relatively common when the proportion of CAVs in the traffic stream was
low. A CAV’s status could change from leader to follower and back,
depending on lane-changing and merging activity.
• Intraplatoon gaps. A distribution of intervehicle gaps within platoons was
used, based on a field study of driver preferences for gap size that used
vehicles whose adaptive cruise control systems were set to allow short
gaps. The distribution consisted of 0.6 s for 57% of vehicles, 0.7 s for 24%,
0.9 s for 7%, and 1.1 s for 12%, resulting in an average intervehicle gap
within platoons of 0.71 s (12).
• Interplatoon gaps. A CAV that was the leader of a platoon operated in
adaptive cruise control mode, with a gap to the next vehicle of 2.0 s, based
on adaptive cruise control settings in commercial use (13).
• Maximum platoon size. The maximum platoon size was 10 passenger cars,
constrained by the need to accommodate lane-changes and merges at
ramps and the need to maintain reliable communication between the
platoon leader and the vehicles toward the rear of the platoon.
System Reliability
The ability of CAVs to safely operate with short intervehicle gaps requires,
among other things, low communications latency (i.e., information can be
quickly exchanged between vehicles and acted upon), vehicle manufacturers to
build vehicles with reliable components, vehicle owners to promptly repair
components if they do break, and regulatory agencies to provide adequate
bandwidth for vehicle-to-vehicle communication. Consistent with other base-
condition assumptions in the freeway methodology (e.g., standard lane widths,
good weather), a base assumption for CAV analysis is that all necessary
communication elements are in place and working with a high degree of
reliability.
Traffic Stream Composition

A key assumption an analyst will need to make when performing a CAV
analysis is the percentage of CAVs that will be in the traffic stream during the
analysis year(s). Once CAVs become available to consumers, it may take many
years for the vehicle fleet to transition to an all-CAV fleet. In 2018, the average
age of light cars and trucks in the United States was just under 12 years (14), and
it takes even longer for the national fleet to turn over. Furthermore, based on past
adoption rates of new automotive technologies such as automatic transmissions,
airbags, and hybrid vehicles, many people may not choose a CAV the first or
even the second time they replace their vehicle (15). On the other hand, if many
urban dwellers decide not to replace their car and rely instead on mobility
services employing CAVs, adoption of CAVs could occur more rapidly than with
prior automotive technologies. The possibility also exists that high-occupancy
vehicle (HOV) lanes could be converted to CAV-only lanes (i.e., 100% CAVs) even
though overall CAV market penetration might be considerably less than 100%.
The simulation modeling that developed this section’s CAFs assumed a
traffic stream consisting of 100% passenger cars. The percentage of CAVs in the
traffic stream was varied from 0% to 100% in 20% increments. Analysts should
consider the latest available information about CAV adoption rates and the
effects of CAV usage on travel demand when performing an analysis of CAV
effects on freeway capacity.
Addressing Uncertain Assumptions in a CAV Analysis

Any evaluation of future conditions requires assumptions about future
population growth, mode choice, travel demand, and travel patterns, among
other factors, none of which are known with great certainty. Adding assumptions
related to CAVs, particularly when based on simulation that cannot yet be
calibrated to actual operating conditions, only increases the uncertainty in the
analysis inputs.
Because of this uncertainty, it is recommended that the CAV CAFs and service
volume tables presented in this section be applied to the evaluation of “what if”
scenarios, rather than being taken as the final word on what will happen once
CAVs become widespread. In particular, the analyst should consider:
• What if the minimum headway permitted by technology, regulation, or policy, or
the average headway produced by different vehicles’ user settings, is longer than
the modeling assumed? In this case, the capacity increase would be less than
predicted by the CAV CAFs or service volume tables.
• How reliable will the necessary communications and automation technology be?
To the extent that individual CAV-capable vehicles must be driven by a
human at any given time due to equipment malfunction, the proportion
of operating CAVs in the traffic stream will be less than the proportion of
CAV-capable vehicles. (Alternatively, the demand will be lower, in the
situation where only vehicles with functioning systems are allowed on the
facility.)
• How quickly will CAV technology become available and adopted, and how will
CAVs affect travel demand? The assumptions made related to these
questions will determine the assumed volume and proportion of CAVs in
the traffic stream, along with the assumed CAF.
CAPACITY ADJUSTMENT FACTORS

Basic Freeway Segments
Exhibit 26-15 provides CAFs for basic freeway segments where CAVs are
present in the traffic stream, CAFCAV. To determine the CAF value, first calculate
the segment’s initial adjusted capacity cadj using Equation 12-8, applying all other
applicable CAFs. Next, determine the CAFCAV value from Exhibit 26-15 based on
the proportion of CAVs in the traffic stream and the initial adjusted capacity,
interpolating as needed. Finally, determine the segment’s adjusted capacity with
CAVs by multiplying the initial adjusted capacity by CAFCAV.
Proportion of CAVs Adjusted Segment Capacity Exhibit 26-15

in Traffic Stream 2,400 pc/h/ln 2,100 pc/h/ln 1,800 pc/h/ln Capacity Adjustment Factors
0 1.00 1.00 1.00 for CAVs for Basic Freeway
20 1.02 1.02 1.15 and Freeway Diverge
40 1.07 1.10 1.27 Segments
60 1.13 1.25 1.40
80 1.22 1.37 1.60
100 1.33 1.52 1.78
Notes: CAV = connected and automated vehicle, defined here as a vehicle with an operating cooperative adaptive
cruise control system.
Interpolate for other CAV proportions and adjusted segment capacities.
Assumptions: Average intervehicle gap within CAV platoons = 0.71 s based on a distribution (see text),
CAV interplatoon gap = 2.0 s, maximum CAV platoon size = 10 pc, human-driven vehicles operate with
average gaps calibrated to the given adjusted segment capacity.
Freeway Diverge Segments

CAFCAV values for freeway diverge segments are determined similarly to
basic freeway segments. First, calculate the segment’s initial adjusted capacity
cmda using Equation 14-21, applying all other applicable CAFs. Next, determine
the CAFCAV value from Exhibit 26-15 based on the percentage of CAVs in the
traffic stream and the initial adjusted capacity, interpolating as needed. Finally,
determine the segment’s adjusted capacity with CAVs by multiplying the initial
adjusted capacity by CAFCAV.
Freeway Merge Segments

CAFCAV values for freeway merge segments where CAVs are present in the
traffic stream are given in Exhibit 26-16, based on the proportion of CAVs in the
traffic stream and interpolating as needed. The segment’s adjusted capacity is
then determined using Equation 14-21 by applying all applicable CAFs,
including CAFCAV.
Exhibit 26-16 Proportion of CAVs in Traffic Stream CAFCAV

Capacity Adjustment Factors 0 1.00
for CAVs for Freeway Merge 20 1.02
Segments 40 1.07
60 1.16
80 1.33
100 1.45
Interpolate for other CAV proportions and adjusted segment capacities.
average gaps calibrated to 2,200 pc/h/ln.
Freeway Weaving Segments

CAFCAV values for freeway weaving segments where CAVs are present in the
traffic stream are given in Exhibit 26-17. The CAF value is determined from the
proportion of CAVs in the traffic stream and the volume ratio (i.e., the weaving
demand flow rate divided by the total demand flow rate in the weaving
segment), interpolating as needed. The weaving segment method in Chapter 13
offers two definitions of capacity, one based on weaving flows and one based on
density. The CAFs in Exhibit 26-17 are intended to be applied to the latter,
density-based measure. The segment’s adjusted capacity is then determined
using Equation 13-9 by applying all applicable CAFs, including CAFCAV.
Exhibit 26-17 Proportion of CAVs Volume Ratio

Capacity Adjustment Factors in Traffic Stream 0.2 0.3 0.4
for CAVs for Freeway Weaving 0 1.00 1.00 1.00
Segments 20 1.03 1.04 1.05
40 1.08 1.08 1.09
60 1.15 1.15 1.13
80 1.23 1.22 1.20
100 1.37 1.37 1.34
Interpolate for other CAV proportions and volume ratios.
The volume ratio is the weaving demand flow rate divided by the total demand flow rate in the segment.
average gaps calibrated to 2,200 pc/h/ln.
SERVICE VOLUME TABLES

Exhibit 26-18 presents daily maximum service volumes at LOS E for basic
freeway segments with CAVs present in the traffic stream. Values in the exhibit
represent 2-way average annual daily traffic (AADT) divided by the total
number of lanes in both directions. Exhibit 26-19 presents hourly maximum
volumes per lane at LOS E for basic freeway segments with CAVs present in the
traffic stream. Assumptions used in creating these exhibits are listed below the
exhibit; see Chapter 12 for definitions of these terms.
Area Proportion of CAVs in Traffic Stream Exhibit 26-18

Type Terrain 0% 20% 40% 60% 80% 100% Daily Maximum Service
Urban Level 19,900 20,500 21,800 24,600 26,800 29,700 Volumes for Basic Freeway
Urban Rolling 19,000 19,900 21,400 24,500 26,800 29,700 Segments with CAV Presence
(2-way veh/day/ln)
Rural Level 16,800 17,900 19,300 22,000 24,400 26,800
Rural Rolling 15,200 17,200 19,100 21,600 24,400 26,800
Values represent the maximum annual average daily traffic per lane at LOS E.
Urban assumptions: Free-flow speed = 70 mph, 5% trucks, PHF = 0.94, K-factor = 0.09, D-factor = 0.60.
Rural assumptions: Free-flow speed = 70 mph, 12% trucks, PHF = 0.94, K-factor = 0.10, D-factor = 0.60.
CAV assumptions: Average intervehicle gap within CAV platoons = 0.71 s based on a distribution (see
text), CAV interplatoon gap = 2.0 s, maximum CAV platoon size = 10 pc, human-driven vehicles operate
with average gaps calibrated to 2,400 pc/h/ln.
Area Proportion of CAVs in Traffic Stream Exhibit 26-19

Type Terrain 0% 20% 40% 60% 80% 100% Hourly Maximum Service
Urban Level 2,150 2,210 2,350 2,660 2,900 3,200 Volumes for Basic Freeway
Urban Rolling 2,050 2,150 2,310 2,640 2,900 3,200 Segments with CAV Presence
(veh/h/ln)
Rural Level 2,010 2,140 2,310 2,640 2,900 3,200
Rural Rolling 1,820 2,060 2,290 2,580 2,900 3,200
Values represent the maximum analysis hour volume per lane at LOS E.
Urban assumptions: Free-flow speed = 70 mph, 5% trucks, PHF = 0.94, K-factor = 0.09, D-factor = 0.60.
Rural assumptions: Free-flow speed = 70 mph, 12% trucks, PHF = 0.94, K-factor = 0.10, D-factor = 0.60.
CAV assumptions: Average intervehicle gap within CAV platoons = 0.71 s based on a distribution (see
text), CAV interplatoon gap = 2.0 s, maximum CAV platoon size = 10 pc, human-driven vehicles operate
with average gaps calibrated to 2,400 pc/h/ln.
7. FREEWAY AND MULTILANE HIGHWAY

EXAMPLE PROBLEMS
Exhibit 26-20 lists the seven example problems provided in this section. The
problems demonstrate the computational steps involved in applying the
automobile methodology to basic freeway and multilane highway segments. All
the freeway example problems address urban freeway situations.
Exhibit 26-20 Example

List of Freeway and Multilane Problem Description Application
Highway Example Problems 1 Four-lane freeway LOS Operational analysis
2 Number of lanes required for target LOS Design analysis
3 Six-lane freeway LOS and capacity Operational and planning analysis
4 LOS on a five-lane highway with a two-way
Operational analysis
left-turn lane
5 Mixed-flow operational performance Operational analysis
6 Severe weather effects on a basic freeway
Operational analysis
segment
7 Basic managed lane segment Operational analysis
EXAMPLE PROBLEM 1: FOUR-LANE FREEWAY LOS

The Facts
• Four-lane freeway (two lanes in each direction)
• Lane width = 11 ft
• Right-side lateral clearance = 2 ft
• Commuter traffic (regular users)
• Peak hour, peak direction demand volume = 2,000 veh/h
• Traffic composition: 5% trucks
• Peak hour factor (PHF) = 0.92
• One cloverleaf interchange per mile
• Level terrain
• Facility operates under ideal conditions (no incidents, work zones, or
weather events).
Comments
The task is to find the expected LOS for this freeway during the worst 15 min
of the peak hour. With one cloverleaf interchange per mile, the total ramp
density will be 4 ramps/mi.
Step 1: Input Data

Freeway and Multilane Highway Example Problems Chapter 26/Freeway and Highway Segments: Supplemental
Step 2: Estimate and Adjust FFS

The FFS of the freeway is estimated from Equation 12-2 as follows:
𝐹𝐹𝑆 = 75.4 − 𝑓𝐿𝑊 − 𝑓𝑅𝐿𝐶 − 3.22 × 𝑇𝑅𝐷 0.84
The adjustment for lane width is selected from Exhibit 12-20 for 11-ft lanes
(1.9 mi/h). The adjustment for right-side lateral clearance is selected from Exhibit
12-21 for a 2-ft clearance on a freeway with two lanes in one direction (2.4 mi/h).
The total ramp density is 4 ramps/mi. Then
𝐹𝐹𝑆 = 75.4 − 1.9 − 2.4 − 3.22 × 40.84 = 60.8 mi/h
Because the facility is operating under ideal conditions, the SAF used in
Equation 12-5 is 1, and FFSadj = FFS.
Step 3: Estimate and Adjust Capacity

The capacity of the freeway is estimated from Equation 12-6 as follows:
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (60.8 − 50) = 2,308 pc/h/ln
Because the facility is operating under ideal conditions, the CAF used in
Equation 12-8 is 1, and cadj = c.
Step 4: Adjust Demand Volume

The demand volume must be adjusted to a flow rate that reflects passenger
cars per hour per lane under equivalent base conditions by using Equation 12-9.
𝑉
𝑣𝑝 =
𝑃𝐻𝐹 × 𝑁 × 𝑓𝐻𝑉
The demand volume is given as 2,000 veh/h. The PHF is specified to be 0.92,
and there are two lanes in each direction. The driver population consists of
regular users (commuters). Trucks make up 5% of the traffic stream, so a heavy-
vehicle adjustment factor must be determined.
From Exhibit 12-25, the PCE for trucks is 2.0 for level terrain. The heavy-
vehicle adjustment factor is then computed with Equation 12-10.
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1)
1
𝑓𝐻𝑉 = = 0.952
1 + 0.05(2.0 − 1)
then
2,000
𝑣𝑝 = = 1,142 pc/h/ln
0.92 × 2 × 0.952
Because this value is less than the base capacity of 2,308 pc/h/ln for a freeway
with FFS = 60.8 mi/h, LOS F does not exist, and the analysis continues to Step 5.
Chapter 26/Freeway and Highway Segments: Supplemental Freeway and Multilane Highway Example Problems
Step 5: Estimate Speed and Density

The FFS of the basic freeway segment is now estimated along with the
demand flow rate (in passenger cars per hour per lane) under equivalent base
conditions. Using the equations provided in Exhibit 12-6, the breakpoint for a
60.8-mi/h FFS speed–flow curve is
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 𝐹𝐹𝑆𝑎𝑑𝑗 )] × 𝐶𝐴𝐹 2 = 1,568 pc/h/ln
As the flow rate of 1,142 pc/h/ln is less than the breakpoint value of 1,568
pc/h/ln, the freeway operates within the constant-speed portion of the speed–
flow curve, so S = 60.8 mi/h. The density of the traffic stream may now be
computed from Equation 12-11.
𝑣𝑝 1,142
𝐷= = = 18.8 pc/mi/ln
𝑆 60.8
Step 6: Determine LOS

From Exhibit 12-15, a density of 18.8 pc/mi/ln corresponds to LOS C, but it is
close to the boundary for LOS B, which is a maximum of 18 pc/mi/ln. This
solution could also be calculated graphically from Exhibit 12-16, as illustrated in
Exhibit 26-21.
Exhibit 26-21
Example Problem 1: Graphical
Solution
Discussion
This basic freeway segment of a four-lane freeway is expected to operate at
LOS C during the worst 15 min of the peak hour. It is important to note that the
operation, although at LOS C, is close to the LOS B boundary. In most
jurisdictions, this operation would be considered to be quite acceptable.
EXAMPLE PROBLEM 2: NUMBER OF LANES REQUIRED FOR TARGET LOS

The Facts
• Demand volume = 4,000 veh/h (one direction)
• Level terrain
• Traffic composition: 8% SUTs and buses
• Provision of 12-ft lanes
• Provision of 6-ft right-side lateral clearance
• PHF = 0.85
• Ramp density = 3 ramps/mi
• Target LOS = D
weather events).
Comments
This example problem is a classic design application of the methodology.
The number of lanes needed to provide LOS D during the worst 15 min of the
peak hour is to be determined.
Step 1: Input Data


FFS is estimated by using Equation 12-2. Because the lane width and lateral
clearance to be provided on the new freeway will be 12 ft and 6 ft, respectively,
there are no adjustments for these features from Exhibit 12-20 or Exhibit 12-21.
The total ramp density is given as 3 ramps/mi. Then
𝐹𝐹𝑆 = 75.4 − 𝑓𝐿𝑊 − 𝑓𝑅𝐿𝐶 − 3.22 × 𝑇𝑅𝐷 0.84
𝐹𝐹𝑆 = 75.4 − 0 − 0 − 3.22 × 30.84 = 67.3 mi/h
Because the facility is operating under ideal conditions, the SAF used in
Equation 12-5 is 1, and FFSadj = FFS.

The capacity of the freeway is estimated from Equation 12-6.
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (67.3 − 50) = 2,373 pc/h/ln
Step 4: Estimate Number of Lanes Needed

Because this is a design analysis, Step 4 of the operational analysis
methodology is modified. Equation 12-23 may be used directly to determine the
number of lanes needed to provide at least LOS D.
𝑉
𝑁=
𝑀𝑆𝐹𝑖 × 𝑃𝐻𝐹 × 𝑓𝐻𝑉
A value of the maximum service flow rate must be selected from Exhibit 12-
37 for an FFS of 65 mi/h and LOS D. Note that this exhibit only provides these
values in 5-mi/h increments; therefore, FFS is rounded to 65 mi/h. The
corresponding maximum service flow rate is 2,060 pc/h/ln.
The PHF is given as 0.85. A heavy-vehicle factor for 8% trucks must be
determined by using Exhibit 12-25 for level terrain. The PCE of trucks on level
terrain is 2.0, so the heavy-vehicle adjustment based on Equation 12-10 is
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1)
1
𝑓𝐻𝑉 = = 0.926
1 + 0.08(2 − 1)
and
4,000
𝑁= = 2.5 ln
2,060 × 0.85 × 0.926
It is not possible to build 2.5 lanes. To provide a minimum of LOS D, it will
be necessary to provide three lanes in each direction, or a six-lane freeway.
At this point, the design application ends. It is possible, however, to consider
what speed, density, and LOS will prevail when three lanes are actually
provided. Therefore, the example problem continues with Steps 5 and 6.

In pursuing additional information, the problem now reverts to an
operational analysis of a three-lane basic freeway segment with a demand
volume of 4,000 pc/h.
Equation 12-9 is used to compute the actual demand flow rate per lane under
equivalent base conditions.
𝑉
𝑣𝑝 =
4,000
𝑣𝑃 = = 1,694 pc/h/ln
0.85 × 3 × 0.926
From Exhibit 12-6, the breakpoint for a speed–flow curve with FFS equal to
67.3 is
In this case, the demand flow rate of 1,694 pc/h/ln exceeds the breakpoint
value of 1,308 pc/h/ln, and the average speed will be less than the FFS.
The expected speed of the traffic stream may be estimated by using either
Exhibit 12-7 (for a graphical solution) or Equation 12-1 as follows:
𝑐𝑎𝑑𝑗 𝑎
(𝐹𝐹𝑆𝑎𝑑𝑗 − 𝐷 ) (𝑣𝑝 − 𝐵𝑃)
𝑐
𝑆 = 𝐹𝐹𝑆𝑎𝑑𝑗 − 𝑎
(𝑐𝑎𝑑𝑗 − 𝐵𝑃)
2,373
(67.3 − 45 ) (1,694 − 1,308)2
𝑆 = 67.3 − = 65.4 mi/h
(2,373 − 1,308)2
The density may now be computed from Equation 12-11.
𝑣𝑝 1,694
𝐷= = = 25.9 pc/mi/ln
𝑆 65.4

Entering Exhibit 12-15 with a density of 25.9 pc/mi/ln, the LOS is C, but that
density is very close to the boundary of LOS D, which is 26 pc/mi/ln.
Discussion
The resulting LOS is C, which represents a better performance than the target
design. Although the minimum number of lanes needed was 2.5, which would
have produced a minimal LOS D, providing three lanes yields a density that is
close to the LOS C boundary. In any event, the target LOS of the design will be
met by providing a six-lane basic freeway segment.
EXAMPLE PROBLEM 3: SIX-LANE FREEWAY LOS AND CAPACITY

The Facts
• Volume of 5,000 veh/h (one direction, existing)
• Volume of 5,788 veh/h (one direction, in 3 years)
• Rolling terrain
• Three lanes in each direction
• FFS = 70 mi/h (measured)
• PHF = 0.96
• Traffic growth = 5% per year
weather events).
Comments
This example consists of two operational analyses, one for the present
demand volume of 5,000 pc/h and one for the demand volume of 5,788 pc/h
expected in 3 years. In addition, a planning element is introduced: Assuming
traffic grows as expected, when will the capacity of the roadway be exceeded?
This analysis requires that capacity be determined in addition to the normal
output of operational analyses.
Step 1: Input Data


Step 2 is not needed, as the FFS was directly measured (70 mi/h). Because the
facility is operating under ideal conditions, the SAF used in Equation 12-5 is 1,
and FFSadj = FFS.

The capacity of the freeway is estimated from Equation 12-6.
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (70 − 50) = 2,400 pc/h/ln

In this case, two demand volumes will be adjusted by using Equation 12-9.
𝑉
𝑣𝑝 =
The PHF is given as 0.96, and there are three lanes in each direction. The
heavy-vehicle factor will reflect 4% trucks in rolling terrain. From Exhibit 12-25,
the PCE for trucks in rolling terrain is 3.0. Equation 12-10 then gives
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1)
1
𝑓𝐻𝑉 = = 0.926
1 + 0.04(3.0 − 1)
Two values of vp are computed: one for present conditions and one for
conditions in 3 years.
5,000
𝑣𝑝 (present) = = 1,875 pc/h
0.96 × 3 × 0.926
5,788
𝑣𝑝 (future) = = 2,171 pc/h
0.96 × 3 × 0.926

Two values of speed and density will be estimated, one each for the present
and future conditions. Equation 12-1 will be used to estimate speeds. First, the
breakpoint for the speed–flow curve is computed from Exhibit 12-6.
One equation applies to both cases; a 70-mi/h FFS with a flow rate over 1,200
pc/h/ln is used.
𝑐
2,400
(70 − 45 ) (1,875 − 1,200)2
𝑆(present) = 70 − = 64.7 mi/h
(2,400 − 1,200)2
2,400
(70 − 45 ) (2,171 − 1,200)2
𝑆(future) = 70 − = 59.1 mi/h
(2,400 − 1,200)2
The corresponding densities may now be estimated from Equation 12-11.
𝑣𝑃
𝐷=
𝑆
1,875
𝐷(present) = = 29.0 pc/mi/ln
64.7
2,171
𝐷(future) = = 36.7 pc/mi/ln
59.1

From Exhibit 12-15, the LOS for the present situation is D, and the LOS for
the future scenario (in 3 years) is E due to the increase in density.
Step 7: Determine When Capacity Will Be Reached

Step 7 is an additional step for this problem. To determine when capacity
will be reached, the capacity of the basic freeway segment must be estimated.
From Exhibit 12-37, the maximum service flow rate for LOS E on a basic freeway
segment with a 70-mi/h FFS is 2,400 pc/h/ln. This flow rate is synonymous with
capacity.
The analyst must be sure the capacity and demand flow rates compared in
Step 7 are measured on the same basis. The 2,400 pc/h/ln is a flow rate under
equivalent base conditions. The demand flow rate in 3 years was estimated to be
2,171 pc/h/ln on this basis. These two values, therefore, may be compared. As an
alternative, the capacity could be computed for prevailing conditions with
Equation 12-24.
𝑆𝐹𝐸 = 𝑀𝑆𝐹𝐸 × 𝑁 × 𝑓𝐻𝑉
𝑆𝐹𝐸 = 2,400 × 3 × 0.926 = 6,667 veh/h
This capacity, however, is stated as a flow rate. The demand volume is stated
as an hourly volume. Thus, a service volume for LOS E is needed as estimated
𝑆𝑉𝐸 = 𝑆𝐹𝐸 × 𝑃𝐻𝐹 = 6,667 × 0.96 = 6,400 veh/h
The problem may be solved either by comparing the demand volume of
5,788 veh/h (in 3 years) with the hourly capacity of 6,400 veh/h or by comparing
the demand flow rate under equivalent base conditions of 2,171 pc/h/ln with the
base capacity of 2,400 pc/h/ln. With the hourly demand volume and capacity,
6,400 = 5,788 × (1.05)𝑛
𝑛 = 2.06 years
On the basis of the forecasts of traffic growth, the basic freeway segment
described will reach capacity within 5 years. The demand value of 5,788 veh/h
occurs 3 years from the present per the problem description, and the calculation
above shows capacity is reached after an additional 2 years. If this result is added
to the 3-year planning horizon, capacity will be reached within 5 years of the
time of the analysis.
Discussion
The LOS on this segment will reach LOS E within 3 years due to the increase
in density. The demand is expected to exceed capacity within 5 years. Given the
normal lead times for planning, design, and approvals before the start of
construction, it is probable that planning and preliminary design for an
improvement should be started immediately.
EXAMPLE PROBLEM 4: LOS ON A FIVE-LANE HIGHWAY WITH

A TWO-WAY LEFT-TURN LANE
The Facts
• Lane width: 12 ft
• Lateral clearance, both sides of the roadway: 12 ft
• Traffic composition: 6% trucks, with default truck mix (30% SUTs, 70%
TTs)
• Access points per mile: eastbound = 10; westbound = 0
• PHF = 0.90
• Median type: two-way left-turn lane
• Peak hour demand: 1,500 veh/h
• The upgrade occurs in the westbound direction
• Posted speed limit = 45 mi/h
Comments
A 6,600-ft segment of a five-lane highway (two travel lanes in each direction
plus a two-way left-turn lane) is on a 3.5% grade. At what LOS is the facility
expected to operate in each direction?
There is one segment in each direction. The upgrade and downgrade
segments on the 3.5% grade must be analyzed separately. This example is more
complex than the previous examples because the segment characteristics are not
all the same, particularly the number of access points. Because no base FFS is
given, it will be estimated as the speed limit plus 7 mi/h, or 45 + 7 = 52 mi/h.
Step 1: Input Data

All input data are given above.

FFS is estimated by using Equation 12-3.
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑓𝐿𝑊 − 𝑓𝑇𝐿𝐶 − 𝑓𝑀 − 𝑓𝐴
In this case, the base FFS is estimated to be 52 mi/h. The lane width is 12 ft, which
is the base condition; therefore, fLW = 0.0 mi/h (Exhibit 12-20). The lateral
clearance is 12 ft at each roadside, but a maximum value of 6 ft may be used. A
two-way left-turn lane is considered to have a median lateral clearance of 6 ft.
Thus, the total lateral clearance is 6 + 6 = 12 ft, which is also a base condition.
Therefore, fTLC = 0.0 mi/h (Exhibit 12-22). The median-type adjustment fM is also
0.0 mi/h (Exhibit 12-23).
For this example problem, only the access-point density produces a nonzero
adjustment to the base FFS. The eastbound (EB) segment (3.5% downgrade) has
10 access points/mi. From Exhibit 12-24, the corresponding FFS adjustment is 2.5
mi/h. The westbound (WB) segment (3.5% upgrade) has 0 access points/mi and a
corresponding FFS adjustment of 0.0 mi/h. Therefore,
𝐹𝐹𝑆𝐸𝐵 = 52.0 − 0.0 − 0.0 − 0.0 − 2.5 = 49.5 mi/h
𝐹𝐹𝑆𝑊𝐵 = 52.0 − 0.0 − 0.0 − 0.0 − 0.0 = 52.0 mi/h

The capacity of the multilane highway segment is estimated as follows from
Equation 12-7.
𝑐 = 1,900 + 20 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 45)
𝑐𝐸𝐵 = 1,900 + 20 × (49.5 − 45) = 1,990 pc/h/ln
𝑐𝑊𝐵 = 1,900 + 20 × (52.0 − 45) = 2,040 pc/h/ln

Demand volume is adjusted by using Equation 12-9.
𝑉
𝑣𝑝 =
To compute the heavy-vehicle adjustment factor fHV, PCEs for trucks are
needed for (a) the 3.5%, 6,600-ft upgrade and (b) the 3.5%, 6,600-ft downgrade.
The segment is 1.25 mi (6,600/5,280 ft) long. The following values are obtained
from Exhibit 12-26:
• Eastbound: 2.24 (using 6% trucks, a 2% downgrade, and 1.25-mi grade
length). Note that all downgrades exceeding 2% use the PCE values for a
2% downgrade.
• Westbound: 3.97 (using 6% trucks, a 3.5% upgrade, and a 1.25-mi grade
length).
The heavy-vehicle adjustment factors fHV for each segment are calculated
1
𝑓𝐻𝑉,𝐸𝐵 = = 0.93
1 + 0.06 × (2.24 − 1)
1
𝑓𝐻𝑉,𝑊𝐵 = = 0.85
1 + 0.06 × (3.97 − 1)
The segments’ flow rates are then calculated as
1,500
𝑣𝑝,𝐸𝐵 = = 896 pc/h/ln
0.90 × 2 × 0.93
1,500
𝑣𝑝,𝑊𝐵 = = 980 pc/h/ln
0.90 × 2 × 0.85

Speed is estimated with Equation 12-1 or the graph in Exhibit 12-7. With
Equation 12-1, both demand flow rates are less than the multilane highway
breakpoint value of 1,400 pc/h/ln. Therefore, the speeds S are equal to FFS. The
densities are computed from Equation 12-11.
𝑣𝑝,𝐸𝐵 896
𝐷𝐸𝐵 = = = 18.1 pc/mi/ln
𝑆𝐸𝐵 49.5
𝑣𝑝,𝑊𝐵 980
𝐷𝑊𝐵 = = = 18.8 pc/mi/ln
𝑆𝑊𝐵 52

LOS is found by comparing the densities of the segments with the criteria in
Exhibit 12-15. As both densities are greater than 18 pc/mi/ln, both upgrade and
downgrade segments operate at LOS C.
Discussion
Even though the upgrade and downgrade segments operate at LOS C, they
are very close to the LOS B boundary (18.0 pc/mi/ln). Both directions of the
multilane highway on this grade operate well.
EXAMPLE PROBLEM 5: MIXED-FLOW FREEWAY OPERATIONS

This example illustrates the application of the mixed-flow model for an
extended single grade on a six-lane rural freeway.
The Facts
• 2-mi basic segment on a 5% upgrade
• Traffic composition: 5% SUTs and 10% TTs
• FFS = 65 mi/h
• Mixed-traffic flow rate = 1,500 veh/h/ln
Comments
The task is to estimate the segment’s speed and density. Given the significant
truck presence (15%) and the 5%, 2-mi grade, the mixed-flow model should be
applied.
Step 1: Input Data

Step 2: Compute Mixed-Flow Capacity Adjustment Factor

Capacity is computed with Equation 26-1.
𝐶𝐴𝐹mix = 𝐶𝐴𝐹𝑎𝑜 − 𝐶𝐴𝐹𝑇,mix − 𝐶𝐴𝐹𝑔,mix
There are three terms in the equation. The CAF for auto-only CAFao is 1.00, as
no driver population, weather, incident, or work zone adjustments are specified
in the problem statement.
The truck effect term is computed with Equation 26-2.
𝐶𝐴𝐹𝑇,mix = 0.53 × 𝑃𝑇 0.72 = 0.53 × 0.150.72 = 0.135

The grade effect term is computed with Equation 26-3 and Equation 26-4.
𝐶𝐴𝐹𝑔,mix = 𝜌𝑔,mix × max[0, 0.69 × (𝑒 12.9𝑔 − 1)]
× max [0, 1.72 × (1 − 1.71𝑒 −3.16𝑑 )]
𝜌𝑔,mix = 0.126 − 0.03𝑃𝑇 = 0.126 − (0.03)(0.15) = 0.1215
𝐶𝐴𝐹𝑔,mix = 0.1215 × max[0, 0.69 × (𝑒 (12.9)(0.05) − 1)]
× max[0, 1.72 × (1 − 1.71𝑒 (−3.16)(2) )] = 0.131
then
𝐶𝐴𝐹mix = 1 − 0.135 − 0.131 = 0.734
The mixed-flow capacity is then computed from Equation 26-5.
𝐶mix = 𝐶𝑎𝑜 × 𝐶𝐴𝐹mix
The auto-only capacity 𝐶𝑎𝑜 is computed from Exhibit 12-6.
𝐶𝑎𝑜 = 2,200 + 10(𝐹𝐹𝑆 − 50) = 2,200 + 10 × (65 − 50) = 2,350 pc/h/ln
then
𝐶mix = 2,350 × 0.734 = 1,725 veh/h/ln
As the mixed-traffic flow rate of 1,500 veh/h/ln is less than the mixed-flow
capacity of 1,725 veh/h/ln, the analysis can proceed.
Step 3: Compute Mixed-Flow FFS and FFS Adjustment Factor

Equation 26-6 through Equation 26-8 compute the free-flow travel rates for
SUTs, TTs, and automobiles, respectively. The FFS of this basic freeway segment
is 65 mi/h. Truck performance curves for free-flow speeds other than 70 ± 2.5
mi/h are provided in Appendix A. The 65-mi/h curves for SUTs and TTs are
found in Exhibit 26-A4 and Exhibit 26-A9, respectively.
The travel time for a SUT TSUT at a point 10,000 ft along the upgrade can be
read directly from Exhibit 26-A4 by observing where the 5% upgrade curve
intersects 10,000 ft: 134 s. Similarly, the travel time for a TT TTT is 173 s.
As the grade is 2 mi (10,560 ft) long and the performance curves only provide
values up to 10,000 ft, Equation 26-12 is used to determine the travel time rates
for the upgrade as a whole. The slope of the travel time versus distance curve δ,
which is used in Equation 26-12, can be determined from Exhibit 26-7 for SUTs
and Exhibit 26-8 for TTs. The δ values are 0.0146 and 0.0202, respectively.
Then
𝑇𝑆𝑈𝑇,10000𝑓𝑡 10,000
𝜏𝑘𝑖𝑛,𝑆𝑈𝑇 = + 𝛿 (1 − ) × 5,280
𝑑 5280𝑑
134 10,000
𝜏𝑘𝑖𝑛,𝑆𝑈𝑇 = + 0.0146 (1 − ) × 5,280 = 71.1 s/mi
2 10,560
173 10,000
𝜏𝑘𝑖𝑛,𝑇𝑇 = + 0.0202 (1 − ) × 5,280 = 92.2 s/mi
2 10,560
As this step’s objective is to compute the FFS of the mixed-traffic stream, the
traffic interaction term ΔτTI is zero, and the mixed-flow rate is set to 1 veh/h/ln.
The SUT, TT, and auto travel time rates are then computed using Equation 26-6
through Equation 26-8.
𝜏𝑆𝑈𝑇,𝐹𝐹𝑆 = 71.1 + 0 = 71.1 s/mi
𝜏𝑇𝑇,𝐹𝐹𝑆 = 92.2 + 0 = 92.2 s/mi
3,600
𝜏𝑎,𝐹𝐹𝑆 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆
2.76
𝑣mix 0.46 𝜏𝑘𝑖𝑛,𝑆𝑈𝑇 3,600
+100.42 × ( ) × 𝑃𝑆𝑈𝑇 0.68 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
1.81
𝑣mix 1.36 𝜏𝑘𝑖𝑛,𝑇𝑇 3,600
+110.64 × ( ) × 𝑃𝑇𝑇 0.62 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
3,600
𝜏𝑎,𝐹𝐹𝑆 = +0
65
1 0.46 71.1 3,600 2.76
+100.42 × ( ) × 0.050.68 × max [0, − ]
1,000 100 (65 × 100)
1 1.36 92.2 3,600 1.81
+110.64 × ( ) × 0.10.62 × max [0, − ]
1,000 100 (65 × 100)
𝜏𝑎,𝐹𝐹𝑆 = 55.4 s/mi

Mixed-flow travel rates and speeds are computed with Equation 26-13 and
Equation 26-14.
𝜏mix,𝐹𝐹𝑆 = 𝑃𝑎 𝜏𝑎,𝐹𝐹𝑆 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇,𝐹𝐹𝑆 + 𝑃𝑇𝑇 𝜏𝑇𝑇,𝐹𝐹𝑆
𝜏mix,𝐹𝐹𝑆 = (0.85)(55.4) + (0.05)(71.1) + (0.1)(92.2) = 59.87 s/mi
3,600 3,600
𝐹𝐹𝑆mix = = = 60.1 mi/h
𝜏mix,𝐹𝐹𝑆 59.87
Finally, the segment’s SAF is estimated with Equation 26-15.
𝐹𝐹𝑆mix 60.1
𝑆𝐴𝐹mix = = = 0.92
𝐹𝐹𝑆 65
Step 4: Compute the Mixed-Flow Rate at the Breakpoint

The breakpoint is calculated from Equation 26-16.
𝐵𝑃mix = max[0, 𝐵𝑃𝑎𝑜 (1 − 0.4𝑃𝑇 0.1 × max[0, 𝑒 30𝑔 + 1] × 𝑑 0.01 )]
where the auto-only breakpoint is calculated by using an equation given in
Exhibit 12-6.
𝐵𝑃𝑎𝑜 = [1,000 + 40 × (75 − 𝐹𝐹𝑆)] × 𝐶𝐴𝐹 2
𝐵𝑃𝑎𝑜 = [1,000 + 40 × (75 − 65)] × 12 = 1,400 veh/h/ln
then
𝐵𝑃mix = max[0, (1,400)(1 − 0.4(0.15)0.1 × max[0, 𝑒 30×0.05 + 1] × 20.01 )]
𝐵𝑃mix = 0 veh/h/ln
This result implies that speeds drop immediately at zero flow (i.e., the
mixed-flow FFS cannot be sustained even at low flows).
Step 5: Compute Modal and Mixed-Flow Speeds at Capacity and 90%

of Capacity
The speeds and densities for each mode at capacity and 90% of capacity are
calculated in this step. Equation 26-6 through Equation 26-8 are applied twice
more, once for a flow rate equal to the mixed-flow capacity of 1,725 veh/h/ln
calculated in Step 2, and again for a flow rate equal to 90% of capacity. Applying
these equations requires determining the traffic interaction term ΔτTI, which in
turn requires determining the equivalent auto-only speed Sao.
The calculation process will be demonstrated for conditions at capacity. The
value of Cmix determined in Step 2 (1,725 veh/h/ln) will be used as vmix in the
calculations.
The auto-only speed at capacity is computed by Equation 26-10.
𝑣mix
𝐶𝐴𝐹mix
2
𝑆𝑎𝑜 = 𝑐 𝑣
(𝐹𝐹𝑆 − ) ( mix − 𝐵𝑃𝑎𝑜) 𝑣mix
𝐷𝑐 𝐶𝐴𝐹mix
{ (𝑐 − 𝐵𝑃𝑎𝑜)2 𝐶𝐴𝐹mix
The value of vmix/CAFmix is 1,725/0.734 = 2,350 veh/h/ln, which is greater than
the auto-only breakpoint of 1,400 veh/h/ln calculated in Step 4. Therefore, the
second of the two equations is applied.
2
2,350 1,725
(65 − ) (0.734 − 1,400)
𝑆𝑎𝑜,𝑐𝑎𝑝 = 65 − 45 = 52.2 mi/h
(2,350 − 1,400)2
The traffic interaction term can now be computed with Equation 26-9.
3,600 3,600 1
𝛥𝜏𝑇𝐼,𝑐𝑎𝑝 = ( − ) × (1 + 3 [ − 1])
𝑆𝑎𝑜,𝑐𝑎𝑝 𝐹𝐹𝑆 𝐶𝐴𝐹mix
3,600 3,600 1
𝛥𝜏𝑇𝐼,𝑐𝑎𝑝 = ( − ) × (1 + 3 [ − 1]) = 28.3 s/mi
52.2 65 0.734
Equation 26-6 through Equation 26-8 are now applied to find the modal
travel time rates at capacity.
𝜏𝑆𝑈𝑇,𝑐𝑎𝑝 = 𝜏𝑆𝑈𝑇,𝑘𝑖𝑛 + 𝛥𝜏𝑇𝐼 = 71.1 + 28.3 = 99.4 s/mi
𝜏𝑇𝑇,𝑐𝑎𝑝 = 𝜏𝑇𝑇,𝑘𝑖𝑛 + 𝛥𝜏𝑇𝐼 = 92.2 + 28.3 = 120.5 s/mi
3,600
𝜏𝑎,𝑐𝑎𝑝 = + 𝛥𝜏𝑇𝐼
𝐹𝐹𝑆
2.76
𝑣mix 0.46 𝜏𝑆𝑈𝑇,𝑘𝑖𝑛 3,600
+100.42 × ( ) × 𝑃𝑆𝑈𝑇 0.68 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
1.81
𝑣mix 1.36 𝜏𝑇𝑇,𝑘𝑖𝑛 3,600
+110.64 × ( ) × 𝑃𝑇𝑇 0.62 × max [0, − ]
1,000 100 (𝐹𝐹𝑆 × 100)
3,600
𝜏𝑎,𝑐𝑎𝑝 = + 28.3
65
1,725 0.46 71.1 3,600 2.76
+100.42 × ( ) × 0.050.68 × max [0, − ]
1,000 100 (65 × 100)
1,725 1.36 92.2 3,600 1.81
+110.64 × ( ) × 0.10.62 × max [0, − ]
1,000 100 (65 × 100)
𝜏𝑎,𝑐𝑎𝑝 = 92.9 s/mi

Based on these travel rates, the overall mixed-traffic space mean speed at
capacity can be calculated with Equation 26-19.
3,600
𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝 =
𝑃𝑎 𝜏𝑎 + 𝑃𝑆𝑈𝑇 𝜏𝑆𝑈𝑇 + 𝑃𝑇𝑇 𝜏𝑇𝑇
3,600
𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝 = = 37.5 mi/h
(0.85)(92.9) + (0.05)(99.4) + (0.1)(120.5)
The same process is used to calculate the mixed-traffic speed at 90% of
capacity (vmix = 0.9 × 1,725 = 1,553 veh/h/ln). The resulting calculation results are
2
2,350 1,553
(65 − ) (0.734 − 1,400)
𝑆𝑎𝑜,90𝑐𝑎𝑝 = 65 − 45 = 57.7 mi/h
(2,350 − 1,400)2
3,600 3,600 1
𝛥𝜏𝑇𝐼,90𝑐𝑎𝑝 = ( − ) × (1 + 3 [ − 1]) = 14.6 s/mi
57.7 65 0.734
𝜏𝑆𝑈𝑇,90𝑐𝑎𝑝 = 71.1 + 14.6 = 85.7 s/mi
𝜏𝑇𝑇,90𝑐𝑎𝑝 = 92.2 + 14.6 = 106.8 s/mi
3,600
𝜏𝑎,90𝑐𝑎𝑝 = + 14.6
65
1,553 0.46 71.1 3,600 2.76
+100.42 × ( ) × 0.050.68 × max [0, − ]
1,000 100 (65 × 100)
1,553 1.36 92.2 3,600 1.81
+110.64 × ( ) × 0.10.62 × max [0, − ]
1,000 100 (65 × 100)
𝜏𝑎,90𝑐𝑎𝑝 = 78.0 s/mi

3,600
𝑆𝑐𝑎𝑙𝑖𝑏,90𝑐𝑎𝑝 = = 44.3 mi/h
(0.85)(78.0) + (0.05)(85.7) + (0.1)(106.8)
Step 6: Compute the Exponent for the Speed–Flow Curve

The exponent for the speed–flow curve is computed from Equation 26-20.
𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,90𝑐𝑎𝑝
ln ( 𝐹𝐹𝑆 − 𝑆 )
mix 𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝
𝜙mix = 1.195 ×
0.9𝐶 − 𝐵𝑃
ln ( 𝐶 mix− 𝐵𝑃 mix )
mix mix
60.1 − 44.3
ln (
𝜙mix = 1.195 × 60.1 − 37.5) = 4.07
1,553 − 0
ln (
1,725 − 0)
Step 7: Compute the Mixed-Flow Speed Under Mixed-Flow Conditions

The mixed-flow speed under mixed-flow conditions is computed by
Equation 26-21.
𝐹𝐹𝑆mix 𝑣mix ≤ 𝐵𝑃mix
𝑆mix =
𝑣mix − 𝐵𝑃mix 𝜙mix
𝐹𝐹𝑆mix − (𝐹𝐹𝑆mix − 𝑆𝑐𝑎𝑙𝑖𝑏,𝑐𝑎𝑝 ) ( ) 𝑣mix > 𝐵𝑃mix
{ 𝐶mix − 𝐵𝑃mix
The mixed-flow rate is 1,500 veh/h/ln, which is greater than the breakpoint.
Therefore,
1,500 − 0 4.07
𝑆mix = 60.1 − (60.1 − 37.5) ( ) = 47.3 mi/h
1,725 − 0
Step 8: Compute the Mixed-Flow Density Under Mixed-Flow Conditions

The final step is to compute the mixed-flow density by using Equation 26-22.
𝑣mix 1,500
𝐷mix = = = 31.7 veh/mi/ln
𝑆mix 47.3
Comparison with the PCE-Based Approach

For comparison purposes, the following procedure show the results for this
case if the PCE-based approach explained in Chapter 12 is applied.
Step 1: Input Data


For basic freeway segments, Equation 12-2 can be used to estimate FFS.
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑓𝐿𝑊 − 𝑓𝑅𝐿𝐶 − 3.22 × 𝑇𝑅𝐷 0.84
For the purposes of comparing the two methods with respect to truck effects
on FFS, the lane width, lateral clearance, and ramp density adjustment factors
can be neglected. Then,
𝐹𝐹𝑆 = 65 − 0 − 0 − 3.22 × 00.84 = 65 mi/h
The adjusted FFS is computed from Equation 12-5, assuming no weather or
incident effects.
𝐹𝐹𝑆𝑎𝑑𝑗 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹
𝐹𝐹𝑆𝑎𝑑𝑗 = 65 × 1 = 65 mi/h

Equation 12-6 is used to compute the capacity of a basic freeway segment.
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (65 − 50) = 2,350 pc/h/ln
Assuming no adverse weather conditions or incidents, the adjusted capacity
from Equation 12-8 is then
𝑐𝑎𝑑𝑗 = 𝑐 × 𝐶𝐴𝐹 = 2,350 × 1 = 2,350 pc/h/ln

This basic freeway segment is in a rural area with more TTs than SUTs.
Therefore, the PCE table for 30% SUTs and 70% TTs (Exhibit 12-26) will be used.
As stated in the Facts section of the example problem, the grade is 5% for 2 mi.
There are no values specifically for a 5% grade in Exhibit 12-26; therefore, PCE
values will be interpolated from the values for 4.5% and 5.5%. As the maximum
grade length provided in the exhibit is 1 mi for these two grades, values for a 1-
mi grade will also apply to longer grades. For a 1-mi, 4.5% grade, the PCE value
for 15% trucks is 3.11; and the PCE value for a 1-mi, 5.5% grade with 15% trucks
is 3.51. Interpolating between these two values for a 5% grade results in a PCE of
3.31.
The heavy-vehicle factor can be computed with Equation 12-10.
1 1
𝑓𝐻𝑉 = = = 0.743
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.15 × (3.31 − 1)
Equation 12-9 is used to adjust the demand volume to account for truck
presence. The freeway is a three-lane facility and the driver population is
assumed to be all local drivers.
𝑉 1,500 × 3
𝑣𝑝 = = = 2,019 pc/h/ln
𝑃𝐻𝐹 × 𝑁 × 𝑓𝐻𝑉 1 × 3 × 0.743

The speed can be read directly from Exhibit 12-7 for a demand flow rate of
2,019 pc/h/ln. Under base conditions, the mean speed of the traffic stream is 59.6
mi/h as calculated from Equation 26-1.
Equation 12-11 is used to compute density.
𝑣𝑃 2,019
𝐷= = = 33.9 pc/mi/ln
𝑆 59.6
If the density above is multiplied by the heavy-vehicle factor, then the
mixed-flow density Dmix can be estimated as follows:
𝐷mix = 𝐷 × 𝑓𝐻𝑉 = 33.9 × 0.743 = 25.2 veh/mi/ln
The PCE-based density of 25.2 veh/mi/ln is about 22% lower than 32.6
veh/mi/ln, which is the density predicted in Step 8 of the mixed-flow model.
Dmix is the mixed-flow density, not an auto-only flow density. As such, it cannot
be used to derive LOS.
EXAMPLE PROBLEM 6: SEVERE WEATHER EFFECTS ON A BASIC

FREEWAY SEGMENT
The Facts
• Four-lane freeway (two lanes in each direction)
• Peak hour, peak direction demand volume = 2,000 veh/h
• PHF = 0.92
• Rolling terrain
• Facility operates under heavy snow conditions (CAF = 0.78; SAF = 0.86).
Comments
The task is to find the expected LOS for this freeway during the worst 15 min
of the peak hour under heavy snow conditions. With one cloverleaf interchange
per mile, the total ramp density will be 4 ramps/mi. This example problem is
similar to Example Problem 1, with the only change being the presence of heavy
snow.
Step 1: Input Data


The FFS of the freeway is estimated from Equation 12-2 as follows:
𝐹𝐹𝑆 = 75.4 − 𝑓𝐿𝑊 − 𝑓𝑅𝐿𝐶 − 3.22 × 𝑇𝑅𝐷 0.84
The adjustment for lane width is selected from Exhibit 12-20 for 11-ft lanes
(1.9 mi/h). The adjustment for right-side lateral clearance is selected from Exhibit
12-21 for a 2-ft clearance on a freeway with two lanes in one direction (2.4 mi/h).
The total ramp density is 4 ramps/mi. Then
𝐹𝐹𝑆 = 75.4 − 1.9 − 2.4 − 3.22 × 40.84 = 60.8 mi/h
A free-flow speed adjustment factor (SAF) for heavy snow conditions can be
obtained from Exhibit 11-5 in Chapter 11, Freeway Reliability Analysis, by
interpolating between the values for 60 and 65 mi/h (0.86 and 0.85, respectively),
resulting in a SAF of 0.86. No other speed adjustments are made, as no incidents
were specified in the problem statement and because the driver population was
specified to be commuters. The SAF is applied through Equation 12-5.
𝐹𝐹𝑆𝑎𝑑𝑗 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 = 60.8 × 0.86 = 52.3 mi/h

Exhibit 11-5 also provides a CAF of 0.78 for heavy snow conditions,
applicable to all FFS values. As with the SAF in Step 2, no other capacity
adjustments apply in this situation. The freeway’s capacity is then estimated
𝑐 = 𝐶𝐴𝐹(2,200 + 10 × [𝐹𝐹𝑆𝑎𝑑𝑗 − 50])
𝑐 = 0.78 × (2,200 + 10 × [52.3 − 50]) = 1,734 pc/h/ln

The demand volume is adjusted by using Equation 12-9 to a flow rate that
reflects passenger cars per hour per lane under equivalent base conditions.
𝑉
𝑣𝑝 =
The demand volume is given as 2,000 veh/h. The PHF is specified to be 0.92,
and there are two lanes in each direction. Trucks make up 5% of the traffic
stream, so a heavy-vehicle adjustment factor must be determined.
From Exhibit 12-25, the PCE for trucks is 3.0 for rolling terrain. The heavy-
vehicle adjustment factor is then computed by using Equation 12-10.
1 1
𝑓𝐻𝑉 = = = 0.909
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.05(3 − 1)
then
2,000
𝑣𝑝 = = 1,195 pc/h/ln
0.92 × 2 × 0.91
Because this value is less than the base capacity of 1,743 pc/h/ln for a freeway
with an FFS of 52.3 mi/h, LOS F conditions do not exist, and the analysis
continues to Step 5.

The FFS of the basic freeway segment is now estimated along with the
conditions. Using the equations provided in Exhibit 12-6, the breakpoint for a
53.5-mi/h FFS speed–flow curve is
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 𝐹𝐹𝑆𝑎𝑑𝑗 )] × (𝐶𝐴𝐹)2
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 52.3)] × (0.78)2 = 1,161 pc/h/ln
Because the flow rate is greater than the breakpoint value, the operating
speed of the segment is estimated from Equation 12-1, by using a value of 2 for
the exponent calibration parameter a from Exhibit 12-6.
𝑐
1,734
(52.3 − 45 ) (1,195 − 1,161)2
𝑆 = 52.3 − = 52.3 mi/h
(1,734 − 1,161)2
The density may now be computed from Equation 12-11.

𝑣𝑝 1,195
𝐷= = = 22.8 pc/mi/ln
𝑆 52.3

From Exhibit 12-15, a density of 22.8 pc/mi/ln corresponds to LOS C.
Discussion
This basic freeway segment of a four-lane freeway is expected to operate at
LOS C during the worst 15 min of the peak hour under heavy snow conditions,
with an average speed of 52.3 mi/h and a density of 22.8 pc/mi/ln. By contrast,
the same facility under no adverse weather conditions would be expected to
operate at an FFS of 60.8 mi/h and a density of 19.7 pc/mi/ln, but still at LOS C.
Although the segment’s performance is affected by the snow, the overall LOS is
unchanged.
However, the segment’s capacity is reduced from 2,308 to 1,734 pc/h/ln,
which means the snow effect would be more severe at elevated volume-to-
capacity ratios, particularly as the segment approached capacity. For elevated
flow rates, the snow condition is expected to result in further deterioration of
speed and breakdown at lower flow rates.
EXAMPLE PROBLEM 7: BASIC MANAGED LANE SEGMENT

The Facts
• Six-lane freeway with two general purpose lanes and one managed lane
in each direction
• Peak hour, peak direction demand volume in the general purpose lanes =
2,000 veh/h (Case 1) or 3,800 veh/h (Case 2)
• Peak hour, peak direction demand volume in the managed lane (both
cases) = 1,300 veh/h
• Continuous access separation between the managed and general purpose
lanes
• FFS = 60 mi/h for both the managed and general purpose lanes
• Traffic composition: 7.5% trucks, using the default truck mix for both the
managed and general purpose lanes
• PHF = 0.92
• Level terrain
weather events).
Comments
The task is to find the expected LOS for this freeway for both the managed
and general purpose lanes during the worst 15 min of the peak hour for the two
described cases. With one cloverleaf interchange per mile, the total ramp density
will be 4 ramps/mi.
Step 1: Input Data


The facility’s FFS is given as 60 mi/h for both the managed and general
purpose lanes. Because the facility is operating under ideal conditions, the SAF
used in Equation 12-5 is 1.

The capacity of the freeway general purpose lanes is estimated from
Equation 12-6 as follows:
𝑐 = 2,200 + 10 × (𝐹𝐹𝑆𝑎𝑑𝑗 − 50)
𝑐 = 2,200 + 10 × (60 − 50) = 2,300 pc/h/ln
As the freeway is operating under ideal conditions, no capacity adjustment is
made for the general purpose lanes (i.e., CAF = 1 in Equation 12-8).
The capacity of the managed lane is calculated with Equation 12-14.
𝑐𝑎𝑑𝑗 = 𝐶𝐴𝐹 × (𝑐75 − 𝜆𝑐 × [75 − 𝐹𝐹𝑆𝑎𝑑𝑗 ])
As with the general purpose lanes, CAF = 1 for the managed lane. The values
of the parameters C75 and λc are obtained from Exhibit 12-30, and are 1,800 and
10, respectively, for continuous access separation. Then
𝑐𝑎𝑑𝑗 = 1.00 × (1,800 − 10 × [75 − 60]) = 1,650 pc/h/ln

The demand volume is adjusted by using Equation 12-9 to a flow rate that
reflects passenger cars per hour per lane under equivalent base conditions.
𝑉
𝑣𝑝 =
The demand volume is given as 2,000 veh/h and 3,800 veh/h for Cases 1 and
2, respectively. The PHF is specified to be 0.92, and there are two lanes in each
direction. Trucks make up 5% of the traffic stream, so a heavy-vehicle adjustment
factor must be determined.
From Exhibit 12-25, the PCE for trucks is 2.0 for level terrain. The heavy-
vehicle adjustment factor is then computed using Equation 12-10.
1 1
𝑓𝐻𝑉 = = 0.93
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.075(2.0 − 1)
Then for Case 1,
2,000
𝑣𝑝,𝐺𝑃,Case1 = = 1,169 pc/h/ln
0.92 × 2 × 0.93
and for Case 2,

3,800
𝑣𝑝,𝐺𝑃,Case2 = = 2,221 pc/h/ln
0.92 × 2 × 0.93
The flow rate on the managed lane is
1,300
𝑣𝑝,𝑀𝐿 = = 1,519 pc/h/ln
0.92 × 1 × 0.93
Because all the flow rates are less than their corresponding capacities, LOS F
conditions do not exist, and the analysis continues to Step 5.

The FFS of the basic freeway segment is now estimated, along with the
conditions. Based on the equations provided in Exhibit 12-6, the breakpoint for a
60-mi/h FFS speed–flow curve is
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 𝐹𝐹𝑆𝑎𝑑𝑗 )] × (𝐶𝐴𝐹)2
𝐵𝑃𝑎𝑑𝑗 = [1,000 + 40 × (75 − 60)] × (1.00)2 = 1,600 pc/h/ln
In Case 1, the flow rate is less than the breakpoint value of 1,600 pc/h/ln. As
this flow rate is in the constant-speed portion of the curve, SGP,Case1 = 60 mi/h. The
density of the traffic stream is computed from Equation 12-11.
𝑣𝑝 1,169
𝐷𝐺𝑃,Case1 = = = 19.5 pc/mi/ln
𝑆 60
In Case 2, the flow rate is higher than the breakpoint. Therefore, the speed is
computed with Equation 12-1, by using a value of 2 for the exponent calibration
parameter a from Exhibit 12-6, as follows:
(𝐹𝐹𝑆𝑎𝑑𝑗 − ) (𝑣𝑝 − 𝐵𝑃)
𝐷𝑐
2,300
(60 − 45 ) (2,221 − 1,600)2
𝑆𝐺𝑃,Case2 = 60 − = 53.0 mi/h
(2,300 − 1,600)2
Density is computed with Equation 12-11.
𝑣𝑝 2,221
𝐷𝐺𝑃,Case2 = = = 41.9 pc/mi/ln
𝑆 53
To compute the managed lane speed, the breakpoint first needs to be
computed by using Equation 12-13 and values for the parameters BP75 and λBP
from Exhibit 12-30.
𝐵𝑃𝑀𝐿 = [𝐵𝑃75 + 𝜆𝐵𝑃 × (75 − 𝐹𝐹𝑆𝑎𝑑𝑗 )] × 𝐶𝐴𝐹2
𝐵𝑃𝑀𝐿 = [500 + 0 × (75 − 60)] × (1.00)2 = 500 pc/h/ln
Because the managed lane flow rate is higher than the breakpoint, three
speeds, S1, S2, and S3, need to be computed by using Equations 12-15, 12-17, and
12-19, respectively (with parameters from Exhibit 12-30), as follows:
𝑆1 = 𝐹𝐹𝑆𝑎𝑑𝑗 − 𝐴1 × min(𝑣𝑝 , 𝐵𝑃) = 60 − 0 × min(1,519, 500) = 60 mi/h

𝑐𝑎𝑑𝑗
(𝑆1,𝐵𝑃 − 𝑛𝑓 )
𝐾𝑐 𝐴2
𝑆2 = 𝐴2 (𝑣𝑝 − 𝐵𝑃)
1,650
(60 − 30 )
𝑆2 = (1,519 − 500)2.5 = 3.7 mi/h
(1,650 − 500)2.5
𝑐𝑎𝑑𝑗 𝑐𝑎𝑑𝑗
( 𝑛𝑓 ) − ( 𝑓 )
𝐾𝑐 𝐾𝑐 2
𝑆3 = 2 (𝑣𝑝 − 𝐵𝑃)
1,650 1,650
( 30 ) − ( )
𝑆3 = 45 (1,519 − 500)2 = 14.4 mi/h
(1,650 − 500) 2
The space mean speed of the managed lane is given by Equation 12-12.
𝑆1 𝑣𝑝 ≤ 𝐵𝑃
𝑆𝑀𝐿 = {
𝑆1 − 𝑆2 − 𝐼𝑐 × 𝑆3 𝐵𝑃 < 𝑣𝑝 ≤ 𝑐
Because the managed lane’s demand flow of 1,519 pc/h/ln is greater than the
breakpoint value of 500 pc/h/ln calculated in Step 4, the second of the two
equations applies. To apply this equation, the value of the indicator variable Ic
must first be determined from Equation 12-18.
0 𝐾𝐺𝑃 ≤ 35 pc/mi/ln
𝐼𝑐 = { or segment type is Buffer 2, Barrier 1, or Barrier 2
1 otherwise
In Case 1, the density of the adjacent general purpose lane is less than 35
pc/mi/ln, as determined in Step 5. As a result, the indicator variable 𝐼𝑐 will have a
value of zero. Thus, the managed lane speed in Case 1 will be
𝑆𝑀𝐿,Case1 = 60 − 3.7 − (0 × 14.4) = 56.3 mi/h
In Case 2, the density of the adjacent general purpose lane is greater than 35
pc/ln/mi, and therefore the indicator variable 𝐼𝑐 will have a value of 1. The
managed lane speed in Case 2 will be
𝑆𝑀𝐿,Case2 = 60 − 3.7 − (1 × 14.4) = 41.9 mi/h
The managed lane density for the two cases is given by Equation 12-11.
𝑣𝑝 1,519
𝐷𝑀𝐿,Case1 = = = 27.0 pc/mi/ln
𝑆 56.3
𝑣𝑝 1,519
𝐷𝑀𝐿,Case2 = = = 36.3 pc/mi/ln
𝑆 41.9

The managed lane facility’s density of 27.0 pc/mi/ln under Case 1
corresponds to LOS D, but it is close to the LOS C boundary, which has a
maximum value of 26 pc/mi/ln. In Case 2, the density of 36.3 pc/mi/ln
corresponds to LOS E.
Discussion
In this example, the managed lane’s operating speed and density have been
investigated for two operating conditions in the general purpose lanes. When
high-density conditions exist in the general purpose lanes, the managed lane’s
operational speed is reduced and, as a consequence, the managed lane operates
at a worse LOS than when lower-density conditions exist in the general purpose
lanes.
8. TWO-LANE HIGHWAY EXAMPLE PROBLEMS
Exhibit 26-22 lists the five example problems provided in this section. The
problems demonstrate the computational steps involved in applying the two-
lane highway automobile and bicycle methodologies.
Exhibit 26-22 Problem

List of Two-Lane Highway Number Description Type of Analysis
Example Problems 1 Level, straight, Passing Constrained segment Operational analysis
2 Passing Constrained segment with horizontal curves Operational analysis
3 Facility analysis in level terrain Operational analysis
4 Facility analysis in mountainous terrain Operational analysis
5 Bicycle LOS Planning analysis
EXAMPLE PROBLEM 1: LEVEL, STRAIGHT, PASSING CONSTRAINED

SEGMENT
This example problem illustrates the computation of the LOS in a single
direction on a straight, 0.75-mi-long Passing Constrained segment in level terrain.
This example problem follows the flowchart of analysis steps outlined in Exhibit
15-9.
The Facts
The segment has the following known characteristics:
• Segment length = 3,960 ft (0.75 mi);
• Segment type = Passing Constrained;
• No upstream passing lanes;
• Vehicle count in the analysis direction = 752 veh/h;
• PHF = 0.94;
• Posted speed limit: 50 mi/h;
• Percent heavy vehicles (%HV) = 5%;
• Percent grade = 0%;
• Horizontal curvature = none;
• Lane width = 12 ft;
• Shoulder width = 6 ft; and
• Access points = 0.
Objective
Estimate the LOS in the subject direction on the two-lane highway segment
as described.
Step 1: Identify Facility Study Boundaries and Segmentation

The limits of the segment were identified following the guidance given in
Step 1 on page 15-15. The characteristics of this segment were determined by
examination to be essentially homogenous. These characteristics included the
Two-Lane Highway Example Problems Chapter 26/Freeway and Highway Segments: Supplemental
ability to pass, lane geometry, grades, lane and shoulder widths, posted speed
limit, traffic demands, adjacent land uses, and driveways.
A field examination of the segment determined that it met the definition of a
Passing Constrained segment, being a segment in which “passing in the
oncoming lane is either prohibited or is effectively negligible due to geometric or
sight distance limitations.”
Step 2: Determine Demand Flow Rates, Capacity, and d/c Ratio

In this step, the hourly demand volume at the upstream entrance of the
directional segment being evaluated is converted to a peak 15-min flow rate by
applying the peak hour factor (PHF) using Equation 15-1:
𝑉𝑑 752
𝑣𝑑 = = = 800 veh/h
𝑃𝐻𝐹 0.94
The capacity of a Passing Constrained segment is 1,700 veh/h, as stated in the
description of Step 2 on page 15-18. The demand flow rate is less than capacity;
therefore the calculation process proceeds to Step 3. Note that it is only necessary
to compute the actual opposing flow rate for Passing Zone segments.
Step 3: Determine Vertical Alignment Classification

The segment is assigned a vertical alignment classification of 1, based on
Exhibit 15-11 for a level (0% grade), 0.75-mi-long segment. From Exhibit 15-10,
the segment length of 0.75 mi is between the minimum (0.25 mi) and maximum
(3.0 mi) lengths for a Passing Constrained segment of vertical class 1, and
therefore no adjustment is needed to the segment length.
Step 4: Determine the Free-Flow Speed

As stated on page 15-18, direct field measurement of FFS is preferred. In this
case it is not feasible to measure the FFS, so it will be estimated using the
procedure given in Step 4.
Because this is a Passing Constrained segment, the opposing flow rate vo is
set at 1,500 veh/h in Equation 15-4 for the purposes of computing FFS, regardless
of the actual opposing flow rate. First, the base free-flow speed BFFS is estimated
using Equation 15-2. Next, Equation 15-4 through Equation 15-6 are used to
determine factors relating to lane and shoulder width, access-point density, and
heavy-vehicle percentage, which are used in the estimation of FFS. Finally, the
FFS is estimated by Equation 15-3.
𝐵𝐹𝐹𝑆 = 1.14 × 𝑆𝑝𝑙 = 1.14 × 50 = 57.0 mi/h
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)

𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 6) = 0
𝐴𝑃𝐷 0
𝑓𝐴 = min ( , 10) = min ( , 10) = 0
4 4
𝑎 = max[0.0333, 0] = 0.0333 Coefficients a0 through a5 are
all 0 for vertical class 1 (Exhibit
15-12), so the right side of
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑎(𝐻𝑉%) − f LS − f A Equation 15-6 reduces to 0.
Chapter 26/Freeway and Highway Segments: Supplemental Two-Lane Highway Example Problems
𝐹𝐹𝑆 = 57.0 − (0.0333)(5) − 0 − 0 = 56.83 mi/h

There are no geometry-related FFS adjustments for 12-ft lanes and 6-ft
shoulders. There are also no adjustments for access points (driveways or streets),
because this segment has no access points.
Step 5: Estimate the Average Speed

Because the demand flow rate in the subject direction is greater than 100
veh/h, the equations given in Step 5 are used to estimate the average speed.
Step 5a: Calculate the Slope Coefficient

The slope coefficient m determines how rapidly the average speed is
estimated to decrease as a function of the entering flow rate. It is computed as a
function of six coefficients b0 to b5, which are obtained from Exhibit 15-13 for a
Passing Constrained segment. For a segment of vertical class 1, these coefficients
are 0.0558, 0.0542, 0.3278, 0.1029, 0, and 0, respectively. Equation 15-8 is then
used to determine the slope coefficient.
𝑣𝑜
𝑚 = max [𝑏5 , 𝑏0 + 𝑏1 × 𝐹𝐹𝑆 + 𝑏2 × √ + max(0, 𝑏3 ) × √𝐿
1,000
+ max(0, 𝑏4 ) × √𝐻𝑉% ]
1,500
𝑚 = max [0,0.0558 + 0.0542 × 56.83 + 0.3278 × √
1,000
+ max(0, 0.1029) × √0.75 + max(0, 0) × √5 ]
𝑚 = 3.626
Step 5b: Calculate the Power Coefficient

The power coefficient p is used to estimate how fast the average speed
decreases at higher flow rates. The equation uses nine coefficients f0 to f8, which
are obtained from Exhibit 15-19 for a Passing Constrained segment. For a
segment of vertical class 1, all of these coefficients take on values of 0, except for
f0 (0.67576), f3 (0.12060), and f4 (−0.35919). Equation 15-11 is then used to
determine the power coefficient.
𝑣𝑜 𝑣𝑜
𝑝 = max [𝑓8 , 𝑓0 + 𝑓1 × 𝐹𝐹𝑆 + 𝑓2 × 𝐿 + 𝑓3 × + 𝑓4 × √ + 𝑓5 × 𝐻𝑉%
1,000 1,000
+ 𝑓6 × √𝐻𝑉% + 𝑓7 × (𝐿 × 𝐻𝑉%)]
1,500
𝑝 = max [0,0.67576 + 0 × 56.83 + 0 × 0.75 + 0.12060 ×
1,000
1,500
− 0.35919 × √ + 0 × 5 + 0 × √5 + 0 × (0.75 × 5)]
1,000
𝑝 = 0.41676
Step 5c: Calculate Average Speed for the Segment

The average speed for the given entry flow rate is computed in Step 5-3
using Equation 15-7. The previously computed flow rate vd, slope coefficient m,
and power coefficient p are used in this equation.
𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.41676
800
𝑆 = 56.83 − 3.626 ( − 0.1)
1,000
𝑆 = 53.7 mi/h
Step 5d: Adjust Speed for Horizontal Alignment

Because this is a straight segment, no adjustment to the speed estimate is
required for horizontal alignment.
Assessment of Speed Results

The average speed at a directional flow rate of 800 veh/h is estimated to be
approximately 3 mi/h (about 5%) lower than the FFS, but about 4 mi/h (about
7%) higher than the posted speed limit.
Step 6: Estimate the Percent Followers

The service measure percent followers is estimated in Step 6 for the given flow
rate vd. under the prevailing geometric conditions and percent heavy vehicles.
First, the percent followers at 100% of capacity is estimated. Next, the percent
followers at 25% of capacity is estimated. Then, the slope and power coefficients
for an exponential curve fitting those two points (percent followers at 25% and
100% capacity) are estimated. Finally, the fitted curve is used to estimate the
percent followers for the given flow rate.
Step 6a: Compute Percent Followers at Capacity

Percent followers at capacity is calculated using Equation 15-18, applying
eight parameters b0 to b7 obtained from Exhibit 15-24.
𝑃𝐹𝑐𝑎𝑝 = 𝑏0 + 𝑏1 (𝐿) + 𝑏2 (√𝐿) + 𝑏3 (𝐹𝐹𝑆) + 𝑏4 (√𝐹𝐹𝑆) + 𝑏5 (𝐻𝑉%) +

𝑜𝑣 𝑣
0
𝑏6 (𝐹𝐹𝑆 × 1,000 ) + 𝑏7 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(0.75) − 7.90866(√0.75) − 0.94321(56.83) +

1,500 1,500
13.64266(√56.83) − 0.00050(5) − 0.05500 (56.83 × 1,000) + 7.1376 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 86.41%
Step 6b: Compute Percent Followers at 25% Capacity

Percent followers at 25 percent of capacity is calculated using Equation 15-20,
applying eight parameters c0 to c7 obtained from Exhibit 15-26.
𝑃𝐹25𝑐𝑎𝑝 = 𝑐0 + 𝑐1 (𝐿) + 𝑐2 (√𝐿) + 𝑐3 (𝐹𝐹𝑆) + 𝑐4 (√𝐹𝐹𝑆) + 𝑐5 (𝐻𝑉%)

𝑣𝑜 𝑣0
+ 𝑐6 (𝐹𝐹𝑆 × ) + 𝑐7 (√ )
1,000 1,000
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(0.75) − 21.60000(√0.75) − 0.97853(56.83)

1,500
+ 12.05214(√56.83) − 0.00750(5) − 0.06700 (56.83 × )
1,000
1,500
+ 11.6041 (√ )
1,000
𝑃𝐹25𝑐𝑎𝑝 = 50.52%
Step 6c: Calculate the Slope Coefficient

Equation 15-22 is used to compute the slope coefficient m for an exponential
curve fitted between percent following at capacity and percent following at 25%
capacity. It employs two parameters d1 and d2 obtained from Exhibit 15-28; the
parameters for Passing Constrained segments are used.
𝑃𝐹25𝑐𝑎𝑝 𝑃𝐹𝑐𝑎𝑝
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = 𝑑1 ( 𝑐𝑎𝑝 ) + 𝑑2 ( 𝑐𝑎𝑝 )
0.25 [1,000] [1,000]
50.52 86.41
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( )
1,700 1,700
0.25 [1,000] [1,000]
𝑚 = −1.337
Step 6d: Calculate the Power Coefficient

Equation 15-23 is used to compute the power coefficient p for an exponential
capacity. It employs five parameters e0 through e4 obtained from Exhibit 15-29;
the parameters for Passing Constrained segments are used.
0 − ln [1 − ] 0 − ln [1 − ]
100 100
𝑝 = 𝑒0 + 𝑒1 ( 𝑐𝑎𝑝 ) + 𝑒2 ( 𝑐𝑎𝑝 )
0.25 [ ] [ ]
1,000 1,000
0 − ln (1 − ) 0 − ln (1 − )
100 100
+ 𝑒3 √ 𝑐𝑎𝑝 + 𝑒4
√
𝑐𝑎𝑝
0.25 [ ] [ ]
1,000 1,000
50.52 86.41
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
50.52 86.41
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.7524
Step 6e: Calculate Percent Followers

Equation 15-17 is used to compute percent followers PF.
𝑣𝑑 𝑝
(𝑚 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
800 0.7524
(−1.337 ×{ } )
1,000
𝑃𝐹 = 100 × [1 − 𝑒 ]
𝑃𝐹 = 67.7%
Step 7: Calculate Additional Performance Measure Values for a Passing

Lane Segment
This step is only applicable to passing lane segments. Therefore, Step 7 is
skipped.
Step 8: Calculate Follower Density

Follower density FD is estimated using Equation 15-35.
𝑃𝐹 𝑣𝑑 67.7 800
𝐹𝐷 = × = × = 10.1 followers/mi/ln
100 𝑆 100 53.7
Step 9: Determine Potential Adjustment to Follower Density

There is no passing lane upstream of the analysis segment. Therefore, no
adjustment is needed to follower density and Step 9 is skipped.

The segment’s LOS is determined from Exhibit 15-6, using the column for a
higher-speed highway (posted speed limit equal to or greater than 50 mi/h). With
10.1 followers/mi, the subject direction of travel on the segment operates at
LOS D.
Discussion
The estimated FFS and average speed for a flow rate of 800 veh/h are both
above the posted speed limit. This result is reasonable for a flat, straight segment
in this volume range. However, the follower density produces LOS D operations.
This flow rate is large enough to produce fairly high levels of platooning, but not
so high as to cause significant reductions in speed. The combination of a
moderately high flow rate and moderately high level of platooning will result in
travelers perceiving a relatively poor level of service.
EXAMPLE PROBLEM 2: PASSING CONSTRAINED SEGMENT WITH

HORIZONTAL CURVES
This example problem illustrates the computation of LOS in a single direction
of a curving, 0.75-mi-long Passing Constrained segment in level terrain.
The Facts
The segment to be evaluated has the same general demand and geometric
characteristics as the segment evaluated in Example Problem 1. The difference is
that this segment has horizontal curvature instead of being straight; otherwise,
the same inputs are used as for Example Problem 1.
The segment is split into 11 subsegments, with each subsegment being either
straight (tangent) or curved. Horizontal curvature data for each subsegment is
Exhibit 26-23
Example Problem 2:
Horizontal Curve Inputs
Super- Central
Length elevation Radius Angle Horizontal
Subsegment Type (ft)a (%) (ft) (deg) Classb
1 Tangent 280 -- -- -- --
2 Horizontal curve 432 3 450 55 3
3 Tangent 260 -- -- -- --
4 Horizontal curve 366.5 2 300 70 4
5 Tangent 250 -- -- -- --
7 Tangent 275.6 -- -- -- --
9 Tangent 285 -- -- -- --
10 Horizontal curve 767.9 4 1,100 40 1
11 Tangent 369 -- -- -- --
Total 3,960
a
Length for horizontal curves = radius × central angle × π/180.
b
Determined from Exhibit 15-22, with radius and superelevation as inputs.
Objective
Estimate the average speed in the subject direction on the two-lane highway
segment, taking into account the effects of horizontal alignment on the average
speed.

This step was completed in Example Problem 1. The horizontal alignment
does not affect the selection of study boundaries and segmentation.

does not affect the computation of the demand flow rates.
Step 3: Determine the Vertical Alignment Classification

does not affect the determination of the vertical alignment classification.

This step was completed in Example Problem #1. The computed base FFS
applies to the tangent subsegments.

Steps 5a, 5b, and 5c were applied in Example Problem 1 to obtain the
following speed results for the tangent subsegments:
• Base free-flow speed BFFS = 57.0 mi/h,
• Tangent free-flow speed FFS = 56.8 mi/h, and
• Average speed for tangent subsegments S = 53.7 mi/h.

In this step, the average speed for each subsegment with a horizontal curve
is determined. There are three substeps: (a) identifying the horizontal alignment
classification for each subsegment with a horizontal curve, (b) calculating the
average speed for each subsegment with a horizontal curve, and (c) calculating
the adjusted average speed for the segment.
Step 5d.1: Identify all Horizontal Curves Within the Segment

In this step the tangent (straight) length, curve radius, and superelevation are
identified for each horizontal curve within the segment. Each curve is assigned a
horizontal alignment classification on the basis of its radius and percent
superelevation, using Exhibit 15-22. The resulting horizontal classes were
indicated previously in Exhibit 26-23. Note that in typical designs, the crown of
the roadway (designed to shed water from the paved way) will cause the
superelevation to vary by direction of travel. Therefore, a curve’s horizontal class
may also vary by direction of travel.
Step 5d.2: Calculate Average Speed for each Horizontal Curve Within the
Segment
The average speed for a subsegment with horizontal curvature is determined
using Equation 15-12 though Equation 15-15. The process is demonstrated for
Subsegment 2.
Subsegment 2 has a horizontal alignment class of 3 and the BFFS for the
preceding tangent section is 57.0 mi/h. Equation 15-14 is applied to compute the
base free-flow speed for Subsegment 2:
𝐵𝐹𝐹𝑆𝐻𝐶2 = min(𝐵𝐹𝐹𝑆𝑇 , 44.32 + 0.3728 × 𝐵𝐹𝐹𝑆𝑇 − 6.868 × 𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠2 )
𝐵𝐹𝐹𝑆𝐻𝐶2 = min(57.0, 44.32 + 0.3728 × 57.0 − 6.868 × 3)
𝐵𝐹𝐹𝑆𝐻𝐶2 = min(57.0, 44.9656) = 44.9656 mi/h

Subsegment 2’s FFS is computed using Equation 15-13:
𝐹𝐹𝑆𝐻𝐶2 = 𝐵𝐹𝐹𝑆𝐻𝐶2 − 0.0255 × 𝐻𝑉%
𝐹𝐹𝑆𝐻𝐶2 = 44.9656 − 0.0255 × 5
𝐹𝐹𝑆𝐻𝐶2 = 44.8381 𝑚𝑖/ℎ
The slope coefficient m used in the determination of average speed is
computed using Equation 15-15 as follows:
𝑚 = max(0.277, −25.8993 − 0.7756 × 𝐹𝐹𝑆𝐻𝐶2 + 10.6294 × √𝐹𝐹𝑆𝐻𝐶2
+ 2.4766 × 𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠2 − 9.8238 × √𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠2 )
𝑚 = max(0.277, −25.8993 − 0.7756 × 44.8381 + 10.6294 × √44.8381

+ 2.4766 × 3 − 9.8238 × √3)
𝑚 = max(0.277, 0.9145) = 0.9145
Finally, the average speed of Subsegment 2 is computed by Equation 15-12.
𝑣𝑑
𝑆𝐻𝐶2 = min (𝑆, 𝐹𝐹𝑆𝐻𝐶2 − 𝑚 × √ − 0.1)
1,000
800
𝑆𝐻𝐶2 = min (53.7, 44.8381 − 0.9145 × √ − 0.1)
1,000
𝑆𝐻𝐶2 = min(53.7, 44.0730) = 44.1 mi/h

Similar computations are performed for the other subsegments with
horizontal curves. The results are presented in Exhibit 26-24.
Exhibit 26-24 Subsegment BFFSHci (mi/h) FFSHci (mi/h) m SHci (mi/h)

Example Problem 2: 2 44.9656 44.8381 0.9145 44.1
Horizontal Curve Average 4 38.0976 37.9701 0.4081 37.6
Speed Results 6 31.2296 31.1021 0.2770 30.9
8 51.8336 51.7061 1.4905 50.5
10 57.0000 56.8725 2.8036 53.7
Step 5d.3: Calculate Adjusted Average Speed for the Segment

The speed results for all subsegments are summarized in Exhibit 26-25.
Exhibit 26-25 Subsegment Type Speed (mi/h) Length (ft)

Example Problem 2: Average 1 Tangent 53.7 280
Speeds by Subsegment 2 Horizontal curve 44.1 432
3 Tangent 53.7 260
4 Horizontal curve 37.6 366.5
5 Tangent 53.7 250
6 Horizontal curve 30.9 216
7 Tangent 53.7 275.6
8 Horizontal curve 50.5 458
9 Tangent 53.7 285
10 Horizontal curve 53.7 767.9
11 Tangent 53.7 369
Total 3,960
Equation 15-16 is used to calculate the segment’s adjusted average speed by

taking a length-weighted average of the subsegment speeds.
∑11
𝑖=1(𝑆𝑢𝑏𝑠𝑒𝑔𝑆𝑝𝑒𝑒𝑑𝑖 × 𝑆𝑢𝑏𝑠𝑒𝑔𝐿𝑒𝑛𝑔𝑡ℎ𝑖 )
𝑆=
𝐿
(53.7 × 280) + (44.1 × 432) + (53.7 × 260) + (37.6 × 366.5)
[+(53.7 × 250) + (30.9 × 216) + (53.7 × 275.6) + (50.5 × 458)]
+(53.7 × 285) + (53.7 × 767.9) + (53.7 × 369)
𝑆=
3,960
𝑆 = 49.5 mi/h
Discussion
Compared to the straight segment studied in Example Problem 1, the
horizontal curvature in the segment studied in Example Problem 2 reduces the
average speed from 53.7 mi/h to 49.5 mi/h, which is close to the segment’s posted
speed limit of 50 mi/h.
EXAMPLE PROBLEM 3: FACILITY ANALYSIS—LEVEL TERRAIN

This example problem illustrates the computation of LOS in the eastbound
direction of a straight, 5.5-mi-long two-lane highway in level terrain. The study
facility includes a 1.5-mi-long passing lane and a 0.5-mi-long passing zone.
The Facts
The input data for the eastbound direction of the facility are provided in
Exhibit 26-26.
Exhibit 26-26
Example Problem 3:
Input Data
Posted Directional Opposing Peak Heavy

Horizontal Vertical Speed Limit Volume Volume Hour Vehicles
Segment Class Class (mi/h) (veh/h) (veh/h) Factor (%)
1 1 1 55 850 * 0.94 8
2 1 1 55 825 * 0.95 8
3 1 1 55 820 * 0.95 8
4 1 1 55 800 500 0.94 7.5
5 1 1 55 795 * 0.935 8
Note: *Only required for Passing Zone segments.
The facility has the following additional characteristics:

• Facility length = 29,040 ft (5.5 mi);
• No upstream passing lanes;
• Percent grade = 0%;
• Horizontal curvature = none;
• Lane width = 12 ft in all segments;
• Paved shoulder width = 6 ft in all segments; and

• Access points = 0 in all segments (for simplification of this problem,
despite variation in segment volume).
Objective
Estimate the LOS in the subject direction on the two-lane highway segment,
taking into account the effects of the passing lane and the passing zone.

The facility was divided into homogeneous segments following the guidance
given in Step 1. The characteristics considered when segmenting the facility
included the ability to pass, lane geometry, grades, lane and shoulder widths,
posted speed limit, traffic demands, adjacent land uses, and driveways. Each
segment was designated as a Passing Constrained, Passing Zone, or Passing Lane
segment following the guidance on Segmentation given on page 15-4. The
resulting segment lengths and designations were shown in Exhibit 26-26.
Steps 2–9 of the two-lane highway analysis procedure are now followed for
each of the facility’s five segments, starting with the most upstream segment
(Segment 1) and proceeding in sequence to the downstream segments.
Segment 1: Passing Constrained Segment

Equation 15-1 is used to convert the segment’s hourly demand volume to a
peak 15-min flow rate:
𝑉𝑑 850
𝑣𝑑 = = = 904 veh/h
𝑃𝐻𝐹 0.94
description of Step 2 on page 15-18. The demand flow rate is less than capacity;
therefore the calculation process proceeds to Step 3.

According to Exhibit 15-11, a segment with a level grade is assigned a
vertical alignment classification of 1. From Exhibit 15-10, the segment length of
0.75 mi is between the minimum (0.25 mi) and maximum (3.0 mi) lengths for a
Passing Constrained segment of vertical class 1, and therefore no adjustment is
needed to the segment length.

Because Segment 1 is a Passing Constrained segment, the opposing flow rate
vo is set at 1,500 veh/h in Equation 15-4 for the purposes of computing FFS. First,
the base free-flow speed BFFS is estimated using Equation 15-2. Next, Equation
15-4 through Equation 15-6 are used to determine factors relating to lane and
shoulder width, access-point density, and heavy-vehicle percentage, which are
used in the estimation of FFS. Finally, the FFS is estimated by Equation 15-3.
𝐵𝐹𝐹𝑆 = 1.14 × 𝑆𝑝𝑙 = 1.14 × 55 = 62.7 mi/h
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)

𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 6) = 0
𝐴𝑃𝐷 0
𝑓𝐴 = min ( , 10) = min ( , 10) = 0
4 4
𝑎 = max[0.0333, 0] = 0.0333 Coefficients a0 through a5 are
all 0 for vertical class 1 (Exhibit
15-12), so the right side of
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑎(𝐻𝑉%) − f LS − f A Equation 15-6 reduces to 0.
𝐹𝐹𝑆 = 62.7 − (0.0333)(8) − 0 − 0 = 62.43 mi/h

veh/h, the equations given in Step 5 are used to estimate the average speed.

The slope coefficient m is computed using six coefficients b0 to b5, which are
obtained from Exhibit 15-13 for a Passing Constrained segment. For a segment of
vertical class 1, these coefficients are 0.0558, 0.0542, 0.3278, 0.1029, 0, and 0,
respectively. Equation 15-8 is then used to determine the slope coefficient.
𝑣𝑜
1,000
+ max(0, 𝑏4 ) × √𝐻𝑉% ]
1,500
𝑚 = max [0,0.0558 + 0.0542 × 62.43 + 0.3278 × √
1,000
+ max(0, 0.1029) × √0.75 + max(0, 0) × √8 ]
𝑚 = 3.930

The power coefficient p is computed using nine coefficients f0 to f8, which are
obtained from Exhibit 15-19 for a Passing Constrained segment. For a segment of
vertical class 1, all of these coefficients take on values of 0, except for f0 (0.67576),
f3 (0.12060), and f4 (−0.35919). Equation 15-11 is then used to calculate p.
𝑣𝑜 𝑣𝑜
1,000 1,000
+ 𝑓6 × √𝐻𝑉% + 𝑓7 × (𝐿 × 𝐻𝑉%)]
1,500
𝑝 = max [0,0.67576 + 0 × 62.43 + 0 × 0.75 + 0.12060 ×
1,000
1,500
− 0.35919 × √ + 0 × 8 + 0 × √8 + 0 × (0.75 × 8)]
1,000
𝑝 = 0.417

The average speed for Segment 1 is calculated using Equation 15-7.
𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.417
904
𝑆 = 62.43 − 3.930 ( − 0.1)
1,000
𝑆 = 58.8 mi/h

Because Segment 1 is straight, no adjustment to the speed estimate is

Percent followers at capacity for a Passing Constrained segment is calculated
using Equation 15-18, applying eight parameters b0 to b7 obtained from Exhibit
15-24.
𝑣
𝑜 𝑣
0
𝑏6 (𝐹𝐹𝑆 × 1,000 ) + 𝑏7 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(0.75) − 7.90866(√0.75) − 0.94321(62.43) +

1,500 1,500
13.64266(√62.43) − 0.00050(8) − 0.05500 (62.43 × 1,000) + 7.1376 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 85.62%

Percent followers at 25 percent of capacity for a Passing Constrained segment
is calculated using Equation 15-20, applying eight parameters c0 to c7 obtained
from Exhibit 15-26.
𝑣𝑜 𝑣0
+ 𝑐6 (𝐹𝐹𝑆 × ) + 𝑐7 (√ )
1,000 1,000
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(0.75) − 21.60000(√0.75) − 0.97853(62.43)

1,500
+ 12.05214(√62.43) − 0.00750(8) − 0.06700 (62.43 × )
1,000
1,500
+ 11.6041 (√ )
1,000
𝑃𝐹25𝑐𝑎𝑝 = 48.83%

0 − ln [1 − 0 − ln [1 −
100 ] 100 ]
𝑚 = 𝑑1 ( 𝑐𝑎𝑝 ) + 𝑑2 ( 𝑐𝑎𝑝 )
0.25 [1,000] [1,000]
48.83 85.62
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( )
1,700 1,700
0.25 [1,000] [1,000]
𝑚 = −1.289

0 − ln [1 − ] 0 − ln [1 − ]
100 100
𝑝 = 𝑒0 + 𝑒1 ( 𝑐𝑎𝑝 ) + 𝑒2 ( 𝑐𝑎𝑝 )
0.25 [ ] [ ]
1,000 1,000
0 − ln (1 − ) 0 − ln (1 − )
100 100
+ 𝑒3 √ 𝑐𝑎𝑝 + 𝑒4
√
𝑐𝑎𝑝
0.25 [ ] [ ]
1,000 1,000
48.83 85.62
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
48.83 85.62
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.767

𝑣𝑑 𝑝
(𝑚 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
904 0.767
(−1.289 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
𝑃𝐹 = 69.7%
Step 7: Calculate Additional Performance Measure Values for a Passing Lane

Segment
skipped.

𝑃𝐹 𝑣𝑑 69.7 904
100 𝑆 100 58.8

There is no passing lane upstream of Segment 1. Therefore, no adjustment is
needed to follower density and Step 9 is skipped.

The segment’s LOS is determined from Exhibit 15-6, using the column for a
higher-speed highway (posted speed limit equal to or greater than 50 mi/h). With
10.7 followers/mi/ln, Segment 1 operates at LOS D in the analysis direction of
travel.
Segment 2: Passing Lane Segment

𝑉𝑑 825
𝑣𝑑 = = = 868 veh/h
𝑃𝐻𝐹 0.95
From Exhibit 15-5, the capacity of a Passing Lane segment of vertical class 1
with 8% heavy vehicles is 1,500 veh/h. The demand flow rate is less than
capacity; therefore the calculation process proceeds to Step 3.

Passing Lane segment of vertical class 1, and therefore no adjustment is needed
to the segment length.

Because Segment 2 is a Passing Lane segment, the opposing flow rate vo is set
at 0 veh/h in Equation 15-4 for the purposes of computing FFS. Otherwise, the
computations are the same as for Segment 1.
𝐵𝐹𝐹𝑆 = 1.14 × 𝑆𝑝𝑙 = 1.14 × 55 = 62.7 mi/h
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)

𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 6) = 0
𝐴𝑃𝐷 0
𝑓𝐴 = min ( , 10) = min ( , 10) = 0
4 4
𝑎 = max[0.0333, 0] = 0.0333
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑎(𝐻𝑉%) − f LS − f A
𝐹𝐹𝑆 = 62.7 − (0.0333)(8) − 0 − 0 = 62.43 mi/h

The slope coefficient m is computed by Equation 15-8, applying six
coefficients b0 to b5, which are obtained from Exhibit 15-14 for a Passing Lane
segment. This exhibit references Equation 15-9 to calculate b3 and Equation 15-10
to calculate b4, and directly provides numerical values for the other coefficients.
The calculation of the b3 coefficient requires four additional coefficients, c0
through c3, which are obtained from Exhibit 15-16 for Passing Lane segments.
The segment has vertical class 1; therefore c1 takes a value of 0.2667, while the
other coefficients are 0.
𝑏3 = 𝑐0 + 𝑐1 × √𝐿 + 𝑐2 × 𝐹𝐹𝑆 + 𝑐3 × (𝐹𝐹𝑆 × √𝐿)
𝑏3 = 0 + 0.2667 × √1.5 + 0 × 62.43 + 0 × (62.43 × √1.5)
𝑏3 = 0.3266
The calculation of the b4 coefficient requires four additional coefficients, d0
through d3, which are obtained from Exhibit 15-18 for Passing Lane segments.
The segment has vertical class 1; therefore d1 takes a value of 0.1252, while the
other coefficients are 0.
𝑏4 = 𝑑0 + 𝑑1 × √𝐻𝑉% + 𝑑2 × 𝐹𝐹𝑆 + 𝑑3 × (𝐹𝐹𝑆 × √𝐻𝑉%)
𝑏4 = 0 + 0.1252 × √8 + 0 × 62.43 + 0 × (62.43 × √8)
𝑏4 = 0.354
𝑣𝑜
1,000
+ max(0, 𝑏4 ) × √𝐻𝑉% ]
0
𝑚 = max [0, −1.138 + 0.094 × 62.43 + 0.0000 × √
1,000
+ max(0, 0.3266) × √1.5 + max(0, 0.354) × √8 ]
𝑚 = 6.139

The power coefficient p is computed by Equation 15-11, applying nine
coefficients f0 to f8, which are obtained from Exhibit 15-20 for a Passing Lane
segment.
𝑣𝑜 𝑣𝑜
1,000 1,000
+ 𝑓6 × √𝐻𝑉% + 𝑓7 × (𝐿 × 𝐻𝑉%)]
0
𝑝 = max [0, 0.91793 − 0.00557 × 62.43 + 0.36862 × 1.5 + 0 ×
1,000
0
+0×√ + 0.00611 × 8 + 0 × √8 − 0.00419 × (1.5 × 8)]
1,000
𝑝 = 1.122

𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
1.122
868
𝑆 = 62.43 − 6.139 ( − 0.1)
1,000
𝑆 = 57.9 mi/h


Percent followers at capacity for a Passing Lane segment is calculated using
Equation 15-19, applying eight parameters b0 to b7 obtained from Exhibit 15-25.

𝑏6 (√𝐻𝑉%) + 𝑏7 (𝐹𝐹𝑆 × 𝐻𝑉%)
𝑃𝐹𝑐𝑎𝑝 = 61.73075 + 6.73922(1.5) − 23.68853(√1.5) − 0.84126(62.43) +

11.44533(√62.43) − 1.05124(8) + 1.5039(√8) + 0.00491(62.43 × 8)
𝑃𝐹𝑐𝑎𝑝 = 79.04%

Percent followers at 25 percent of capacity for a Passing Lane segment is
calculated using Equation 15-21, applying eight parameters c0 to c7 obtained from
Exhibit 15-27.
+ 𝑐6 (√𝐻𝑉% ) + 𝑐7 (𝐹𝐹𝑆 × 𝐻𝑉%)
𝑃𝐹25𝑐𝑎𝑝 = 80.37105 + 14.44997(1.5) − 46.41831(√1.5) − 0.23367(62.43)

+ 0.84914(√62.43) − 0.56747(8) + 0.89427(√8 )
+ 0.00119(62.43 × 8)
𝑃𝐹25𝑐𝑎𝑝 = 35.90%

parameters for Passing Lane segments are used.
0 − ln [1 − 0 − ln [1 −
100 ] 100 ]
𝑚 = 𝑑1 ( 𝑐𝑎𝑝 ) + 𝑑2 ( 𝑐𝑎𝑝 )
0.25 [1,000] [1,000]
35.90 79.04
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.15808 ( ) − 0.83732 ( )
1,500 1,500
0.25 [1,000] [1,000]
𝑚 = −1.060

the parameters for Passing Lane segments are used.
0 − ln [1 − ] 0 − ln [1 − ]
100 100
𝑝 = 𝑒0 + 𝑒1 ( 𝑐𝑎𝑝 ) + 𝑒2 ( 𝑐𝑎𝑝 )
0.25 [ ] [ ]
1,000 1,000
0 − ln (1 − ) 0 − ln (1 − )
100 100
+ 𝑒3 √ 𝑐𝑎𝑝 + 𝑒4
√
𝑐𝑎𝑝
0.25 [ ] [ ]
1,000 1,000
35.90 79.04
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = −1.63246 + 1.6496 ( 100 ) − 4.45823 ( 100 )
1,500 1,500
0.25 [ ] [ ]
1,000 1,000
35.90 79.04
0 − ln (1 − ) 0 − ln (1 − )
− 4.89119√ 100 + 10.33057√ 100
1,500 1,500
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.897

𝑣𝑑 𝑝
(𝑚 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
868 0.897
(−1.060 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
𝑃𝐹 = 60.7%

Segment
Step 7a: Calculate the Flow Rate in Each Lane of the Passing Lane Segment
Equation 15-24 through Equation 15-27 are applied as follows.
𝐻𝑉% 8
𝑁𝑢𝑚𝐻𝑉 = 𝑣𝑑 × = 868 × = 69 veh
100 100
𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 0.92183 − 0.05022 × ln(𝑣𝑑 ) − 0.00030 × 𝑁𝑢𝑚𝐻𝑉

𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 0.92183 − 0.05022 × ln(868) − 0.00030 × 69
𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 0.561
𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 𝑣𝑑 × 𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 = 868 × 0.561 = 487 veh/h/ln
𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝑆𝐿 = 𝑣𝑑 × (1 − 𝑃𝑟𝑜𝑝𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 ) = 868 × (1 − 0.561) = 381 veh/h/ln
Step 7b: Calculate the Percentage of Heavy Vehicles in Each Lane of the
Passing Lane Segment
Equation 15-28 through Equation 15-30 are applied as follows.
𝐻𝑉%𝐹𝐿 = 𝐻𝑉% × 𝐻𝑉𝑃𝑟𝑜𝑝𝑀𝑢𝑙𝑡𝑖𝑝𝑙𝑖𝑒𝑟𝐹𝐿 = 8 × 0.4 = 3.2%
𝐻𝑉%𝐹𝐿 3.2
𝑁𝑢𝑚𝐻𝑉𝑆𝐿 = 𝑁𝑢𝑚𝐻𝑉 − (𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 × ) = 69 − (487 × ) = 54 veh
100 100
𝑁𝑢𝑚𝐻𝑉𝑆𝐿 54
𝐻𝑉%𝑆𝐿 = × 100 = × 100 = 14.2%
𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝑆𝐿 381
Step 7c: Calculate the Average Speed in Each Lane of the Passing Lane
Segment
Applying the equations and passing lane coefficient tables of Step 5
(Estimate the Average Speed), with the corresponding flow rate and heavy
vehicle percentage for each lane yields:
Sinit_FL = 60.7 mi/h; and
Sinit_SL = 60.6 mi/h.
The average speed lane differential adjustment is calculated with Equation
15-31.
𝐻𝑉%
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 = 2.750 + 0.00056 × 𝑣𝑑 + 3.8521 ×
100
8
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 = 2.750 + 0.00056 × 868 + 3.8521 ×
100
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 = 3.54 mi/h
Next, the average speed for each lane at the passing lane segment midpoint
is calculated with Equation 15-32 and Equation 15-33.
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 3.54
𝑆𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 = 𝑆𝑖𝑛𝑖𝑡_𝐹𝐿 + = 60.7 + = 62.5 mi/h
2 2
𝐴𝑣𝑔𝑆𝑝𝑒𝑒𝑑𝐷𝑖𝑓𝑓𝐴𝑑𝑗 3.54
𝑆𝑃𝐿𝑚𝑖𝑑_𝑆𝐿 = 𝑆𝑖𝑛𝑖𝑡_𝑆𝐿 − = 60.6 − = 58.8 mi/h
2 2
Step 7d: Calculate the Percent Followers in Each Lane of the Passing Lane
Segment
Applying the equations and passing lane coefficient tables of Step 6
(Estimate the Percent Followers), with the corresponding flow rate and heavy
vehicle percentage for each lane yields
PFPLmid_FL = 44.5%, and
PFPLmid_SL = 35.6%.

The follower density, FD, for the midpoint of the passing lane segment is
estimated using Equation 15-34.
𝑃𝐹𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 𝑃𝐹 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝑆𝐿
( 100 × 𝑆 ) + ( 𝑃𝐿𝑚𝑖𝑑_𝑆𝐿
100 × 𝑆 )
𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 𝑃𝐿𝑚𝑖𝑑_𝑆𝐿
𝐹𝐷𝑃𝐿𝑚𝑖𝑑 =
2
44.5 487 35.6 381
( 100 × 62.5) + ( 100 × 58.8)
2
𝐹𝐷𝑃𝐿𝑚𝑖𝑑 = 2.9 followers/mi/ln
The follower density, FD, for the endpoint of the passing lane segment is
𝑃𝐹 𝑣𝑑 60.7 868
100 𝑆 100 57.9

There is no passing lane upstream of segment 2. Therefore, no adjustment is
needed to follower density for this segment. However, before proceeding to the
analysis of the downstream segments, the effective length of this passing lane
segment is determined. The first criterion for identifying the passing lane effective
length, that is, the point at which the percent improvement to percent followers
goes to zero, is applied by testing different values of DownstreamDistance in
Equation 15-36 until the value of %ImprovePF goes to zero. The resulting effective
length is 14.4 mi.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 𝐷𝑜𝑤𝑛𝑠𝑡𝑟𝑒𝑎𝑚𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒)]
+ 0.1 × max[0, 𝑃𝐹 − 30] + 3.5 × ln[max(0.3, 𝑃𝑎𝑠𝑠𝐿𝑎𝑛𝑒𝐿𝑒𝑛𝑔𝑡ℎ)]
− 0.01 × 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[14.4] + 0.1 × max[0, 69.7 − 30]
+ 3.5 × ln[max(0.3,1.5)] − 0.01 × 904) = 0
The second criterion for identifying the passing lane effective length, that is,
the point at which the downstream adjusted follower density reaches 95% of the
follower density immediately prior to the start of the passing lane, is applied by
testing different values of DownstreamDistance in Equation 15-36 through
Equation 15-38. The target value for FDadj is 95% of the follower density for
segment 1, or 0.95 × 10.7 = 10.17 followers/mi/ln. This target value is reached at a
downstream distance of 8.1 mi.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[8.1] + 0.1 × max[0, 69.7 − 30]
+ 3.5 × ln[max(0.3,1.5)] − 0.01 × 904) = 5.0
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 𝐷𝑜𝑤𝑛𝑠𝑡𝑟𝑒𝑎𝑚𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒

+ 0.1 × max[0, 𝑃𝐹 − 30] + 0.75 × 𝑃𝑎𝑠𝑠𝐿𝑎𝑛𝑒𝐿𝑒𝑛𝑔𝑡ℎ
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 8.1 + 0.1 × max[0, 69.7 − 30] + 0.75 × 1.5
− 0.005 × 904)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, −2.9) = 0
𝑃𝐹 %𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒
𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 %𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆
𝑆 × (1 + )
100
69.7 5.0 904
𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 0
58.8 × (1 + 100)
𝐹𝐷𝑎𝑑𝑗 = 10.18 followers/mi/ln
The passing lane effective length is taken as the shorter of the two values, in
this case 8.1 mi. The remaining downstream segments (3, 4, and 5) are all within
the passing lane’s effective length; therefore, an adjusted follower density will be
calculated for each of these segments.

From Exhibit 15-6, for a higher-speed highway, the LOS is B.

𝑉𝑑 820
𝑣𝑑 = = = 863 veh/h
𝑃𝐻𝐹 0.95
The demand flow rate is less than the capacity of a Passing Constrained
segment (1,700 veh/h); therefore, the calculation process proceeds to Step 3.


Segment 3 is a Passing Constrained segment similar to Segment 1. The only
aspect in which it differs from Segment 1 is its length, which is used in Equation
15-4 to determine the a coefficient. However, the length is multiplied by zero in
Equation 15-4 and therefore does not affect the final result. The calculated FFS is
the same as Segment 1, 62.43 mi/h.

The slope coefficient m is computed using Equation 15-8. The calculation is
the same as for Segment 1, except that the segment length is different.
1,500
𝑚 = max [0,0.0558 + 0.0542 × 62.43 + 0.3278 × √
1,000
+ max(0, 0.1029) × √1 + max(0, 0) × √8 ] = 3.944

The power coefficient p is computed using Equation 15-11. The calculation is
the same as for Segment 1, except that the segment length is different.
1,500
𝑝 = max [0,0.67576 + 0 × 62.43 + 0 × 1.00 + 0.12060 ×
1,000
1,500
− 0.35919 × √ + 0 × 8 + 0 × √8 + 0 × (1.00 × 8)]
1,000
𝑝 = 0.417

𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.417
863
𝑆 = 62.43 − 3.944 ( − 0.1)
1,000
𝑆 = 58.9 mi/h

Because segment 3 is straight, no adjustment to the speed estimate is

using Equation 15-18. The calculation is the same as for Segment 1, except that
the segment length is different.
𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(1.00) − 7.90866(√1.00) − 0.94321(62.43) +
1,500 1,500
13.64266(√62.43) − 0.00050(8) − 0.05500 (62.43 × ) + 7.1376 (√1,000)
1,000
𝑃𝐹𝑐𝑎𝑝 = 85.32%

is calculated using Equation 15-20. The calculation is the same as for Segment 1,
except that the segment length is different.
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(1.00) − 21.60000(√1.00) − 0.97853(62.43)
1,500
+ 12.05214(√62.43) − 0.00750(8) − 0.06700 (62.43 × )
1,000
1,500
+ 11.6041 (√ ) = 48.43%
1,000

Equation 15-22 is used to compute the slope coefficient m.
48.43 85.32
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑚 = −1.275

Equation 15-23 is used to compute the power coefficient p.
48.43 85.32
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
48.43 85.32
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.768

863 0.768
(−1.275 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ] = 68.0%

Segment
skipped.

68.0 863
𝐹𝐷 = × = 10.0 followers/mi/ln
100 58.9

As previously determined, this segment is within the effective length of the
upstream passing lane; therefore, an adjusted follower density value is
calculated. The downstream distance used in these calculations is 2.5 mi (passing
lane segment length + subject segment length).
Equation 15-36 is used to determine the percentage improvement to percent
followers.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 2.5)] + 0.1 × max[0, 69.7 − 30]
+ 3.5 × ln[max(0.3,1.5)] − 0.01 × 863)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = 15.7%
Equation 15-37 is used to determine the percentage improvement to average
speed.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 2.5 + 0.1 × max[0, 69.7 − 30] + 0.75 × 1.5
− 0.005 × 863)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = 1.8%
Equation 15-38 is used to determine the adjusted follower density.

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
𝑆 × (1 + 100 )
68.0 15.7 863

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 1.8
58.9 × (1 +
100)
This adjusted follower density result is used in Steps 10 and 11.

From Exhibit 15-6, for a higher-speed highway, the LOS is D (barely).
Segment 4: Passing Zone Segment

𝑉𝑑 800
𝑣𝑑 = = = 851 veh/h
𝑃𝐻𝐹 0.94
The demand flow rate is less than the capacity of a Passing Zone segment
(1,700 veh/h) and the calculation process can proceed to Step 3. However,
because Segment 4 is a Passing Zone segment, the demand flow rate in the
opposing direction is also required and is calculated as follows:
𝑉𝑜 500
𝑣𝑜 = = = 532 veh/h
𝑃𝐻𝐹 0.94

Passing Zone segment of vertical class 1, and therefore no adjustment is needed
to the segment length.

The determination of the FFS for Segment 4 is similar to that of the previous
segments, as they have similar geometric characteristics. It differs only in terms
of length (which does not play a role in determining FFS for level segments) and
percent heavy vehicles. Therefore, from Equation 15-3:
𝐹𝐹𝑆 = 62.7 − (0.0333)(7.5) − 0 − 0 = 62.45 mi/h

The slope coefficient m is computed using Equation 15-8. Passing Zone
segments use the same coefficients in this equation as do Passing Constrained
segments. However, the values for FFS, opposing flow rate, segment length, and
percent heavy vehicles are different for Segment 4.
532
𝑚 = max [0,0.0558 + 0.0542 × 62.45 + 0.3278 × √
1,000
+ max(0, 0.1029) × √0.50 + max(0, 0) × √7.5 ] = 3.752

The power coefficient p is computed using Equation 15-11.
532
𝑝 = max [0,0.67576 + 0 × 62.45 + 0 × 0.5 + 0.12060 ×
1,000
532
− 0.35919 × √ + 0 × 7.5 + 0 × √7.5 + 0 × (0.5 × 7.5)]
1,000
𝑝 = 0.478

𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.478
851
𝑆 = 62.45 − 3.752 ( − 0.1)
1,000
𝑆 = 59.2 mi/h

Because segment 4 is straight, no adjustment to the speed estimate is

Percent followers at capacity for a Passing Zone segment is calculated using
Equation 15-18. The coefficients used in the equation are the same ones used for
Passing Constrained segments.
𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(0.50) − 7.90866(√0.50) − 0.94321(62.45) +
532 532
13.64266(√62.45) − 0.00050(7.5) − 0.05500 (62.45 × ) + 7.1376 (√1,000)
1,000
𝑃𝐹𝑐𝑎𝑝 = 85.90%

is calculated using Equation 15-20.
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(0.50) − 21.60000(√0.50) − 0.97853(62.45)

+ 12.05214(√62.45) − 0.00750(7.5)
532 532
− 0.06700 (62.45 × ) + 11.6041 (√ ) = 48.06%
1,000 1,000

48.06 85.90
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( ) = −1.287
1,700 1,700
0.25 [ ] [ ]
1,000 1,000

48.06 85.90
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
48.06 85.90
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.791

851 0.791
(−1.287 ×{ } )
1,000
𝑃𝐹 = 100 × [1 − 𝑒 ] = 67.8%

Segment
skipped.

67.8 851
100 59.2

calculated. The downstream distance used in these calculations is the sum of the
lengths of Segments 2, 3, and 4, which is 3.0 mi.
followers.

%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 3.0)] + 0.1 × max[0, 69.7 − 30]
+ 3.5 × ln[max(0.3,1.5)] − 0.01 × 851)
speed.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 3.0 + 0.1 × max[0, 69.7 − 30] + 0.75 × 1.5
− 0.005 × 851)

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
𝑆 × (1 + 100 )
67.8 14.3 851

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 1.4
59.2 × (1 + 100)

From Exhibit 15-6, for a higher-speed highway, the LOS is D (barely).

Step 2: Determine Demand Flow Rates, Capacity and d/c Ratio
𝑉𝑑 795
𝑣𝑑 = = = 850 veh/h
𝑃𝐻𝐹 0.935
The demand flow rate is less than the capacity of a Passing Constrained
segment (1,700 veh/h); therefore, the calculation process proceeds to Step 3.


Segment 5 is a Passing Constrained segment similar to Segments 1 and 3. The
only aspect in which it differs from those segments is its length, which is used in
Equation 15-4 to determine the a coefficient, where it is multiplied by zero and
therefore does not affect the final result. The calculated FFS is therefore the same
as Segments 1 and 3, 62.43 mi/h.

The slope coefficient m is computed using Equation 15-8.
1,500
𝑚 = max [0,0.0558 + 0.0542 × 62.43 + 0.3278 × √
1,000
+ max(0, 0.1029) × √1.75 + max(0, 0) × √8 ] = 3.977

The power coefficient p is computed using Equation 15-11.
1,500
𝑝 = max [0,0.67576 + 0 × 62.43 + 0 × 1.75 + 0.12060 ×
1,000
1,500
− 0.35919 × √ + 0 × 8 + 0 × √8 + 0 × (1.75 × 8)]
1,000
𝑝 = 0.417

0.417
850
𝑆 = 62.43 − 3.977 ( − 0.1)
1,000
𝑆 = 58.9 mi/h


𝑃𝐹𝑐𝑎𝑝 = 37.68080 + 3.05089(1.75) − 7.90866(√1.75) − 0.94321(62.43) +
1,500 1,500
13.64266(√62.43) − 0.00050(8) − 0.05500 (62.43 × 1,000) + 7.1376 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 85.05%

is calculated using Equation 15-20.
𝑃𝐹25𝑐𝑎𝑝 = 18.01780 + 10.00000(1.75) − 21.60000(√1.75) − 0.97853(62.43)
1,500
+ 12.05214(√62.43) − 0.00750(8) − 0.06700 (62.43 × )
1,000
1,500
+ 11.6041 (√ ) = 48.96%
1,000

48.96 85.05
0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = −0.29764 ( ) − 0.71917 ( )
1,700 1,700
0.25 [1,000] [1,000]
𝑚 = −1.275

48.96 85.05
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
48.96 85.05
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.750

850 0.750
(−1.275 ×{ } )
1,000
𝑃𝐹 = 100 × [1 − 𝑒 ] = 67.7%

Segment
skipped.

67.7 850
100 58.9

calculated. The downstream distance used in these calculations is the sum of the
lengths of Segments 2–5, 4.75 mi.
followers.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 4.75)]
+ 0.1 × max[0, 69.7 − 30] + 3.5 × ln[max(0.3,1.5)]
− 0.01 × 850)
speed.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 4.75 + 0.1 × max[0, 69.7 − 30] + 0.75 × 1.5
− 0.005 × 850)
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = 0%

𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
𝑆 × (1 + )
100
67.7 10.2 850
𝐹𝐷𝑎𝑑𝑗 = × (1 − )×
100 100 0
58.9 × (1 + 100)

From Exhibit 15-6, for a higher-speed highway, the LOS is D.
Step 11: Facility Analysis

The average follower density and the average LOS for the facility are computed
by averaging the length-weighted segment densities using Equation 15-39.
∑5𝑖=1 𝐹𝐷𝑖 × 𝐿𝑖
𝐹𝐷𝐹 =
∑5𝑖=1 𝐿𝑖
10.7 × 0.75 + 2.9 × 1.5 + 8.2 × 1.0 + 8.2 × 0.5 + 8.8 × 1.75
𝐹𝐷𝐹 =
0.75 + 1.5 + 1.0 + 0.5 + 1.75
40.1
𝐹𝐷𝐹 = = 7.3 followers/mi/ln
5.5
From Exhibit 15-6, for a higher-speed highway, the facility LOS is C. Exhibit
26-27 summarizes LOS results for each segment and the facility as a whole.
Length Follower Density Exhibit 26-27

Segment (mi) Type (followers/mi) LOS Example Problem 3:
1 0.75 Passing Constrained 10.7 D LOS Results
2 1.50 Passing Lane 2.9 B
3 1.00 Passing Constrained 8.2 D
4 0.50 Passing Zone 8.2 D
5 1.75 Passing Constrained 8.8 D
5.50 Facility 7.3 C
Discussion
The non–passing lane segments all operate at LOS D and the passing lane
segment operates at LOS B. The operating conditions within the passing lane
segment and its benefits several miles downstream result in an overall facility
LOS of C.
EXAMPLE PROBLEM 4: FACILITY ANALYSIS – MOUNTAIN ROAD

This example problem illustrates the computation of the LOS in the eastbound
direction of a 5.1-mi-long two-lane highway in mountainous terrain with two 0.5-
mi-long passing lanes. Grades reach 6% and there are a series of reverse curves
that constrain speeds.
This highway is a popular summer and winter recreational area access route.
It experiences heavy snowfall each winter. The road is extensively plowed
during winter and chains are required during winter storms. The Chapter 15
method is not appropriate for the analysis of winter operations.
This analysis focuses on summer, dry weather operations. Summer volumes
are higher than winter volumes.
The Facts
A diagram of the study facility, showing its six segments, is provided in
Exhibit 26-28. The study direction (eastbound) starts in the top left of the figure
and proceeds to the bottom center.
Exhibit 26-28
Example Problem 4:
Facility Diagram
Exhibit 26-29 provides volume and speed data for the facility, while Exhibit
26-30 provides grade and horizontal curve data. Tangent sections are either
straight or have horizontal curves with radii greater than 2,550 ft. Segment 2 has
a series of seven reverse curves of similar radius and central angle. To reduce
repetitive computations that would produce the same estimated speed, they have
been combined into a single long, curved subsegment. Segment 3 contains a short
passing lane 910 ft (0.17 mi) in length. Because this length is shorter than the
minimum passing lane lengths given in Exhibit 15-10 (i.e., is too short to be
effectively used as a passing lane), the passing lane is ignored. Segment 3 is treated
as a Passing Constrained segment instead, following the guidance on page 15-17.
Exhibit 26-29 Posted Directional Peak Heavy

Example Problem 4: Length Speed Limit Volume Hour Vehicles
Facility Volume and Segment Segment Type (mi) (mi/h) (veh/h) Factor (%)
Speed Data 1 Passing Constrained 1.3 55 1,100 0.90 8
2 Passing Constrained 1.0 55 1,100 0.90 8
5 Passing Lane 0.5 55 1,100 0.90 8
Exhibit 26-30 Horizontal Curvature

Example Problem 4: Segment Super-
Facility Grade and Horizontal Length Grade Sub- Horizontal Length elevation Curve
Curve Data Segment (mi) (%) segment Alignment (ft) (%) Radius (ft)
a Tangent 5,964 — —
1 1.3 4
b Curve 900 2 350
2 1.0 6
b 7 curves 4,280 2 500
3 0.5 6 Tangent 2,640 — —
4 1.3 4
b Curve 3,000 2 850
5 0.5 −3 Tangent 2,640 — —
6 0.5 −3 Tangent 2,640 — —
The facility has the following additional characteristics:

• Facility length = 26,928 ft (5.1 mi);
• Not within the effective length of upstream passing lanes;
• No turnouts;
• Lane width = 12 ft in all segments;
• Paved shoulder width = 6 ft in all segments; and
• Access points = 0 in all segments.
Objective
Estimate the LOS in the eastbound direction of the two-lane highway facility,
taking into account the effects of the passing lanes.

The facility was divided into homogeneous segments following the guidance
given in Step 1. The characteristics considered when segmenting the facility
included the ability to pass, lane geometry, grades, lane and shoulder widths,
posted speed limit, traffic demands, adjacent land uses, and driveways. Each
segment was designated as a Passing Constrained, Passing Zone, or Passing Lane
segment following the guidance on Segmentation given on page 15-4. The
resulting segment lengths and designations were shown in Exhibit 26-29.
Steps 2–10 of the two-lane highway analysis procedure are now followed for
each of the facility’s five segments, starting with the most upstream segment
(Segment 1) and proceeding in sequence to the downstream segments.

Equation 15-1 is used to convert the facility’s hourly demand volume to a
peak 15-min flow rate. Each segment has the same demand volume and thus the
same flow rate.
𝑉𝑑 1,100
𝑣𝑑 = = = 1,222 veh/h
𝑃𝐻𝐹 0.90
description of Step 2 on page 15-18, while the capacity of a Passing Lane segment
with vertical class 1 and 8% heavy vehicles is 1,500 veh/h, from Exhibit 15-5. The
demand flow rate is less than these capacities; therefore the calculation process
proceeds to Step 3.

Each segment is assigned a vertical alignment classification on the basis of
Exhibit 15-11. The results are shown in Exhibit 26-31.
Exhibit 26-31 Segment Length (mi) Grade (%) Vertical Class

Example Problem 4: 1 1.3 4 4
Vertical Alignment 2 1.0 6 5
Classifications by Segment 3 0.5 6 4
4 1.3 4 4
5 0.5 −3 1
6 0.5 −3 1
From Exhibit 15-10, all segment lengths lie between the minimum and
maximum lengths for their respective segment types. Therefore, no adjustment is
needed to any segment length.

The FFS is computed using Equation 15-2 through Equation 15-6. The
process will be demonstrated for Segment 1 and the results presented for the
remaining segments.
Because Segment 1 is a Passing Constrained segment, the opposing flow rate
vo is set at 1,500 veh/h in Equation 15-4 for the purposes of computing FFS. First,
the base free-flow speed BFFS is estimated using Equation 15-2. Next, Equation
15-5 and Equation 15-6 are used to determine factors relating to lane and shoulder
width and access-point density, which are used in the estimation of FFS.
𝐵𝐹𝐹𝑆 = 1.14 × 𝑆𝑝𝑙 = 1.14 × 55 = 62.7 mi/h
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)

𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 6) = 0
𝐴𝑃𝐷 0
𝑓𝐴 = min ( , 10) = min ( , 10) = 0
4 4
Unlike the previous example problems, Segment 1 is not level and therefore
the value of the coefficient a does not reduce to 0.0333. Equation 15-4 is used to
calculate a. This equation uses six coefficients a0 to a5, which are obtained from
Exhibit 15-12.
𝑎 = max [0.0333, 𝑎0 + 𝑎1 × 𝐵𝐹𝐹𝑆 + 𝑎2 × 𝐿

𝑣𝑜
+ max(0, 𝑎3 + 𝑎4 × 𝐵𝐹𝐹𝑆 + 𝑎5 × 𝐿) × ]
1,000
𝑎 = max [0.0333, −0.40902 + 0.00975 × 62.7 + 0.00767 × 1.3
1,500
+ max(0, −0.18363 + 0.00423 × 62.7 + 0 × 1.3) × ]
1,000
𝑎 = 0.335
Finally, the FFS is estimated by Equation 15-3.
𝐹𝐹𝑆 = 𝐵𝐹𝐹𝑆 − 𝑎(𝐻𝑉%) − f LS − f A

𝐹𝐹𝑆 = 62.7 − (0.335)(8) − 0 − 0 = 60.0 mi/h
Exhibit 26-32 summarizes the FFS results for all segments.
Segment Exhibit 26-32

Variable 1 2 3 4 5 6 Example Problem 4:
BFFS (mi/h) 62.7 62.7 62.7 62.7 62.7 62.7 Free-Flow Speed Results
fLS (mi/h) 0 0 0 0 0 0
fA (mi/h) 0 0 0 0 0 0
a0 −0.40902 −0.38360 −0.40902 −0.40902 0.00000 0.00000
a1 0.00975 0.01074 0.00975 0.00975 0.00000 0.00000
a2 0.00767 0.01945 0.00767 0.00767 0.00000 0.00000
a3 −0.18363 −0.69848 −0.18363 −0.18363 0.00000 0.00000
a4 0.00423 0.01069 0.00423 0.00423 0.00000 0.00000
a5 0.00000 0.12700 0.00000 0.00000 0.00000 0.00000
a 0.335 0.457 0.329 0.335 0.0333 0.0333
FFS (mi/h) 60.0 59.0 60.1 60.0 62.4 62.4

veh/h, the equations given in Step 5 are used to estimate the average speed. The
process is demonstrated for Segment 1, with results summarized for all segments
afterwards.

The slope coefficient m is computed using Equation 15-8. This equation
requires six coefficients b0 to b5, which are obtained from Exhibit 15-13 for a
Passing Constrained segment. The exhibit references Equation 15-9 to calculate b3
and Equation 15-10 to calculate b4, and directly provides numerical values for the
other coefficients.
Equation 15-9 is used to determine the segment length coefficient b3. It uses
four coefficients c0 to c3 , which are obtained from Exhibit 15-15 for a Passing
Constrained segment.
𝑏3 = 𝑐0 + 𝑐1 × √𝐿 + 𝑐2 × 𝐹𝐹𝑆 + 𝑐3 × (𝐹𝐹𝑆 × √𝐿)
𝑏3 = −12.5113 + 0 × √1.3 + 0.2656 × 60.0 + 0 × (60 × √1.3)
𝑏3 = 3.4247
Equation 15-10 is used to determine the segment length coefficient b3. It uses
four coefficients d0 to d3, which are obtained from Exhibit 15-17 for a Passing
Constrained segment.
𝑏4 = 𝑑0 + 𝑑1 × √𝐻𝑉% + 𝑑2 × 𝐹𝐹𝑆 + 𝑑3 × (𝐹𝐹𝑆 × √𝐻𝑉%)
𝑏4 = −5.7775 + 0 × √8 + 0.1373 × 60.0 + 0 × (60.0 × √8)
𝑏4 = 2.4605
With all the coefficients now determined, the slope coefficient m can be
calculated:
𝑣𝑜
1,000
+ max(0, 𝑏4 ) × √𝐻𝑉% ]
1,500
𝑚 = max [3.2685,9.0115 − 0.1994 × 60.0 + 1.8252 × √
1,000
+ max(0,3.4247) × √1.3 + max(0,2.4605) × √8 ]
𝑚 = 10.147

The power coefficient p is computed using Equation 15-11. This equation
requires nine coefficients f0 to f8, which are obtained from Exhibit 15-19 for a
Passing Constrained segment.
𝑣𝑜 𝑣𝑜
1,000 1,000
+ 𝑓6 × √𝐻𝑉% + 𝑓7 × (𝐿 × 𝐻𝑉%)]
𝑝 = max [0.33950,0.67689 + 0.00534 × 60.0 − 0.13037 × 1.3
1,500 1,500
+ 0.25699 × − 0.68465 × √ − 0.00709 × 8
1,000 1,000
+ 0.07087 × √8 + 0 × (1.3 × 8)]
𝑝 = 0.519

The average speed is calculated using Equation 15-7.
𝑣𝑑 𝑝
𝑆 = 𝐹𝐹𝑆 − 𝑚 ( − 0.1)
1,000
0.519
1,222
𝑆 = 60.0 − 10.147 ( − 0.1)
1,000
𝑆 = 49.2 mi/h
Exhibit 26-33 summarizes the average speed results for all segments.

b0 9.0115 23.9144 9.0115 9.0115 −1.1379 0.0558 Unadjusted Average Speed
b1 −0.1994 −0.6925 −0.1994 −0.1994 0.0941 0.0542 Results
b2 1.8252 1.9473 1.8252 1.8252 0.0000 0.3278
c0 −12.5113 −14.8961 −12.5113 −12.5113 0.0000 0.1029
c1 0.0000 0.0000 0.0000 0.0000 0.2667 0.0000
c2 0.2656 0.4370 0.2656 0.2656 0.0000 0.0000
c3 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
b3 3.4247 10.8869 3.4513 3.4247 0.1886 0.1029
d0 −5.7775 −18.2910 −5.7775 −5.7775 0.0000 0.0000
d1 0.0000 2.3875 0.0000 0.0000 0.1252 0.0000
d2 0.1373 0.4494 0.1373 0.1373 0.0000 0.0000
d3 0.0000 −0.0520 0.0000 0.0000 0.0000 0.0000
b4 2.4605 6.2989 2.4742 2.4605 0.3541 0.0000
b5 3.2685 3.5115 3.2685 3.2685 0.0000 0.0000
m 10.147 14.145 8.702 10.147 5.869 3.912
f0 0.67689 1.13262 0.67689 0.67689 0.91793 0.67576
f1 0.00534 0.00000 0.00534 0.00534 −0.00557 0.00000
f2 −0.13037 −0.26367 −0.13037 −0.13037 0.36862 0.00000
f3 0.25699 0.18811 0.25699 0.25699 0.00000 0.12060
f4 −0.68465 −0.64304 −0.68465 −0.68465 0.00000 −0.35919
f5 −0.00709 −0.00867 −0.00709 −0.00709 0.00611 0.00000
f6 0.07087 0.08675 0.07087 0.07087 0.00000 0.00000
f7 0.00000 0.00000 0.00000 0.00000 −0.00419 0.00000
f8 0.33950 0.30590 0.33950 0.33950 0.00000 0.00000
p 0.519 0.540 0.623 0.519 0.787 0.417
S (mi/h) 49.2 43.9 50.8 49.2 56.0 58.3

In this step, the average speed for each subsegment with a horizontal curve
is determined. There are three substeps: (a) identifying the horizontal alignment
classification for each subsegment with a horizontal curve, (b) calculating the
average speed for each subsegment with a horizontal curve, and (c) calculating
the adjusted average speed for the segment. The process will be demonstrated
for Segment 1, with results summarized for the other segments afterwards. Only
Segments 1, 2, and 4 contain horizontal curves.
Step 5d.1: Identify all Horizontal Curves Within the Segment

From the facts given previously, the horizontal curve in subsegment 1b has a
radius of 350 ft and a superelevation of 2%. From Exhibit 15-22, it is assigned a
horizontal alignment class of 4.
Step 5d.2: Calculate Average Speed for each Horizontal Curve Within the
Segment
The average speed for a subsegment with horizontal curvature is determined
using Equation 15-12 though Equation 15-15. First, Equation 15-14 is applied to
compute the base free-flow speed:
𝐵𝐹𝐹𝑆𝐻𝐶1𝑏 = min(𝐵𝐹𝐹𝑆𝑇 , 44.32 + 0.3728 × 𝐵𝐹𝐹𝑆𝑇 − 6.868 × 𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠1𝑏 )
𝐵𝐹𝐹𝑆𝐻𝐶1𝑏 = min(62.7,44.32 + 0.3728 × 62.7 − 6.868 × 4)
𝐵𝐹𝐹𝑆𝐻𝐶1𝑏 = min(62.7, 40.22) = 40.22 mi/h
Next, the FFS is computed using Equation 15-13:
𝐹𝐹𝑆𝐻𝐶1𝑏 = 𝐵𝐹𝐹𝑆𝐻𝐶1𝑏 − 0.0255 × 𝐻𝑉%
𝐹𝐹𝑆𝐻𝐶1𝑏 = 40.22 − 0.0255 × 8

𝐹𝐹𝑆𝐻𝐶1𝑏 = 40.02 mi/h
The slope coefficient m used in the determination of average speed is
computed using Equation 15-15 as follows:
𝑚 = max(0.277, −25.8993 − 0.7756 × 𝐹𝐹𝑆𝐻𝐶1𝑏 + 10.6294 × √𝐹𝐹𝑆𝐻𝐶1𝑏
+ 2.4766 × 𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠1𝑏 − 9.8238 × √𝐻𝑜𝑟𝑖𝑧𝐶𝑙𝑎𝑠𝑠1𝑏 )
𝑚 = max(0.277, −25.8993 − 0.7756 × 40.02 + 10.6294 × √40.02 + 2.4766 × 4

− 9.8238 × √4)
𝑚 = max(0.277, 0.563) = 0.563
Finally, the average speed of subsegment 1b is computed by Equation 15-12.
𝑣𝑑
𝑆𝐻𝐶1𝑏 = min (𝑆, 𝐹𝐹𝑆𝐻𝐶1𝑏 − 𝑚 × √ − 0.1)
1,000
1,222
𝑆𝐻𝐶1𝑏 = min (49.2, 40.02 − 0.563 × √ − 0.1)
1,000
𝑆𝐻𝐶1𝑏 = min(49.2, 39.4) = 39.4 mi/h
Step 5d.3: Calculate Adjusted Average Speed for the Segment

Equation 15-16 is used to calculate Segment 1’s adjusted average speed by
taking a length-weighted average of the subsegment speeds.
∑2𝑖=1(𝑆𝑢𝑏𝑠𝑒𝑔𝑆𝑝𝑒𝑒𝑑𝑖 × 𝑆𝑢𝑏𝑠𝑒𝑔𝐿𝑒𝑛𝑔𝑡ℎ𝑖 )
𝑆=
𝐿
(49.2 × 5,964) + (39.4 × 900)
𝑆=
6,864
𝑆 = 47.9 mi/h
Exhibit 26-34 presents the calculation results for adjusted average speed for
each segment.
Exhibit 26-34 Segment

Example Problem 4: Variable 1 2 3 4 5 6
Adjusted Average Speed Horizontal class 4 3 — 2 — —
Results BFFSHC (mi/h) 40.22 47.09 — 53.96 — —
FFSHC (mi/h) 40.0 46.9 — 53.8 — —
m 0.563 0.933 — 1.401 — —
SHC (mi/h) 39.4 43.9 — 49.2 — —
Adjusted S (mi/h) 47.9 43.9 50.8 49.2 56.0 58.3
Note: — = not applicable, no horizontal curve in the segment.

The calculation of percent followers is demonstrated for Segment 1, with
results for all segments presented afterwards.

using Equation 15-18, applying eight parameters b0 to b7 obtained from Exhibit
15-24.
𝑜𝑣 𝑣0
𝑏6 (𝐹𝐹𝑆 × 1,000 ) + 𝑏7 (√1,000)
𝑃𝐹𝑐𝑎𝑝 = 58.29978 − 0.53611(1.3) + 7.35076(√1.3) − 0.27046(60.0) +

1,500 1,500
4.49850(√60.0) − 0.01100(8) − 0.02968 (60.0 × ) + 8.8968 (√1,000)
1,000
𝑃𝐹𝑐𝑎𝑝 = 92.74%

is calculated using Equation 15-20, applying eight parameters c0 to c7 obtained
from Exhibit 15-26.
𝑣𝑜 𝑣0
+ 𝑐6 (𝐹𝐹𝑆 × ) + 𝑐7 (√ )
1,000 1,000
𝑃𝐹25𝑐𝑎𝑝 = 103.13534 + 14.68459(1.3) − 23.72704(√1.3) + 0.664436(60.0)

1,500
− 11.95763(√60.0) − 0.10000(8) + 0.00172 (60.0 × )
1,000
1,500
+ 14.7007 (√ )
1,000
𝑃𝐹25𝑐𝑎𝑝 = 59.77%

0 − ln [1 − 100 ] 0 − ln [1 − 100 ]
𝑚 = 𝑑1 ( 𝑐𝑎𝑝 ) + 𝑑2 ( 𝑐𝑎𝑝 )
0.25 [1,000] [1,000]
59.77 92.74
0 − ln [1 − ] 0 − ln [1 − ]
𝑚 = −0.29764 ( 100 ) − 0.71917 ( 100 )
1,700 1,700
0.25 [1,000] [1,000]
𝑚 = −1.747

0 − ln [1 − ] 0 − ln [1 − ]
100 100
𝑝 = 𝑒0 + 𝑒1 ( 𝑐𝑎𝑝 ) + 𝑒2 ( 𝑐𝑎𝑝 )
0.25 [ ] [ ]
1,000 1,000
0 − ln (1 − ) 0 − ln (1 − )
100 100
+ 𝑒3 √ 𝑐𝑎𝑝 + 𝑒4
√
𝑐𝑎𝑝
0.25 [ ] [ ]
1,000 1,000
59.77 92.74
0 − ln [1 − ] 0 − ln [1 − ]
𝑝 = 0.81165 + 0.37920 ( 100 ) − 0.49524 ( 100 )
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
59.77 92.74
0 − ln (1 − ) 0 − ln (1 − )
− 2.11289√ 100 + 2.41146√ 100
1,700 1,700
0.25 [ ] [ ]
1,000 1,000
𝑝 = 0.762

𝑣𝑑 𝑝
(𝑚 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
1,222 0.762
(−1.747 ×{ } )
𝑃𝐹 = 100 × [1 − 𝑒 1,000 ]
𝑃𝐹 = 86.9%
Exhibit 26-35 presents the percent followers results for all segments.
Step 7: Calculate Additional Performance Measure Values for a Passing

Lane Segment
This step applies only to Segment 5, the passing lane segment. Equation 15-24
through Equation 15-33 are applied, as was demonstrated for the passing lane
segment in Example Problem 3. The results that will be used in the segment
midpoint follower density calculation are as follows:
FlowRateFL = 654 veh/h/ln;
FlowRateSL = 568 veh/h/ln;

SPLmid_FL = 61.1 mi/h;
SPLmid_SL = 56.8 mi/h;
PFPLmid_FL = 63.1%; and
PFPLmid_SL = 55.9%.

Follower density, FD, is estimated using Equation 15-35. For Segment 1, the
calculation is performed as follows:
𝑃𝐹 𝑣𝑑 86.9 1,222
100 𝑆 100 47.9
The follower density, FD, for the midpoint of the passing lane segment is
𝑃𝐹𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝐹𝐿 𝑃𝐹 𝐹𝑙𝑜𝑤𝑅𝑎𝑡𝑒𝑆𝐿
( × ) + ( 𝑃𝐿𝑚𝑖𝑑_𝑆𝐿 × )
100 𝑆𝑃𝐿𝑚𝑖𝑑_𝐹𝐿 100 𝑆𝑃𝐿𝑚𝑖𝑑_𝑆𝐿
2
63.1 654 55.9 568
( 100 × 61.1) + ( 100 × )
𝐹𝐷𝑃𝐿𝑚𝑖𝑑 = 56.8
2
𝐹𝐷𝑃𝐿𝑚𝑖𝑑 = 6.2 followers/mi/ln
Exhibit 26-35 presents the follower density results for all segments.

b0 58.29978 3.32968 58.29978 58.29978 61.73075 37.68080 Percent Follower and
b1 −0.53611 −0.84377 −0.53611 −0.53611 6.73922 3.05089 Unadjusted Follower Density
b2 7.35076 7.08952 7.35076 7.35076 −23.68853 −7.90866 Results
b3 −0.27046 −1.32089 −0.27046 −0.27046 −0.84126 −0.94321
b4 4.49850 19.98477 4.49850 4.49850 11.44533 13.64266
b5 −0.01100 −0.01250 −0.01100 −0.01100 −1.05124 −0.00050
b6 −0.02968 −0.02960 −0.02968 −0.02968 1.50390 −0.05500
b7 8.89680 9.99450 8.89680 8.89680 0.00491 7.13760
PFcap (%) 92.74 94.67 89.98 92.74 84.56 86.12
c0 103.13534 89.00000 103.13534 103.13534 80.37105 18.01780
c1 14.68459 19.02642 14.68459 14.68459 14.44997 10.00000
c2 −23.72704 −34.54240 −23.72704 −23.72704 −46.41831 −21.60000
c3 0.66444 0.29792 0.66444 0.66444 −0.23367 −0.97853
c4 −11.95763 −6.62528 −11.95763 −11.95763 0.84914 12.05214
c5 −0.10000 −0.16000 −0.10000 −0.10000 −0.56747 −0.00750
c6 0.00172 0.00480 0.00172 0.00172 0.89427 −0.06700
c7 14.70074 17.56610 14.70074 14.70074 0.00119 11.60410
PF25cap (%) 59.77 60.83 58.29 59.77 45.48 49.77
d1 −0.29764 −0.29764 −0.29764 −0.29764 −0.15808 −0.29764
d2 −0.71917 −0.71917 −0.71917 −0.71917 −0.83732 −0.71917
m −1.747 −1.897 −1.586 −1.747 −1.299 −1.317
e0 0.81165 0.81165 0.81165 0.81165 −1.63246 0.81165
e1 0.37920 0.37920 0.37920 0.37920 1.64960 0.37920
e2 −0.49524 −0.49524 −0.49524 −0.49524 −4.45823 −0.49524
e3 −2.11289 −2.11289 −2.11289 −2.11289 −4.89119 −2.11289
e4 2.41146 2.41146 2.41146 2.41146 10.33057 2.41146
p 0.762 0.823 0.696 0.762 0.791 0.760
PF (%) 86.9 89.3 83.9 86.9 78.2 78.5
FD (followers/mi) 22.2 24.9 20.2 21.6 17.1 16.5

Segment 5 is a passing lane. Therefore, an adjustment may be needed to
follower density in the downstream segment, Segment 6. Following the same
process outlined in Example Problem 3, the effective length is determined to be
4.4 mi (the adjusted follower density within 95% of the upstream follower
density controls the effective length in this case).
Because the end of the downstream, and last, segment is within the effective
length of the passing lane, Equation 15-36 through Equation 15-38 are used to
calculate the adjusted follower density for Segment 6. The downstream distance
used in these calculations is 1.0 mi (Segment 5 length + Segment 6 length).
followers in Segment 6 as a result of the upstream passing lane.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑃𝐹 = max(0, 27 − 8.75 × ln[max(0.1, 1.0)] + 0.1 × max[0,86.9 − 30]
+ 3.5 × ln[max(0.3,0.5)] − 0.01 × 1,222)
speed in Segment 6 as a result of the upstream passing lane, and Equation 15-38
is used to determine the adjusted follower density.
%𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑆 = max(0, 3 − 0.8 × 1.0 + 0.1 × max[0, 86.9 − 30] + 0.75 × 0.5
− 0.005 × 1,222)
𝐹𝑜𝑙𝑙𝑜𝑤𝑒𝑟𝐷𝑒𝑛𝑠𝑖𝑡𝑦𝑎𝑑𝑗 = × (1 − )×
𝑆 × (1 + 100 )
78.5 18.0 1,222
𝐹𝑜𝑙𝑙𝑜𝑤𝑒𝑟𝐷𝑒𝑛𝑠𝑖𝑡𝑦𝑎𝑑𝑗 = × (1 − )×
100 100 2.2
58.3 × (1 +
100)
𝐹𝑜𝑙𝑙𝑜𝑤𝑒𝑟𝐷𝑒𝑛𝑠𝑖𝑡𝑦𝑎𝑑𝑗 = 13.2 followers/mi

Each segment’s LOS is determined from Exhibit 15-6, using the column for a
higher-speed highway (posted speed limit equal to or greater than 50 mi/h). The
follower density in all segments exceeds 12 followers/mi; therefore, all segments
operate at LOS E.
Step 11: Facility Analysis

The average follower density and the average LOS for the facility are computed
by averaging the length-weighted segment densities using Equation 15-39.
∑6𝑖=1 𝐹𝐷𝑖 × 𝐿𝑖
𝐹𝐷𝐹 =
∑6𝑖=1 𝐿𝑖
22.2 × 1.3 + 24.9 × 1.0 + 20.2 × 0.5 + 21.6 × 1.3 + 6.2 × 0.5 + 13.2 × 0.5
𝐹𝐷𝐹 =
1.3 + 1.0 + 0.5 + 1.3 + 0.5 + 0.5
𝐹𝐷𝐹 = 20.0 followers/mi/ln

From Exhibit 15-6, for a higher-speed highway, the facility LOS is E. Exhibit
26-36 summarizes LOS results for each segment and the facility as a whole.
Length Follower Density Exhibit 26-36

Segment (mi) Type (followers/mi) LOS Example Problem 4:
1 1.3 Passing Constrained 22.2 E LOS Results
2 1.0 Passing Constrained 24.9 E
5 0.5 Passing Lane 6.2 C
5.1 Facility 20.0 E
Discussion
The main conclusions of this analysis are:
• The tight 350-ft radius horizontal curve at the end of Segment 1
significantly reduces speeds at the end of that segment. The other
segments are comparatively unaffected by their horizontal curvature.
• The 6% upgrade significantly affects speeds and percent followers. The
Passing Lane segment significantly reduces percent followers on the
downstream segment. However, the improvement is not large enough to
change the LOS from E to D for this mountainous highway.
• The percent followers and the follower density are high on this facility,
resulting in LOS E. The last segment of the facility (a downgrade
following a 0.5-mi passing lane) is slightly over the threshold for LOS E.
Average speeds for the non-passing lane segments range from 44 mi/h to
56 mi/h. Within the passing lane, the average speeds range from 56 to 61
mi/h across the two lanes. The demand/capacity ratio varies from 0.72 to
0.87 during the summer peak hour.
The long upgrade extending across several segments generates interactions
across the segments that are not well modeled by this macroscopic analysis
method for evaluating facilities. Consequently, microsimulation would be
recommended to verify and potentially refine the results.
As noted in The Facts section at the start of this example problem, Segment 3
contains a 910-ft passing lane that is too short to provide a substantial
operational benefit. One could analyze the effect of extending the passing lane
length to 0.5 mi, assuming it is actually feasible to extend the passing lane in this
mountainous terrain. In that case, Segment 3 would be analyzed as a Passing
Lane segment, and the follower density in Segment 4 would be adjusted to reflect
the effects of the passing lane. Another passing lane starts in Segment 5;
therefore, the analysis of the effects of the Segment 3 passing lane would not be
carried past Segment 4. With the improved passing lane, the follower density for
the facility would improve to 17.6 followers/mi/ln, although this still yields LOS E.
EXAMPLE PROBLEM 5: TWO-LANE HIGHWAY BICYCLE LOS

A segment of two-lane highway (without passing lanes) is being evaluated
for potential widening, realigning, and repaving. Analyze the impacts of the
proposed project on the bicycle LOS (BLOS) in the peak direction.
The Facts
The roadway currently has the following characteristics:
• Lane width = 12 ft,
• Shoulder width = 2 ft,
• Pavement rating = 3 (fair),
• Posted speed limit = 50 mi/h,
• Hourly directional volume = 500 veh/h (no growth is expected),
• Percentage of heavy vehicles = 5%,
• PHF = 0.90, and
• No on-highway parking.
The proposed roadway design has the following characteristics:
• Shoulder width = 6 ft,
• Pavement rating = 5 (very good),
• Posted speed limit = 55 mi/h, and
• No on-highway parking.

All data needed to perform the analysis are listed above.
Step 2: Calculate the Directional Flow Rate in the Outside Lane

Using the hourly directional volume and the PHF, calculate the directional
demand flow rate with Equation 15-40. Because this is a two-lane highway
segment without a passing lane, the number of directional lanes N is 1. Because
traffic volumes are not expected to grow over the period of the analysis, vOL is the
same for both current and future conditions.
𝑉 500
𝑣𝑂𝐿 = = = 556 veh/h
𝑃𝐻𝐹 × 𝑁 0.90 × 1
Step 3: Calculate the Effective Width

For current conditions, the hourly directional demand V is greater than 160
veh/h and the paved shoulder width is 2 ft; therefore, Equation 15-43 and
Equation 15-44 are used to determine the effective width of the outside lane.
Under future conditions, the paved shoulder width will increase to 6 ft;
therefore, Equation 15-42 and Equation 15-44 are used.
For current conditions,
𝑊𝑣 = 𝑊𝑂𝐿 + 𝑊𝑠 = 12 + 2 = 14 ft
𝑊𝑒 = 𝑊𝑣 − (%𝑂𝐻𝑃[2 ft + 𝑊𝑠 ]) = 14 − (0 × [2 + 2]) = 14 ft
Under the proposed design,
𝑊𝑣 = 𝑊𝑂𝐿 + 𝑊𝑠 = 12 + 6 = 18 ft
𝑊𝑒 = 𝑊𝑣 + 𝑊𝑠 − 2 × (%𝑂𝐻𝑃[2 ft + 𝑊𝑠 ]) = 18 + 6 − 2 × (0 × [2 + 6]) = 24 ft
Step 4: Calculate the Effective Speed Factor

Equation 15-46 is used to calculate the effective speed factor. Under current
conditions,
𝑆𝑡 = 1.1199 ln(𝑆𝑝𝑙 − 20) + 0.8103 = 1.1199 ln (50 − 20) + 0.8103 = 4.62
Under the proposed design,
𝑆𝑡 = 1.1199 ln (55 − 20) + 0.8103 = 4.79
Step 5: Determine the LOS

Equation 15-47 is used to calculate the BLOS score, which is then used in
Exhibit 15-7 to determine the LOS. Under existing conditions,
𝐵𝐿𝑂𝑆 = 0.507 ln(𝑣𝑂𝐿 ) + 0.1999𝑆𝑡 (1 + 10.38𝐻𝑉)2 + 7.066(1/𝑃)2 − 0.005(𝑊𝑒 )2
+ 0.760
𝐵𝐿𝑂𝑆 = 0.507 ln(556) + 0.1999(4.62)(1 + 10.38 × 0.05)2 + 7.066(1/3)2
− 0.005(14)2 + 0.760
𝐵𝐿𝑂𝑆 = 3.205 + 2.131 + 0.785 − 0.980 + 0.760 = 5.90
Therefore, the BLOS for existing conditions is LOS F. Use of the same process
for the proposed design results in the following:
𝐵𝐿𝑂𝑆 = 0.507 ln(556) + 0.1999(4.79)(1 + 10.38 × 0.05)2 + 7.066(1/5)2
− 0.005(24)2 + 0.760
𝐵𝐿𝑂𝑆 = 3.205 + 2.209 + 0.283 − 2.880 + 0.760 = 3.58
The corresponding LOS for the proposed design is LOS D, close to the
boundary of LOS C (BLOS = 3.50).
Discussion
Although the posted speed would increase as a result of the proposed
design, this negative impact on bicyclists would be more than offset by the
proposed shoulder widening, as indicated by the improvement from LOS F to
LOS D.
9. REFERENCES
Some of these references can 1. Zegeer, J. D., M. A. Vandehey, M. Blogg, K. Nguyen, and M. Ereti. NCHRP
Reference Library in Volume 4. Report 599: Default Values for Highway Capacity and Level of Service Analyses.
D.C., 2008.
2. Dowling, R., G. F. List, B. Yang, E. Witzke, and A. Flannery. NCFRP Report
41: Incorporating Truck Analysis into the Highway Capacity Manual.
D.C., 2014.
3. Washburn, S. S., and S. Ozkul. Heavy Vehicle Effects on Florida Freeways and
Multilane Highways. Report TRC-FDOT-93817-2013. Florida Department of
Transportation, Tallahassee, 2013.
4. Ozkul, S., and Washburn, S. S. Updated Commercial Truck Speed versus
Distance-Grade Curves for the Highway Capacity Manual. In Transportation
D.C., 2015, pp. 91–101.
5. Hu, J., B. Schroeder, and N. Rouphail. Rationale for Incorporating Queue
Discharge Flow into Highway Capacity Manual Procedure for Analysis of
Freeway Facilities. In Transportation Research Record: Journal of the
Transportation Research Board, No. 2286, Transportation Research Board of the
National Academies, Washington, D.C., 2012, pp. 76–83.
6. Elefteriadou, L., A. Kondyli, and B. St. George. Estimation of Capacities on
Florida Freeways. Final Report. Transportation Research Center, University of
Florida, Gainesville, Sept. 2014.
7. Brilon, W., J. Geistefeldt, and M. Regler. Reliability of Freeway Traffic Flow:
A Stochastic Concept of Capacity. In Proceedings of the 16th International
Symposium on Transportation and Traffic Theory, College Park, Md., July 2005,
pp. 125–144.
8. SAE International. Taxonomy and Definitions for Terms Related to Driving
Automation Systems for On-Road Motor Vehicles. Recommended Practice J3016.
Warrendale, Pa., June 2018.
9. Adebisi, A., Y. Liu, B. Schroeder, J. Ma, B. Cesme, A. Jia, and A. Morgan.
Developing Highway Capacity Manual Capacity Adjustment Factors for
Connected and Automated Traffic on Freeway Segments. In Transportation
D.C., 2020, pp. 401–415.
10. Jones, S. Cooperative Adaptive Cruise Control: Human Factors Analysis. Report
FHWA-HRT-13-045. Federal Highway Administration, Washington, D.C.,
Oct. 2013.
References Chapter 26/Freeway and Highway Segments: Supplemental

11. Krechmer, D., K. Blizzard, M.G. Cheung, R. Campbell, V. Alexiadis, J. Hyde,

J. Osborne, M. Jensen, S. Row, A. Tudela, E. Flanigan, and J. Bitner. Connected
Vehicle Impacts on Transportation Planning. Primer and Final Report. Report
FHWA-JPO-16-420. Federal Highway Administration, Washington, D.C.,
June 2016.
12. Nowakowski, C., J. O’Connell, S.E. Shladover, and D. Cody. Cooperative
Adaptive Cruise Control: Driver Acceptance of Following Gap Settings Less
than One Second. Proceedings, Human Factors and Ergonomics Society
Annual Meeting, San Francisco, Calif., 2010.
13. Goñi-Ros, B., W.J. Schakel, A.E. Papacharalampous, M. Wang, V.L. Knoop, I.
Sakata, B. van Arem, and S.P. Hoogendoorn. Using Advanced Adaptive
Cruise Control Systems to Reduce Congestion at Sags: An Evaluation Based
on Microscopic Traffic Simulation. Transportation Research Part C: Emerging
Technologies, Vol. 102, 2019, pp. 411–426.
14. Davis, S.C., and R.G. Boundy. Transportation Energy Data Book, Edition 37.
Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge,
Tenn., Aug. 2019.
15. Litman, T. Autonomous Vehicle Implementation Predictions: Implications for
Transport Planning. Victoria Transport Policy Institute, Victoria, B.C., Oct.
2019.
Chapter 26/Freeway and Highway Segments: Supplemental References

APPENDIX A: TRUCK PERFORMANCE CURVES
This appendix provides travel time versus distance curves for SUTs and TTs
for 50-, 55-, 60-, 65-, and 75-mi/h free-flow speeds (FFS). Curves for SUTs and TTs
for a 70-mi/h FFS are presented in Section 3 as Exhibit 26-5 and Exhibit 26-6,
respectively.
Exhibit 26-A1
FFS

Exhibit 26-A2
FFS

Appendix A: Truck Performance Curves Chapter 26/Freeway and Highway Segments: Supplemental
Exhibit 26-A3
FFS

Exhibit 26-A4
FFS

Chapter 26/Freeway and Highway Segments: Supplemental Appendix A: Truck Performance Curves
Exhibit 26-A5
FFS
Exhibit 26-A6
FFS

Exhibit 26-A7
FFS

Exhibit 26-A8
FFS

Chapter 26/Freeway and Highway Segments: Supplemental Appendix A: Truck Performance Curves
Exhibit 26-A9
FFS

Exhibit 26-A10
FFS
APPENDIX B: WORK ZONES ON TWO-LANE HIGHWAYS
This appendix presents a method for estimating the capacity and operation
of work zones on two-lane highways when one of the two lanes is closed. This
method is based on research conducted by National Cooperative Highway
Research Program (NCHRP) Project 03-107 (B-1).
Work zones along two-lane highways can take three forms:
1. Shoulder closure. Work activity is limited to the shoulder of one direction
of travel and does not require lane reconfiguration. In this case, only the
direction of travel adjacent to the work zone is slightly affected.
2. Lane shift. Work activity extends beyond the shoulder, but both directions
of travel can be accommodated with a lane shift that utilizes the opposite
paved shoulder.
3. Lane closure. Work activity requires the closure of one of the two lanes.
Flaggers or temporary traffic signals are used to alternately serve one
direction of travel at a time. Both directions of travel can be significantly
affected.
The method presented in this appendix addresses the third scenario—lane This method addresses a one-
lane closure on a two-lane
closure—as it has the greatest impact on traffic operations. highway. Other types of work
zones, such as shoulder
closures or lane shifts, are not
CONCEPTS addressed.
A lane closure on a two-lane highway converts traffic flow from an
uninterrupted to an interrupted condition. With traffic control devices (flaggers
or signals) provided at each end, the operation of the lane closure can be
described in terms similar to those used for a signalized intersection:
• Capacity is the number of vehicles that can be processed through the work
zone per cycle or per hour. It can be determined based on the saturation
flow rate at the control points and the traffic control “cycle length.”
• Cycle length is determined by the flagging operations or signal timing at
each control point and the time required to travel through the work zone.
Travel time is dependent on the average travel speed of the platoons
traveling through the work zone. Factors that may influence travel speed
include posted speed limit, use of a pilot car, heavy-vehicle percentage,
grade, intensity of construction activity, lane width, lateral distance to the
work activity, and lighting conditions (day versus night).
Performance measures, including delay and queue length, can be calculated
by using capacity and cycle length.
WORK ZONE CAPACITY

The methodology for estimating the capacity of a work zone on a two-lane The work zone capacity
methodology is analogous to
highway with one lane closed is analogous to the capacity calculation for a two- the capacity calculation for a
phase signalized intersection. Average travel speed is estimated from a regression two-phase traffic signal.
model developed through observations of two directions of travel at three work

zones (B-1).
Chapter 26/Freeway and Highway Segments: Supplemental Appendix B: Work Zones on Two-Lane Highways
Step 1: Collect Data

For a typical capacity calculation, the analyst must specify traffic information
(including traffic demands, travel speed, and heavy-vehicle percentage), roadway
geometric configuration (e.g., lane width, lateral clearance, speed limit), and
work zone data (including work zone length, signal green time, and traffic
control plan).
A basic traffic flagger control process for a two-lane highway work zone
involving a lane closure is shown in Exhibit 26-B1. Direction 1 refers to the travel
direction whose lane is blocked by the work zone; Direction 2 refers to the travel
direction with the open lane.
Exhibit 26-B1
Traffic Control for a Two-Lane
Highway Work Zone Involving
a Lane Closure
Source: Schoen et al. (B-1).
Some data, such as average travel speed, saturation flow rate, and green
interval length, may be difficult to collect in the field. In Steps 2–4, the
mathematical models that can be used to estimate these data are presented.
Analysts must note that, for capacity calculations, field data are always more
desirable to use when available.
Measuring two-lane highway A procedure is given in Section 6 of Chapter 31, Signalized Intersections:
work zone saturation flow rates
requires a longer data Supplemental, for determining the saturation flow rate of a signalized
collection time than for a intersection. This procedure involves counting and timing the number of queue
signalized intersection because
of the longer cycle lengths discharge vehicles that pass through an intersection to determine the saturated
involved. vehicle headway. As two-lane highway work zone traffic control typically has a
much longer cycle length than a typical signalized intersection, the time period
for gathering saturation flow data is recommended to be 30–60 min. Of course, a
longer time period is generally more desirable when possible. The work zone
capacity can then be determined from the measured saturation flow rate and the
effective green–to–cycle length ratio.
Unlike the core two-lane highway procedure described in Chapter 15, the
work zone procedure requires that demand volumes be adjusted for the effects of
heavy vehicles and grades, using Equation 26-B1.
Appendix B: Work Zones on Two-Lane Highways Chapter 26/Freeway and Highway Segments: Supplemental
𝑉𝑖 Equation 26-B1
𝑣𝑖 =
𝑃𝐻𝐹 × 𝑓𝑔 × 𝑓𝐻𝑉
where
vi = demand flow rate (pc/h);
i = “d” (analysis direction) or “o” (opposing direction);
Vi = demand volume for direction i (veh/h);
PHF = peak hour factor (decimal);
fg = grade adjustment factor, from Exhibit 26-B2 or Exhibit 26-B3; and
fHV = heavy vehicle adjustment factor, from Equation 26-B2.
Grade Adjustment Factor Calculation

Exhibit 26-B2 shows grade adjustment factors for extended segments of level
and rolling terrain, as well as for specific downgrades. Exhibit 26-B2 is entered
with the one-direction demand flow rate vvph, in vehicles per hour. If demand is
expressed as an hourly volume, it must be divided by the PHF (vvph = V/PHF) to
obtain the appropriate factor. Other adjustment factor tables associated with
Equation 26-B1 are entered with this value as well.
One-Direction Adjustment Factor Exhibit 26-B2

Demand Flow Rate, vvph Level Terrain and Two-Lane Highway Work
(veh/h) Specific Downgrades Rolling Terrain Zone Grade Adjustment
≤100 1.00 0.67 Factor (fg) for Level Terrain,
200 1.00 0.75 Rolling Terrain, and Specific
300 1.00 0.83 Downgrades
400 1.00 0.90
500 1.00 0.95
600 1.00 0.97
700 1.00 0.98
800 1.00 0.99
≥900 1.00 1.00
Note: Interpolation to the nearest 0.01 is recommended.
Exhibit 26-B3 shows grade adjustment factors for specific upgrades. The
negative impact of upgrades on two-lane highway speeds increases as both the
severity of the upgrade and its length increase. The impact declines as demand
flow rate increases. At higher demand flow rates, lower speeds would already
result, and the additional impact of the upgrades is less severe.
Heavy Vehicle Adjustment Factor Calculation

Determining the heavy vehicle adjustment factor is a two-step process:
1. Passenger car equivalents are found for trucks (ET) and recreational
vehicles (RVs) (ER) under prevailing conditions.
2. A heavy vehicle adjustment factor is computed from the passenger car
equivalents with Equation 26-B2.
Exhibit 26-B3 Grade

Two-Lane Highway Work Directional Demand Flow Rate, vvph (veh/h)
Grade Length
Zone Grade Adjustment (%) (mi) 100 200 300 400 500 600 700 800 900
Factor (fg) for Specific 0.25 0.78 0.84 0.87 0.91 1.00 1.00 1.00 1.00 1.00
Upgrades 0.50 0.75 0.83 0.86 0.90 1.00 1.00 1.00 1.00 1.00
0.75 0.73 0.81 0.85 0.89 1.00 1.00 1.00 1.00 1.00
1.00 0.73 0.79 0.83 0.88 1.00 1.00 1.00 1.00 1.00
3 <3.5
1.50 0.73 0.79 0.83 0.87 0.99 0.99 1.00 1.00 1.00
2.00 0.73 0.79 0.82 0.86 0.98 0.98 0.99 1.00 1.00
3.00 0.73 0.78 0.82 0.85 0.95 0.96 0.96 0.97 0.98
4.00 0.73 0.78 0.81 0.85 0.94 0.94 0.95 0.95 0.96
0.25 0.75 0.83 0.86 0.90 1.00 1.00 1.00 1.00 1.00
0.50 0.72 0.80 0.84 0.88 1.00 1.00 1.00 1.00 1.00
0.75 0.67 0.77 0.81 0.86 1.00 1.00 1.00 1.00 1.00
1.00 0.65 0.73 0.77 0.81 0.94 0.95 0.97 1.00 1.00
3.5 <4.5
1.50 0.63 0.72 0.76 0.80 0.93 0.95 0.96 1.00 1.00
2.00 0.62 0.70 0.74 0.79 0.93 0.94 0.96 1.00 1.00
3.00 0.61 0.69 0.74 0.78 0.92 0.93 0.94 0.98 1.00
4.00 0.61 0.69 0.73 0.78 0.91 0.91 0.92 0.96 1.00
0.25 0.71 0.79 0.83 0.88 1.00 1.00 1.00 1.00 1.00
0.50 0.60 0.70 0.74 0.79 0.94 0.95 0.97 1.00 1.00
0.75 0.55 0.65 0.70 0.75 0.91 0.93 0.95 1.00 1.00
1.00 0.54 0.64 0.69 0.74 0.91 0.93 0.95 1.00 1.00
4.5 <5.5
1.50 0.52 0.62 0.67 0.72 0.88 0.90 0.93 1.00 1.00
2.00 0.51 0.61 0.66 0.71 0.87 0.89 0.92 0.99 1.00
3.00 0.51 0.61 0.65 0.70 0.86 0.88 0.91 0.98 0.99
4.00 0.51 0.60 0.65 0.69 0.84 0.86 0.88 0.95 0.97
0.25 0.57 0.68 0.72 0.77 0.93 0.94 0.96 1.00 1.00
0.50 0.52 0.62 0.66 0.71 0.87 0.90 0.92 1.00 1.00
0.75 0.49 0.57 0.62 0.68 0.85 0.88 0.90 1.00 1.00
1.00 0.46 0.56 0.60 0.65 0.82 0.85 0.88 1.00 1.00
5.5 <6.5
1.50 0.44 0.54 0.59 0.64 0.81 0.84 0.87 0.98 1.00
2.00 0.43 0.53 0.58 0.63 0.81 0.83 0.86 0.97 0.99
3.00 0.41 0.51 0.56 0.61 0.79 0.82 0.85 0.97 0.99
4.00 0.40 0.50 0.55 0.61 0.79 0.82 0.85 0.97 0.99
0.25 0.54 0.64 0.68 0.73 0.88 0.90 0.92 1.00 1.00
0.50 0.43 0.53 0.57 0.62 0.79 0.82 0.85 0.98 1.00
0.75 0.39 0.49 0.54 0.59 0.77 0.80 0.83 0.96 1.00
1.00 0.37 0.45 0.50 0.54 0.74 0.77 0.81 0.96 1.00
6.5
1.50 0.35 0.45 0.49 0.54 0.71 0.75 0.79 0.96 1.00
2.00 0.34 0.44 0.48 0.53 0.71 0.74 0.78 0.94 0.99
3.00 0.34 0.44 0.48 0.53 0.70 0.73 0.77 0.93 0.98
4.00 0.33 0.43 0.47 0.52 0.70 0.73 0.77 0.91 0.95
Note: Straight-line interpolation of fg,ATS for length of grade and demand flow permitted to the nearest 0.01.
1
Equation 26-B2 𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1) + 𝑃𝑅 (𝐸𝑅 − 1)
where
fHV = heavy vehicle adjustment factor,
PT = proportion of trucks in the traffic stream (decimal),
PR = proportion of RVs in the traffic stream (decimal),
ET = passenger car equivalent for trucks, and
ER = passenger car equivalent for RVs.
The passenger car equivalent is the number of passenger cars displaced from
the traffic stream by one truck or RV. Passenger car equivalents are defined for:
• Extended sections of general level or rolling terrain,
• Specific upgrades, and
• Specific downgrades.
Exhibit 26-B4 contains passenger car equivalents for trucks and RVs in
general terrain segments and for specific downgrades, which are treated as level
terrain in the lower-speed situations associated with work zones.
Exhibit 26-B4
Directional Demand Level Terrain and
Two-Lane Highway Work
Vehicle Type Flow Rate, vvph (veh/h) Specific Downgrades Rolling Terrain
Zone Passenger Car
≤100 1.9 2.7 Equivalents for Trucks (ET)
200 1.5 2.3
and RVs (ER) for Level
300 1.4 2.1 Terrain, Rolling Terrain, and
400 1.3 2.0
Specific Downgrades
Trucks, ET 500 1.2 1.8
600 1.1 1.7
700 1.1 1.6
800 1.1 1.4
≥900 1.0 1.3
RVs, ER All flows 1.0 1.1
Note: Interpolation to the nearest 0.1 is recommended.
Exhibit 26-B5 and Exhibit 26-B6 show passenger car equivalents for trucks
and RVs, respectively, on specific upgrades.
Step 2: Estimate Average Travel Speed

A simple estimation of average travel speed is obtained as follows. Speeds
for Directions 1 and 2 are calculated by Equation 26-B3 and Equation 26-B4,
respectively. Research on two-lane highway work zones (B-1) found that
Direction 2 (i.e., the direction whose lane is not closed) consistently had higher
average speeds than Direction 1.
𝑆1 = 0.615 × 𝑆𝑝𝑙 − 𝑓𝐿𝑆 − 𝑓𝐴 − 2.4 Equation 26-B3
𝑆2 = 0.692 × 𝑆𝑝𝑙 − 𝑓𝐿𝑆 − 𝑓𝐴 − 2.4 Equation 26-B4
where
Si = average travel speed in direction i (mi/h),
Spl = posted speed limit for the two-lane highway segment (mi/h),
fLS = adjustment for lane and shoulder width from Equation 15-5 (mi/h),
and
fA = adjustment for access-point density from Equation 15-6 (mi/h).
For two-lane highway work zones, these equations provide a constant speed
reduction of 2.4 mi/h in all conditions.
Exhibit 26-B5 Grade

Grade Length
Zone Passenger Car (%) (mi) 100 200 300 400 500 600 700 800 900
Equivalents for Trucks (ET) on 0.25 2.6 2.4 2.3 2.2 1.8 1.8 1.7 1.3 1.1
Specific Upgrades 0.50 3.7 3.4 3.3 3.2 2.7 2.6 2.6 1.9 1.6
0.75 4.6 4.4 4.3 4.2 3.7 3.6 3.4 2.4 1.9
3, 1.00 5.2 5.0 4.9 4.9 4.4 4.2 4.1 3.0 2.3
<3.5 1.50 6.2 6.0 5.9 5.8 5.3 5.0 4.8 3.6 2.9
2.00 7.3 6.9 6.7 6.5 5.7 5.5 5.3 4.1 3.5
3.00 8.4 8.0 7.7 7.5 6.5 6.2 6.0 4.6 3.9
4.00 9.4 8.8 8.6 8.3 7.2 6.9 6.6 4.8 3.7
0.25 3.8 3.4 3.2 3.0 2.3 2.2 2.2 1.7 1.5
0.50 5.5 5.3 5.1 5.0 4.4 4.2 4.0 2.8 2.2
0.75 6.5 6.4 6.5 6.5 6.3 5.9 5.6 3.6 2.6
3.5, 1.00 7.9 7.6 7.4 7.3 6.7 6.6 6.4 5.3 4.7
<4.5 1.50 9.6 9.2 9.0 8.9 8.1 7.9 7.7 6.5 5.9
2.00 10.3 10.1 10.0 9.9 9.4 9.1 8.9 7.4 6.7
3.00 11.4 11.3 11.2 11.2 10.7 10.3 10.0 8.0 7.0
4.00 12.4 12.2 12.2 12.1 11.5 11.2 10.8 8.6 7.5
0.25 4.4 4.0 3.7 3.5 2.7 2.7 2.7 2.6 2.5
0.50 6.0 6.0 6.0 6.0 5.9 5.7 5.6 4.6 4.2
0.75 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5
4.5, 1.00 9.2 9.2 9.1 9.1 9.0 9.0 9.0 8.9 8.8
<5.5 1.50 10.6 10.6 10.6 10.6 10.5 10.4 10.4 10.2 10.1
2.00 11.8 11.8 11.8 11.8 11.6 11.6 11.5 11.1 10.9
3.00 13.7 13.7 13.6 13.6 13.3 13.1 13.0 11.9 11.3
4.00 15.3 15.3 15.2 15.2 14.6 14.2 13.8 11.3 10.0
0.25 4.8 4.6 4.5 4.4 4.0 3.9 3.8 3.2 2.9
0.50 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2
0.75 9.1 9.1 9.1 9.1 9.1 9.1 9.1 9.1 9.1
5.5, 1.00 10.3 10.3 10.3 10.3 10.3 10.3 10.3 10.2 10.1
<6.5 1.50 11.9 11.9 11.9 11.9 11.8 11.8 11.8 11.7 11.6
2.00 12.8 12.8 12.8 12.8 12.7 12.7 12.7 12.6 12.5
3.00 14.4 14.4 14.4 14.4 14.3 14.3 14.3 14.2 14.1
4.00 15.4 15.4 15.3 15.3 15.2 15.1 15.1 14.9 14.8
0.25 5.1 5.1 5.0 5.0 4.8 4.7 4.7 4.5 4.4
0.50 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8
0.75 9.8 9.8 9.8 9.8 9.8 9.8 9.8 9.8 9.8
1.00 10.4 10.4 10.4 10.4 10.4 10.4 10.4 10.3 10.2
6.5
1.50 12.0 12.0 12.0 12.0 11.9 11.9 11.9 11.8 11.7
2.00 12.9 12.9 12.9 12.9 12.8 12.8 12.8 12.7 12.6
3.00 14.5 14.5 14.5 14.5 14.4 14.4 14.4 14.3 14.2
4.00 15.4 15.4 15.4 15.4 15.3 15.3 15.3 15.2 15.1
Note: Interpolation for length of grade and demand flow rate to the nearest 0.1 is recommended.
Exhibit 26-B6
Grade Grade
Zone Passenger Car (%) Length (mi) 100 200 300 400 500 600 700 800 900
Equivalents for RVs (ER) on 0.25 1.1 1.1 1.1 1.0 1.0 1.0 1.0 1.0 1.0
Specific Upgrades >0.25, 0.75 1.2 1.2 1.1 1.1 1.0 1.0 1.0 1.0 1.0
3,
>0.75, 1.25 1.3 1.2 1.2 1.1 1.0 1.0 1.0 1.0 1.0
<3.5
>1.25, 2.25 1.4 1.3 1.2 1.1 1.0 1.0 1.0 1.0 1.0
>2.25 1.5 1.4 1.3 1.2 1.0 1.0 1.0 1.0 1.0
0.75 1.3 1.2 1.2 1.1 1.0 1.0 1.0 1.0 1.0
3.5,
>0.75, 3.50 1.4 1.3 1.2 1.1 1.0 1.0 1.0 1.0 1.0
<4.5
>3.50 1.5 1.4 1.3 1.2 1.0 1.0 1.0 1.0 1.0
4.5, 2.50 1.5 1.4 1.3 1.2 1.0 1.0 1.0 1.0 1.0
<5.5 >2.50 1.6 1.5 1.4 1.2 1.0 1.0 1.0 1.0 1.0
0.75 1.5 1.4 1.3 1.1 1.0 1.0 1.0 1.0 1.0
5.5, >0.75, 2.50 1.6 1.5 1.4 1.2 1.0 1.0 1.0 1.0 1.0
<6.5 >2.50, 3.50 1.6 1.5 1.4 1.3 1.2 1.1 1.0 1.0 1.0
>3.50 1.6 1.6 1.6 1.5 1.5 1.4 1.3 1.2 1.1
2.50 1.6 1.5 1.4 1.2 1.0 1.0 1.0 1.0 1.0
6.5 >2.50, 3.50 1.6 1.5 1.4 1.2 1.3 1.3 1.3 1.3 1.3
>3.50 1.6 1.6 1.6 1.5 1.5 1.5 1.4 1.4 1.4
Note: Interpolation in this exhibit is not recommended.
Step 3: Estimate Saturation Flow Rate

If the saturation flow rate is not measured in the field, a directional
saturation flow rate can be estimated by using Equation 26-B5 with Equation
26-B6 and Equation 26-B7.
3,600 Equation 26-B5
𝑠𝑖 =
ℎ̂𝑖
with
ℎ̂𝑖 = ℎ0 × 𝑓speed,𝑖 Equation 26-B6
𝑓speed,𝑖 = 1 − 0.005(min[𝑆𝑖 , 45] − 45) Equation 26-B7
where
si = saturation flow rate for direction i (pc/h);
ĥi = adjusted time headway for direction i (s);
h0 = base saturation headway (s/pc) = 3,600/1,900 = 1.89 s/pc;
fspeed,i = speed adjustment for direction i (decimal); and
Si = average travel speed in direction i (mi/h).
Step 4: Estimate Green Time

The length of the green interval can be applied directly if a fixed-time signal
is applied at the work zone site. However, most work zones apply flagger
control, for which the green time in each cycle is not fixed. For flagger control
under relatively balanced directional demand conditions, a simple estimation of
optimal directional effective green time can be found by using Equation 26-B8.
20 0.0375𝑙 < 20
𝐺𝑜𝑝𝑡 = {0.0375𝑙 20 ≤ 0.0375𝑙 ≤ 60 Equation 26-B8
60 0.0375𝑙 > 60
where
Gopt = optimal effective green time for one direction (s), and
l = work zone length (ft).

To ensure traffic can be fully discharged in two directions, directional
effective green-time lengths must satisfy Equation 26-B9 with Equation 26-B10.
𝑣𝑖
𝐺𝑖 ≥ 𝐺𝑖,𝑚𝑖𝑛 = (𝐶 − 𝐺𝑖 ) Equation 26-B9
𝑠𝑖 − 𝑣𝑖
with
𝑙 𝑙
𝐶= + + 𝐺1 + 𝐺2 + 2𝐿𝑆 Equation 26-B10
𝑆1,𝑓𝑝𝑠 𝑆2,𝑓𝑝𝑠
where
Gi = effective green time for direction i (s),
Gi,min = minimum effective green time for direction i (s),
si = saturation flow rate for direction i (pc/h),
vi = demand flow rate for direction i (pc/h),

C = cycle length (s),
Si,fps = average travel speed in direction i (ft/s) = (Si × 5,280 ft/mi)/(3,600 s/h),
Si = average travel speed in direction i (mi/h), and
LS = start-up lost time (s).
Step 5: Calculate Capacity

Directional capacity is calculated by Equation 26-B11.
𝑠𝑖 𝐺𝑖
Equation 26-B11 𝑐𝑖 =
𝐶
where
ci = capacity for direction i (pc/h),
si = saturation flow rate for direction i (pc/h),
Gi = effective green time for direction i (s), and
C = cycle length (s).
The start-up lost time, the elapsed time between the last vehicle in the
opposing direction exiting the work zone and the entry of the first queued
vehicle traveling in the subject direction, is assumed to be independent of traffic
direction, as the two directions follow the same traffic control plan. A default
value of 2 s for each direction is recommended.
The total capacity ctotal (in passenger cars per hour) can be calculated by
summing the two directional capacities, as shown in Equation 26-B12.
𝑠1 𝐺1 + 𝑠2 𝐺2
Equation 26-B12 𝑐total = 𝑐1 + 𝑐2 =
𝐶
QUEUING AND DELAY ANALYSIS

The previous steps provide a simple procedure to check two-lane highway
work zone capacity. In practice, it might also be useful to have performance data
such as delay and queuing. Users can apply the model to determine the optimal
control plan while minimizing the vehicle delay and queuing data.
A simple way to estimate vehicle delay and queue length is by assuming
deterministic traffic flow for both directions. Exhibit 26-B7 shows the deterministic
queuing diagram for a two-lane highway work zone. Although more accurate
estimates can be calculated from microscopic simulations that incorporate
random processes, these estimates might be difficult to accomplish in practice
because of the extra time and resources required. Therefore, by a similar
procedure to that used in Chapter 19 for signalized intersection control delay
estimation, the incremental delay caused by random arrivals is added to the
deterministic queuing delay associated with the work zone. The interval gi
shown in the exhibit is the portion of the green time with saturated departures.
The maximum queue length for each direction Qi,max (in passenger cars) is the
height of the triangles in the queue length area of the exhibit. These lengths can
be calculated by Equation 26-B13 and Equation 26-B14 for Directions 1 and 2,

respectively.
𝑣1 𝑙 𝑙 Equation 26-B13
𝑄1,𝑚𝑎𝑥 = ( + + 𝐺2 + 2𝐿𝑆 )
3,600 𝑆1,𝑓𝑝𝑠 𝑆2,𝑓𝑝𝑠
𝑣2 𝑙 𝑙
𝑄2,𝑚𝑎𝑥 = ( + + 𝐺1 + 2𝐿𝑆 ) Equation 26-B14
Exhibit 26-B7
Directional Queueing Diagram
for a Two-Lane Highway
Lane-Closure Work Zone
Source: Schoen et al. (B-1).
For undersaturated conditions, directional vehicle delay caused by a two-lane

highway work zone with one lane closed can be represented by Equation 26-B15
𝑑 = 𝑑1 + 𝑑2 Equation 26-B15
where
d = control delay per passenger car (s/pc),
d1 = uniform control delay assuming uniform traffic arrivals (s/pc), and
d2 = incremental delay resulting from random arrivals and oversaturation
queues (s/pc).
For each direction i, the total directional uniform control delay per cycle Di,1 (in
seconds) is the triangle area in the queue length diagram (Exhibit 26-B7). It is
calculated as one-half the queue length multiplied by the queueing duration. Di,1
is given by Equation 26-B16.
𝑠𝑖 𝑣𝑖
𝐷1,𝑖 = (𝐶 − 𝐺𝑖 )2 Equation 26-B16
2(𝑠𝑖 − 𝑣𝑖 )
The average uniform delay in direction i is given by Equation 26-B17.
𝐷1,𝑖 𝑠𝑖 (𝐶 − 𝐺𝑖 )2
𝑑1,𝑖 = = Equation 26-B17
𝑣𝑖 𝐶 2(𝑠𝑖 − 𝑣𝑖 )𝐶
Finally, by following Equation 19-26 in Chapter 19, the average incremental

delay in direction i is given by Equation 26-B18.
8𝑘𝐼𝑋𝑖
Equation 26-B18 𝑑2,𝑖 = 900 𝑇 [(𝑋𝑖 − 1) + √(𝑋𝑖 − 1)2 + ]
𝑐𝑖 𝑇
where
T = analysis period duration (h),
k = incremental delay factor (decimal),
I = upstream filtering adjustment factor (decimal),
ci = directional capacity (pc/h) from Equation 26-B11, and
Xi = directional volume-to-capacity ratio or degree of saturation (unitless).
Values for k can be calculated with Equation 19-22 in Chapter 19. For fixed-
time control, k = 0.5. Because the purpose of calculating delay in a work zone
context is to identify the optimal effective green time, which is assumed to repeat
every cycle, a value for k of 0.5 is recommended for use in Equation 26-B18. It
incorporates the effects of metered arrivals from upstream signals or work zones.
If the work zone is isolated, then I = 1.0.
The average delay per passenger car is the sum of the directional total
delays, divided by the total number of passenger cars, as shown in Equation 26-
B19. Note that the traffic flow rates used in the equation are in units of passenger
cars per hour; therefore, vehicle delay is calculated in terms of seconds per
passenger car.
(𝑑1,1 + 𝑑2,1 )𝑣1 + (𝑑1,2 + 𝑑2,2 )𝑣2
Equation 26-B19 𝑑=
𝑣1 + 𝑣2
In equations calculating queue length and vehicle delay, all variables are
given by roadway or traffic parameters, except that directional effective green
time Gi should be determined by users. Thus users can change the traffic control
plan to optimize the result. Users must note, however, that they should not
arbitrarily choose an effective green-time value.
EXAMPLE CALCULATION
This subsection presents an example application of the methodology. An
isolated 1,000-ft-long work zone will be located on a rural two-lane highway.
Known peak hour roadway and traffic parameters are summarized in Exhibit 26-
B8 and Exhibit 26-B9.
Lane Width Shoulder Width No. of Access General Terrain Exhibit 26-B8
Direction (ft) (ft) Points per Mile Type Example Calculation: Work
1 12 3 0 Rolling Zone Roadway Parameters
2 12 3 0 Rolling
Speed Limit Traffic Demand Truck RV Exhibit 26-B9

Direction (mi/h) (veh/h) PHF Percentage Percentage Example Calculation: Work
1 45 300 0.88 10.0 10.0 Zone Traffic Parameters
2 45 300 0.88 10.0 10.0
Step 1: Collect Data

Most of the necessary data are provided in the problem statement. However,
the traffic demand Vi (in vehicles per hour) must be converted into a traffic flow
rate vi (in passenger cars per hour).
𝑉𝑖
𝑣𝑖 =
𝑃𝐻𝐹 × 𝑓𝑔 × 𝑓𝐻𝑉
This equation requires determining both an adjustment factor for grade (in
this case, general terrain) and an adjustment factor for heavy vehicles (which also
includes terrain effects). In addition, a peak hour factor (PHF) is applied; this was
given in the problem statement as 0.88 for each direction.
From Exhibit 26-B2, the grade adjustment factor fg for rolling terrain is 0.83,
while from Exhibit 26-B4, the truck PCE is 2.1 and the RV PCE is 1.1. The heavy
vehicle adjustment factor fHV can then be calculated from Equation 26-B2.
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1) + 𝑃𝑅 (𝐸𝑅 − 1)
1
𝑓𝐻𝑉 =
1 + (0.10)(2.1 − 1) + (0.10)(1.1 − 1)
𝑓𝐻𝑉 = 0.89
then
300
𝑣1 = 𝑣2 = = 461 pc/h
0.88 × 0.83 × 0.89
Step 2: Estimate Average Travel Speed

Average travel speed through the work zone is calculated with Equation 26-
B3 and Equation 26-B4 for Directions 1 and 2, respectively.
𝑆1 = 0.615 × 𝑆𝑝𝑙 − 𝑓𝐿𝑆 − 𝑓𝐴 − 2.4
𝑆2 = 0.692 × 𝑆𝑝𝑙 − 𝑓𝐿𝑆 − 𝑓𝐴 − 2.4
The speed limit Spl is given. From Equation 15-5, the adjustment for lane and
shoulder width fLS for 12-ft lane widths and 3-ft shoulder widths is as follows:
𝑓𝐿𝑆 = 0.6 × (12 − 𝐿𝑊) + 0.7 × (6 − 𝑆𝑊)
𝑓𝐿𝑆 = 0.6 × (12 − 12) + 0.7 × (6 − 3)
𝑓𝐿𝑆 = 2.1 mi/h
Finally, from Equation 15-6, the adjustment for access point density is 0.0
mi/h when no access points are present. Then
𝑆1 = 0.615 × 45 − 2.1 − 0 − 2.4 = 23.2 mi/h

𝑆2 = 0.692 × 45 − 2.1 − 0 − 2.4 = 26.6 mi/h
Step 3: Estimate Saturation Flow Rate

Equation 26-B5 through Equation 26-B7 are used to estimate the saturation
flow rate through the work zone.
First, the speed adjustment factor is calculated for each direction as follows:
𝑓speed,𝑖 = 1 − 0.005(min[𝑆𝑖 , 45] − 45)
𝑓speed,1 = 1 − 0.005(min[23.2,45] − 45) = 1.11
𝑓speed,2 = 1 − 0.005(min[26.6,45] − 45) = 1.09
Next, an adjusted time headway is calculated for each direction as follows:
ℎ̂𝑖 = ℎ0 × 𝑓speed,𝑖
̂1 = 1.89 × 1.11 = 2.10 s
ℎ
̂2 = 1.89 × 1.09 = 2.06 s
ℎ
where the base saturation headway of 1.89 s/pc is as given in the text following
Equation 26-B6.
Finally, the saturation flow rate for each direction is calculated as
3,600
𝑠𝑖 =
ℎ̂𝑖
3,600
𝑠1 = = 1,714 pc/h/ln
2.10
3,600
𝑠2 = = 1,748 pc/h/ln
2.06
Step 4: Estimate Green Time

In Step 4, the effective green time length is determined. It may be difficult to
choose a green time value without knowing the traffic performance parameters,
but an estimate of the optimal value can be obtained with Equation 26-B8.
20 0.0375𝑙 < 20
𝐺𝑜𝑝𝑡 = {0.0375𝑙 20 ≤ 0.0375𝑙 ≤ 60
60 0.0375𝑙 > 60
As the work zone will be 1,000 ft long, the value 0.0375l computes to 37.5 s.
As 37.5 is between 20 and 60, it can be used directly; however, this value should
be checked to make sure it is long enough to discharge the vehicle queues.
Equation 26-B9 provides this check.
𝑣𝑖
𝐺𝑖 ≥ 𝐺𝑖,𝑚𝑖𝑛 = (𝐶 − 𝐺𝑖 )
𝑠𝑖 − 𝑣𝑖
The cycle length C is computed from Equation 26-B10, incorporating a
default value of 2.0 s for the start-up lost time.
𝑙 𝑙
𝐶= + + 𝐺1 + 𝐺2 + 2𝐿𝑆
𝑆1,𝑓𝑝𝑠 𝑆2,𝑓𝑝𝑠
1,000 1,000
𝐶= + + 37.5 + 37.5 + 2(2.0)
23.2 × 5,280/3,600 26.6 × 5,280/3,600
𝐶 = 134.0 s
then
461
𝐺1,𝑚𝑖𝑛 = (134.0 − 37.5) = 35.5 s
1,714 − 461
461
𝐺2,𝑚𝑖𝑛 = (134.0 − 37.5) = 34.6 s
1,748 − 461
As the optimal effective green time of 37.5 s is greater than the minimum
required time for each direction, it is accepted, and the process continues to
Step 5.

Directional capacity is calculated with Equation 26-B11.
𝑠𝑖 𝐺𝑖
𝑐𝑖 =
𝐶
(1,714)(37.5)
𝑐1 = = 480 pc/h
134.0
(1,748)(37.5)
𝑐2 = = 489 pc/h
134.0
As v1 < c1 and v2 < c2, this 1,000-ft work zone can serve the traffic demand
without accumulating vehicle queues when the effective green time is 37.5 s for
both directions.
Queuing and Delay

If desired, the maximum queue length and average vehicle delay can be
calculated for both directions. The maximum queue length is calculated from
Equation 26-B13 and Equation 26-B14 for Directions 1 and 2, respectively.
𝑣1 𝑙 𝑙
𝑄1,𝑚𝑎𝑥 = ( + + 𝐺2 + 2𝐿𝑆 )
461
𝑄1,𝑚𝑎𝑥 = (29.4 + 25.6 + 37.5 + 4.0) = 13 veh
3,600
𝑣2 𝑙 𝑙
𝑄2,𝑚𝑎𝑥 = ( + + 𝐺1 + 2𝐿𝑆 )
461
𝑄2,𝑚𝑎𝑥 = (29.4 + 25.6 + 37.5 + 4.0) = 13 veh
3,600
The average uniform delay by direction is calculated with Equation 26-B17.
𝑠𝑖 (𝐶 − 𝐺𝑖 )2
𝑑1,𝑖 =
2(𝑠𝑖 − 𝑣𝑖 )𝐶
(1,714)(134.0 − 37.5)2
𝑑1,1 = = 47.5 s/veh
(2)(1,714 − 461)(134.0)
(1,748)(134.0 − 37.5)2
𝑑1,2 = = 47.2 s/veh
(2)(1,748 − 461)(134.0)
The average incremental delay by direction is calculated from Equation 26-

B18 The recommended value of 0.5 is used for the incremental delay factor k, and
as the work zone is isolated, a value of 1.0 is used for the upstream filtering
adjustment factor I.
8𝑘𝐼𝑋𝑖
𝑑2,𝑖 = 900 𝑇 [(𝑋𝑖 − 1) + √(𝑋𝑖 − 1)2 + ]
𝑐𝑖 𝑇
461
2 (8)(0.5)(1.0) (
𝑑2,1 = (900)(1) [(
461 √
− 1) + (
461
− 1) + 480)] = 52.4 s
480 480 (480)(1)
461
2 (8)(0.5)(1.0) (
𝑑2,2 = (900)(1) [(
461 √
− 1) + (
461
− 1) + 489)] = 42.8 s
489 489 (489)(1)
Finally, the average delay per passenger car is given by Equation 26-B19.
(47.5 + 52.4)(461) + (47.2 + 42.8)(461)
𝑑= = 95.0 s
461 + 461
REFERENCE
B-1. Schoen, J. M., J. A. Bonneson, C. Safi, B. Schroeder, A. Hajbabaie, C. H.
Yeom, N. Rouphail, Y. Wang, W. Zhu, and Y. Zou. Work Zone Capacity
Methods for the Highway Capacity Manual. National Cooperative Highway
Research Program Project 3-107 final report, preliminary draft. Kittelson &
Associates, Inc., Tucson, Ariz., April 2015.
HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 27
FREEWAY WEAVING: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 27-1
2. EXAMPLE PROBLEMS......................................................................................... 27-2

Example Problem 1: LOS of a Major Weaving Segment ................................ 27-2
Example Problem 2: LOS for a Ramp Weave .................................................. 27-7
Example Problem 3: LOS of a Two-Sided Weaving Segment ..................... 27-12
Example Problem 4: Design of a Major Weaving Segment for a
Desired LOS ................................................................................................ 27-16
Example Problem 5: Constructing a Service Volume Table for a
Weaving Segment ....................................................................................... 27-22
Example Problem 6: LOS of an ML Access Segment with Cross-
Weaving ....................................................................................................... 27-27
Example Problem 7: ML Access Segment with Downstream Off-Ramp ... 27-32
3. ALTERNATIVE TOOL EXAMPLES FOR WEAVING SEGMENTS ......... 27-37

Determining the Weaving Segment Capacity ............................................... 27-38
Effect of Demand on Performance .................................................................. 27-39
Effect of Queue Backup from a Downstream Signal on the Exit Ramp ..... 27-40
Chapter 27/Freeway Weaving: Supplemental Contents

LIST OF EXHIBITS
Exhibit 27-1 List of Example Problems for Weaving Segment Analysis .............27-2
Exhibit 27-2 Example Problem 1: Major Weaving Segment Data ........................27-2
Exhibit 27-3 Example Problem 1: Determination of Configuration
Variables ...............................................................................................................27-4
Exhibit 27-4 Example Problem 1: Capacity of Entry and Exit Roadways ...........27-5
Exhibit 27-5 Example Problem 2: Ramp-Weave Segment Data ............................27-7
Exhibit 27-6 Example Problem 2: Configuration Characteristics .........................27-9
Exhibit 27-7 Example Problem 2: Capacity of Entry and Exit Legs ...................27-10
Exhibit 27-8 Example Problem 3: Two-Sided Weaving Segment Data..............27-12
Exhibit 27-9 Example Problem 3: Configuration Characteristics .......................27-14
Exhibit 27-10 Example Problem 3: Capacity of Entry and Exit Legs .................27-15
Exhibit 27-11 Example Problem 4: Major Weaving Segment Data ....................27-17
Exhibit 27-12 Example Problem 4: Trial Design 1 ................................................27-18
Exhibit 27-13 Example Problem 4: Trial Design 2 ................................................27-20
Exhibit 27-14 Example Problem 5: Maximum Density Thresholds for LOS
A–D ......................................................................................................................27-23
Exhibit 27-15 Example Problem 5: Service Flow Rates (pc/h) Under Ideal
Conditions (SFI) .................................................................................................27-25
Exhibit 27-16 Example Problem 5: Service Flow Rates (veh/h) Under
Prevailing Conditions (SF) ...............................................................................27-25
Exhibit 27-17 Example Problem 5: Service Volumes (veh/h) Under
Prevailing Conditions (SV) ...............................................................................27-26
Exhibit 27-18 Example Problem 5: Daily Service Volumes (veh/day)
Under Prevailing Conditions (DSV)................................................................27-26
Exhibit 27-19 Example Problem 6: ML Access Segment with Cross-
Weaving ..............................................................................................................27-27
Exhibit 27-20 Example Problem 6: Hourly Flow Rates After PHF Is
Applied ...............................................................................................................27-29
Exhibit 27-21 Example Problem 6: Configuration Characteristics .....................27-29
Exhibit 27-22 Example Problem 6: Capacity of Entry and Exit Legs .................27-31
Exhibit 27-23 Example Problem 7: ML Access Segment Data ............................27-32
Exhibit 27-24 Example Problem 7: Weaving Flows for Managed Lane
Segment ...............................................................................................................27-33
Exhibit 27-25 Link–Node Structure for the Simulated Weaving Segment ........27-37
Exhibit 27-26 Input Data for Various Demand Levels (veh/h) ...........................27-37
Exhibit 27-27 Determining the Capacity of a Weaving Segment by
Simulation ...........................................................................................................27-38
Contents Chapter 27/Freeway Weaving: Supplemental

Exhibit 27-28 Simulated Effect of Demand Volume on Weaving Segment

Capacity and Speed ........................................................................................... 27-39
Exhibit 27-29 Exit Ramp Signal Operating Parameters ....................................... 27-40
Exhibit 27-30 Deterioration of Weaving Segment Operation due to Queue
Backup from a Traffic Signal ............................................................................ 27-41
Exhibit 27-31 Effect of Demand on Weaving Segment Throughput with
Exit Ramp Backup ............................................................................................. 27-41
Exhibit 27-32 Effect of Demand on Exit Ramp Throughput with Signal
Queuing .............................................................................................................. 27-42
Chapter 27/Freeway Weaving: Supplemental Contents

1. INTRODUCTION
Chapter 27 is the supplemental chapter for Chapter 13, Freeway Weaving VOLUME 4: APPLICATIONS
GUIDE
Segments, which is found in Volume 2 of the Highway Capacity Manual (HCM). 25. Freeway Facilities:
Section 2 provides seven example problems demonstrating the application of the Supplemental
Chapter 13 core methodology and its extension to freeway managed lanes. Segments: Supplemental
Section 3 presents examples of applying alternative tools to the analysis of 27. Freeway Weaving:
Supplemental
freeway weaving sections to address limitations of the Chapter 13 methodology. 28. Freeway Merges and
Supplemental
30. Urban Street Segments:
Supplemental
Supplemental
32. STOP-Controlled
Intersections:
Supplemental
33. Roundabouts:
Supplemental
Terminals: Supplemental
Supplemental
36. Concepts: Supplemental
Chapter 27/Freeway Weaving: Supplemental Introduction

2. EXAMPLE PROBLEMS
The example problems in this section illustrate various applications of the

freeway weaving segment methodology detailed in Chapter 13. Exhibit 27-1 lists
the example problems included. Example problem results from intermediate and
final calculations were derived by using a handheld scientific calculator with 12-
digit precision. For displaying equation results in text, the results were
appropriately rounded. Users may obtain slightly different results if rounded
parameters are used in intermediate and final calculations.

List of Example Problems for Problem Description Application
Weaving Segment Analysis 1 LOS of a major weaving segment Operational analysis
2 LOS for a ramp weave Operational analysis
3 LOS of a two-sided weaving segment Operational analysis
4 Design of a major weaving segment for a desired LOS Design analysis
5 Service volume table construction Planning analysis
6 LOS of an ML access segment with cross-weaving Operational analysis
7 ML access segment with downstream off-ramp Operational analysis
EXAMPLE PROBLEM 1: LOS OF A MAJOR WEAVING SEGMENT

The Weaving Segment
The subject of this operational analysis is a major weaving segment on an
urban freeway under nonsevere weather conditions and without incidents, as
shown in Exhibit 27-2. The short length of the weaving segment LS is 1,500 ft.
Exhibit 27-2
Example Problem 1: Major
Weaving Segment Data
What is the level of service (LOS) and capacity of the weaving segment shown in
Exhibit 27-2?
The Facts
In addition to the information contained in Exhibit 27-2, the following
characteristics of the weaving segment are known:
PHF = 0.91 (for all movements);
Heavy vehicles = 5% trucks;
Example Problems Chapter 27/Freeway Weaving: Supplemental

Free-flow speed (FFS) = 65 mi/h; ramp FFS = 50 mi/h;

ID = 0.8 interchange/mi; and
Terrain = level.
Note that the ideal freeway capacity per lane cIFL is the capacity of a basic
freeway segment, where the FFS is 65 mi/h. It is drawn from the methodology of
Chapter 12, Basic Freeway and Multilane Highway Segments.
Comments
Chapter 12, Basic Freeway and Multilane Highway Segments, must be
consulted to find appropriate values for the heavy-vehicle adjustment factor fHV.
Chapter 26, Section 2, should be consulted if the driver population includes a
significant proportion of noncommuters.
All input parameters have been specified, so default values are not needed.
Demand volumes are given in vehicles per hour under prevailing conditions.
These must be converted to passenger cars per hour under equivalent ideal
conditions for use with the weaving methodology. The weaving segment length
must be compared with the maximum length for weaving analysis to determine
whether the Chapter 13 methodology is applicable. The capacity of the weaving
segment is estimated and compared with the total demand flow to determine
whether LOS F exists. Lane-changing rates are calculated to allow estimations of
speed for weaving and nonweaving flows. Average overall speed and density
are computed and compared with the criteria of Exhibit 13-6 to determine LOS.
Without specific information to the contrary, it is assumed that good weather
conditions prevail and that there are no incidents during the analysis period.
Step 1: Input Data

All inputs have been specified in Exhibit 27-2 and the Facts section of the
problem statement.
Step 2: Adjust Volume

Equation 13-1 is used to convert the four component demand volumes to
flow rates under equivalent ideal conditions. Chapter 12 is consulted to obtain a
value of ET (2.0 for level terrain). From Chapter 12, the heavy-vehicle adjustment
factor is computed as
1 1
𝑓𝐻𝑉 = = = 0.952
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.05(2 − 1)
Equation 13-1 is now used to convert all demand volumes:
𝑉𝑖
𝑣𝑖 =
𝑃𝐻𝐹 × 𝑓𝐻𝑉
1,815
𝑣𝐹𝐹 = = 2,094 pc/h
0.91 × 0.952
692
𝑣𝐹𝑅 = = 798 pc/h
0.91 × 0.952
Chapter 27/Freeway Weaving: Supplemental Example Problems

1,037
𝑣𝑅𝐹 = = 1,197 pc/h
0.91 × 0.952
1,297
𝑣𝑅𝑅 = = 1,497 pc/h
0.91 × 0.952
Then
𝑣𝑊 = 798 + 1,197 = 1,995 pc/h
𝑣𝑁𝑊 = 2,094 + 1,497 = 3,591 pc/h
𝑣 = 1,995 + 3,591 = 5,586 pc/h
1,995
𝑉𝑅 = = 0.357
5,586
Step 3: Determine Configuration Characteristics

The configuration is examined to determine the values of LCRF, LCFR, and
NWL. These determinations are illustrated in Exhibit 27-3. From these values, the
minimum number of lane changes by weaving vehicles, LCMIN, is then computed
Exhibit 27-3
Example Problem 1:
Determination of
Configuration Variables
Exhibit 27-3 indicates that ramp-to-freeway vehicles can execute their

weaving maneuver without making a lane change (if they so desire). Thus, LCRF =
0. Freeway-to-ramp vehicles must make at least one lane change to complete
their desired maneuver. Thus, LCFR = 1. If optional lane changes are considered,
weaving movements can be accomplished with one or no lane changes from both
entering ramp lanes and from the rightmost freeway lane. Thus, NWL = 3.
Equation 13-2 can now be applied:
𝐿𝐶𝑀𝐼𝑁 = (𝐿𝐶𝑅𝐹 × 𝑣𝑅𝐹 ) + (𝐿𝐶𝐹𝑅 × 𝑣𝐹𝑅 )
𝐿𝐶𝑀𝐼𝑁 = (0 × 1,197) + (1 × 798) = 798 lc/h
Step 4: Determine Maximum Weaving Length

The maximum length over which weaving movements may exist is
determined by Equation 13-4. The determination is case-specific, and the result is
valid only for the case under consideration:
𝐿𝑀𝐴𝑋 = [5,728(1 + 𝑉𝑅)1.6 ] − (1,566𝑁𝑊𝐿 )
𝐿𝑀𝐴𝑋 = [5,728(1 + 0.357)1.6 ] − (1,566 × 3) = 4,639 ft
Since the maximum length is significantly greater than the actual segment
length of 1,500 ft, weaving operations do exist, and the analysis may continue
with the weaving analysis methodology.

Step 5: Determine Weaving Segment Capacity

Capacity may be controlled by one of two factors: operations reaching a
maximum density of 43 pc/mi/ln or by the weaving demand flow rate reaching
3,500 pc/h (for a weaving segment with NWL = 3). Equations 13-5 through 13-10
are used to make these determinations.
Capacity Controlled by Density

𝑐𝐼𝑊𝐿 = 𝑐𝐼𝐹𝐿 − [438.2(1 + 𝑉𝑅)1.6] + (0.0765𝐿𝑆 ) + (119.8𝑁𝑊𝐿 )
𝑐𝐼𝑊𝐿 = 2,350 − [438.2(1 + 0.357)1.6 ] + (0.0765 × 1,500) + (119.8 × 3)
𝑐𝐼𝑊𝐿 = 2,110 pc/h/ln
𝑐𝑊 = 𝑐𝐼𝑊𝐿 × 𝑁 × 𝑓𝐻𝑉
𝑐𝑊 = 2,110 × 4 × 0.952 = 8,038 veh/h
Capacity Controlled by Maximum Weaving Flow Rate

3,500 3,500
𝑐𝐼𝑊 = = = 9,800 pc/h
𝑉𝑅 0.357
𝑐𝑊 = 9,800 × 0.952 × 1 = 9,333 veh/h
Note that the methodology computes the capacity controlled by density in
passenger cars per hour per lane, while the capacity controlled by maximum
weaving flow rate is computed in passenger cars per hour. After conversion,
however, both are in units of vehicles per hour.
The controlling value is the smaller of the two, or 8,038 veh/h. Since the total
demand flow rate is only 5,320 veh/h, the capacity is clearly sufficient, and this
situation will not result in LOS F.
Capacity of Input and Output Roadways

The capacity of the entry and exit roadways should also be checked,
although this is rarely a factor in weaving segment operation. Basic capacities for
the freeway entry and exit legs (with FFS = 65 mi/h) are taken from Chapter 12,
while the capacity for the two-lane entry and exit ramps (with ramp FFS = 50
mi/h) is taken from Chapter 14. The comparisons are shown in Exhibit 27-4.
Leg Demand Flow (pc/h) Capacity (pc/h) Exhibit 27-4

Freeway entry 2,094 + 798 = 2,892 2 × 2,350 = 4,700 Example Problem 1: Capacity
Freeway exit 1,197 + 2,094 = 3,291 3 × 2,350 = 7,050 of Entry and Exit Roadways
Ramp entry 1,197 + 1,497 = 2,694 4,200
Ramp exit 798 + 1,497 = 2,295 4,200
As can be seen, capacity is sufficient on each of the entry and exit roadways
and will therefore not affect operations within the weaving segment.
Step 6: Determine Lane-Changing Rates

Equations 13-11 through 13-17 are used to estimate the lane-changing rates of
weaving and nonweaving vehicles in the weaving segment. In turn, these will be
used to estimate weaving and nonweaving vehicle speeds.

Weaving Vehicle Lane-Changing Rate

𝐿𝐶𝑊 = 𝐿𝐶𝑀𝐼𝑁 + 0.39[(𝐿𝑆 − 300)0.5 𝑁 2 (1 + 𝐼𝐷)0.8 ]
𝐿𝐶𝑊 = 798 + 0.39[(1,500 − 300)0.5 (42 )(1 + 0.8)0.8] = 1,144 lc/h
Nonweaving Vehicle Lane-Changing Rate

𝐿𝑆 × 𝐼𝐷 × 𝑣𝑁𝑊
𝐼𝑁𝑊 =
10,000
1,500 × 0.8 × 3,591
𝐼𝑁𝑊 = = 431 < 1,300
10,000
𝐿𝐶𝑁𝑊 = 𝐿𝐶𝑁𝑊1 = (0.206𝑣𝑁𝑊 ) + (0.542𝐿𝑆 ) − (192.6𝑁)
𝐿𝐶𝑁𝑊 = (0.206 × 3,591) + (0.542 × 1,500) − (192.6 × 4) = 782 lc/h
Total Lane-Changing Rate

𝐿𝐶𝐴𝐿𝐿 = 𝐿𝐶𝑊 + 𝐿𝐶𝑁𝑊 = 1,144 + 782 = 1,926 lc/h
Step 7: Determine Average Speeds of Weaving and Nonweaving

Vehicles
The average speeds of weaving and nonweaving vehicles are computed from
Equation 13-18 through Equation 13-21:
𝐿𝐶𝐴𝐿𝐿 0.789
𝑊 = 0.226 ( )
𝐿𝑆
1,926 0.789
𝑊 = 0.226 ( ) = 0.275
1,500
Then
𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 15
𝑆𝑊 = 15 + ( )
1+𝑊
𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 15 65 × 1 − 15
𝑆𝑊 = 15 + ( ) = 15 + ( ) = 54.2 mi/h
1+𝑊 1 + 0.275
and
𝑣
𝑆𝑁𝑊 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (0.0072𝐿𝐶𝑀𝐼𝑁 ) − (0.0048 )
𝑁
5,586
𝑆𝑁𝑊 = 65 × 1 − (0.0072 × 798) − (0.0048 ) = 52.5 mi/h
4
Equation 13-22 is now used to compute the average speed of all vehicles in
the segment:
𝑣𝑊 + 𝑣𝑁𝑊
𝑆= 𝑣 𝑣
(𝑆𝑊 ) + (𝑆𝑁𝑊 )
𝑊 𝑁𝑊
3,591 + 1,995
𝑆= = 53.1 mi/h
3,591 1,995
( 52.5 ) + ( 54.2 )


Equation 13-23 is used to convert the average speed of all vehicles in the
segment to an average density:
(𝑣/𝑁) (5,586/4)
𝐷= = = 26.3 pc/mi/ln
𝑆 53.1
The resulting density of 26.3 pc/mi/ln is compared with the LOS criteria of
Exhibit 13-6. The LOS is C, since the density is within the specified range of 20 to
28 pc/h/ln for that level.
Discussion
As indicated by the results, this weaving segment operates at LOS C, with an
average speed of 53.1 mi/h for all vehicles. Weaving vehicles travel a bit faster
than nonweaving vehicles, primarily because the configuration favors weaving
vehicles and many weaving maneuvers can be made without a lane change. In
turn, the method estimates that nonweaving vehicles are affected by the weave
turbulence, which results in a drop in speed of those movements. The demand
flow rate of 5,320 veh/h is considerably less than the capacity of the segment,
8,038 veh/h. In other words, demand can grow significantly before reaching the
capacity of the segment.
EXAMPLE PROBLEM 2: LOS FOR A RAMP WEAVE

The Weaving Segment
The weaving segment that is the subject of this operational analysis, under
nonsevere weather conditions and without incidents, is shown in Exhibit 27-5. It
is a typical ramp-weave segment.
Exhibit 27-5
Example Problem 2: Ramp-
Weave Segment Data
What is the capacity of the weaving segment of Exhibit 27-5, and at what
LOS is it expected to operate with the demand flow rates as shown?
The Facts
In addition to the information given in Exhibit 27-5, the following facts are
known about the subject weaving segment:
PHF = 1.00 (demands stated as flow rates);
Heavy vehicles = 0%; demand given in passenger car equivalents;


FFS = 75 mi/h; SFR = 40 mi/h;
ID = 1.0 int/mi; and
Terrain = level.
Comments
Because the demands have been specified as flow rates in passenger cars per
hour under equivalent ideal conditions, Chapter 12 does not have to be consulted
to obtain appropriate adjustment factors.
Several of the computational steps related to converting demand volumes to
flow rates under equivalent ideal conditions are unnecessary, since demands are
already specified in that form. Lane-changing characteristics will be estimated.
The maximum length for weaving operations in this case will be estimated and
compared with the actual length of the segment. The capacity of the segment will
be estimated and compared with the demand to determine whether LOS F exists.
If it does not, component flow speeds will be estimated and averaged. A density
will be estimated and compared with the criteria of Exhibit 13-6 to determine the
expected LOS.
Step 1: Input Data

All input data are stated in Exhibit 27-5 and the Facts section.

Because all demands are stated as flow rates in passenger cars per hour
under equivalent ideal conditions, no further conversions are necessary. Key
volume parameters are as follows:
𝑣𝐹𝐹 = 4,000 pc/h
𝑣𝐹𝑅 = 300 pc/h
𝑣𝑅𝐹 = 600 pc/h
𝑣𝑅𝑅 = 100 pc/h
𝑣𝑊 = 300 + 600 = 900 pc/h
𝑣𝑁𝑊 = 4,000 + 100 = 4,100 pc/h
𝑣 = 4,100 + 900 = 5,000 pc/h
900
𝑉𝑅 = = 0.180
5,000

The configuration is examined to determine the values of LCRF, LCFR, and
NWL. These determinations are illustrated in Exhibit 27-6. From these values, the
minimum number of lane changes by weaving vehicles LCMIN is then computed

Exhibit 27-6
Example Problem 2:
Configuration Characteristics
From Exhibit 27-6, it is clear that all ramp-to-freeway vehicles must make at
least one lane change (LCRF = 1) and that all freeway-to-ramp vehicles must make
at least one lane change (LCFR = 1). It is also clear that a weaving maneuver can
only be completed with a single lane change from the right lane of the freeway or
the auxiliary lane (NWL = 2). Then, by using Equation 13-2, LCMIN is computed as
𝐿𝐶𝑀𝐼𝑁 = (1 × 600) + (1 × 300) = 900 lc/h

The maximum length over which weaving operations may exist for the
segment described is found by using Equation 13-4:
𝐿𝑀𝐴𝑋 = [5,728(1 + 𝑉𝑅)1.6 ] − (1,566𝑁𝑊𝐿 )
𝐿𝑀𝐴𝑋 = [5,728(1 + 0.180)1.6 ] − (1,566 × 2) = 4,333 ft > 1,000 ft
Since the maximum length for weaving operations significantly exceeds the
actual length, this is a weaving segment, and the analysis continues.

The capacity of the weaving segment is controlled by one of two limiting
factors: density reaches 43 pc/mi/ln or weaving demand reaches 2,400 pc/h for
the configuration of Exhibit 27-5 (a ramp weave with NWL = 2).
Capacity Limited by Density

The capacity limited by reaching a density of 43 pc/mi/ln is estimated by
using Equation 13-5 and Equation 13-6:
𝑐𝐼𝑊𝐿 = 2,400 − [438.2(1 + 0.180)1.6 ] + (0.0765 × 1,000) + (119.8 × 2)
𝑐𝑊 = 2,145 × 4 = 8,580 pc/h
Capacity Limited by Weaving Demand Flow

The capacity limited by the weaving demand flow is estimated by using
Equation 13-7 and Equation 13-8:
2,400 2,400
𝑐𝐼𝑊 = = = 13,333 pc/h
𝑉𝑅 0.180
𝑐𝑊 = 𝑐𝐼𝑊 × 𝑓𝐻𝑉
𝑐𝑊 = 13,333 × 1 = 13,333 pc/h

The controlling capacity is the smaller value, or 8,580 pc/h. At this point, the
value is usually stated as vehicles per hour. In this case, because inputs were
already adjusted and were stated in passenger cars per hour, conversions back to
vehicles per hour are not possible.
Since the capacity of the weaving segment is larger than the demand flow
rate of 5,000 pc/h, LOS F does not exist, and the analysis may continue.

Although it is rarely a factor in weaving operations, the capacity of input and
output roadways should be checked to ensure that no deficiencies exist. There
are three input and output freeway lanes (with FFS = 75 mi/h) and one lane on
the entrance and exit ramps (with ramp FFS = 35 mi/h). The criteria of Chapter 12
and Chapter 14, respectively, are used to determine the capacity of freeway legs
and ramps. Demand flows and capacities are compared in Exhibit 27-7.
Exhibit 27-7 Leg Demand Flow (pc/h) Capacity (pc/h)

Example Problem 2: Capacity Freeway entry 4,000 + 300 = 4,300 3 × 2,400 = 7,200
of Entry and Exit Legs Freeway exit 4,000 + 600 = 4,600 3 × 2,400 = 7,200
Ramp entry 600 + 100 = 700 2,000
Ramp exit 300 + 100 = 400 2,000
The capacity of all input and output roadways is sufficient to accommodate

the demand flow rates.

Equation 13-11 through Equation 13-17 are used to estimate the lane-
changing rates of weaving and nonweaving vehicles in the weaving segment. In
turn, these will be used to estimate weaving and nonweaving vehicle speeds.

𝐿𝐶𝑊 = 𝐿𝐶𝑀𝐼𝑁 + 0.39[(𝐿𝑆 − 300)0.5 𝑁 2 (1 + 𝐼𝐷)0.8 ]
𝐿𝐶𝑊 = 900 + 0.39[(1,000 − 300)0.5 (42 )(1 + 1)0.8 ] = 1,187 lc/h

𝐼𝑁𝑊 =
10,000
1,000 × 1 × 4,100
𝐼𝑁𝑊 = = 410 < 1,300
10,000
𝐿𝐶𝑁𝑊 = 𝐿𝐶𝑁𝑊1 = (0.206𝑣𝑁𝑊 ) + (0.542𝐿𝑆 ) − (192.6𝑁)
𝐿𝐶𝑁𝑊 = (0.206 × 4,100) + (0.542 × 1,000) − (192.6 × 4) = 616 lc/h


Step 7: Determine Average Speeds of Weaving and Nonweaving Vehicles

𝑊 = 0.226 ( )
𝐿𝑆
1,803 0.789
𝑊 = 0.226 ( ) = 0.360
1,000
Then
𝑆𝑊 = 15 + ( )
1+𝑊
75 × 1 − 15
𝑆𝑊 = 15 + ( ) = 59.1 mi/h
1 + 0.360
and
𝑣
𝑆𝑁𝑊 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (0.0072𝐿𝐶𝑀𝐼𝑁 ) − (0.0048 )
𝑁
5,000
𝑆𝑁𝑊 = 75 × 1 − (0.0072 × 900) − (0.0048 ) = 62.5 mi/h
4
the segment:
𝑆= 𝑣 𝑣
(𝑆𝑊 ) + (𝑆𝑁𝑊 )
𝑊 𝑁𝑊
4,100 + 900
𝑆= = 61.9 mi/h
4,100 900
( )+( )
62.5 59.1

The average density in the weaving segment is estimated by using Equation
13-23.
(𝑣/𝑁) (5,000/4)
𝐷= = = 20.2 pc/mi/ln
𝑆 61.9
From Exhibit 13-6, this density is within the stated boundaries of LOS C (20
to 28 pc/mi/ln). However, it is very close to the LOS B boundary condition.
Discussion
As noted, the segment is operating well (LOS C) and is close to the LOS B
boundary. Weaving and nonweaving speeds are relatively high, suggesting a
stable flow. The demand flow rate of 5,000 pc/h is well below the capacity of the
segment (8,580 pc/h). Weaving vehicles travel somewhat more slowly than
nonweaving vehicles, which is typical of ramp-weave segments, where the vast
majority of nonweaving vehicles are running from freeway to freeway.

EXAMPLE PROBLEM 3: LOS OF A TWO-SIDED WEAVING SEGMENT

The Weaving Segment
The weaving segment that is the subject of this example problem is shown in
Exhibit 27-8. The analysis assumes no adverse weather effects or incidents in the
segment.
Exhibit 27-8
Example Problem 3: Two-
Sided Weaving Segment Data
What is the expected LOS and capacity for the weaving segment of Exhibit
27-8?
The Facts
In addition to the information contained in Exhibit 27-8, the following facts
concerning the weaving segment are known:
PHF = 0.94 (all movements);
Heavy vehicles = 11% trucks;
FFS = 60 mi/h; ramp FFS = 30 mi/h;
ID = 2 int/mi; and
Terrain = rolling.
Comments
Because this example illustrates the analysis of a two-sided weaving
segment, several key parameters are different from those for a more typical one-
side weaving segment.
In a two-sided weaving segment, only the ramp-to-ramp flow is considered
to be a weaving flow. While the freeway-to-freeway flow technically weaves
with the ramp-to-ramp flow, the operation of freeway-to-freeway vehicles more
closely resembles that of nonweaving vehicles. These vehicles generally make
few lane changes as they move through the segment in a freeway lane. This
segment is in a busy urban corridor with a high interchange density and a
relatively low FFS for the freeway.

Solution steps are the same as in the first two example problems. However,
since the segment is a two-sided weaving segment, some of the key values will
be computed differently, as described in the methodology.
Component demand volumes will be converted to equivalent flow rates in
passenger cars per hour under ideal conditions, and key demand parameters will
be calculated. A maximum weaving length will be estimated to determine
whether a weaving analysis is appropriate. The capacity of the weaving segment
will be estimated to determine whether LOS F exists. In addition, the segment
density will be estimated to evaluate whether LOS F exists. If it does not, lane-
changing parameters, speeds, density, and LOS will be estimated.
Step 1: Input Data

All information concerning this example problem is given in Exhibit 27-8 and
the Facts section.

To convert demand volumes to flow rates under equivalent ideal conditions,
Chapter 12 must be consulted to obtain the following values:
ET = 3.0 (for rolling terrain)
Then
1 1
𝑓𝐻𝑉 = = = 0.82
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.11(3 − 1)
Component demand volumes may now be converted to flow rates under
equivalent ideal conditions:
𝑉𝑖
𝑣𝑖 =
3,500
𝑣𝐹𝐹 = = 4,541 pc/h
0.94 × 0.82
250
𝑣𝐹𝑅 = = 324 pc/h
0.94 × 0.82
100
𝑣𝑅𝐹 = = 130 pc/h
0.94 × 0.82
300
𝑣𝑅𝑅 = = 389 pc/h
0.94 × 0.82
Because this is a two-sided weaving segment, the only weaving flow is the
ramp-to-ramp flow. All other flows are treated as nonweaving. Then
𝑣𝑊 = 389 pc/h
𝑣𝑁𝑊 = 4,541 + 324 + 130 = 4,995 pc/h
𝑣 = 4,995 + 389 = 5,384 pc/h
𝑉𝑅 = 389/5,384 = 0.072


The determination of configuration characteristics is also affected by the
existence of a two-sided weaving segment. Exhibit 27-9 illustrates the
determination of LCRR, the key variable for two-sided weaving segments. For
such segments, NWL = 0 by definition.
Exhibit 27-9
Example Problem 3:
From Exhibit 27-9, ramp-to-ramp vehicles must make two lane changes to
complete their desired weaving maneuver. Then
𝐿𝐶𝑀𝐼𝑁 = (𝐿𝐶𝑅𝑅 × 𝑣𝑅𝑅 ) = 2 × 389 = 778 lc/h

The maximum length of a weaving segment for this configuration and
demand scenario is estimated by using Equation 13-4:
𝐿𝑀𝐴𝑋 = [5,728(1 + 𝑉𝑅)1.6 ] − (1,566𝑁𝑊𝐿 )
𝐿𝑀𝐴𝑋 = [5,728(1 + 0.072)1.6 ] − (1,566 × 0) = 6,405 ft > 750 ft
In this two-sided configuration, the impacts of weaving on operations could
be felt at lengths as long as 6,405 ft. Since this is significantly greater than the
actual length of 750 ft, the segment clearly operates as a weaving segment, and
therefore the methodology of this chapter should be applied.

The capacity of a two-sided weaving segment can only be estimated when a
density of 43 pc/h/ln is reached. This estimation is made by using Equation 13-5
and Equation 13-6:
𝑐𝐼𝑊𝐿 = 2,300 − [438.2(1 + 0.072)1.6 ] + (0.0765 × 750) + (119.8 × 0)
𝑐𝑊 = 1,867 × 3 × 0.820 = 4,593 veh/h > 4,150 veh/h
Because the capacity of the segment exceeds the demand volume (in vehicles
per hour), LOS F is not expected, and the analysis may be continued.

The capacity of input and output roadways must also be checked. The
freeway input and output roadways have three lanes and a capacity of 2,300 × 3 =
6,900 pc/h (Chapter 12). The one-lane ramps (with ramp FFS = 30 mi/h) have a
capacity of 1,900 pc/h (Chapter 14). Exhibit 27-10 compares these capacities with
the demand flow rates (in pc/h).

Freeway entry 4,541 + 324 = 4,865 6,900 Example Problem 3: Capacity
Freeway exit 4,541 + 130 = 4,671 6,900 of Entry and Exit Legs
Ramp entry 130 + 389 = 519 1,900
Ramp exit 324 + 389 = 713 1,900
All demands are below their respective capacities.


𝐿𝐶𝑊 = 𝐿𝐶𝑀𝐼𝑁 + 0.39[(𝐿𝑆 − 300)0.5 𝑁 2 (1 + 𝐼𝐷)0.8 ]
𝐿𝐶𝑊 = 778 + 0.39[(750 − 300)0.5 (32 )(1 + 2)0.8 ] = 960 lc/h

𝐼𝑁𝑊 =
10,000
750 × 2 × 5,015
𝐼𝑁𝑊 = = 752 < 1,300
10,000
𝐿𝐶𝑁𝑊 = 𝐿𝐶𝑁𝑊1 = (0.206𝑣𝑁𝑊 ) + (0.542𝐿𝑆 ) − (192.6𝑁)
𝐿𝐶𝑁𝑊 = (0.206 × 5,015) + (0.542 × 750) − (192.6 × 3) = 861 lc/h

𝐿𝐶𝐴𝐿𝐿 = 𝐿𝐶𝑊 + 𝐿𝐶𝑁𝑊 = 960 + 861 = 1,821 lc/h

Vehicles
𝑊 = 0.226 ( )
𝐿𝑆
1,821 0.789
𝑊 = 0.226 ( ) = 0.455
750
Then
𝑆𝑊 = 15 + ( )
1+𝑊
60 × 1 − 15
𝑆𝑊 = 15 + ( ) = 45.9 mi/h
1 + 0.455

and
𝑣
𝑆𝑁𝑊 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (0.0072𝐿𝐶𝑀𝐼𝑁 ) − (0.0048 )
𝑁
5,384
𝑆𝑁𝑊 = 60 × 1 − (0.0072 × 778) − (0.0048 ) = 45.8 mi/h
3
the segment:
𝑆= 𝑣 𝑣
(𝑆𝑊 ) + (𝑆𝑁𝑊 )
𝑊 𝑁𝑊
389 + 4,995
𝑆= = 45.8 mi/h
389 4,995
( )+( )
45.9 45.8

The average density in this two-sided weaving segment is estimated by
using Equation 13-23:
(𝑣/𝑁) (5,384/3)
𝐷= = = 39.2 pc/mi/ln
𝑆 45.8
From Equation 13-12, this density is clearly in LOS E. It is not far from the 43
pc/h/ln that would likely cause a breakdown.
Discussion
This two-sided weaving segment operates at LOS E, not far from the LOS E/F
boundary. The v/c ratio is 4,150/4,573 = 0.91. The major problem is that 300 veh/h
crossing the freeway from ramp to ramp creates a great deal of turbulence in the
traffic stream and limits capacity. The speeds estimated for weaving and
nonweaving vehicles are effectively the same in this example. Two-sided
weaving segments do not operate well with such large numbers of ramp-to-ramp
vehicles. If this were a basic freeway segment, the per lane flow rate of 5,405/3 =
1,802 pc/h/ln would not be considered excessive and would be well within a
basic freeway segment’s capacity of 2,300 pc/h/ln.
EXAMPLE PROBLEM 4: DESIGN OF A MAJOR WEAVING SEGMENT

FOR A DESIRED LOS
The Weaving Segment
A weaving segment is to be designed between two major junctions in which
two urban freeways join and then separate, as shown in Exhibit 27-11. The
analysis assumes no adverse weather effects or incidents in the segment. Entry
and exit legs have the numbers of lanes shown. The maximum length of the
weaving segment is 1,000 ft, based on the location of the junctions. The FFS of all
entry and exit legs is 75 mi/h. All demands are shown as flow rates under
equivalent ideal conditions.

Exhibit 27-11
Example Problem 4: Major
Weaving Segment Data
What design would be appropriate to deliver LOS C for the demand flow
rates shown?
The Facts
In addition to the information contained in Exhibit 27-11, the following facts
are known concerning this weaving segment:
PHF = 1.00 (all demands stated as flow rates),
Heavy vehicles = 0% trucks (all demands in pc/h),
Driver population = regular commuters,
FFS = 75 mi/h (all legs and weaving segment),
cIFL = 2,400 pc/h/ln (for FFS = 75 mi/h),
ID = 1 int/mi, and
Terrain = level.
Comments
As is the case in any weaving segment design, considerable constraints are
imposed. The problem states that the maximum length is 1,000 ft, no doubt
limited by locational issues for the merge and diverge junctions. Shorter lengths
are probably not worth investigating, and the maximum should be assumed for
all trial designs. The simplest design merely connects entering lanes with exit
lanes in a straightforward manner, producing a section of five lanes. A section
with four lanes could be considered by merging two lanes into one at the entry
gore and separating it into two again at the exit gore. In any event, the design is
limited to a section of four or five lanes. No other widths would work without
major additions to input and output legs. The configuration cannot be changed
without adding a lane to at least one of the entry or exit legs. Thus, the initial trial
will be at a length of 1,000 ft, with the five entry lanes connected directly to the
five exit lanes, with no changes to the exit or entry leg designs. If this does not
produce an acceptable operation, changes will be considered.
While the problem clearly states that all legs are freeways, no feasible
configuration produces a two-sided weaving section. Thus, to fit within the one-
sided analysis methodology, the right-side entry and exit legs will be classified as
ramps in the computational analysis. Note that by inspection, the capacity of all

entry and exit legs is more than sufficient to handle the demand flow rates
indicated.
Step 1: Input Data—Trial 1

All input information is given in Exhibit 27-11 and in the accompanying
Facts section for this example problem.
Step 2: Adjust Volume—Trial 1

All demands are already stated as flow rates in passenger cars per hour
under equivalent ideal conditions. No further adjustments are needed. Critical
demand values are as follows:
𝑣𝐹𝐹 = 2,000 pc/h
𝑣𝐹𝑅 = 1,450 pc/h
𝑣𝑅𝐹 = 1,500 pc/h
𝑣𝑅𝑅 = 2,000 pc/h
𝑣𝑊 = 1,500 + 1,450 = 2,950 pc/h
𝑣𝑁𝑊 = 2,000 + 2,000 = 4,000 pc/h
𝑣 = 2,950 + 4,000 = 6,950 pc/h
𝑉𝑅 = 2,950/6,950 = 0.424
Step 3: Determine Configuration Characteristics—Trial 1

Exhibit 27-12 illustrates the weaving segment formed under the assumed
design discussed previously.
Exhibit 27-12
Example Problem 4: Trial
Design 1
The direct connection of entry and exit legs produces a weaving segment in
which the ramp-to-freeway movement can be made without a lane change (LCRF
= 0). However, freeway-to-ramp vehicles must make two lane changes (LCFR = 2).
With regard to the lane-changing pattern, there are no lanes on the entering
freeway leg from which a weaving maneuver can be made with one or no lane
changes. However, ramp drivers wishing to weave can enter on either of the two
left ramp lanes and weave with one or no lane changes. Thus, NWL = 2.
By using Equation 13-2, LCMIN is computed as
𝐿𝐶𝑀𝐼𝑁 = (0 × 1,500) + (2 × 1,450) = 2,900 lc/h

Step 4: Determine Maximum Weaving Length—Trial 1

𝐿𝑀𝐴𝑋 = [5,728(1 + 𝑉𝑅)1.6 ] − (1,566𝑁𝑊𝐿 )
𝐿𝑀𝐴𝑋 = [5,728(1 + 0.424)1.6 ] − (1,566 × 2) = 6,950 ft > 1,000 ft
Since the maximum length is much greater than the actual length of 1,000 ft,
analysis of the segment with this chapter’s methodology is appropriate.
Step 5: Determine Weaving Segment Capacity—Trial 1

the configuration of Exhibit 27-12.

𝑐𝐼𝑊𝐿 = 2,400 − [438.2(1 + 0.424)1.6 ] + (0.0765 × 1,000) + (119.8 × 2)
𝑐𝑊 = 1,944 × 5 × 1 = 9,721 pc/h

2,400 2,400
𝑐𝐼𝑊 = = = 5,654 pc/h
𝑉𝑅 0.424
𝑐𝑊 = 𝑐𝐼𝑊 × 𝑓𝐻𝑉 = 5,654 × 1 = 5,654 pc/h
In this case, the capacity of the segment is limited by the maximum weaving
flow rate, which limits total capacity of the segment to 5,654 pc/h, which is smaller
than the total demand flow rate of 6,950 pc/h. Thus, this section is expected to
operate at LOS F. No further analysis is possible with this methodology.
Discussion: Trial 1
This weaving segment would be expected to fail under the proposed design.
The critical feature appears to be the configuration. Note that the capacity is
limited by the maximum weaving flows that can be sustained, not by a density
expected to produce queuing. This is primarily due to the freeway-to-ramp flow,
which must make two lane changes. The number of lane changes can be reduced
to one by adding one lane to the “ramp” at the exit gore area. This not only
reduces the number of lane changes made by 1,450 freeway-to-ramp vehicles but
also increases the value of NW from 2 to 3. In turn, the segment’s capacity (as
limited by weaving flow rate) is effectively increased to 3,500/VR = 3,500/0.424 =

8,255 pc/h, which is well in excess of the demand flow rate of 6,950 pc/h. Another
analysis (Trial 2) will be conducted by using this approach.
Steps 1 and 2: Input Data and Adjust Volume—Trial 2

Steps 1 and 2 are the same as for Trial 1. They are not repeated here. The new
configuration affects the results beginning with Step 3.
Step 3: Determine Configuration Characteristics—Trial 2

Exhibit 27-13 illustrates the new configuration that will result from the
changes discussed above. The addition of a lane to the exit-ramp leg allows the
freeway-to-ramp movement to be completed with only one lane change (LCFR =
1). The value of LCRF is not affected and remains 0. The right lane of the freeway-
entry leg can also be used by freeway-to-ramp drivers to make a weaving
maneuver with a single lane change, increasing NWL to 3.
Exhibit 27-13
Example Problem 4:
Trial Design 2
Then
𝐿𝐶𝑀𝐼𝑁 = (0 × 1,500) + (1 × 1,450) = 1,450 lc/h
Step 4: Determine Maximum Weaving Length—Trial 2

𝐿𝑀𝐴𝑋 = [5,728(1 + 𝑉𝑅)1.6 ] − (1,566𝑁𝑊𝐿 )
𝐿𝑀𝐴𝑋 = [5,728(1 + 0.424)1.6 ] − (1,566 × 3) = 5,391 ft > 1,000 ft
Since the maximum length is much greater than the actual length of 1,000 ft,
analyzing the segment by using this chapter’s methodology is appropriate.
Step 5: Determine Weaving Segment Capacity—Trial 2

the configuration of Exhibit 27-13.

𝑐𝐼𝑊𝐿 = 2,400 − [438.2(1 + 0.424)1.6 ] + (0.0765 × 1,000) + (119.8 × 3)

𝑐𝑊 = 2,064 × 5 × 1 = 10,320 pc/h

3,500 3,500
𝑐𝐼𝑊 = = = 8,255 pc/h
𝑉𝑅 0.424
𝑐𝑊 = 𝑐𝐼𝑊 × 𝑓𝐻𝑉 × 𝑓𝑝 = 8,255 × 1 × 1 = 8,255 pc/h
Once again, the capacity of the segment is limited by the maximum weaving
flow rate: the difference is that now the capacity is 8,255 pc/h. This is larger than
the total demand flow rate of 6,950 pc/h. Thus, this section is expected to operate
without breakdown, and the analysis may continue.
Step 6: Determine Lane-Changing Rates—Trial 2


𝐿𝐶𝑊 = 𝐿𝐶𝑀𝐼𝑁 + 0.39[(𝐿𝑆 − 300)0.5 𝑁 2 (1 + 𝐼𝐷)0.8 ]
𝐿𝐶𝑊 = 1,450 + 0.39[(1,000 − 300)0.5 (52)(1 + 1)0.8] = 1,899 lc/h

𝐼𝑁𝑊 =
10,000
1,000 × 1 × 4,000
𝐼𝑁𝑊 = = 400 < 1,300
10,000
𝐿𝐶𝑁𝑊 = (0.206𝑣𝑁𝑊 ) + (0.542𝐿𝑆 ) − (192.6𝑁)
𝐿𝐶𝑁𝑊 = (0.206 × 4,000) + (0.542 × 1,000) − (192.6 × 5) = 403 lc/h


Vehicles—Trial 2
Equation 13-18 through Equation 13-21.
𝑊 = 0.226 ( )
𝐿𝑆
2,302 0.789
𝑊 = 0.226 ( ) = 0.436
1,000

Then
𝑆𝑊 = 15 + ( )
1+𝑊
75 × 1 − 15
𝑆𝑊 = 15 + ( ) = 56.8 mi/h
1 + 0.436
and
𝑣
𝑆𝑁𝑊 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (0.0072𝐿𝐶𝑀𝐼𝑁 ) − (0.0048 )
𝑁
6,950
𝑆𝑁𝑊 = 75 × 1 − (0.0072 × 1,450) − (0.0048 ) = 57.9 mi/h
5
the segment:
𝑆= 𝑣 𝑣
(𝑆𝑊 ) + (𝑆𝑁𝑊 )
𝑊 𝑁𝑊
4,000 + 2,950
𝑆= = 57.4 mi/h
4,000 2,950
( )+( )
57.9 56.8
Step 8: Determine the Level of Service—Trial 2

13-23:
(𝑣/𝑁) (6,950/5)
𝐷= = = 24.2 pc/mi/ln
𝑆 57.4
From Exhibit 13-12, this density is within the stated boundaries of LOS C
(20 to 28 pc/mi/ln). Since the design target was LOS C, the second trial design is
acceptable.
Discussion: Trial 2
The relatively small change in the configuration makes all the difference in
this design. LOS C can be achieved by adding a lane to the right exit leg; without
it, the section fails because of excessive weaving turbulence. If the extra lane is not
needed on the departing freeway leg, it will be dropped somewhere downstream,
perhaps as part of the next interchange. The extra lane would have to be carried
for several thousand feet to be effective. An added lane generally will not be
fully utilized by drivers if they are aware that it will be immediately dropped.
EXAMPLE PROBLEM 5: CONSTRUCTING A SERVICE VOLUME TABLE FOR

A WEAVING SEGMENT
This example shows how a table of service flow rates or service volumes or
both can be constructed for a weaving section with certain specified
characteristics. The methodology of this chapter does not directly yield service
flow rates or service volumes, but they can be developed by using spreadsheets
or more sophisticated computer programs.

The key issue is the definition of the threshold values for the various levels of
service. For weaving sections on freeways, levels of service are defined as
limiting densities, as shown in Exhibit 27-14:
LOS Maximum Density (pc/mi/ln) Exhibit 27-14

A 10 Example Problem 5: Maximum
B 20 Density Thresholds for LOS
C 28 A–D
D 35
By definition, the service flow rate at LOS E is the capacity of the weaving
section, which may or may not be keyed to a density.
Before the construction of such a table is illustrated, several key definitions
should be reviewed:
• Service flow rate (under ideal conditions): The maximum rate of flow under
equivalent ideal conditions that can be sustained while maintaining the
designated LOS (SFI, pc/h).
• Service flow rate (under prevailing conditions): The maximum rate of flow
under prevailing conditions that can be sustained while maintaining the
designated LOS (SF, veh/h).
• Service volume: The maximum hourly volume under prevailing conditions
that can be sustained while maintaining the designated LOS in the worst
15 min of the hour (SV, veh/h).
• Daily service volume: The maximum annual average daily traffic under
prevailing conditions that can be sustained while maintaining the
designated LOS in the worst 15 min of the peak hour (DSV, veh/day).
Note that flow rates are for a 15-min period, often a peak 15 min within the
analysis hour, or the peak hour. These values are related as follows:
𝑆𝐹𝑖 = 𝑆𝐹𝐼𝑖 × 𝑓𝐻𝑉
𝑆𝑉𝑖 = 𝑆𝐹𝑖 × 𝑃𝐻𝐹
𝑆𝑉𝑖
𝐷𝑆𝑉𝑖 =
𝐾×𝐷
This chapter’s methodology estimates both the capacity and the density
expected in a weaving segment of given geometric and demand characteristics.
Conceptually, the approach to generating values of SFI is straightforward: for
any given situation, keep increasing the input flow rates until the boundary
density for the LOS is reached; the input flow rate is the SFI for that situation and
LOS. This obviously involves many iterations. A spreadsheet can be
programmed to do this, either semiautomatically with manual input of demands,
or fully automatically, with the spreadsheet automatically generating solutions
until a density match is found. The latter method is not very efficient and
involves a typical spreadsheet program running for several hours. A program
could, of course, be written to automate the entire process.

An Example
While all of the computations cannot be shown, demonstration results for a
specific case can be illustrated. A service volume table is desired for a weaving
section with the following characteristics:
• One-sided major weaving section
• Demand splits as follows:
o vFF = 65% of v
o vRF = 15% of v
o vFR = 12% of v
o vRR = 8% of v
• Trucks = 5%
• Level terrain
• PHF = 0.93
• Regular commuters in the traffic stream
• ID = 1 interchange/mi
• FFS = 65 mi/h
For these characteristics, a service volume table can be constructed for a
range of lengths and widths and for configurations in which NW is 2 and 3. For
illustrative purposes, lengths of 500, 1,000, 1,500, 2,000, and 2,500 ft and widths of
three, four, or five lanes will be used. In a major weaving section, one weaving
flow does not have to make a lane change. In this example, the ramp-to-freeway
movement is assumed to have this characteristic. The freeway-to-ramp movement
would require one or two lane changes, on the basis of the value of NWL.
First Computations
Initial computations will be aimed at establishing values of SFI for the
situations described. A spreadsheet will be constructed in which the first column
is the flow rate to be tested (in passenger cars per hour under ideal conditions),
and the last column produces a density. Each line will be iterated (manually in
this case) until each threshold density value is reached. Intermediate columns
will be programmed to produce the intermediate results needed to get to this
result. Because maximum length and capacity are decided at intermediate points,
the applicable results will be manually entered before continuing. Such a
procedure is less difficult than it seems once the basic computations are
programmed. Manual iteration using the input flow rate is efficient; the operator
will observe how fast the results are converging to the desired threshold and will
change the inputs accordingly.
The results of a first computation are shown in Exhibit 27-15. They represent
service flow rates under ideal conditions, SFI. Consistent with the HCM’s results
presentation guidelines (Chapter 7, Interpreting HCM and Alternative Tool
Results), all hourly service flow rates and volumes in these exhibits have been
rounded down to the nearest 100 passenger cars or vehicles for presentation.

Length of Weaving Section (ft) Exhibit 27-15

LOS 500 1,000 1,500 2,000 2,500 500 1,000 1,500 2,000 2,500 Example Problem 5: Service
N = 3; NWL = 2 N = 3; NWL = 3 Flow Rates (pc/h) Under Ideal
Conditions (SFI)
A 1,700 1,700 1,700 1,700 1,700 1,800 1,800 1,800 1,800 1,800
B 3,200 3,200 3,200 3,200 3,200 3,300 3,300 3,400 3,400 3,400
C 4,200 4,200 4,300 4,300 4,300 4,400 4,500 4,500 4,500 4,500
D 5,000 5,100 5,100 5,100 5,100 5,300 5,400 5,400 5,500 5,500
E 5,900 6,000 6,100 6,300 6,400 6,300 6,400 6,500 6,600 6,700
N = 4; NWL = 2 N = 4; NWL = 3
A 2,200 2,300 2,300 2,300 2,300 2,300 2,300 2,300 2,300 2,300
B 4,100 4,200 4,200 4,200 4,200 4,300 4,400 4,400 4,400 4,400
C 5,400 5,500 5,500 5,500 5,600 5,800 5,900 5,900 5,900 5,900
D 6,300 6,500 6,500 6,600 6,600 6,900 7,000 7,100 7,100 7,100
E 7,900 8,000 8,200 8,400 8,500 8,400 8,500 8,700 8,800 9,000
N = 5; NWL = 2 N = 5; NWL = 3
A 2,800 2,800 2,800 2,800 2,800 2,900 2,900 2,900 2,900 2,900
B 5,000 5,100 5,100 5,100 5,100 5,400 5,400 5,400 5,500 5,500
C 6,500 6,600 6,700 6,700 6,700 7,100 7,200 7,200 7,300 7,300
D 7,600 7,800 7,900 7,900 7,900 8,400 8,600 8,700 8,700 8,700
E 8,800 8,800 8,800 8,800 8,800 10,500 10,700 10,900 11,100 11,200
Exhibit 27-16 shows service flow rates under prevailing conditions, SF. Each
value in Exhibit 27-15 (before rounding) is multiplied by
1 1
𝑓𝐻𝑉 = = = 0.952
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.05(2 − 1)
Length of Weaving Section (ft) Exhibit 27-16

LOS 500 1,000 1,500 2,000 2,500 500 1,000 1,500 2,000 2,500 Example Problem 5: Service
N = 3; NWL = 2 N = 3; NWL = 3 Flow Rates (veh/h) Under
Prevailing Conditions (SF)
A 1,600 1,600 1,600 1,600 1,600 1,700 1,700 1,700 1,700 1,700
B 3,000 3,000 3,100 3,100 3,100 3,100 3,200 3,200 3,200 3,200
C 4,000 4,000 4,100 4,100 4,100 4,200 4,300 4,300 4,300 4,300
D 4,700 4,800 4,900 4,900 4,900 5,100 5,100 5,200 5,200 5,200
E 5,600 5,700 5,800 5,900 6,100 6,000 6,100 6,200 6,200 6,400
N = 4; NWL = 2 N = 4; NWL = 3
A 2,100 2,100 2,200 2,200 2,200 2,200 2,200 2,200 2,200 2,200
B 3,900 4,000 4,000 4,000 4,000 4,100 4,200 4,200 4,200 4,200
C 5,100 5,200 5,200 5,300 5,300 5,500 5,600 5,600 5,600 5,600
D 5,900 6,200 6,200 6,300 6,300 6,600 6,700 6,700 6,800 6,800
E 7,500 7,700 7,800 7,900 8,100 8,000 8,100 8,200 8,400 8,500
N = 5; NWL = 2 N = 5; NWL = 3
A 2,600 2,700 2,700 2,700 2,700 2,700 2,700 2,800 2,800 2,800
B 4,700 4,800 4,900 4,900 4,900 5,100 5,100 5,200 5,200 5,200
C 6,200 6,300 6,300 6,400 6,400 6,700 6,800 6,900 6,900 6,900
D 7,300 7,400 7,500 7,500 7,500 8,000 8,200 8,200 8,300 8,300
E 8,400 8,400 8,400 8,400 8,400 10,000 10,200 10,300 10,500 10,700
Exhibit 27-17 shows service volumes, SV. Each value in Exhibit 27-16 (before
rounding) is multiplied by a PHF of 0.93.

Exhibit 27-17 Length of Weaving Section (ft)

Example Problem 5: Service LOS 500 1,000 1,500 2,000 2,500 500 1,000 1,500 2,000 2,500
Volumes (veh/h) Under N = 3; NWL = 2 N = 3; NWL = 3
Prevailing Conditions (SV)
A 1,500 1,500 1,500 1,500 1,500 1,500 1,500 1,500 1,500 1,500
B 2,800 2,800 2,800 2,800 2,900 2,900 2,900 3,000 3,000 3,000
C 3,700 3,700 3,800 3,800 3,800 3,900 4,000 4,000 4,000 4,000
D 4,400 4,500 4,500 4,500 4,500 4,700 4,800 4,800 4,800 4,800
E 5,200 5,300 5,400 5,500 5,600 5,500 5,600 5,700 5,800 5,900
N = 4; NWL = 2 N = 4; NWL = 3
A 2,000 2,000 2,000 2,000 2,000 2,000 2,100 2,100 2,100 2,100
B 3,600 3,700 3,700 3,700 3,700 3,800 3,900 3,900 3,900 3,900
C 4,700 4,800 4,900 4,900 4,900 5,100 5,200 5,200 5,200 5,200
D 5,500 5,700 5,800 5,800 5,800 6,100 6,200 6,300 6,300 6,300
E 7,000 7,100 7,300 7,400 7,500 7,400 7,500 7,700 7,800 7,900
N = 5; NWL = 2 N = 5; NWL = 3
A 2,400 2,500 2,500 2,500 2,500 2,500 2,500 2,600 2,600 2,600
B 4,400 4,500 4,500 4,500 4,500 4,700 4,800 4,800 4,800 4,800
C 5,700 5,800 5,900 5,900 5,900 6,200 6,400 6,400 6,400 6,400
D 6,700 6,900 7,000 7,000 7,000 7,500 7,600 7,700 7,700 7,700
E 7,800 7,800 7,800 7,800 7,800 9,300 9,400 9,600 9,800 9,900
Exhibit 27-18 shows daily service volumes, DSV. An illustrative K-factor of

0.08 (typical of a large urban area) and an illustrative D-factor of 0.55 (typical of
an urban route without strong peaking by direction) are used. Each nonrounded
value used to generate Exhibit 27-17 was divided by both of these numbers.
Exhibit 27-18 Length of Weaving Section (ft)

Example Problem 5: Daily LOS 500 1,000 1,500 2,000 2,500 500 1,000 1,500 2,000 2,500
Service Volumes (veh/day) N = 3; NWL = 2 N = 3; NWL = 3
Under Prevailing Conditions
A 35,200 35,200 35,400 35,500 35,600 36,200 36,300 36,300 36,300 36,300
(DSV)
B 64,300 65,300 65,500 65,700 66,100 67,600 68,000 68,400 68,400 68,400
C 84,700 86,100 86,700 87,200 87,500 89,700 90,900 91,500 91,700 91,900
D 100,800 102,800 103,600 104,000 104,400 107,800 109,600 110,200 110,600 110,800
E 119,800 122,100 124,400 126,700 129,100 127,000 129,400 131,600 132,800 136,300
N = 4; NWL = 2 N = 4; NWL = 3
A 45,800 46,200 46,600 46,600 46,600 47,600 47,800 47,800 47,900 47,900
B 83,300 84,700 85,100 85,500 85,700 88,300 89,300 89,500 89,700 89,900
C 108,600 110,800 111,600 112,200 112,600 117,100 118,700 119,500 120,100 120,300
D 126,700 131,300 132,400 133,200 133,600 140,000 142,400 143,600 144,000 144,400
E 159,800 162,800 165,900 169,000 172,100 169,400 172,500 175,400 178,600 181,700
N = 5; NWL = 2 N = 5; NWL = 3
A 56,300 57,100 57,300 57,500 57,500 58,700 58,900 59,300 59,400 59,400
B 101,400 103,000 103,600 104,200 104,400 108,600 109,600 110,000 110,600 110,800
C 131,300 133,800 135,000 135,800 136,200 142,800 145,400 146,200 146,800 147,400
D 154,500 157,700 159,100 159,900 160,300 170,600 173,600 175,000 175,800 175,800
E 178,800 178,800 178,800 178,800 178,800 211,800 215,600 219,500 223,300 227,200
This example problem illustrates how service volume tables may be created
for a given set of weaving parameters. So many variables affect the operation of a
weaving segment that “typical” service volume tables are not recommended.
They may be significantly misleading when they are applied to segments with
different parameters.

EXAMPLE PROBLEM 6: LOS OF AN ML ACCESS SEGMENT WITH CROSS-

WEAVING
The ML Access Segment

Exhibit 27-19 shows a freeway facility that includes both general purpose
and managed lanes. The analysis assumes no adverse weather effects or
incidents in the segment. A freeway with an adjacent managed lane facility is
evaluated as two parallel lane groups, as discussed in more detail in Chapter 10,
Freeway Facilities Core Methodology. The example below shows two segments,
each with two adjacent lane groups. Lane Group Pair 1 in the first segment
includes a general purpose (GP) merge segment and a managed lane (ML) basic
segment. Lane Group Pair 2 consists of GP and ML access segments.
Lane Group Lane Group

Exhibit 27-19
Pair 1 Pair 2 Example Problem 6:
ML Access Segment with
Cross-Weaving
ML Basic ML Access
540 veh/h 810 veh/h
2,970 veh/h 3,600 veh/h

3,330 veh/h
GP Merge GP Access
1,000 ft 1,500 ft
360 veh/h travel
to ML
Note: GP = general purpose, ML = managed lane.
What is the capacity reduction in the GP merge segment due to cross-

weaving, and what is the expected LOS for the ML access segment with the
demand flow rates shown?
The Facts
In addition to the information given in Exhibit 27-19, the following facts are
known about the subject weaving segment:
PHF = 0.90;
Heavy vehicles = 0% single-unit trucks, 0% tractor-trailer;
FFS = 65 mi/h (for both managed and general purpose lanes);
ID = 1.0 interchange/mi; and
Terrain = level.

Comments
Lane-changing characteristics will be estimated for Lane Group Pair 2. The
maximum length for weaving operations in the access segments will be
estimated and compared with the segment’s actual length. The access segment’s
capacity will be estimated and compared with demand to determine whether
LOS F exists. If it does not, component flow speeds will be estimated and
averaged. Finally, the access segment density will be estimated and Exhibit 13-6
used to determine the expected LOS.
Capacity Reduction in GP Merge Segment (Lane Group Pair 1)

The capacity reduction due to the cross-weave effect is evaluated for Lane
Group Pair 1. On the basis of the facility configuration provided in Exhibit 27-19,
the Lcw-min and Lcw-max values are 1,000 ft and 2,500 ft, respectively. The cross-weave
demand volume is 360/0.9 = 400 veh/h. The number of general purpose lanes NGP
is 3. Thus the capacity reduction factor CRF will be
𝐶𝑅𝐹 = −0.0897 + 0.0252 ln(𝐶𝑊) − 0.00001453𝐿𝑐𝑤-𝑚𝑖𝑛 + 0.002967𝑁𝐺𝑃
𝐶𝑅𝐹 = 0.056
Performance of ML Access Segment (Lane Group Pair 2)

The following steps illustrate the computations in the ML access segment,
which is described above as Lane Group Pair 2.
Step 1: Input Data

All input data are stated in Exhibit 27-19 and the Facts section.

The flow rates are computed on the basis of the hourly demand flow rates by
using the specified PHF.
3,060
𝑣𝐹𝐹 = =3,400 pc/h
0.9
540
𝑣𝐹𝑅 = = 600 pc/h
0.9
270
𝑣𝑅𝐹 = = 300 pc/h
0.9
270
𝑣𝑅𝑅 = = 300 pc/h
0.9
𝑣𝑊 = 600 + 300 = 900 pc/h
𝑣𝑁𝑊 = 3,400 + 300 = 3,700 pc/h
𝑣 = 3,700 + 900 = 4,600 pc/h
900
𝑉𝑅 = = 0.196
4,600
Exhibit 27-20 summarizes the hourly flow rates computed on the basis of
hourly demand flow rates.

Lane Group Lane Group Exhibit 27-20

Pair A Pair B Example Problem 6: Hourly
Flow Rates After PHF Is
Applied
ML Basic ML Access
600 veh/h 900 veh/h
3,300 veh/h 4,000 veh/h

3,700 veh/h
GP Basic GP Access
1,000 ft 1,500 ft
400 veh/h travel
to ML

The configuration of the ML access segment is examined to determine the
values of LCRF, LCFR, and NWL. The lane geometry is illustrated in Exhibit 27-21.
From these values, the minimum number of lane changes by weaving vehicles
LCMIN is computed.
Lane Group Exhibit 27-21

Pair 2 Example Problem 6:
ML Access
GP Access
1,500 ft
From Exhibit 27-21, it is clear that all ramp-to-freeway vehicles must make at
least one lane change (LCRF = 1). Similarly, all freeway-to-ramp vehicles must
make at least one lane change (LCFR = 1). In addition, a weaving maneuver can
only be completed with a single lane change from the leftmost lane of the
freeway or the auxiliary lane (NWL = 2). Then, by using Equation 13-2, LCMIN is
computed as
𝐿𝐶𝑀𝐼𝑁 = (1 × 300) + (1 × 600) = 900 lc/h

The maximum length over which weaving operations may exist for the
segment described is found by using Equation 13-4:

𝐿𝑀𝐴𝑋 = [5,728(1 + 𝑉𝑅)1.6 ] − (1,566𝑁𝑊𝐿 )

𝐿𝑀𝐴𝑋 = [5,728(1 + 0.196)1.6 ] − (1,566 × 2) = 4,495 ft > 1,500 ft
Because the maximum length for weaving operations significantly exceeds
the actual length, the segment qualifies as a weaving segment, and the analysis
continues.

factors: density reaching 43 pc/mi/ln or weaving demand reaching 2,350 pc/h for
the configuration of Exhibit 27-19 (a ramp-weave with NWL = 2).

𝑐𝐼𝑊𝐿 = 2,350 − [438.2(1 + 0.196)1.6 ] + (0.0765 × 1,500) + (119.8 × 2)
𝑐𝑊 = 2,121 × 4 × 1 = 8,483 pc/h

2,400 2,400
𝑐𝐼𝑊 = = = 12,245 pc/h
𝑉𝑅 0.196
𝑐𝑊 = 𝑐𝐼𝑊 × 𝑓𝐻𝑉
𝑐𝑊 = 12,245 × 1 = 12,245 pc/h
The controlling capacity is the smaller of the two values, or 8,483 pc/h. At
this point, the value is usually stated as vehicles per hour. In this case, because
inputs were already adjusted and were stated in passenger cars per hour,
conversions back to vehicles per hour are not possible.
Since the capacity of the weaving segment is larger than the demand flow
rate of 4,600 pc/h, LOS F does not exist, and the analysis may continue.

Although it is rarely a factor in weaving operations, the capacity of input and
output roadways should be checked to ensure that no deficiencies exist. There
are three input and output freeway lanes (with FFS = 65 mi/h). The capacities of
the entry and exit ramps are determined for a basic managed lane segment with
a free-flow speed of 65 mi/h, separated by markings. The criteria of Chapter 12
are used to determine the capacity of the freeway legs and the managed lane
entry and exit lanes. Demand flows and capacities are compared in Exhibit 27-22.


Freeway entry 4,000 3 × 2,350 = 7,050 Example Problem 6: Capacity
Freeway exit 4,000 + 300 – 600 = 3,700 3 × 2,350 = 7,050 of Entry and Exit Legs
Ramp entry 600 1,700
Ramp exit 600 – 300 + 600 = 900 1,700
The capacities of all input and output roadways are sufficient to

accommodate the demand flow rates.

changing rates of weaving and nonweaving vehicles in the access segment. These
rates will be used in Step 7 to estimate the weaving and nonweaving vehicle
speeds.

𝐿𝐶𝑊 = 𝐿𝐶𝑀𝐼𝑁 + 0.39[(𝐿𝑆 − 300)0.5 𝑁 2 (1 + 𝐼𝐷)0.8 ]
𝐿𝐶𝑊 = 900 + 0.39[(1,500 − 300)0.5 (42 )(1 + 1)0.8 ] = 1,276 lc/h

𝐼𝑁𝑊 =
10,000
1,500 × 1 × 3,700
𝐼𝑁𝑊 = = 555 < 1,300
10,000
𝐿𝐶𝑁𝑊 = 𝐿𝐶𝑁𝑊1 = (0.206𝑣𝑁𝑊 ) + (0.542𝐿𝑆 ) − (192.6𝑁)
𝐿𝐶𝑁𝑊 = (0.206 × 3,700) + (0.542 × 1,500) − (192.6 × 4) = 805 lc/h

Step 7: Determine Average Speeds of Weaving and Nonweaving Vehicles

𝑊 = 0.226 ( )
𝐿𝑆
2,081 0.789
𝑊 = 0.226 ( ) = 0.293
1,500
Then
𝑆𝑊 = 15 + ( )
1+𝑊
65 × 1 − 15
𝑆𝑊 = 15 + ( ) = 53.7 mi/h
1 + 0.293
and
𝑣
𝑆𝑁𝑊 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (0.0072𝐿𝐶𝑀𝐼𝑁 ) − (0.0048 )
𝑁

4,600
𝑆𝑁𝑊 = 65 × 1 − (0.0072 × 900) − (0.0048 ) = 53.0 mi/h
4
the segment:
𝑆= 𝑣 𝑣
(𝑆𝑊 ) + (𝑆𝑁𝑊 )
𝑊 𝑁𝑊
900 + 3,700
𝑆= = 53.1 mi/h
900 3,700
( )+( )
53.7 53.0

13-23.
(𝑣/𝑁) (4,600/4)
𝐷= = = 21.7 pc/mi/ln
𝑆 53.1
From Exhibit 13-6, this density is within the stated boundaries of LOS C (20
to 28 pc/mi/ln).
Discussion
As noted, the access segment is operating at LOS C. Weaving and
nonweaving speeds are relatively high, suggesting a nearly stable flow. The
demand flow rate of 4,600 pc/h is well below the access segment’s capacity of
8,483 pc/h.
EXAMPLE PROBLEM 7: ML ACCESS SEGMENT WITH DOWNSTREAM

OFF-RAMP
An ML access segment is illustrated in Exhibit 27-23. The movements in and
out of the managed lane may be considered to be analogous to a ramp-weave
segment and analyzed accordingly. The impact of cross-weaving traffic between
the managed lane and the nearby off-ramp must also be analyzed to determine
its impact on capacity of the general purpose lanes.
Exhibit 27-23
Example Problem 7:
ML Access Segment Data

The FFS of the segment is 70 mi/h and the interchange density, ID, is 1
interchange per mile. Demand flow rates for this segment are shown in Exhibit
27-24. Note that all demand flows are stated in passenger car equivalents and
represent the flow rate in the worst 15-min period of the hour.
Exhibit 27-24
Example Problem 7: Weaving
Flows for Managed Lane
Segment
Part 1: Analysis of the Weaving Between Managed Lanes and General

Purpose Lanes
The first major issue to consider is the weaving segment created by
movements into and out of the managed lane in the 1,000-ft access segment. This
segment is treated as a ramp-weave configuration with a total of three lanes
(including the managed lane). This is a bit of an approximation, given that the
geometry of the managed lane is better than that of typical ramps in a ramp-
weave segment. Speeds of weaving vehicles are likely to be underestimated,
since approach speeds on the managed lane are considerably higher than what
would be expected on a typical ramp.
Weaving Movements and Parameters

The primary weaving activity is between vehicles entering and leaving the
managed lane in the 1,000-ft access segment. This may be treated as a three-lane
ramp-weave segment and is analyzed with the basic methodology of this chapter.
Because of the simplicity of this case, certain parameters may be established
by inspection:
NWL = 2 lanes,
LCMIN = 100 + 200 = 300 lc/h, and
VR = 300 / 4,300 = 0.07.
All ramp weaves have two weaving lanes, and each weaving vehicle in a
ramp weave must execute one lane change.
Maximum Weaving Length

The maximum weaving length is determined with Equation 13-4.
𝐿𝑀𝐴𝑋 = [5,728(1 + 𝑉𝑅)1.6 ] − (1,566𝑁𝑊𝐿 )
𝐿𝑀𝐴𝑋 = [5,728(1 + 0.07)1.6 ] − (1,566 × 2) = 3,251 ft > 1,000 ft

The result is significantly longer than the actual weaving length of 1,000 ft.
Thus, the access segment may be treated by using the weaving procedure.
Weaving Segment Capacity

The capacity of the ML access segment (a weaving segment) may be based
on density limits (43 pc/mi/ln) or on the maximum weaving flow that can be
accommodated by the ramp-weave configuration (2,400 pc/h).
The former is estimated by using Equations 13-5 and 13-6.
𝑐𝐼𝑊𝐿 = 2,400 − [438.2(1 + 0.07)1.6 ] + (0.0765 × 1,000) + (119.8 × 2)
𝑐𝑊 = 2,228 × 3 × 1 = 6,684 pc/h
The capacity limited by maximum weaving flow is computed by using
Equations 13-7 and 13-8.
2,400 2,400
𝑐𝐼𝑊 = = = 34,286 pc/h
𝑉𝑅 0.07
𝑐𝑊 = 𝑐𝐼𝑊 × 𝑓𝐻𝑉 = 34,286 × 1 = 34,286 pc/h
Obviously, the capacity is controlled by maximum density and is established
as 6,684 pc/h. Since the total flow in the segment is 900 + 100 + 200 + 3,100 = 4,300
pc/h, failure (LOS F) is not expected, and the analysis of the weaving area
continues. By inspection and comparison with Chapter 12 criteria, demand does
not exceed capacity on any of the entry or exit roadways.
Estimate Lane-Changing Rates

To estimate total lane-changing rates, the total number of lane changes made
by weaving and nonweaving vehicles (within the 1,000-ft access segment) must
be estimated.
The total lane-changing rate for weaving vehicles is determined by using
Equation 13-11.
𝐿𝐶𝑊 = 𝐿𝐶𝑀𝐼𝑁 + 0.39[(𝐿𝑆 − 300)0.5 𝑁 2 (1 + 𝐼𝐷)0.8 ]
𝐿𝐶𝑊 = 300 + 0.39[(1,000 − 300)0.5 (32 )(1 + 1)0.8] = 462 lc/h
The total lane-changing rate for nonweaving vehicles is found by using
Equation 13-13 or 13-14, depending on the value of the nonweaving vehicle
index computed with Equation 13-12.
𝐼𝑁𝑊 =
10,000
1,000 × 1 × 4,000
𝐼𝑁𝑊 = = 400 < 1,300
10,000
Since this value is less than 1,300, Equation 13-13 is applied.
𝐿𝐶𝑁𝑊 = 𝐿𝐶𝑁𝑊1 = (0.206𝑣𝑁𝑊 ) + (0.542𝐿𝑆 ) − (192.6𝑁)
𝐿𝐶𝑁𝑊 = (0.206 × 4,000) + (0.542 × 1,000) − (192.6 × 3) = 788 lc/h

The total lane-changing rate for the ML access segment is

𝐿𝐶𝐴𝐿𝐿 = 𝐿𝐶𝑊 + 𝐿𝐶𝑁𝑊 = 462 + 788 = 1,250 lc/h
Estimate Speed of Weaving and Nonweaving Vehicles

The speed of weaving vehicles in the ML access segment is estimated by
using Equations 13-19 and 13-20.
𝑊 = 0.226 ( )
𝐿𝑆
1,250 0.789
𝑊 = 0.226 ( ) = 0.2695
1,000
𝑆𝑊 = 15 + ( )
1+𝑊
70 × 1 − 15
𝑆𝑊 = 15 + ( ) = 58.3 mi/h
1 + 0.2695
The speed of nonweaving vehicles is estimated by using Equation 13-21.
𝑣
𝑆𝑁𝑊 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (0.0072𝐿𝐶𝑀𝐼𝑁 ) − (0.0048 )
𝑁
4,300
𝑆𝑁𝑊 = 70 × 1 − (0.0072 × 300) − (0.0048 ) = 61.0 mi/h
3
The average speed of all vehicles is found by using Equation 13-22.
𝑆= 𝑣 𝑣
(𝑆𝑊 ) + (𝑆𝑁𝑊 )
𝑊 𝑁𝑊
300 + 4,000
𝑆= = 60.8 mi/h
300 4,000
(58.3) + ( 61.0 )
Estimate the Density in the ML Access Segment and Determine the LOS
The density in the segment is found by using Equation 13-23.
(𝑣/𝑁) (4,300/3)
𝐷= = = 23.6 pc/mi/ln
𝑆 60.8
From Exhibit 13-12, this is LOS B but close to the LOS B/C boundary of 24
pc/mi/ln.
Part 2: Estimate the Impact of Cross-Weaving Vehicles on the Capacity

of the General Purpose Lanes
The capacity of the two general purpose lanes (with FFS = 70 mi/h) is
expected to be 2,400 × 2 = 4,800 pc/h. However, there are 100 pc/h executing
cross-weaving movements to access the off-ramp that is 1,500 ft downstream of
the ML access segment.
Equation 13-24 describes the impact that these cross-weaving vehicles are
expected to have on general purpose lane capacity.
𝐶𝑅𝐹 = −0.0897 + 0.0252 ln(𝐶𝑊) − 0.00001453𝐿𝑐𝑤-𝑚𝑖𝑛 + 0.002967𝑁𝐺𝑃
𝐶𝑅𝐹 = −0.0897 + 0.0252 ln(100) − 0.00001453(1,500) + 0.002967(2)
𝐶𝑅𝐹 = 0.0105

𝐶𝐴𝐹 = 1 − 𝐶𝑅𝐹 = 1 − 0.0105 = 0.9895

Therefore, the remaining capacity of the general purpose lanes is
𝑐𝐺𝑃𝐴 = 𝑐𝐺𝑃 × 𝐶𝐴𝐹 = 4,800 × 0.9895 = 4,750 pc/h
Discussion
In this case, the ML access segment is expected to work well. The actual
weaving involving vehicles entering and leaving the segment results in an
overall LOS B designation. The impact of cross-weaving vehicles using the off-
ramp is negligible.

3. ALTERNATIVE TOOL EXAMPLES FOR

WEAVING SEGMENTS
Chapter 13, Freeway Weaving Segments, described a methodology for

analyzing freeway weaving segments to estimate their capacity, speed, and
density as a function of traffic demand and geometric configuration.
Supplemental problems involving the use of alternative tools for freeway
weaving sections to address limitations of the Chapter 13 methodology are
presented here. All of these examples are based on Example Problem 1 in this
chapter, shown in Exhibit 27-2.
Three questions are addressed by using a typical microscopic traffic
simulation tool that is based on the link–node structure:
1. Can weaving segment capacity be estimated realistically by simulation by
varying the demand volumes up to and beyond capacity?
2. How does demand affect performance in terms of speed and density in
the weaving segment, on the basis of the default model parameters for
vehicle and behavioral characteristics?
3. How would the queue backup from a signal at the end of the off-ramp
affect weaving operation?
The first step is to identify the link–node structure, as shown in Exhibit 27-25.
Exhibit 27-25
1 2 3 4 Link–Node Structure for the
Simulated Weaving Segment
5 6
The next step is to develop input data for various demand levels. Several
demand levels ranging from 80% to 180% of the original volumes were analyzed
by simulation. The demand data, adjusted for a peak hour factor of 0.91, are
given in Exhibit 27-26.
Percent of Specified Demand Exhibit 27-26

Type of Demand 80 100 120 140 160 180 Input Data for Various
Freeway-to-freeway demand, VFF 1,596 1,995 2,393 2,792 3,191 3,590 Demand Levels (veh/h)
Ramp-to-freeway demand, VRF 912 1,140 1,367 1,595 1,823 2,051
Freeway-to-ramp demand, VFR 608 760 913 1,065 1,217 1,369
Ramp-to-ramp demand, VRR 1,140 1,425 1,710 1,995 2,280 2,565
Total demand 4,256 5,320 6,384 7,448 8,512 9,576
Total freeway entry 2,204 2,755 3,306 3,857 4,408 4,959
Total freeway exit 2,507 3,134 3,761 4,388 5,015 5,641
Total ramp entry 2,052 2,565 3,078 3,591 4,104 4,617
Total ramp exit 1,749 2,186 2,623 3,060 3,497 3,934
Thirty simulation runs were made for each demand level. The results are
discussed in the following sections. The need to determine performance
measures from an analysis of vehicle trajectories was emphasized in Chapter 7,
Interpreting HCM and Alternative Tool Results. Specific procedures for defining
measures in terms of vehicle trajectories were proposed to guide the future
Chapter 27/Freeway Weaving: Supplemental Alternative Tool Examples for Weaving Segments
development of alternative tools. Pending further development, the examples

presented in this chapter have applied existing versions of alternative tools and
therefore do not reflect the trajectory-based measures described in Chapter 7.
DETERMINING THE WEAVING SEGMENT CAPACITY

Simulation tools do not produce capacity estimates directly. The traditional
way to estimate the capacity of a given system element is to overload it and
determine the maximum throughput under the overloaded conditions. Care
must be taken in this process because a severe overload can reduce the
throughput by introducing self-aggravating phenomena upstream of the output
point.
Exhibit 27-27 shows the relationship between demand volume and
throughput, represented by the output of the weaving segment. As expected,
throughput tracks demand precisely up to the point where no more vehicles can
be accommodated. After that point it levels off and reaches a constant value that
indicates the capacity of the segment. In this case, capacity was reached at
approximately the same value as the HCM estimate. However, this degree of
agreement between the two estimation techniques should not be expected as a
general rule because of differences in the treatment of vehicle and geometric
characteristics.
On the basis of observation, it is reasonable to conclude that the capacity of
this weaving segment can be determined by overloading the facility and that the
results are in general agreement with those of the HCM. In comparing capacity
estimates, the analyst should remember that the HCM expresses results in
passenger car equivalent vehicles, while simulation tools express results in actual
vehicles. The results will diverge as the proportion of trucks increases.
Exhibit 27-27 12,000

Determining the Capacity of a
Weaving Segment by 10,000
Simulation
Volume (veh/h)
8,000
6,000
4,000
2,000
0
80 100 120 140 160 180
Percent of Original Demand
Demand Volume Throughput HCM Capacity
Alternative Tool Examples for Weaving Segments Chapter 27/Freeway Weaving: Supplemental
EFFECT OF DEMAND ON PERFORMANCE

Exhibit 27-28 shows the effect of demand on density and speed. Density
increases with demand volume up to the segment capacity and then levels off at
a constant value of approximately 75 veh/mi/ln, which represents very dense
conditions. The speed remains close to the free-flow speed at lower demand
volumes. It then drops in a more or less linear fashion and eventually levels off
when capacity is reached. The minimum speed is approximately 26 mi/h.
Exhibit 27-28
Simulated Effect of Demand
Volume on Weaving Segment
Capacity and Speed
At the originally specified demand volume level of 5,320 veh/h (peak hour
adjusted), the estimated speed was 62.0 mi/h and the density was 21.4 veh/mi/ln.
The corresponding values from simulation were 53.1 mi/h and 26.3 pc/ln/mi.
Because of differences in definition, these results are not easy to compare. These
differences illustrate the pitfalls of applying LOS thresholds to directly simulated
density to determine the segment LOS.
The densities produced when demand exceeded capacity were greater than
70 veh/ln/mi. This level of density is usually associated with queues that back up
from downstream bottlenecks; however, in this case, no such bottlenecks were
present. Inspection of the animated graphics suggests that the increase in density
within the weaving segment is caused by vehicles that are not able to get into the
required lane for their chosen exit. Some vehicles were forced to stop and wait
for a lane-changing opportunity, and the reduction in average speed produced a
corresponding increase in the average density.
For purposes of illustration, this example focuses on a single link containing
the weaving segment. The overloading of demand prevented all of the vehicles
from entering the link and would have increased the delay substantially if the
vehicles denied entry were considered. For this reason, the delay measures from
the simulation were not included in this discussion.
EFFECT OF QUEUE BACKUP FROM A DOWNSTREAM SIGNAL ON THE

EXIT RAMP
The operation of a weaving segment may be expected to deteriorate when
congestion on the exit ramp causes a queue to back up into the weaving segment.
This condition was one of the stated limitations of the methodology in
Chapter 13, Freeway Weaving Segments.
Signal Operation
To create this condition, a pretimed signal with a slightly oversaturated
operation is added 700 ft from the exit point. The operating parameters for the
signal are given in Exhibit 27-29. Note that the right-turn capacity estimated by
the Chapter 19, Signalized Intersections, procedure is slightly lower than the left-
turn capacity because of the adjustment factors applied to turns by that procedure.
Exhibit 27-29 Cycle length 150 s

Exit Ramp Signal Operating Green interval 95 s
Parameters Yellow interval 4s
All-red clearance 1s
Saturation flow rate 1,800 veh/hg/ln
g/C ratio 0.633
Left-turn movement
• Lanes 1
• Capacity (by HCM Chapter 19) 1,083 veh/h
Right-turn movement
• Lanes 1
• Capacity (by HCM Chapter 19) 969 veh/h
Link capacity (by HCM Chapter 19) 2,052 veh/h
Capacity Calibration
To ensure that the simulation model is properly calibrated to the HCM, the
simulation tool’s operating parameters for the link were modified by trial and
error to match the HCM estimate of the link capacity by overloading the link to
determine its throughput. With a start-up lost time of 2.0 s and a steady-state
headway of 1.8 s/veh, the simulated capacity for the link was 2,040 veh/h, which
compares well with the HCM’s estimate of 2,052 veh/h.
Results with the Specified Demand

An initial run with the demand levels specified in the original example
problem indicated severe problems on the freeway caused by the backup of
vehicles from the signal. Two adverse conditions are observed in the graphics
capture shown in Exhibit 27-30:
1. Some vehicles in the freeway mainline through lanes were unable to
access the auxiliary lane for the exit ramp because of blockage in the lane.
2. The resulting use of the exit ramp lanes prevented the signal operation
from reaching its full capacity. This caused a self-aggravating condition in
which the queue backed up farther onto the freeway.
Exhibit 27-30
Deterioration of Weaving
Segment Operation due to
Queue Backup from a Traffic
Signal
Mainline vehicles unable to
reach the exit lanes
Wasted time on the

signal approach
A reasonable conclusion is that the weaving segment would not operate

properly at the specified demand levels. The logical solution to the problem
would be to improve signal capacity. To support a recommendation for such an
improvement, varying the demand levels to gain further insight into the
operation might be desirable. Since it has already been discovered that the
specified demand is too high, the original levels of 80% to 180% of the specified
demand are clearly inappropriate. The new demand range will therefore be
reduced to a level of 80% to 105%.
Effect of Reducing Demand on Throughput

Exhibit 27-31 illustrates the self-aggravating effect of too much demand.
Throughput is generally expected to increase with demand up to the capacity of
the facility and to level off at that point. Notice that the anticipated relationship
was observed without the signal, as was shown in Exhibit 27-27.
When the signal was added, the situation changed significantly. The
throughput peaked at about 95% of the specified demand and declined
noticeably as more vehicles were allowed to enter the freeway. Another useful
observation is that the peak throughput of approximately 4,560 veh/h is
considerably below the estimated capacity of nearly 8,000 veh/h.
4,600 Exhibit 27-31

Effect of Demand on Weaving
Segment Throughput with Exit
4,500 Ramp Backup
Throughput (veh/h)
4,400
4,300
4,200
4,100
4,000
80 85 90 95 100 105
Percent of Specified Demand
The same phenomenon is observed on the exit ramp approach to the signal,
as shown in Exhibit 27-32. The throughput declined with added demand after
reaching its peak value of about 1,835 veh/h. Note that the peak throughput is
also well below the capacity of 2,040 to 2,050 veh/h estimated by both the HCM
and the simulation tool in the absence of upstream congestion.
Exhibit 27-32 1,840

Effect of Demand on Exit
Throughput from Signal (veh/h)

Ramp Throughput with Signal
Queuing
1,820
1,800
1,780
1,760
1,740
1,720
1,700
80 85 90 95 100 105
Percent of Specified Demand
This example illustrates the potential benefits of using simulation tools to

address conditions that are beyond the scope of the HCM methodology. It also
points out the need to consider conditions outside of the facility under study in
making a performance assessment. Finally, it demonstrates that care must be
taken in estimating the capacity of a facility through an arbitrary amount of
demand overload.
HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 28
FREEWAY MERGES AND DIVERGES: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 28-1

Example Problem 1: Isolated One-Lane, Right-Hand On-Ramp to a
Four-Lane Freeway ...................................................................................... 28-2
Example Problem 2: Two Adjacent Single-Lane, Right-Hand Off-
Ramps on a Six-Lane Freeway .................................................................... 28-4
Example Problem 3: One-Lane On-Ramp Followed by a One-Lane
Off-Ramp on an Eight-Lane Freeway ........................................................ 28-9
Example Problem 4: Single-Lane, Left-Hand On-Ramp on a Six-Lane
Freeway........................................................................................................ 28-14
Example Problem 5: Service Flow Rates and Service Volumes for an
Isolated On-Ramp on a Six-Lane Freeway .............................................. 28-17
3. ALTERNATIVE TOOL EXAMPLES FOR FREEWAY RAMPS ................... 28-22

Problem 1: Ramp-Metering Effects ................................................................. 28-22
Problem 2: Conversion of Leftmost Lane to an HOV Lane ......................... 28-25
Chapter 28/Freeway Merges and Diverges: Supplemental Contents

LIST OF EXHIBITS
Exhibit 28-1 List of Example Problems ....................................................................28-2

Exhibit 28-2 Example Problem 2: Capacity Checks ................................................28-7
Exhibit 28-3 Example Problem 3: Capacity Checks ..............................................28-12
Exhibit 28-4 Example Problem 5: Illustrative Service Flow Rates and
Service Volumes Based on Approaching Freeway Demand .......................28-19
Exhibit 28-5 Example Problem 5: Illustrative Service Flow Rates and
Service Volumes Based on a Fixed Freeway Demand ..................................28-21
Exhibit 28-6 Graphics Capture of the Ramp Merge with Ramp Metering ........28-23
Exhibit 28-7 Density as a Function of Ramp-Metering Headways ....................28-23
Exhibit 28-8 Capacity at a Ramp Junction as a Function of Ramp-
Metering Headways ..........................................................................................28-24
Exhibit 28-9 Queue Length on the Ramp as a Function of Ramp-Metering
Headways ...........................................................................................................28-24
Exhibit 28-10 Graphics Capture of the Segment with an HOV Lane .................28-25
Exhibit 28-11 Density of a Ramp Junction as a Function of the Carpool
Percentage ...........................................................................................................28-25
Exhibit 28-12 Capacity of a Ramp Junction as a Function of the Carpool
Percentage ...........................................................................................................28-26
Exhibit 28-13 Density of a Ramp Junction as a Function of the HOV
Violation Percentage .........................................................................................28-26
Exhibit 28-14 Capacity of a Ramp Junction as a Function of the HOV
Violation Percentage .........................................................................................28-27
Exhibit 28-15 Density of a Ramp Junction as a Function of the Distance at
Which Drivers Begin to React ..........................................................................28-27
Exhibit 28-16 Capacity of a Ramp Junction as a Function of the Distance
at Which Drivers Begin to React ......................................................................28-28
Exhibit 28-17 Density of a Ramp Junction as a Function of the Percentage
of HOV Usage ....................................................................................................28-28
Exhibit 28-18 Capacity of a Ramp Junction as a Function of the
Percentage of HOV Usage ................................................................................28-29
Contents Chapter 28/Freeway Merges and Diverges: Supplemental

1. INTRODUCTION
Chapter 28 is the supplemental chapter for Chapter 14, Freeway Merge and VOLUME 4: APPLICATIONS
GUIDE
Diverge Segments, which is found in Volume 2 of the Highway Capacity Manual 25. Freeway Facilities:
(HCM). Section 2 provides five example problems demonstrating the application Supplemental
of the Chapter 14 methodology and its extension to freeway managed lanes. Segments: Supplemental
Section 3 presents examples of applying alternative tools to the analysis of 27. Freeway Weaving:
Supplemental
freeway merge and diverge segments to address limitations of the Chapter 14 28. Freeway Merges and
methodology. Diverges: Supplemental
Supplemental
Supplemental
Supplemental
32. STOP-Controlled
Intersections:
Supplemental
33. Roundabouts:
Supplemental
Supplemental
Chapter 28/Freeway Merges and Diverges: Supplemental Introduction

2. EXAMPLE PROBLEMS
Exhibit 28-1 lists the example problems presented in this section.

List of Example Problems Problem Title Type of Analysis
Isolated One-Lane, Right-Hand On-Ramp to a Four-Lane Operational analysis
1
Freeway
Two Adjacent Single-Lane, Right-Hand Off-Ramps on a Operational analysis
2
Six-Lane Freeway
One-Lane On-Ramp Followed by a One-Lane Off-Ramp Operational analysis
3
on an Eight-Lane Freeway
4 Single-Lane, Left-Hand On-Ramp on a Six-Lane Freeway Special case
Service Flow Rates and Service Volumes for an Isolated Service flow rates and
5
On-Ramp on a Six-Lane Freeway service volumes
EXAMPLE PROBLEM 1: ISOLATED ONE-LANE, RIGHT-HAND ON-RAMP

TO A FOUR-LANE FREEWAY
The Facts
The following data are available to describe the traffic and geometric
characteristics of this location. The example assumes no impacts of inclement
weather or incidents.
1. Isolated location (no adjacent ramps to consider);
2. One-lane ramp roadway and junction;
3. Four-lane freeway (two lanes in each direction);
4. Upstream freeway demand volume = 2,500 veh/h;
5. Ramp demand volume = 535 veh/h;
6. 5% trucks throughout;
7. Acceleration lane = 740 ft;
8. FFS, freeway = 60 mi/h;
9. FFS, ramp = 45 mi/h;
10. Level terrain for freeway and ramp;
11. Peak hour factor (PHF) = 0.90; and
12. Drivers are regular commuters.
Comments
All input parameters are known, so no default values are needed or used.
Adjustment factors for heavy vehicles and driver population are found in
Chapter 12, Basic Freeway and Multilane Highway Segments.
Example Problems Chapter 28/Freeway Merges and Diverges: Supplemental

Step 1: Specify Inputs and Convert Demand Volumes to Demand

Flow Rates
Input parameters were specified in the Facts section above. Equation 14-1 is
used to convert demand volumes to flow rates under equivalent ideal conditions:
𝑉𝑖
𝑣𝑖 =
Demand volumes are given for the freeway and the ramp. The PHF is
specified. The driver population adjustment factors for commuters are 1.00
(Chapter 12), while the heavy vehicle adjustment factor is computed as follows:
1
𝑓𝐻𝑉 =
(𝐸
1 + 𝑃𝑇 𝑇 − 1)
Truck presence is given. The value of ET for level terrain is 2.0 (Chapter 12).
On the basis of these values, the freeway and ramp demand volumes are
converted as follows:
For the freeway,
1 1
𝑓𝐻𝑉 = = = 0.952
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.05(2.0 − 1)
2,500
𝑣𝐹 = = 2,918 pc/h
0.90 × 0.952
For the ramp, the calculations are identical:
1
𝑓𝐻𝑉 = = 0.952
1 + 0.05(2.0 − 1)
535
𝑣𝑅 = = 625 pc/h
0.90 × 0.952
Step 2: Estimate the Approaching Flow Rate in Lanes 1 and 2 of the

Freeway Immediately Upstream of the Ramp Influence Area
The demand flow in Lanes 1 and 2 immediately upstream of the ramp
influence area is computed by using Equation 14-2.
𝑣12 = 𝑣𝐹 × 𝑃𝐹𝑀
The freeway flow rate was computed in Step 1. The value of PFM is found in
Exhibit 14-8. For a four-lane freeway, the value is 1.00. Then
𝑣12 = 2,918 × 1.00 = 2,918 pc/h
Because there are no outer lanes on a four-lane freeway, there is no need to
check this result for reasonableness.
Step 3: Estimate the Capacity of the Ramp–Freeway Junction and

Compare with Demand Flow Rates
The critical capacity checkpoint for a single-lane on-ramp is the downstream
freeway segment:
𝑣𝐹𝑂 = 𝑣𝐹 + 𝑣𝑅 = 2,918 + 625 = 3,543 pc/h
Chapter 28/Freeway Merges and Diverges: Supplemental Example Problems

The capacity of a four-lane freeway (two lanes in one direction) with an FFS
of 60 mi/h is given in Exhibit 14-10. The capacity is 4,600 pc/h, which is more
than the demand flow of 3,543 pc/h. The capacity of a one-lane ramp with an FFS
of 45 mi/h is given in Exhibit 14-12 as 2,100 pc/h, which is well in excess of the
ramp demand flow of 625 pc/h. The maximum desirable flow rate entering the
ramp influence area is also 4,600 pc/h, again more than 3,543. Thus, the operation
of the segment is expected to be stable. LOS F does not exist. Note that there
were no adjustments to speed (SAF) or capacity (CAF) due to inclement weather,
incidents, or other impacts for this case.
Step 4: Estimate Density in the Ramp Influence Area and Determine

the Prevailing LOS
The estimated density in the ramp–freeway junction is estimated by using
Equation 14-22:
𝐷𝑅 = 5.475 + 0.00734𝑣𝑅 + 0.0078𝑣12 − 0.00627𝐿𝐴
𝐷𝑅 = 5.475 + 0.00734(625) + 0.0078(2,918) − 0.00627(740)
𝐷𝑅 = 28.2 pc/mi/ln
From Exhibit 14-3, this is LOS D, but the result is close to the LOS C
boundary.
Step 5: Estimate Speeds in the Vicinity of Ramp–Freeway Junctions

Since there are no outer lanes on a four-lane freeway, only the speed within
the ramp influence area should be computed, by using the equations given in
Exhibit 14-13:
𝑀𝑆 = 0.321 + 0.0039𝑒 (𝑣𝑅12 /1,000) − 0.002(𝐿𝐴 × 𝑆𝐹𝑅 × 𝑆𝐴𝐹/1,000)
𝑀𝑆 = 0.321 + 0.0039𝑒 (3,543/1,000) − 0.002(740 × 45 × 1.00/1,000) = 0.389
𝑆𝑅 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 42)𝑀𝑆
𝑆𝑅 = 60 × 1.00 − (60 × 1.00 − 42)(0.389) = 53.0 mi/h
Note that the speed adjustment factor, SAF, is 1.00, since this is not a case
where inclement weather or other factors would necessitate a correction.
Discussion
The results indicate that the merge area operates in a stable fashion, with
some deterioration in density and speed due to merging operations.
EXAMPLE PROBLEM 2: TWO ADJACENT SINGLE-LANE, RIGHT-HAND

OFF-RAMPS ON A SIX-LANE FREEWAY
The Facts
The following information concerning demand volumes and geometries is
available for this problem. The example assumes no impacts of inclement
weather or incidents.
1. Two consecutive one-lane, right-hand off-ramps;
2. Six-lane freeway with FFS = 60 mi/h;
3. Level terrain for freeway and both ramps;

4. 7.5% trucks on freeway and both ramps;

5. First-ramp FFS = 40 mi/h;
6. Second-ramp FFS = 25 mi/h;
7. Drivers are regular commuters;
8. Freeway demand volume = 4,500 veh/h (immediately upstream of the
first off-ramp);
9. First-ramp demand volume = 300 veh/h;
10. Second-ramp demand volume = 500 veh/h;
11. Distance between ramps = 750 ft;
12. First-ramp deceleration lane length = 500 ft;
13. Second-ramp deceleration lane length = 300 ft; and
14. Peak hour factor = 0.95.
Comments
The solution will use adjustment factors for heavy vehicle presence and
driver population selected from Chapter 12, Basic Freeway and Multilane
Highway Segments. All input parameters are specified, so no default values are
needed or used.

Flow Rates
𝑉𝑖
𝑣𝑖 =
In this case, three demand volumes must be converted: the freeway volume
immediately upstream of the first ramp and the two ramp demand volumes.
Since all demands include 7.5% trucks, only a single heavy vehicle adjustment
factor will be needed. From Chapter 12, the appropriate value of ET for level
terrain is 2.0.
Then
1 1
𝑓𝐻𝑉 = = = 0.930
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.075(2 − 1)
and
4,500
𝑣𝐹 = = 5,093 pc/h
0.95 × 0.930
300
𝑣𝑅1 = = 340 pc/h
0.95 × 0.930
500
𝑣𝑅2 = = 566 pc/h
0.95 × 0.930


Because two consecutive off-ramps are under consideration, the first will
have to consider the impact of the second on its operations, and the second will
have to consider the impact of the first.
First Off-Ramp
From Exhibit 14-9, flow in Lanes 1 and 2 of the freeway is estimated by using
Equation 14-11 or Equation 14-9, depending on whether the impact of the
downstream off-ramp is significant. This is determined by computing the
equivalence distance by using Equation 14-13:
𝑣𝐷
𝐿𝐸𝑄 =
1.15 − 0.000032𝑣𝐹 − 0.000369𝑣𝑅
566
𝐿𝐸𝑄 = = 657 ft
1.15 − 0.000032(5,093) − 0.000369(340)
Since the actual distance between ramps, 750 ft, is greater than the
equivalence distance of 657 ft, the ramp may be treated as if it were isolated, with
Equation 14-9:
𝑃𝐹𝐷 = 0.760 − 0.000025𝑣𝐹 − 0.000046𝑣𝑅
𝑃𝐹𝐷 = 0.760 − 0.000025(5,093) − 0.000046(340) = 0.617
Then from Equation 14-8,
𝑣12 = 𝑣𝑅 + (𝑣𝐹 − 𝑣𝑅 )𝑃𝐹𝐷
𝑣12 = 340 + (5,093 − 340)(0.617) = 3,273 pc/h
Because a six-lane freeway includes one lane in addition to the ramp
influence areas (the innermost lane, Lane 3), the reasonableness of the predicted
lane distribution of arriving freeway vehicles should be checked. The flow rate in
Lane 3 is 5,093 – 3,273 = 1,820 pc/h. The average flow per lane in Lanes 1 and 2 is
3,273/2 = 1,637 pc/h (rounded to the nearest pc). Then:
Is v3 > 2,700 pc/h/ln? No
Is v3 > 1.5 × (1,637) = 2,456 pc/h/ln? No
Since both checks for reasonable lane distribution are passed, the computed
value of v12 for the first off-ramp is accepted as 3,273 pc/h.
Second Off-Ramp
From Exhibit 14-9, the second off-ramp should be analyzed by using
Equation 14-9, which is for an isolated off-ramp. Adjacent upstream off-ramps do
not affect the lane distribution of arriving vehicles at a downstream off-ramp.
The freeway flow approaching Ramp 2, however, includes the freeway flow
approaching Ramp 1, less the flow rate of vehicles exiting the freeway at Ramp 1.
Therefore, the freeway flow rate approaching Ramp 2 is as follows:
𝑣𝐹2 = 5,093 − 340 = 4,753 pc/h

Then
𝑃𝐹𝐷 = 0.760 − 0.000025𝑣𝐹 − 0.000046𝑣𝑅
𝑃𝐹𝐷 = 0.760 − 0.000025(4,753) − 0.000046(566) = 0.615
𝑣12 = 566 + (4,753 − 566)(0.615) = 3,141 pc/h
Again, because there is an outer lane on a six-lane freeway, the
reasonableness of this estimate must be checked. The flow rate in the innermost
lane v3 is 4,753 – 3,141 = 1,612 pc/h. The average flow rate in Lanes 1 and 2 is
3,141/2 = 1,571 pc/h (rounded). Then:
Is v3 > 2,700 pc/h/ln? No
Is v3 > 1.5 × 1,571 = 2,357 pc/h/ln? No
Once again, the predicted lane distribution of arriving vehicles is reasonable,
and v12 is taken to be 3,141 pc/h.

Because two off-ramps are involved in this segment, there are several
capacity checkpoints:
1. Total freeway flow upstream of the first off-ramp (the point at which
maximum freeway flow exists),
2. Capacity of both off-ramps, and
3. Maximum desirable flow rates entering each of the two off-ramp
influence areas.
These comparisons are shown in Exhibit 28-2. Note that freeway capacity is
based on a freeway with FFS = 60 mi/h. The first ramp capacity is based on a
ramp FFS of 40 mi/h and the second on a ramp FFS of 25 mi/h.
Demand Flow Exhibit 28-2

Capacity (pc/h) from Rate Example Problem 2:
Item Exhibit 14-10 or Exhibit 14-12 (pc/h) Problem? Capacity Checks
Freeway flow rate 6,900 5,093 No
First off-ramp 2,000 340 No
Second off-ramp 1,900 566 No
Max. v12 first ramp 4,400 3,273 No
Max. v12 second ramp 4,400 3,141 No
Note: Max. = maximum.
None of the capacity values are exceeded, so operation of these ramp

junctions will be stable, and LOS F does not occur. Again, there are no situations
that would call for an adjustment to be made to speed (SAF) or capacity (CAF).

the Prevailing LOS
Because there are two off-ramps, two ramp influence areas are involved, and
two ramp influence area densities will be computed with Equation 14-23.
𝐷𝑅 = 4.252 + 0.0086𝑣12 − 0.009𝐿𝐷
𝐷𝑅1 = 4.252 + 0.0086(3,273) − 0.009(500) = 27.9 pc/mi/ln
𝐷𝑅2 = 4.252 + 0.0086(3,141) − 0.009(300) = 28.6 pc/mi/ln

From Exhibit 14-3, both of these ramp influence areas operate close to the
boundary between LOS C and LOS D (28.0 pc/mi/ln). Ramp 1 operates in LOS C,
while Ramp 2 operates in LOS D.
Although it makes virtually no difference in this case, note that the two ramp
influence areas overlap. The influence area of the first off-ramp extends 1,500 ft
upstream. The influence area of the second off-ramp also extends 1,500 ft
upstream. Since the ramps are only 750 ft apart, the second ramp influence area
overlaps the first for 750 ft (immediately upstream of the first diverge point). The
worse of the two levels of service is applied to this 750-ft overlap. In this case, the
levels of service are different, even though the predicted densities are similar.
Thus, the overlapping influence area is assigned LOS D.

Because these ramps are on a six-lane freeway with an outer lane, the speed
within each ramp influence area, the speed in the outer lane adjacent to each
ramp influence area, and the weighted average of the two can be estimated.
First Off-Ramp
The speed within the first ramp influence area is computed by using the
equations given in Exhibit 14-14:
𝐷𝑆 = 0.883 + 0.00009𝑣𝑅 − 0.013𝑆𝐹𝑅 × 𝑆𝐴𝐹
𝐷𝑆 = 0.883 + 0.00009(340) − 0.013(40)(1.00) = 0.394
𝑆𝑅 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 42)𝐷𝑆
𝑆𝑅 = (60)(1.00) − (60 × 1.00 − 42)(0.394) = 52.9 mi/h
The flow rate in the outer lane vOA is 5,093 – 3,273 = 1,820 pc/h/ln. The average
speed in this outer lane is computed as follows, by using the equation given in
Exhibit 14-14:
𝑆𝑂 = 1.097 × 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 0.0039(𝑣𝑂𝐴 − 1,000)
𝑆𝑂 = (1.097)(60)(1.00) − 0.0039(1,820 − 1,000) = 62.6 mi/h
The average speed in Lane 3 is predicted to be slightly higher than the FFS of
the freeway. This is not uncommon, since through vehicles at higher speeds use
Lane 3 to avoid congestion in the ramp influence area. However, the average
speed across all lanes should not be higher than the FFS. In this case, the average
speed across all lanes is computed as follows, by using the appropriate equation
from Exhibit 14-15:
𝑣12 + 𝑣𝑂𝐴 𝑁𝑂 3,273 + (1,820)(1)
𝑆= = = 56.0 mi/h
𝑣12 𝑉𝑂𝐴 𝑁𝑂 3,273 1,820 × 1
( 𝑆 ) + ( 𝑆 ) ( 52.9 ) + ( 62.6 )
𝑅 𝑂
This result is, as expected, less than the FFS of the freeway.
Note that once again the SAF is 1.00, since there are no conditions that would
require an adjustment.

Second Off-Ramp
The speed in the second ramp influence area is computed as follows:
𝐷𝑆 = 0.883 + 0.00009(566) − 0.013(25)(1.00) = 0.609
𝑆𝑅 = (60)(1.00) − (60 × 1.00 − 42)(0.609) = 49.0 mi/h
Lane 3 has a demand flow rate of 4,753 – 3,141 = 1,612 pc/h/ln. The average
speed in this outer lane is computed as follows:
𝑆𝑂 = (1.097)(60)(1.00) − 0.0039(1,612 − 1,000) = 63.4 mi/h
The average speed across all freeway lanes is
𝑣12 + 𝑣𝑂𝐴 𝑁𝑂 3,141 + (1,612)(1)
𝑆= = = 53.1 mi/h
𝑣 𝑉 𝑁𝑂 3,141 1,612 × 1
( 𝑆12 ) + ( 𝑂𝐴
𝑆𝑂 ) ( 49.0 ) +(
63.4 )
𝑅
Discussion
The speed results in this case are interesting. While densities are similar for
both ramps, the density is somewhat higher and the speed somewhat lower in
the second influence area. This is primarily the result of a shorter deceleration
lane and a lower ramp FFS (25 mi/h versus 40 mi/h). In both cases, the average
speed in the outer lane is higher than the FFS, which applies as an average across
all lanes.
Since the operation is stable, there is no special concern here, short of a
significant increase in demand flows. LOS is technically D but falls just over the
LOS C boundary. In this case the step-function LOS assigned may imply
operation poorer than actually exists. It emphasizes the importance of knowing
not only the LOS but also the value of the service measure that produces it.
EXAMPLE PROBLEM 3: ONE-LANE ON-RAMP FOLLOWED BY A

ONE-LANE OFF-RAMP ON AN EIGHT-LANE FREEWAY
The Facts
The following information is available concerning this pair of ramps to be
analyzed. The example assumes no impacts of inclement weather or incidents.
1. Eight-lane freeway with an FFS of 65 mi/h;
2. One-lane, right-hand on-ramp with an FFS of 30 mi/h;
3. One-lane, right-hand off-ramp with an FFS of 25 mi/h;
4. Distance between ramps = 1,300 ft;
5. Acceleration lane on Ramp 1 = 260 ft;
6. Deceleration lane on Ramp 2 = 260 ft;
7. Level terrain on freeway and both ramps;
8. 10% trucks on freeway and off-ramp;
9. 5% trucks on on-ramp;
10. Freeway flow rate (upstream of first ramp) = 5,490 veh/h;
11. On-ramp flow rate = 410 veh/h;
12. Off-ramp flow rate = 600 veh/h;

13. PHF = 0.94; and

14. Drivers are regular commuters.
Comments
As with previous example problems, the conversion of demand volumes to
flow rates requires adjustment factors selected from Chapter 12, Basic Freeway
and Multilane Highway Segments. All pertinent information is given, and no
default values will be applied.

Flow Rates
𝑉𝑖
𝑣𝑖 =
Three demand volumes must be converted to flow rates under equivalent
ideal conditions: the freeway volume immediately upstream of the first ramp
junction, the first ramp volume, and the second ramp volume. Because the
freeway segment under study has level terrain, the value of ET will be 2.0 for all
volumes.
Then, for the freeway demand volume,
1 1
𝑓𝐻𝑉 = = = 0.909
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.10(2 − 1)
5,490
𝑣𝐹 = = 6,425 pc/h
0.94 × 0.909
For the on-ramp demand volume,
1
𝑓𝐻𝑉 = = 0.952
1 + 0.05(2 − 1)
410
𝑣𝑅1 = = 458 pc/h
0.94 × 0.952
For the off-ramp demand volume,
1
𝑓𝐻𝑉 = = 0.909
1 + 0.10(2 − 1)
600
𝑣𝑅2 = = 702 pc/h
0.94 × 0.909
In the remaining computations, these converted demand flow rates are used
as input values.

Once again, the situation involves a pair of adjacent ramps. Analysis of each
ramp must take into account the potential impact of the other on its operations.
Because the ramps are on an eight-lane freeway (four lanes in each direction),
Exhibit 14-8 and Exhibit 14-9 indicate that each ramp is considered as if it were
isolated.

First Ramp: On-Ramp

Exhibit 14-8 applies to on-ramps. Exhibit 14-8 presents two possible
equations for use in estimating v12 on the basis of the value of vF/SFR. In this case,
the value is 6,425/30 = 214.2 > 72. Therefore, the second equation for eight-lane
freeways given in Exhibit 14-8 is used, giving the following:
𝑃𝐹𝑀 = 0.2178 − 0.000125𝑣𝑅 = 0.2178 − 0.000125(458) = 0.16
𝑣12 = (6,425)(0.16) = 1,028 pc/h
Because the eight-lane freeway includes two outer lanes in each direction,
the reasonableness of this prediction must be checked. The average flow per lane
in Lanes 1 and 2 is 1,028/2 = 514 pc/h/ln (rounded). The flow in the two outer
lanes, Lanes 3 and 4, is 6,425 – 1,028 = 5,397 pc/h. The average flow per lane in
Lanes 3 and 4 is, therefore, 5,397/2 ~ 2,699 pc/h/ln. Then:
Is v av 34 > 2,700 pc/h/ln? No
Is v av 34 > 1.5 × 514 = 771 pc/h/ln? Yes
Therefore, the predicted lane distribution is not reasonable. Too many
vehicles are placed in the two outer lanes compared with Lanes 1 and 2. Equation
14-19 is used to produce a more reasonable distribution:
𝑣𝐹 6,425
𝑣12𝑎 = ( )=( ) = 2,570 pc/h
2.50 2.50
On the basis of this adjusted value, the number of vehicles now assigned to
the two outer lanes is 6,425 – 2,570 = 3,855 pc/h.
Second Ramp: Off-Ramp

Equation 14-8 and Exhibit 14-9 apply to off-ramps. Exhibit 14-9 shows that
the value of PFD for off-ramps on eight-lane freeways is a constant: 0.436. Since
the methodology is based on regression analysis of a database, the
recommendation of a constant reflects a small sample size in that database. Note
also that the freeway flow approaching the second ramp is the sum of the
freeway flow approaching the first ramp and the on-ramp flow that is now also
on the freeway, or 6,425 + 458 = 6,883 pc/h. The flow rate in Lanes 1 and 2 is now
easily computed by using Equation 14-8:
𝑣12 = 𝑣𝑅 + (𝑣𝐹 − 𝑣𝑅 )𝑃𝐹𝐷
𝑣12 = 702 + (6,883 − 702)(0.436) = 3,397 pc/h
Because there are two outer lanes on this eight-lane freeway, the
reasonableness of this estimate must be checked. The average flow per lane in
Lanes 1 and 2 is 3,397/2 = 1,699 pc/h/ln. The total flow in Lanes 3 and 4 of the
freeway is 6,883 – 3,397 = 3,486 pc/h, or an average flow rate per lane of 3,486/2 =
1,743 pc/h/ln.
Is vav34 > 2,700 pc/h/ln? No
Is vav34 > 1.5 × 1,699 = 2,549 pc/h/ln? No
Therefore, the estimated value of v12 is deemed reasonable and is carried
forward in the computations.


Because there are two ramps in this segment, there are five capacity
checkpoints to consider:
1. The freeway flow rate at its maximum point—which in this case is
between the on- and off-ramp, since this is the only location where both
on- and off-ramp vehicles are on the freeway.
2. The capacity of the on-ramp.
3. The capacity of the off-ramp.
4. The maximum desirable flow entering the on-ramp influence area.
5. The maximum desirable flow entering the off-ramp influence area.
These comparisons are shown in Exhibit 28-3. The capacity of the freeway is
based on an eight-lane freeway with an FFS of 65 mi/h. The capacity of the on-
ramp is based on an FFS of 30 mi/h, and the capacity of the off-ramp is based on
an FFS of 25 mi/h.
Exhibit 28-3 Capacity (pc/h) from Demand Flow Rate

Example Problem 3: Item Exhibit 14-10 or Exhibit 14-12 (pc/h) Problem?
Capacity Checks Freeway flow rate 9,400 6,883 No
First on-ramp 1,900 458 No
Second off-ramp 1,900 702 No
Max. vR12 first ramp 4,600 2,570 + 458 = 3,028 No
Max. v12 second ramp 4,400 3,397 No
There are no capacity concerns, since all demands are well below the
associated capacities or maximum desirable values. No adjustments to capacity
are required. LOS F is not present in any part of this segment, and operations are
expected to be stable.

the Prevailing LOS
Equation 14-22 is used to find the density in the first on-ramp influence area:
𝐷𝑅 = 5.475 + 0.00734𝑣𝑅 + 0.0078𝑣12 − 0.00627𝐿𝐴
𝐷𝑅 = 5.475 + 0.00734(458) + 0.0078(2,570) − 0.00627(260)
𝐷𝑅 = 27.2 pc/mi/ln
Equation 14-23 is used to find the density in the second off-ramp influence
area:
𝐷𝑅 = 4.252 + 0.0086𝑣12 − 0.009𝐿𝐷
𝐷𝑅 = 4.252 + 0.0086(3,397) − 0.009(260) = 31.1 pc/mi/ln
From Exhibit 14-3, both of these ramp influence areas operate close to the
boundary between LOS C and LOS D (28 pc/mi/ln). Ramp 1 operates in LOS C,
while Ramp 2 operates in LOS D.
Because the on-ramp influence area extends 1,500 ft downstream, the off-
ramp influence area extends 1,500 ft upstream, and the two ramps are only 1,300
ft apart, the distance between the ramps is included in both. Therefore, the lower
LOS D for the off-ramp governs the operation. Note that the additional 200 ft of

the off-ramp influence area is actually upstream of the on-ramp, and the
additional 200 ft of the on-ramp influence area is downstream of the off-ramp.

Because the facility is an eight-lane freeway, speeds should be estimated for
the two ramp influence areas, for the outer lanes (Lanes 3 and 4) adjacent to the
ramp influence areas, and for all vehicles—the weighted average of the other two
speeds.
First Ramp (On-Ramp)

Equations for estimation of average speed in an on-ramp influence area and
in outer lanes adjacent to it are taken from Exhibit 14-13.
𝑀𝑆 = 0.321 + 0.0039𝑒 (𝑣𝑅12 /1,000) − 0.002(𝐿𝐴 × 𝑆𝐹𝑅 × 𝑆𝐴𝐹/1,000)
𝑀𝑆 = 0.321 + 0.0039𝑒 (3,025/1,000) − 0.002(260 × 30 × 1.00/1,000) = 0.385
𝑆𝑅 = (65)(1.00) − (65 × 1.00 − 42)(0.385) = 56.2 mi/h
Since the average outer lane demand flow rate is 3,855/2 = 1,927 pc/h/ln,
which is greater than 500 pc/h/ln and less than 2,300 pc/h/ln, the outer speed is
estimated as follows, by using the appropriate equation from Exhibit 14-13:
𝑆𝑂 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 0.0036(𝑣𝑂𝐴 − 500)
𝑆𝑂 = (65)(1.00) − 0.0036(1,927 − 500) = 59.9 mi/h
Note that the speed adjustment factor (SAF) is 1.00.
The weighted average speed of all vehicles is
𝑣12 + 𝑣𝑂𝐴 𝑁𝑂 3,028 + (1,927)(2)
𝑆= = = 58.8 mi/h
𝑣12 𝑣𝑂𝐴 𝑁𝑂 3,028 1,927 × 2
(𝑆 )+( 𝑆 ) ( )+( )
𝑅 𝑂 56.2 59.9
Second Ramp (Off-Ramp)

For off-ramps, equations for estimation of average speed are drawn from
Exhibit 14-14. At the second ramp, the flow in Lanes 1 and 2 has been computed
as 3,397 pc/h or 1,699 pc/h/ln, while the flow in Lanes 3 and 4 is 3,486 pc/h or
1,743 pc/h/ln. Then
𝐷𝑆 = 0.883 + 0.00009𝑣𝑅 − 0.013𝑆𝐹𝑅 × 𝑆𝐴𝐹
𝐷𝑆 = 0.883 + 0.00009(702) − 0.013(25)(1.00) = 0.621
𝑆𝑅 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − (𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 42)𝐷𝑆
𝑆𝑅 = (65)(1.00) − (65 × 1.00 − 42)(0.621) = 50.7 mi/h
Because the average flow in the outer lanes is greater than 1,000 pc/h/ln, the
average speed of vehicles in the outer lanes (Lanes 3 and 4) is as follows:
𝑆𝑂 = 1.097 × 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 0.0039(𝑣𝑂𝐴 − 1,000)
𝑆𝑂 = (1.097)(65)(1.00) − 0.0039(1,743 − 1,000) = 68.4 mi/h

The weighted average speed of all vehicles is

𝑣12 + 𝑣𝑂𝐴 𝑁𝑂 3,397 + (1,743)(2)
𝑆= = = 58.3 mi/h
𝑣 𝑣 𝑁 3,397 1,743 × 2
( 𝑆12 ) + ( 𝑂𝐴
𝑆𝑂
𝑂
) ( 50.7 ) + ( 68.4 )
𝑅
Discussion
As noted previously, between the ramps, the influence areas of both ramps
fully overlap. Since a higher density is predicted for the off-ramp influence area,
and LOS D results, this density should be applied to the entire area between the
two ramps.
The speed results are also interesting. The slower speeds within the off-ramp
influence area will also control the overlap area. On the other hand, the speed
results indicate a higher average speed for all vehicles associated with the off-
ramp than for those associated with the on-ramp. This is primarily due to the
much larger disparity between speeds within the ramp influence area and in
outer lanes when the off-ramp is considered. The speed differential is more than
20 mi/h for the off-ramp, as opposed to a little more than 3 mi/h for the on-ramp.
This is not entirely unexpected. At diverge junctions, vehicles in outer lanes tend
to face less turbulence than those in outer lanes near merge junctions. All off-
ramp vehicles must be in Lanes 1 and 2 for some distance before exiting the
freeway. On-ramp vehicles, in contrast, can execute as many lane changes as they
wish, and more of them may wind up in outer lanes within 1,500 ft of the
junction point.
Thus, the total operation of this two-ramp segment is expected to be LOS D,
with speeds of approximately 50 mi/h in Lanes 1 and 2 and approximately 70
mi/h in Lanes 3 and 4.
EXAMPLE PROBLEM 4: SINGLE-LANE, LEFT-HAND ON-RAMP ON A

SIX-LANE FREEWAY
The Facts
The following information is available concerning this example problem. The
example assumes no impacts of inclement weather or incidents.
1. One-lane, left-side on-ramp on a six-lane freeway (three lanes in each
direction);
2. Freeway demand volume upstream of ramp = 4,000 veh/h;
3. On-ramp demand volume = 490 veh/h;
4. 7.5% trucks on freeway, 3% trucks on the on-ramp;
5. Freeway FFS = 65 mi/h;
6. Ramp FFS = 30 mi/h;
7. Acceleration lane = 820 ft;
8. Level terrain on freeway and ramp;
9. Drivers are regular commuters; and
10. PHF = 0.90.

Comments
This is a special application of the ramp analysis methodology presented in
Chapter 14. For left-hand ramps, the flow rate in Lanes 1 and 2 (v12) is initially
computed as if it were a right-hand ramp. Exhibit 14-18 is then used to convert
this result to an estimate of the flow in Lanes 2 and 3 (v23), since these are the two
leftmost lanes that will be involved in the merge. In effect, the ramp influence
area is, in this case, Lanes 3 and 4 and the acceleration lane for a distance of 1,500
ft downstream of the merge point.
Step 1: Specify Inputs and Convert Demand Volumes to Demand Flow

Rates
𝑉𝑖
𝑣𝑖 =
From Chapter 12, Basic Freeway and Multilane Highway Segments, the
passenger car equivalent ET for trucks in level terrain is 2.0.
For the freeway demand volume,
1 1
𝑓𝐻𝑉 = = = 0.93
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.075(2 − 1)
4,000
𝑣𝐹 = = 4,779 pc/h
0.90 × 0.93
For the ramp demand volume,
1
𝑓𝐻𝑉 = = 0.971
1 + 0.03(2 − 1)
490
𝑣𝑅 = = 561 pc/h
0.90 × 0.971

To estimate flow in the two left lanes, the flow normally expected in Lanes 1
and 2 for a similar right-hand ramp must first be computed. From Exhibit 14-8,
for an isolated on-ramp on a six-lane freeway, Equation 14-4 is used:
𝑃𝐹𝑀 = 0.5775 + 0.000028𝐿𝐴 = 0.5775 + 0.000028(820) = 0.600
𝑣12 = (4,779)(0.600) = 2,867 pc/h
From Exhibit 14-18, the adjustment factor applied to this result to find the
estimated flow rate in Lanes 2 and 3 is 1.12. Therefore,
𝑣23 = 2,867 × 1.12 = 3,211 pc/h
While, strictly speaking, the reasonableness criteria for lane distribution do
not apply to left-hand ramps, they can be applied very approximately. In this
case, the single “outer lane” (which is now Lane 1) would have a flow rate of
4,779 – 3,211 = 1,568 pc/h. This is not greater than 2,700 pc/h/ln, nor is it greater
than 1.5 times the average flow in Lanes 2 and 3 (1.5 × 3,211/2 = 2,408 pc/h/ln).

Thus, even if the reasonableness criteria were approximately applied in this case,
no violation would exist.
The remaining computations proceed for the left-hand ramp, with the
substitution of v34 for v12 in all algorithms used.

For this case, there are three simple checkpoints:
1. The principal capacity checkpoint is the total demand flow rate
downstream of the merge, 4,779 + 561 = 5,340 pc/h. From Exhibit 14-10, for
a six-lane freeway with an FFS of 65 mi/h, the capacity is 7,050 pc/h, well
over the demand flow rate.
2. The ramp roadway capacity should also be checked by using Exhibit 14-
12. For a single-lane ramp with an FFS of 30 mi/h, the capacity is 1,900
pc/h, which is much greater than the demand flow rate of 561 pc/h.
3. Finally, the maximum flow entering the ramp influence area should be
checked. In this case, a left-hand ramp, the total flow entering the ramp
influence area is the freeway flow remaining in Lanes 2 and 3 plus the
ramp flow rate. Thus, the total flow entering the ramp influence area is
3,211 + 561 = 3,772 pc/h, which is lower than the maximum desirable flow
rate of 4,600 pc/h, shown in Exhibit 14-10.
Thus, there are no capacity problems at this merge point, and stable
operations are expected. LOS F will not result from the stated conditions.

the Prevailing LOS
The density in the ramp influence area is found by using Equation 14-22,
except v23 replaces v12 because of the left-hand ramp placement:
𝐷𝑆 = 5.475 + 0.00734𝑣𝑅 + 0.0078𝑣23 − 0.00627𝐿𝐴
𝐷𝑆 = 5.475 + 0.00734(561) + 0.0078(3,211) − 0.00627(820)
𝐷𝑆 = 29.5 pc/mi/ln
From Exhibit 14-3, this is LOS D.

The speed estimation algorithms were calibrated for right-hand ramps, and
the estimation algorithms for “outer lane(s)” assume that these are the leftmost
lanes. Thus, for a left-hand ramp, these computations must be considered
approximate at best.
By using the equations in Exhibit 14-13, the following results are obtained:
𝑀𝑆 = 0.321 + 0.0039𝑒 (𝑣𝑅23 /1,000) − 0.002(𝐿𝐴 × 𝑆𝐹𝑅 × 𝑆𝐴𝐹/1,000)
𝑀𝑆 = 0.321 + 0.0039𝑒 (3,777/1,000) − 0.002(820 × 30 × 1.00/1,000) = 0.443
𝑆𝑅 = (65)(1.00) − (65 × 1.00 − 42)(0.443) = 54.8 mi/h

𝑆𝑂 = 𝐹𝐹𝑆 × 𝑆𝐴𝐹 − 0.0036(𝑣𝑂𝐴 − 500)

𝑆𝑂 = (65)(1.00) − 0.0036(1,568 − 500) = 61.2 mi/h
𝑣23 + 𝑣𝑂𝐴 𝑁𝑂 3,777 + (1,568)(1)
𝑆= = = 56.5 mi/h
𝑣 𝑉 𝑁𝑂 3,777 1,568 × 1
( 𝑆23 ) + ( 𝑂𝐴
𝑆𝑂 ) ( ) + ( )
𝑅 54.8 61.2
While traffic in the outer lane is predicted to travel somewhat faster than
traffic in the lanes in the ramp influence area (which includes the acceleration
lane), the approximate nature of the speed result for left-hand ramps makes it
difficult to draw any firm conclusions concerning speed behavior.
Discussion
This example problem is typical of the way the situations in the Special Cases
section of Chapter 14 are treated. Modifications as specified are applied to the
standard algorithms used for single-lane, right-hand ramp junctions. In this case,
operations are acceptable, but in LOS D—though not far from the LOS C
boundary. Because the left-hand lanes are expected to carry freeway traffic
flowing faster than right-hand lanes, right-hand ramps are normally preferable
to left-hand ramps when they can be provided without great difficulty.
EXAMPLE PROBLEM 5: SERVICE FLOW RATES AND SERVICE VOLUMES

FOR AN ISOLATED ON-RAMP ON A SIX-LANE FREEWAY
The Facts
The following information is available concerning this example problem. The
example assumes no impacts of inclement weather or incidents.
1. Single-lane, right-hand on-ramp with an FFS of 40 mi/h;
2. Six-lane freeway (three lanes in each direction) with an FFS of 70 mi/h;
3. Level terrain for freeway and ramp;
4. 6.5% trucks on both freeway and ramp segments;
5. Peak hour factor = 0.87;
6. Drivers are regular users of the facility; and
7. Acceleration lane = 1,000 ft.
Comments
This example illustrates the computation of service flow rates and service
volumes for a ramp–freeway junction. The case selected is relatively
straightforward to avoid extraneous complications that have been addressed in
other example problems.
Two approaches will be demonstrated:
1. The ramp demand flow rate will be stated as a fixed percentage of the
arriving freeway flow rate. The service flow rates and service volumes are
expressed as arriving freeway flow rates that result in the threshold
densities within the ramp influence area that define the limits of the
various levels of service. For this computation, the ramp flow is set at 10%
of the approaching freeway flow rate.

2. A fixed freeway demand flow rate will be stated, with service flow rates
and service volumes expressed as ramp demand flow rates that result in
the threshold densities within the ramp influence area that define the
limits of the various levels of service. For this computation, the
approaching freeway flow rate is set at 4,000 veh/h.
For LOS E, density does not define the limiting value of service flow rate,
which is analogous to capacity for ramp–freeway junctions. It is defined as the
flow that results in capacity being reached on the downstream freeway segment
or ramp roadway.
Since all algorithms in this methodology are calibrated for passenger cars per
hour under equivalent ideal conditions, initial computations are made in those
terms. Results are then converted to service flow rates by using the appropriate
heavy vehicle and driver population adjustment factors. Service flow rates are
then converted to service volumes by multiplying by the peak hour factor.
From Exhibit 14-3, the following densities define the limits of LOS A–D:
LOS A: 10 pc/mi/ln
LOS B: 20 pc/mi/ln
LOS C: 28 pc/mi/ln
LOS D: 35 pc/mi/ln
From Exhibit 14-10 and Exhibit 14-12, capacity (or the threshold for LOS E)
occurs when the downstream freeway flow rate reaches 7,200 pc/h (FFS = 70
mi/h) or when the ramp flow rate reaches 2,000 pc/h (ramp FFS = 40 mi/h).
Case 1: Ramp Demand Flow Rate = 0.10 × Freeway Demand Flow Rate
Equation 14-22 defines the density in an on-ramp influence area as follows:
𝐷𝑅 = 5.475 + 0.00734𝑣𝑅 + 0.0078𝑣12 − 0.00627𝐿𝐴
In this case
vR = 0.10 vF
LA = 1,000 ft
Equation 14-22 and Exhibit 14-8 give the following:
𝑃𝐹𝑀 = 0.5775 + 0.000028𝐿𝐴 = 0.5775 + 0.000028(1,000) = 0.6055
𝑣12 = 0.6055𝑣𝐹
Substitution of these values into Equation 14-22 gives
𝐷𝑅 = 5.475 + 0.00734(0.10𝑣𝐹 ) + 0.0078(0.6055𝑣𝐹 ) − 0.00627(1,000)
𝐷𝑅 = 5.475 + 0.000734𝑣𝐹 + 0.00472𝑣𝐹 − 6.27
𝐷𝑅 = 0.005454𝑣𝐹 − 0.795
𝐷𝑅 + 0.795
𝑣𝐹 =
0.005454
This equation can now be solved for threshold values of vF for LOS A
through D by using the appropriate threshold values of density. The results will
be in terms of service flow rates under equivalent ideal conditions:

10 + 0.795
𝑣𝐹 (LOS A) = = 1,979 pc/h
0.005454
20 + 0.795
𝑣𝐹 (LOS B) = = 3,813 pc/h
0.005454
28 + 0.795
𝑣𝐹 (LOS C) = = 5,280 pc/h
0.005454
35 + 0.795
𝑣𝐹 (LOS D) = = 6,563 pc/h
0.005454
At capacity, the limiting flow rate occurs when the downstream freeway
segment is 7,200 pc/h. If the ramp flow rate is 0.10 of the approaching freeway
flow rate, then
𝑣𝐹𝑂 = 7,200 = 𝑣𝐹 + 0.10𝑣𝐹 = 1.10𝑣𝐹
7,200
𝑣𝐹(LOS E) = = 6,545 pc/h
1.10
This must be checked to ensure that the ramp flow rate (0.10 × 6,545 = 655
pc/h) does not exceed the ramp capacity of 2,000 pc/h. Since it does not, the
computation stands.
However, the LOS E (capacity) threshold is lower than the LOS D threshold.
This indicates that LOS D operation can be achieved at this location; however,
the service flow rate for LOS D cannot be achieved. Before densities reach the 35-
pc/h/ln threshold for LOS D, the capacity of the merge junction has been reached.
Thus, there is no service flow rate or service volume for LOS D.
The computed values are in terms of passenger cars per hour under
equivalent ideal conditions. To convert them to service flow rates in vehicles per
hour under prevailing conditions, they must be multiplied by the heavy vehicle
adjustment factor and the driver population factor. The approaching freeway
flow includes 6.5% trucks on both the ramp and the mainline. For level terrain
(Chapter 12, Basic Freeway and Multilane Highway Segments), ET = 2.0. Then
1
𝑓𝐻𝑉 =
1 + 𝑃𝑇 (𝐸𝑇 − 1)
1
𝑓𝐻𝑉 = = 0.939
1 + 0.065(2 − 1)
Service volumes are obtained by multiplying service flow rates by the
specified PHF, 0.87. These computations are illustrated in Exhibit 28-4.
Service Flow Rate, Service Flow Rate, Service Volume Exhibit 28-4
Ideal Conditions Prevailing Conditions (SF) (SV) Example Problem 5:
LOS (pc/h) (veh/h) (veh/h) Illustrative Service Flow Rates
A 1,979 1,979 × 0.939 × 1 = 1,858 1,858 × 0.87 = 1,616 and Service Volumes Based
B 3,813 3,813 × 0.939 × 1 = 3,580 3,580 × 0.87 = 3,115 on Approaching Freeway
C 5,280 5,280 × 0.939 × 1 = 4,958 4,958 × 0.87 = 4,313 Demand
D NA NA NA
E 6,545 6,545 × 0.939 × 1 = 6,146 6,146 × 0.87 = 5,347
The service flow rates and service volumes shown in Exhibit 28-4 are stated
in terms of the approaching hourly freeway demand.

Case 2: Approaching Freeway Demand Volume = 4,000 veh/h

In this case, the approaching freeway demand will be held constant, and
service flow rates and service volumes will be stated in terms of the ramp
demand that can be accommodated at each LOS.
Since the freeway demand is stated in terms of an hourly volume in mixed
vehicles per hour, it will be converted to passenger cars per hour under
equivalent ideal conditions for use in the algorithms of this methodology:
𝑉𝐹 4,000
𝑣𝐹 = = = 4,896 pc/h
𝑃𝐻𝐹 × 𝑓𝐻𝑉 0.87 × 0.939
The density is estimated by using Equation 14-22, and the variable PFM—
which is not dependent on vR—remains 0.6055 as in Case 1. With a fixed value of
freeway demand,
𝑣12 = 0.6055 × 4,896 = 2,965 pc/h
Then, by using Equation 14-22,
𝐷𝑅 = 5.475 + 0.00734𝑣𝑅 + 0.0078𝑣12 − 0.00627𝐿𝐴
𝐷𝑅 = 5.475 + 0.00734𝑣𝑅 + 0.0078(2,965) − 0.00627(1,000)
𝐷𝑅 = 22.33 + 0.00734𝑣𝑅
𝐷𝑅 − 22.33
𝑣𝑅 =
0.00734
It is clear from this equation that neither LOS A (DR = 10 pc/mi/ln) nor LOS B
(DR = 20 pc/mi/ln) can be achieved with a freeway demand flow of 4,896 pc/h.
For LOS C and D,
28 − 22.33
𝑣𝑅 (LOS C) = = 772 pc/h
0.00734
35 − 22.33
𝑣𝑅 (LOS D) = = 1,726 pc/h
0.00734
Capacity, the limit of LOS E, occurs when the downstream freeway flow
reaches 7,200 pc/h. With a fixed freeway demand,
𝑣𝐹𝑂 = 7,200 − 4,896 + 𝑣𝑅
𝑣𝑅 (LOS E) = 7,900 − 4,896 = 3,004 pc/h
This, however, violates the capacity of the ramp roadway, which is 2,000
pc/h. Thus, the limiting ramp flow rate for LOS E is set at 2,000 pc/h.
As in Case 1, these values are all stated in terms of passenger cars per hour
under equivalent ideal conditions. They are converted to service flow rates by
multiplying by the appropriate heavy vehicle factor (0.939 from Case 1). Service
flow rates are converted to service volumes by multiplying by the PHF. These
computations for ramp service volumes are illustrated in Exhibit 28-5.

Service Flow Rate, Service Flow Rate, Exhibit 28-5

Ideal Conditions Prevailing Conditions Ramp Service Volume Example Problem 5:
LOS (pc/h) (veh/h) (veh/h) Illustrative Service Flow Rates
A NA NA NA and Service Volumes Based
B NA NA NA on a Fixed Freeway Demand
C 769 772 × 0.939 × 1 = 725 725 × 0.87 = 631
D 1,723 1,726 × 0.939 × 1 = 1,621 1,621 × 0.87 = 1,410
E 2,000 2,000 × 0.939 × 1 = 1,878 1,878 × 0.87 = 1,633
These service flow rates and service volumes are based on a constant
upstream arriving freeway demand and are stated in terms of limiting on-ramp
demands for that condition.
Discussion
As this illustration shows, many considerations are involved in estimating
service flow rates and service volumes for ramp–freeway junctions, not the least
of which is specifying how such values should be defined. The concept of service
flow rates and service volumes at specific ramp–freeway junctions is of limited
utility. Since many of the details that affect the estimates will not be determined
until final designs are prepared, operational analysis of the proposed design may
be more appropriate.
Case 2 could have applications in considering how to time ramp meters.
Appropriate limiting ramp flows can be estimated by using the same approach
as for service volumes and service flow rates.

3. ALTERNATIVE TOOL EXAMPLES FOR FREEWAY RAMPS
Chapter 14, Freeway Merge and Diverge Segments, described a methodology

for analyzing ramps and ramp junctions to estimate capacity, speed, and density
as a function of traffic demand and geometric configuration. This chapter
includes two supplemental problems that examine situations that are beyond the
scope of the Chapter 14 methodology. A typical microsimulation-based tool is
used for this purpose, and the simulation results are compared, where
appropriate, with those of the HCM.
Both problems are based on this chapter’s Example Problem 3, which
analyzes an eight-lane freeway segment with an entrance and an exit ramp. The
first problem evaluates the effects of the addition of ramp metering, while the
second evaluates the impacts of converting the leftmost lane of the mainline into
a high-occupancy vehicle (HOV) lane.
The need to determine performance measures based on the analysis of
vehicle trajectories was emphasized in Chapter 7, Interpreting HCM and
Alternative Tool Results. Specific procedures for defining measures in terms of
vehicle trajectories were proposed to guide the future development of alternative
tools. Pending further development, the examples presented in this chapter have
applied existing versions of alternative tools and therefore do not reflect the
trajectory-based measures described in Chapter 7.
For purposes of illustration, the default calibration parameters of the
simulation tool (e.g., lane-changing behavioral characteristics) were applied to
these examples. However, most simulation tools offer the ability to adjust these
parameters. The parameter values can have a significant effect on the results,
especially when the operation is close to full saturation.
PROBLEM 1: RAMP-METERING EFFECTS

This problem analyzes the impacts of ramp metering along the segment. The
HCM procedure for ramp-merge junctions cannot estimate the impacts of ramp
metering. These impacts can be approximated to some extent by not allowing the
ramp demand to exceed the ramp-metering rate. To address ramp metering at a
more detailed level, a typical microsimulation tool was used to evaluate the
impacts of ramp metering on the density and capacity of the merge.
The subject segment consists of an on-ramp followed by an off-ramp,
separated by 1,300 ft. The upstream segment is 1 mi long. Each simulation run
was for 1 full hour. It was assumed that the mainline demand was 6,111 veh/h
and that the ramp demand was 444 veh/h. The ramp metering is clock-time
based (i.e., the metering rate does not change as a function of the mainline
demand).
Experiments were conducted to obtain the density and capacity of the subject
segment as a function of the ramp-metering rate. The queue length upstream of
the ramp meter was also obtained as a function of the ramp-metering rate.
Exhibit 28-6 provides a graphics capture of the simulated site.
Alternative Tool Examples for Freeway Ramps Chapter 28/Freeway Merges and Diverges: Supplemental
Exhibit 28-6
Graphics Capture of the Ramp
Merge with Ramp Metering
Exhibit 28-7 provides the density of the segment between the on-ramp and
the off-ramp as a function of the ramp-metering rate (or discharge headway from
the on-ramp). As shown, the density is not much affected by the ramp-metering
rate. As expected, the density of Lane 1 (the rightmost lane) is the highest, while
the density in Lane 4 is the lowest.
50 Exhibit 28-7
Density as a Function of
45
Ramp-Metering Headways
40
Density (veh/mi/ln)
35
30
25
20
15
10
0
4 6 8 10 12
Ramp-Metering Headway (s)
Lane 1 (Connecting Ramp) Lane 2 Lane 3 Lane 4 Link Density
Exhibit 28-8 provides capacity as a function of the ramp-metering headway

and when no ramp metering is implemented. As shown, the simulation model
predicts that capacity is higher when ramp metering is implemented. Capacity in
simulation is typically measured in the form of maximum throughput
downstream of a queued segment and is therefore one of the outputs of the
simulation, as opposed to an input as in the HCM.
Chapter 28/Freeway Merges and Diverges: Supplemental Alternative Tool Examples for Freeway Ramps
Exhibit 28-8 9,500

Capacity at a Ramp Junction
as a Function of Ramp-
Metering Headways 9,000
Capacity (veh/h)
8,500
8,000
7,500
7,000
6,500
4 6 8 10 12
With Ramp Metering Without Ramp Metering
Exhibit 28-9 provides the queue length expected on the ramp as a function of
the ramp-metering headway and when no ramp metering is implemented. As
expected, the queue length is higher when ramp metering is implemented, and it
increases dramatically when the ramp-metering rate exceeds 8 s/veh. The reason
for this increase is that the demand on the ramp is approximately 8 s/veh (444
veh/h corresponds to an average headway of 8.1 s/veh).
Exhibit 28-9 250

Queue Length on the Ramp as
a Function of Ramp-Metering
Headways
200
Ramp Queue Length (veh)
150
100
50
0
4 6 8 10 12
With Ramp Metering Without Ramp Metering
As indicated above, the effects of ramp metering cannot be evaluated with

the HCM. The freeway facilities methodology (HCM Chapter 10) can handle
changes in segment capacity; however, other tools are required to estimate what
the maximum throughput would be under various types of ramp-metering
algorithms and rates. Also, the HCM cannot estimate the queue length on the on-
ramp as a function of ramp metering. An analytical method could be developed
to estimate queue length as a function of demand and service rate at the meter.
PROBLEM 2: CONVERSION OF LEFTMOST LANE TO AN HOV LANE

This problem is also based on this chapter’s Example Problem 3. It evaluates
operating conditions when the leftmost lane of the mainline is converted into an
HOV lane. Exhibit 28-10 provides a graphics capture of the segment.
Exhibit 28-10
Graphics Capture of the
Segment with an HOV Lane
Exhibit 28-11 and Exhibit 28-12 show the density and capacity of the ramp
junction as a function of the percentage of carpools. As shown, when the
percentage of carpools increases, the density of the HOV lane and the overall link
capacity increase. This occurs because for the range of values tested here, the
utilization of the HOV lane increases, which improves the overall link
performance.
50 Exhibit 28-11
Density of a Ramp Junction as
45 a Function of the Carpool
Percentage
40
Density (veh/mi/ln)
35
30
25
20
15
10
0
0 10 20 30
Carpool Percentage (%)
Link Density Average Density of 3 Non-HOV Lanes HOV Lane Density
Exhibit 28-12 9,500

Capacity of a Ramp Junction
as a Function of the Carpool
Percentage
9,000
Capacity (veh/h)
8,500
8,000
7,500
7,000
6,500
0 10 20 30
Carpool Percentage (%)
Exhibit 28-13 presents the density as a function of HOV violators, while

Exhibit 28-14 presents the corresponding capacity. These two graphs assume that
there are 10% carpools in the traffic stream. As shown, density generally
decreases while capacity increases as the percentage of HOV violators increases.
The reason is that under this scenario, the facility is more efficiently utilized as
violations increase with general traffic using the HOV lane.
Exhibit 28-13 50
a Function of the HOV 45
Violation Percentage
40
Density (veh/mi/ln)
35
30
25
20
15
10
0
1 3 5 8 10
HOV Violation Percentage (%)

9,500 Exhibit 28-14

as a Function of the HOV
Violation Percentage
9,000
Capacity (veh/h)
8,500
8,000
7,500
7,000
6,500
1 2 3 4 5 6 7 8 9 10
HOV Violation Percentage (%)
Exhibit 28-15 and Exhibit 28-16 present the density and capacity of the ramp
junction as a function of the distance at which drivers begin to react to the
presence of the HOV lane (i.e., the distance to the regulatory sign). As shown, the
longer that distance, the lower the density of the HOV lane and the higher the
density in the other lanes. The reason is that under this scenario the percentage of
carpools is relatively low (10%). When the HOV lane begins, non-HOVs
congregate in the remaining lanes. Capacity is reduced as the distance at which
drivers begin to react increases, because the HOV lane is not utilized as much
when drivers are given early warning to switch lanes.
50 Exhibit 28-15
45 a Function of the Distance at
Which Drivers Begin to React
40
Density (veh/mi/ln)
35
30
25
20
15
10
0
0.0 0.5 1.0
Distance at Which Drivers Begin to React (mi)
Exhibit 28-16 9,500

as a Function of the Distance
at Which Drivers Begin to
9,000
React
Capacity (veh/h)
8,500
8,000
7,500
7,000
6,500
0.0 0.5 1.0
Distance at Which Drivers Begin to React (mi)
Exhibit 28-17 and Exhibit 28-18 present the density and capacity of the ramp
junction as a function of the percentage of HOV usage. As expected, when usage
of the HOV lane increases, the density of the HOV lane and the overall link
capacity increase.
Exhibit 28-17 50
45
a Function of the Percentage
of HOV Usage
40
35
Density (veh/mi/ln)
30
25
20
15
10
0
30 40 50 60 70 80 90 100
Percentage of HOV Usage (%)
9,500 Exhibit 28-18

as a Function of the
Percentage of HOV Usage
9,000
Capacity (veh/h)
8,500
8,000
7,500
7,000
6,500
30 40 50 60 70 80 90 100
Percentage of HOV Usage (%)
The type of analysis presented in this example cannot be conducted with the
HCM, since the method does not estimate the HOV lane density separately.
Variables such as the impact of the distance of the HOV regulatory sign cannot
be evaluated, since they pertain to driver behavior attributes and their impact on
density and capacity. The impact of the percentage of carpools and the
percentage of violators could perhaps be estimated with appropriate
modifications of the existing HCM method.
HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 29
URBAN STREET FACILITIES: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 29-1
2. SCENARIO GENERATION PROCEDURE ...................................................... 29-2

Weather Event Generation ................................................................................ 29-2
Traffic Demand Variation Generation ............................................................. 29-7
Traffic Incident Generation ............................................................................... 29-8
Scenario Dataset Generation ............................................................................29-16
3. SUSTAINED SPILLBACK PROCEDURE ....................................................... 29-25

Overview of the Procedure ..............................................................................29-25
Computational Steps .........................................................................................29-26
Procedure for Saving Performance Measures ...............................................29-31
Computational Engine Documentation ..........................................................29-33
4. USE OF ALTERNATIVE TOOLS ..................................................................... 29-36

Basic Example Problem Configuration ...........................................................29-36
Signal Timing Plan Design ...............................................................................29-38
Demonstration of Alternative Tool Applications ..........................................29-50
5. EXAMPLE PROBLEMS....................................................................................... 29-60

Example Problem 1: Automobile-Oriented Urban Street ............................29-60
Example Problem 2: Pedestrian and Bicycle Improvements .......................29-68
Example Problem 3: Pedestrian and Parking Improvements ......................29-73
Example Problem 4: Existing Urban Street Reliability .................................29-78
Example Problem 5: Urban Street Strategy Evaluation ................................29-95
6. REFERENCES ..................................................................................................... 29-100
Chapter 29/Urban Street Facilities: Supplemental Contents

LIST OF EXHIBITS
Exhibit 29-1 Weather Event Procedure ....................................................................29-3

Exhibit 29-2 Traffic Demand Variation Procedure .................................................29-8
Exhibit 29-3 Traffic Incident Procedure for Intersection Incidents ......................29-9
Exhibit 29-4 Scenario File Generation Procedure .................................................29-17
Exhibit 29-5 Additional Critical Left-Turn Headway due to Weather ..............29-24
Exhibit 29-6 Spillback Procedure Flowchart .........................................................29-34
Exhibit 29-7 Sustained Spillback Module Routines ..............................................29-35
Exhibit 29-8 Base Configuration for the Examples ...............................................29-37
Exhibit 29-9 Demand Flow Rates and Phasing Plan for Each Intersection .......29-37
Exhibit 29-10 Elements of a Typical Signal Timing Design Tool ........................29-39
Exhibit 29-11 Cycle Length Optimization Results ................................................29-41
Exhibit 29-12 Timing Plan Developed by Split and Offset Optimization .........29-42
Exhibit 29-13 Performance Measures for the Initial Timing Plan.......................29-42
Exhibit 29-14 Progression Quality Measures for the Initial Design ...................29-43
Exhibit 29-15 Progression Quality Measures for the Improved
Progression Design ............................................................................................29-43
Exhibit 29-16 Time–Space Diagram for the Initial Design ..................................29-44
Exhibit 29-17 Time–Space Diagram for the Modified Progression Design .......29-44
Exhibit 29-18 Offset Changes for the Modified Progression Design .................29-44
Exhibit 29-19 Alternative Time–Space Diagram Format .....................................29-45
Exhibit 29-20 Example Illustrating the Use of Flow Profiles...............................29-46
Exhibit 29-21 Composite Flow Profiles for the First Eastbound Segment .........29-47
Exhibit 29-22 Variation of Queue Length Throughout the Signal Cycle for
the First Eastbound Segment ...........................................................................29-47
Exhibit 29-23 Time–Space Diagram with Flows and Queues .............................29-48
Exhibit 29-24 Optimized Phasing Modifications ..................................................29-49
Exhibit 29-25 Time–Space Diagram for the Optimized Phasing Plan ...............29-49
Exhibit 29-26 Time–Space Diagram Showing Ideal Eastbound
Progression .........................................................................................................29-50
Exhibit 29-27 Parameters for the Parking Example ..............................................29-51
Exhibit 29-28 Effect of Parking Activity Level on Travel Time and Delay .......29-51
Exhibit 29-29 Effect of Parking Activity Level on the Percentage of Stops .......29-52
Exhibit 29-30 Roundabout Configuration for Intersection 3 ...............................29-53
Exhibit 29-31 Time–Space Diagrams Showing Simultaneous and
Alternating Platoon Arrivals at the Roundabout ..........................................29-53
Contents Chapter 29/Urban Street Facilities: Supplemental

Exhibit 29-32 Performance Comparison for Simultaneous and

Alternating Platoon Arrivals at a Roundabout ............................................. 29-54
Exhibit 29-33 Queuing Results for the Theoretical Example .............................. 29-56
Exhibit 29-34 Queuing Results for Simultaneous Platoons ................................. 29-56
Exhibit 29-35 Queuing Results for Alternating Platoons .................................... 29-57
Exhibit 29-36 Queuing Results for Isolated TWSC Operation ............................ 29-58
Exhibit 29-37 Effect of Cross-Street Demand Volume on Queue Backup
Beyond 100 ft from the Stop Line .................................................................... 29-59
Exhibit 29-38 Example Problems ............................................................................ 29-60
Exhibit 29-39 Example Problem 1: Urban Street Schematic ................................ 29-60
Exhibit 29-40 Example Problem 1: Segment Geometry ....................................... 29-61
Exhibit 29-41 Example Problem 1: Intersection Turn Movement Counts ......... 29-61
Exhibit 29-42 Example Problem 1: Signal Conditions for Intersection 1 ........... 29-62
Exhibit 29-43 Example Problem 1: Geometric Conditions and Traffic
Characteristics for Signalized Intersection 1 .................................................. 29-63
Exhibit 29-44 Example Problem 1: Access Point Data ......................................... 29-63
Exhibit 29-45 Example Problem 1: Intersection 1 Evaluation ............................. 29-64
Exhibit 29-47 Example Problem 1: Segment 1 Evaluation ................................... 29-66
Exhibit 29-49 Example Problem 1: Facility Evaluation ........................................ 29-67
Exhibit 29-62 Example Problem 4: Urban Street Facility ..................................... 29-79
Exhibit 29-63 Example Problem 4: Input Data Needs and Sources ................... 29-80
Exhibit 29-64 Example Problem 4: Intersection 1 Signal Timing Data .............. 29-80
Exhibit 29-65 Example Problem 4: Sample Weather Data for Lincoln,
Nebraska ............................................................................................................. 29-83
Chapter 29/Urban Street Facilities: Supplemental Contents

Exhibit 29-66 Example Problem 4: Sample Generated Weather Events ............29-84

Exhibit 29-67 Example Problem 4: Sample Demand Profile Calculations ........29-86
Exhibit 29-68 Example Problem 4: Locally Available Crash Frequency
Data ......................................................................................................................29-87
Exhibit 29-69 Example Problem 4: Computation of Crash Frequency by
Weather Type .....................................................................................................29-88
Exhibit 29-70 Example Problem 4: Incident Determination for April 6,
9:00 a.m., for Segment 1-2 .................................................................................29-90
Exhibit 29-71 Example Problem 4: Incident Determination for January 10,
7:00 a.m., for Segment 1-2 .................................................................................29-90
Exhibit 29-72 Example Problem 4: Sample Calculation of Incident
Duration ..............................................................................................................29-91
Exhibit 29-73 Example Problem 4: Reliability Performance Measure
Results .................................................................................................................29-93
Exhibit 29-74 Example Problem 4: Eastbound Travel Time Distribution .........29-94
Exhibit 29-75 Example Problem 4: Confidence Interval Calculation for
Eastbound Direction ..........................................................................................29-94
Exhibit 29-76 Example Problem 4: Annual VHD by Cause ................................29-95
Exhibit 29-77 Example Problem 4: Percentage of Annual VHD by Cause ........29-95
Exhibit 29-78 Example Problem 5: Results for Strategy 1 ....................................29-98
Contents Chapter 29/Urban Street Facilities: Supplemental

1. INTRODUCTION
Chapter 29 is the supplemental chapter for Chapter 16: Urban Street Facilities VOLUME 4: APPLICATIONS
GUIDE
and Chapter 17: Urban Street Reliability and ATDM, which are found in 25. Freeway Facilities:
Volume 3 of the Highway Capacity Manual (HCM). This chapter presents detailed Supplemental
information about the following aspects of urban street facility evaluation: Segments: Supplemental
• The process for generating the scenarios used to evaluate travel time Supplemental
reliability and 28. Freeway Merges and
• The process for evaluating facilities with sustained spillback. 29. Urban Street Facilities:
Supplemental
This chapter also provides details about the computational engine that 30. Urban Street Segments:
Supplemental
implements the sustained spillback procedure and example applications of 31. Signalized Intersections:
alternative tools. Finally, the chapter provides five example problems that Supplemental
32. STOP-Controlled
demonstrate the application of the methodologies to a multimodal evaluation of Intersections:
urban street performance and to the evaluation of urban street reliability. Supplemental
33. Roundabouts:
Supplemental
Supplemental
Chapter 29/Urban Street Facilities: Supplemental Introduction

2. SCENARIO GENERATION PROCEDURE
The methodology for evaluating reliability is described in Section 3 of

Chapter 17, Urban Street Reliability and ATDM. It consists of three stages that
are implemented in the sequence listed below:
• Scenario generation,
• Facility evaluation, and
• Performance summary.
The scenario generation stage is implemented through four sequential
procedures: (a) weather event generation, (b) traffic demand variation generation,
(c) traffic incident generation, and (d) scenario dataset generation. This stage
generates the set of analysis periods that make up the reliability reporting period.
The sequence of computations associated with each procedure is described in
this section.
Details of the facility evaluation stage and the performance summary stage
are provided in Section 3 of Chapter 17.
The combination of demand volume, speed, saturation flow rate, and signal
timing established for each analysis period is assumed to be unique, relative to
the other analysis periods. This assumption recognizes that it is extremely rare in
the urban street environment for two or more analysis periods to have the same
combination of demand volume, capacity, and traffic control for all segments
and intersections making up the facility. Thus, each analysis period is considered
to be one scenario.
WEATHER EVENT GENERATION

The weather event procedure is used to predict weather events that could
occur during the reliability reporting period. The events predicted include
rainfall and snowfall. The time following each event that the pavement remains
wet or covered by snow or ice is also predicted. The presence of these conditions
has been found to influence running speed and intersection saturation flow rate.
The sequence of calculations in the weather event procedure is shown in
Exhibit 29-1. The calculations proceed on a day-by-day basis in chronological
order. If a day is determined to have a weather event, its start time and duration
are recorded for later use in the traffic incident procedure. Thereafter, each
analysis period is evaluated in chronological order for any given day with a
weather event. If the analysis period is associated with a weather event, the event
type (i.e., rain or snow), precipitation rate (i.e., intensity), and pavement status
(i.e., wet or snow covered) are recorded for later use in the scenario file
generation procedure.
The weather event procedure consists of eight calculation steps. The
calculations associated with each step are described in the following paragraphs.
A random number is used in several of the steps. All random numbers have a
real value that is uniformly distributed from 0.0 to 1.0.
Scenario Generation Procedure Chapter 29/Urban Street Facilities: Supplemental

Exhibit 29-1
Weather Event Procedure
Step 1: Precipitation Prediction

The probability of precipitation for any given day is computed by using the
following equation:
𝑁𝑑𝑝𝑚
𝑃(𝑝𝑟𝑒𝑐𝑖𝑝)𝑚 = Equation 29-1
𝑁𝑑𝑚
where
P(precip)m = probability of precipitation in any given day of month m,
Ndpm = number of days with precipitation of 0.01 in. or more in month m (d),
and
Ndm = total number of days in month m (d).
For each day considered in month m, the following rule is checked to
determine whether precipitation occurs:
No precipitation if 𝑅𝑝𝑑,𝑚 ≥ 𝑃(𝑝𝑟𝑒𝑐𝑖𝑝)𝑚
Equation 29-2
Precipitation if 𝑅𝑝𝑑,𝑚 < 𝑃(𝑝𝑟𝑒𝑐𝑖𝑝)𝑚
where Rpd,m is the random number for precipitation for day d of month m. The
rule is applied to each day (on a monthly basis) in the reliability reporting
period.
Step 2: Precipitation Type

If precipitation occurs, the following equation is used to estimate the average
temperature during the weather event for the subject day (1):
𝑇𝑑,𝑚 = normal−1 (𝑝 = 𝑅𝑔𝑑 , 𝜇 = 𝑇̅𝑚 , 𝜎 = 𝑠𝑇 ) Equation 29-3
Chapter 29/Urban Street Facilities: Supplemental Scenario Generation Procedure

where
Td,m = average temperature for day d of month m (˚F),
Rgd = random number for temperature for day d,
—
Tm = normal daily mean temperature in month m (˚F),
sT = standard deviation of daily mean temperature in a month
(= 5.0) (˚F), and
normal (p, μ, σ) = value associated with probability p for a cumulative normal
–1
distribution with mean μ and standard deviation σ.

The average temperature for the day is used to determine whether the
precipitation is in the form of rain or snow. The following rule is checked to
determine the form of the precipitation for that day:
Equation 29-4 Rain if 𝑇𝑑,𝑚 ≥ 32℉
Snow if 𝑇𝑑,𝑚 < 32℉
Step 3: Rain Intensity

The following equation is used to estimate the rainfall rate during a rain
event:
Equation 29-5 𝑟𝑟𝑑,𝑚 = gamma−1 (𝑝 = 𝑅𝑟𝑑 , 𝜇 = 𝑟𝑟
̅̅̅𝑚 , 𝜎 = 𝑠𝜋,𝑚 )
where
rrd,m = rainfall rate for the rain event occurring on day d of month m
(in./h),
Rrd = random number for rainfall rate for day d,
̅̅̅𝑚 = precipitation rate in month m (in./h),
𝑟𝑟
srr,m = standard deviation of precipitation rate in month m (= 1.0 𝑟𝑟
̅̅̅𝑚 )
(in./h), and
gamma–1(p, μ, σ) = value associated with probability p for a cumulative gamma
The average precipitation rate (and its standard deviation) is based on time
periods when precipitation is falling. Thus, the average precipitation rate
represents an average for all hours for which precipitation is falling (and
excluding any hours when precipitation is not falling).
The following equation is used to estimate the total amount of rainfall for a
rain event. Each day with precipitation is assumed to have one rain event.
Equation 29-6 𝑡𝑟𝑑,𝑚 = gamma−1(𝑝 = 𝑅𝑡𝑑 , 𝜇 = 𝑡𝑟
̅𝑚 , 𝜎 = 𝑠𝑡𝑟,𝑚 )
with
𝑡𝑝𝑚
Equation 29-7 ̅𝑚 =
𝑡𝑟
𝑁𝑑𝑝𝑚
Equation 29-8 ̅𝑚 , 0.65)
𝑠𝑡𝑟,𝑚 = min(2.5 𝑡𝑟

where
trd,m = total rainfall for the rain event occurring on day d of month m
(in./event),
Rtd = random number for rainfall total for day d (= Rrd),
—
trm = average total rainfall per event in month m (in./event),
str,m = standard deviation of total rainfall in month m (in./event),
tpm = total normal precipitation for month m (in.), and
Ndpm = number of days with precipitation of 0.01 in. or more in month m (d).
Total rainfall for a rain event is the product of rainfall rate and rain event
duration. Thus, the total rainfall amount is highly correlated with the rainfall
rate. For reliability evaluation, total rainfall is assumed to be perfectly correlated
with rainfall rate such that they share the same random number. This approach
may result in slightly less variability in the estimated total rainfall; however, it
precludes the occasional calculation of unrealistically long or short rain events.
Step 4: Rainfall Duration

The following equation is used to estimate the rainfall duration for a rain
event:
𝑡𝑟𝑑,𝑚
𝑑𝑟𝑑,𝑚 = Equation 29-9
𝑟𝑟𝑑,𝑚
where
drd,m = rainfall duration for the rain event occurring on day d of month m
(h/event),
trd,m = total rainfall for the rain event occurring on day d of month m
(in./event), and
rrd,m = rainfall rate for the rain event occurring on day d of month m (in./h).
The duration computed with Equation 29-9 is used in a subsequent step to
determine whether an analysis period is associated with a rain event. To simplify
the analytics in this subsequent step, it is assumed that no rain event extends
beyond midnight. To ensure this outcome, the duration computed from Equation
29-9 is compared with the duration between the start of the study period and
midnight. The rainfall duration is then set to equal the smaller of these two
values.
Step 5: Start Time of Weather Event

The hour of the day that the rain event starts is determined randomly. The
start hour is computed with the following equation:
𝑡𝑠𝑑,𝑚 = (24 − 𝑑𝑟𝑑,𝑚 )𝑅𝑠,𝑑 Equation 29-10
where
tsd,m = start of rain event on day d of month m (h),
24 = number of hours in a day (h/day),

(h/event), and
Rs,d = random number for rain event start time for day d.
The start time from Equation 29-10 is rounded to the nearest hour for 1-h
analysis periods or to the nearest quarter hour for 15-min analysis periods.
Step 6: Wet Pavement Duration

After a rain event, the pavement typically remains wet for some length of
time. The presence of wet pavement can influence road safety by reducing
surface–tire friction. Research (1) indicates that wet pavement time can be
computed with the following equation:
Equation 29-11 𝑑𝑤𝑑,𝑚 = 𝑑𝑟𝑑,𝑚 + 𝑑𝑜𝑑,𝑚 + 𝑑𝑑𝑑,𝑚
with
Equation 29-12 𝑑𝑑𝑑,𝑚 = 0.888 exp(−0.0070 𝑇𝑑,𝑚 ) + 0.19 𝐼night
where
dwd,m = duration of wet pavement for rain event occurring on day d of
month m (h/event),
(h/event),
dod,m = duration of pavement runoff for rain event occurring on day d of
month m (= 0.083) (h/event),
Td,m = average temperature for day d of month m (˚F),
Inight = indicator variable for night (= 0.0 if rain starts between 6:00 a.m. and
6:00 p.m., 1.0 otherwise), and
ddd,m = duration of drying time for rain event occurring on day d of month m
(h/event).
determine whether an analysis period is associated with wet pavement
conditions. To simplify the analytics in this subsequent step, it is assumed that no
rain event extends beyond midnight. To ensure this outcome, the duration
computed from Equation 29-11 is compared with the duration between the start
of the rain event and midnight. The wet pavement duration is then set to equal
the smaller of these two values.
Step 7: Snow Intensity and Duration

The snowfall rate (i.e., intensity) and duration are computed by using the
calculation sequence in Steps 3 to 6. The equations are the same. The average
snowfall rate and average snow total per event are computed by multiplying the
average precipitation rate and average total rainfall per event, respectively, by
the ratio of snow depth to rain depth. This ratio is estimated at 10 in./in on the
basis of an analysis of weather data reported by the National Climatic Data
Center (2).

In Step 6, the duration of pavement runoff is defined differently for snow

events. Specifically, it is defined as the time after the snow stops falling that
snowpack (or ice) covers the pavement. After this period elapses, the pavement
is exposed and drying begins. A default value for this variable is provided in
Exhibit 17-8 in Chapter 17.
Step 8: Identify Analysis Period Weather

Steps 1 through 7 are repeated for each day of a 2-year period, starting with
the first day of the reliability reporting period. This 2-year record of weather
events is used in the traffic incident procedure to estimate the weather-related
incident frequency.
The days that have weather events are subsequently examined to determine
whether the event occurs during the study period. Specifically, each analysis
period is examined to determine whether it is associated with a weather event. If
the pavement is wet during an analysis period, the precipitation type (i.e., rain or
snow) is recorded for that period. If precipitation is falling, the precipitation rate
is also recorded.
The duration of precipitation and wet pavement from Equation 29-9 and
Equation 29-11, respectively, are rounded to the nearest hour for 1-h analysis
periods or to the nearest quarter hour for 15-min analysis periods. The rounding
ensures the most representative match between event duration and analysis
period start and end times.
TRAFFIC DEMAND VARIATION GENERATION

The traffic demand variation procedure is used to identify the appropriate
traffic demand adjustment factors for each analysis period in the reliability
reporting period. One set of factors accounts for systematic volume variation by
hour of day, day of week, and month of year. Default values for these factors are
provided in Exhibit 17-5 to Exhibit 17-7 in Chapter 17.
The sequence of calculations in the traffic demand variation procedure is
shown in Exhibit 29-2. The calculations proceed on a day-by-day and hour-by-
hour basis in chronological order. Within a given day, the procedure considers
only the hours within the study period. The factors identified in this procedure
are subsequently used in the scenario file generation procedure to compute the
demand volume for the subject urban street facility.
A random variation adjustment factor is also available and can be included,
if desired, by the analyst. It accounts for the random variation in volume that
occurs among 15-min analysis periods. This factor is described in more detail in
the Scenario Dataset Generation section.
The procedure includes two adjustment factors to account for a reduction in
traffic demand during inclement weather. One factor addresses demand change
during periods of rain. The second factor addresses demand change during
periods of snow. Default values for these factors are provided in Exhibit 17-8 in
Chapter 17.

Exhibit 29-2 Start

Traffic Demand Variation
Procedure Demand Compute and save
Variation Day = 1 Hour = 1 volume adj. factors by
Procedure hour for each hour.
Hour =
Hour + 1
Day =
Last hour of day?
Day + 1
Last day of No
reliability reporting
Yes
No period?
Yes
Intersection Incident Procedure
This procedure does not address traffic diversion due to the presence of
work zones or special events. Their accommodation in a reliability evaluation is
discussed in the Analysis Techniques subsection of Section 5 in Chapter 17.
If the traffic volumes provided in the base dataset and the alternative
datasets are computed by using planning procedures, the volumes in the dataset
are assumed to represent the average day of week and month of year. In this
situation, the adjustment factors for day of week and month of year are set to a
value of 1.0.
The factors identified in this procedure are subsequently used in the scenario
dataset generation procedure to compute the demand volume for the subject
urban street facility.
TRAFFIC INCIDENT GENERATION

The traffic incident procedure is used to predict incident date, time, and
duration. It also determines incident event type (i.e., crash or noncrash), severity
level, and location on the facility. Location is defined by the specific intersection
or segment on which the incident occurs and whether the incident occurs on the
shoulder, one lane, or multiple lanes. The procedure uses weather event and
traffic demand variation information from the previous procedures in the
incident prediction process.
The sequence of calculations in the traffic incident procedure is shown in
Exhibit 29-3. The sequence shown is applicable to incidents occurring at
signalized intersections. A similar sequence is followed for incidents occurring at
locations along the urban street between the signalized intersections (i.e.,
midsignal segments).
The traffic incident procedure consists of six calculation steps. The
calculations associated with each step are described in the following paragraphs.
A random number is used in several of the steps. All random numbers have a
real value that is uniformly distributed from 0.0 to 1.0.

Exhibit 29-3
Traffic Incident Procedure for
Intersection Incidents
Step 1: Compute the Equivalent Crash Frequency for Weather

Crash frequency increases when the road is wet, covered by snow, or
covered by ice. The effect of weather on crash frequency is incorporated in the
reliability methodology by converting the input crash frequency data into an
equivalent crash frequency for each type of weather condition. The equivalent
crash frequency for dry pavement conditions is defined with the following
equation:
𝐹𝑐𝑠𝑡𝑟(𝑖) 8,760 𝑁𝑦 Equation 29-13
𝐹𝑐𝑠𝑡𝑟(𝑖),dry =
𝑁ℎdry + 𝐶𝐹𝐴𝐹𝑟𝑓 𝑁ℎ𝑟𝑓 + 𝐶𝐹𝐴𝐹𝑤𝑝 𝑁ℎ𝑤𝑝 + 𝐶𝐹𝐴𝐹𝑠𝑓 𝑁ℎ𝑠𝑓 + 𝐶𝐹𝐴𝐹𝑠𝑝 𝑁ℎ𝑠𝑝
where
Fcstr(i),dry = equivalent crash frequency when every day is dry for street location
i of type str (str = int: intersection, seg: segment) (crashes/year),
Fcstr(i) = expected crash frequency for street location i of type str
(crashes/year),
8,760 = number of hours in a year (h/year),
Ny = total number of years (years),
Nhdry = total number of hours in Ny years with dry conditions (h),
Nhrf = total number of hours in Ny years with rainfall conditions (h),
Nhwp = total number of hours in Ny years with wet pavement and not
raining (h),
Nhsf = total number of hours in Ny years with snowfall conditions (h),

Nhsp = total number of hours in Ny years with snow or ice on pavement and
not snowing (h),
CFAFrf = crash frequency adjustment factor for rainfall,
CFAFwp = crash frequency adjustment factor for wet pavement (not raining),
CFAFsf = crash frequency adjustment factor for snowfall, and
CFAFsp = crash frequency adjustment factor for snow or ice on pavement (not
snowing).
The equivalent crash frequency for nondry conditions is computed with the
following equation. The crash frequency adjustment factor (CFAF) for dry weather
CFAFdry is 1.0.
Equation 29-14 𝐹𝑐𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎 = 𝐹𝑐𝑠𝑡𝑟(𝑖),dry 𝐶𝐹𝐴𝐹𝑤𝑒𝑎
where
Fcstr(i),wea = equivalent crash frequency when every day has weather condition
wea (wea = dry: no precipitation and dry pavement, rf: rainfall, wp:
wet pavement but not raining, sf: snowfall, sp: snow or ice on
pavement but not snowing) for street location i of type str
(crashes/year);
Fcstr(i),dry = equivalent crash frequency when every day is dry for street location
i of type str (crashes/year); and
CFAFwea = crash frequency adjustment factor for weather condition wea.
Equation 29-14 requires the total number of hours for each weather condition
in the vicinity of the subject facility. A weather history that extends for 2 or more
years should be used to reduce the random variability in the data. These hours
can be obtained from available weather records or estimated by using the
weather event procedure.
This step is applied separately to each intersection and segment on the
facility. The expected crash frequency Fc is provided by the analyst for the
subject intersection or the subject segment, whichever is applicable.
The CFAF is the ratio of hourly crash frequency during the weather event to
the hourly crash rate during clear, dry hours. It is computed by using one or
more years of historical weather data and crash data for the region in which the
subject facility is located. Default values for these factors are provided in Exhibit
17-9 in Chapter 17.
Step 2: Establish the CFAFs for Work Zones and Special Events
If the analysis period occurs during a work zone or special event, the CFAF
variable for segments CFAFstr and the CFAF variable for intersections CFAFint are
set equal to the values provided by the analyst. Otherwise, CFAFstr and CFAFint
equal 1.0. This step is repeated for each analysis period of the reliability reporting
period.

Step 3: Determine Whether an Incident Occurs

During this step, each of the 24 h in the subject day is examined to determine
whether an incident occurs. The analysis considers each street location (i.e.,
intersection and segment) separately. At each street location, each of the
following 12 incident types is separately addressed. Each of these types is
separately considered for each hour of the day (whether the hour coincides with
an analysis period is determined in a subsequent step).
• Crash, one lane blocked, fatal or injury;
• Crash, two or more lanes blocked, fatal or injury;
• Crash, shoulder location, fatal or injury;
• Crash, one lane blocked, property damage only;
• Crash, two or more lanes blocked, property damage only;
• Crash, shoulder location, property damage only;
• Noncrash, one lane blocked, breakdown;
• Noncrash, two or more lanes blocked, breakdown;
• Noncrash, shoulder location, breakdown;
• Noncrash, one lane blocked, other; “Other” refers to any kind of
nonbreakdown incident (e.g.,
• Noncrash, two or more lanes blocked, other; and spill, dropped load).
• Noncrash, shoulder location, other.
Initially, the weather event data are checked to determine whether the
subject day and hour are associated with rainfall, wet pavement and not raining,
snowfall, or snow or ice on pavement and not snowing. For a given day, street
location, and hour of day, the average incident frequency is computed with the
following equation on the basis of the weather present at that hour and day.
𝐹𝑐𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎
𝐹𝑖𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑) = 𝐶𝐹𝐴𝐹𝑠𝑡𝑟 Equation 29-15
𝑝𝑐𝑠𝑡𝑟,𝑤𝑒𝑎
where
Fistr(i),wea(h,d) = expected incident frequency for street location i of type str and
weather condition wea(h,d) during hour h and day d
(incidents/year);
CFAFstr = crash frequency adjustment factor for street location type str;
Fcstr(i),wea = equivalent crash frequency when every day has weather condition
wea for street location i of type str (crashes/year); and
pcstr,wea = proportion of incidents that are crashes for street location type str
and weather condition wea.
Default values for the proportion of incidents are provided in the third
column of Exhibit 17-11 in Chapter 17.
The incident frequency is converted to an hourly frequency that is sensitive
to traffic demand variation by hour of day, day of week, and month of year. The
converted frequency is computed with the following equation:

Equation 29-16 𝐹𝑖𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑)

𝑓𝑖𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑),ℎ,𝑑 = (24 𝑓ℎ𝑜𝑑,ℎ,𝑑 )𝑓𝑑𝑜𝑤,𝑑 𝑓𝑚𝑜𝑦,𝑑
8,760
where
fistr(i),wea(h,d),h,d = expected hourly incident frequency for street location i of type
str and weather condition wea(h, d) during hour h and day d
(incidents/h),
Fistr(i),wea(h,d) = expected incident frequency for street location i of type str and
weather condition wea(h, d) during hour h and day d
(incidents/year),
8,760 = number of hours in a year (h/year),
24 = number of hours in a day (h/day),
fhod,h,d = hour-of-day adjustment factor based on hour h and day d,
fdow,d = day-of-week adjustment factor based on day d, and
fmoy,d = month-of-year adjustment factor based on day d.
The hour-of-day adjustment factor includes a day subscript because its
values depend on whether the day occurs during a weekday or weekend. The
day subscript for the day-of-week factor is used to determine which of the 7
weekdays is associated with the subject day. Similarly, the month subscript is
used to determine which of the 12 months is associated with the subject day for
the month-of-year factor. Default values for these adjustment factors are
provided in Exhibit 17-5 to Exhibit 17-7 in Chapter 17.
Incidents for a given day, street location, incident type, and hour of day are
assumed to follow a Poisson distribution. For any given combination of
conditions, the probability of more than one incident of a given type is negligible,
which simplifies the mathematics so that the question of whether an incident
occurs is reduced to whether there are zero incidents or one incident of a given
type. Equation 29-17 is used to compute the probability of no incidents occurring.
Default values for the proportion of incidents are provided in Exhibit 17-11 and
Exhibit 17-12 in Chapter 17.
Equation 29-17 𝑝0𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣,ℎ,𝑑 = exp (−𝑓𝑖𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑),ℎ,𝑑 × 𝑝𝑖𝑠𝑡𝑟,𝑤𝑒𝑎(ℎ,𝑑),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣 )
where
p0str(i),wea(h,d),con,lan,sev,h,d = probability of no incident for street location i of type str,
weather condition wea(h, d) during hour h and day d,
event type con (con = cr: crash, nc: noncrash), lane
location lan (lan = 1L: one lane, 2L: two or more lanes, sh:
shoulder), and severity sev (sev = pdo: property damage
only, fi: fatal or injury, bkd: breakdown, oth: other);
fistr(i),wea(h,d),h,d = expected hourly incident frequency for street location i
of type str and weather condition wea(h, d) during hour h
and day d (incidents/h); and

pistr,wea(h,d),con,lan,sev = proportion of incidents for street location type str,

event type con, lane location lan, and severity sev.
The following rule is checked to determine whether the incident of a specific
type occurs:
No incident if Ristr(i),wea(h,d),con,lan,sev,h,d ≤ p0str(i),wea(h,d),con,lan,sev,h,d Equation 29-18
Incident if Ristr(i),wea(h,d),con,lan,sev,h,d > p0str(i),wea(h,d),con,lan,sev,h,d

where
Ristr(i),wea(h,d),con,lan,sev,h,d = random number for incident for street location i of type
str, weather condition wea(h, d) during hour h and day d,
event type con, lane location lan, and severity sev; and
p0str(i),wea(h,d),con,lan,sev,h,d = probability of no incident for street location i of type str,
weather condition wea(h, d) during hour h and day d, event
type con, lane location lan, and severity sev.
Step 4: Determine Incident Duration

If the result of Step 3 indicates that an incident occurs for a given day, street
location, incident type, and hour of day, the calculations in this step are used to
determine the incident duration. Each hour of the day is considered separately in
this step.
Incident duration includes the incident detection time, response time, and
clearance time. Research (1) indicates that these values can vary by weather
condition, event type, lane location, and severity. Default values for average
incident duration are provided in the text associated with Exhibit 17-10 in
Chapter 17.
The following equation is used to estimate the incident duration for a given
incident:
𝑝 = 𝑅𝑑𝑠𝑡𝑟(𝑖),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣,ℎ,𝑑 ,
𝑑𝑖𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣,ℎ,𝑑 = gamma −1 ̅ 𝑠𝑡𝑟,𝑤𝑒𝑎(ℎ,𝑑),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣 ,)
(𝜇 = 𝑑𝑖 Equation 29-19
𝜎 = 𝑠𝑠𝑡𝑟,𝑤𝑒𝑎(ℎ,𝑑),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣
where
distr(i),wea(h,d),con,lan,sev,h,d = incident duration for street location i of type str, weather
condition wea(h, d) during hour h and day d, event type
con, lane location lan, and severity sev (h);
Rdstr(i),con,lan,sev,h,d = random number for incident duration for street location
i of type str for hour h and day d, event type con, lane
location lan, and severity sev;
—
distr,wea(h,d),con,lan,sev = average incident duration for street location type str,
event type con, lane location lan, and severity sev (h);
sstr,wed(h,d),con,lan,sev = standard deviation of incident duration for street
location type str, weather condition wea(h, d) during

hour h and day d, event type con, lane location lan, and
—
severity sev (= 0.8 distr,wea(h,d),con,lan,sev) (h); and
gamma–1(p, μ, σ) = value associated with probability p for cumulative
gamma distribution with mean μ and standard
deviation σ.
determine whether an analysis period is associated with an incident. To simplify
the analytics in this subsequent step, it is assumed that no incident extends beyond
midnight. To ensure this outcome, the duration computed from Equation 29-19 is
compared with the duration between the start of the study period and midnight.
The incident duration is then set to equal the smaller of these two values.
Step 5: Determine Incident Location

If the result of Step 3 indicates that an incident occurs for a given day, street
location, incident type, and hour of day, the calculations in this step are used to
determine the incident location. For intersections, the location is determined to
be one of the intersection legs. For segments, the location is determined to be one
of the two travel directions. The location algorithm is volume-based so that the
correct location determinations are made when three-leg intersections or one-
way streets are addressed. Each hour of the day is considered separately in this
step.
Intersection Location
When a specific intersection is associated with an incident, the location of the
incident is based on consideration of each intersection leg volume lv. This
volume represents the sum of all movements entering the intersection on the
approach lanes and movements exiting the intersection on the adjacent departure
lanes. In the field, this volume would be measured by establishing a reference
line from outside curb to outside curb on the subject leg (near the crosswalk) and
counting all vehicles that cross the line, regardless of travel direction.
The leg volumes are then summed, starting with the leg associated with
National Electrical Manufacturers Association (NEMA) Phase 2, to produce a
cumulative volume by leg. These volumes are then converted to a proportion by
dividing by the sum of the leg volumes. The calculation of these proportions is
described by the following equations. One set of proportions is determined for
the base dataset and for each work zone and special event dataset.
Equation 29-20 𝑝𝑣𝑖𝑛𝑡(𝑖),2 = 𝑙𝑣𝑖𝑛𝑡(𝑖),2 /(2 𝑡𝑣𝑖𝑛𝑡(𝑖) )
𝑝𝑣𝑖𝑛𝑡(𝑖),4 = 𝑝𝑣𝑖𝑛𝑡(𝑖),2 + 𝑙𝑣𝑖𝑛𝑡(𝑖),4 /(2 𝑡𝑣𝑖𝑛𝑡(𝑖) )
𝑝𝑣𝑖𝑛𝑡(𝑖),6 = 𝑝𝑣𝑖𝑛𝑡(𝑖),4 + 𝑙𝑣𝑖𝑛𝑡(𝑖),6 /(2 𝑡𝑣𝑖𝑛𝑡(𝑖) )
𝑝𝑣𝑖𝑛𝑡(𝑖),8 = 1.0

with
12
𝑡𝑣𝑖𝑛𝑡(𝑖) = ∑ 𝑣input,𝑖𝑛𝑡(𝑖),𝑗 Equation 29-21

𝑗=1
where
pvint(i),n = cumulative sum of volume proportions for leg associated with
NEMA phase n (n = 2, 4, 6, 8) at intersection i,
lvint(i),n = leg volume (two-way total) for leg associated with NEMA phase n at
intersection i (veh/h),
tvint(i) = total volume entering intersection i (veh/h), and
vinput,int(i),j = movement j volume at intersection i (from dataset) (veh/h).
The leg location of the incident is determined by comparing a random
number with the cumulative volume proportions. With this technique, the
likelihood of an incident being assigned to a leg is proportional to its volume
relative to the other leg volumes. The location is determined for a given
intersection i by the following rule:
Incident on Phase 2 if 𝑅𝑣𝑖𝑛𝑡(𝑖),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣 ≤ 𝑝𝑣𝑖𝑛𝑡(𝑖),2 Equation 29-22
Incident on Phase 4 if 𝑝𝑣𝑖𝑛𝑡(𝑖),2 < 𝑅𝑣𝑖𝑛𝑡(𝑖),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣 ≤ 𝑝𝑣𝑖𝑛𝑡(𝑖),4
where
Rvint(i),con,lan,sev = random number for leg volume for intersection i, event type
con, lane location lan, and severity sev; and
pvint(i),n = cumulative sum of volume proportions for leg associated with
NEMA phase n (n = 2, 4, 6, 8) at intersection i.
Segment Location
When a specific segment is associated with an incident, the location of the
incident is based on consideration of the volume in each direction of travel dv.
This volume is computed by using the movement volume at the boundary
intersection that uses NEMA Phase 2 to serve exiting through vehicles. The
volume in the Phase 2 direction is computed as the sum of the movements
exiting the segment at the boundary intersection (i.e., it equals the approach lane
volume). The volume in the Phase 6 direction is computed as the sum of the
movements entering the segment at the boundary intersection (i.e., it equals the
departure lane volume). The two directional volumes are referenced to NEMA
Phases 2 and 6. The sum of these two volumes equals the Phase 2 leg volume
described in the previous subsection.
A cumulative volume proportion by direction is used to determine incident
location. The calculation of these proportions is described by the following
equations. One set of proportions is determined for the base dataset and for each
work zone and special event dataset.

Equation 29-23 𝑝𝑣𝑠𝑒𝑔(𝑖),2 = 𝑑𝑣𝑠𝑒𝑔(𝑖),2 /(𝑑𝑣𝑠𝑒𝑔(𝑖),2 + 𝑑𝑣𝑠𝑒𝑔(𝑖),6 )

𝑝𝑣𝑠𝑒𝑔(𝑖),6 = 1.0
where
pvseg(i),n = volume proportion for the direction of travel served by NEMA phase
n (n = 2, 6) on segment i, and
dvseg(i),n = directional volume for the direction of travel served by NEMA phase
n on segment i (veh/h).
The segment location of the incident is determined by comparing a random
number with the cumulative volume proportions. With this technique, the
likelihood of an incident being assigned to a direction of travel is proportional to
its volume, relative to the volume in the other direction. The location is
determined for a given segment i by the following rule:
Equation 29-24 Incident in Phase 2 direction if 𝑅𝑣𝑠𝑒𝑔(𝑖),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣 ≤ 𝑝𝑣𝑠𝑒𝑔(𝑖),2
Incident in Phase 6 direction if 𝑝𝑣𝑠𝑒𝑔(𝑖),2 < 𝑅𝑣𝑠𝑒𝑔(𝑖),𝑐𝑜𝑛,𝑙𝑎𝑛,𝑠𝑒𝑣 ≤ 𝑝𝑣𝑠𝑒𝑔(𝑖),6
where
Rvseg(i),con lan,sev = random number for volume for segment i, event type con, lane
location lan, and severity sev; and
pvseg(i),n = volume proportion for the direction of travel served by NEMA
phase n (n = 2, 6) on segment i.
Step 6: Identify Analysis Period Incidents

Steps 3 through 5 are repeated for each hour of the subject day. As implied
by the discussion to this point, all incidents are assumed to occur at the start of a
given hour.
During this step, the analysis periods associated with an incident are
identified. Specifically, each hour of the study period is examined to determine
whether it coincides with an incident. If an incident occurs, its event type, lane
location, severity, and street location are identified and recorded. Each
subsequent analysis period coincident with the incident is also recorded.
The incident duration from Equation 29-19 is rounded to the nearest hour for
1-h analysis periods or to the nearest quarter hour for 15-min analysis periods.
This rounding is performed to ensure the most representative match between
event duration and analysis period start and end times.
SCENARIO DATASET GENERATION

The scenario dataset generation procedure uses the results from the
preceding three procedures to develop one HCM dataset for each analysis period
in the reliability reporting period. As discussed previously, each analysis period
is considered to be one scenario.
The sequence of calculations in the scenario file generation procedure is
shown in Exhibit 29-4. The calculations and file generation proceed on a day-by-
day and analysis-period-by-analysis-period basis in chronological order. If a day

is coincident with a work zone or special event, the appropriate alternative

dataset is loaded. Otherwise, the base dataset is loaded.
Exhibit 29-4
Scenario File Generation
Procedure
This procedure creates a new HCM dataset for each analysis period. The
dataset is modified to reflect conditions present during a given analysis period.
Modifications are made to the traffic volumes at each intersection and driveway
and to the saturation flow rate at intersections influenced by an incident or a
weather event. The speed is also adjusted for segments influenced by an incident
or a weather event.
The incident history developed by the traffic incident procedure is consulted
during this procedure to determine whether an incident occurs at an intersection
or on a segment. If an incident occurs at an intersection, the incident lane location
data are consulted to determine which approach and movements are affected. If
the incident occurs on the shoulder, the shoulder in question is assumed to be the
outside shoulder (as opposed to the inside shoulder). If a one-lane incident occurs,
the incident is assumed to occur in the outside lane. If a two-or-more-lane incident
occurs, it is assumed to occur in the outside two lanes. The incident is also

assumed to occur on the intersection approach lanes as opposed to the departure

lanes. These assumptions are consistent with typical intersection crash patterns.
The scenario dataset generation procedure consists of nine calculation steps.
The calculations associated with each step are described in the following
paragraphs.
Step 1: Acquire the Appropriate Dataset

During this step, the appropriate HCM dataset is acquired. This step
proceeds day by day and analysis period by analysis period in chronological
order. The date is used to determine whether a work zone or special event is
present. If one is present, the appropriate alternative dataset is acquired.
Otherwise, the base dataset is acquired. The hour-of-day, day-of-week, and
month-of-year demand adjustment factors associated with each dataset are also
acquired (as identified previously in the traffic demand variation procedure).
Step 2: Compute Weather Adjustment Factors

Signalized Intersections
The following equation is used to compute the saturation flow rate
adjustment factor for analysis periods with poor weather conditions. It is used in
Step 5 to estimate intersection saturation flow rate during weather events.
1.0
Equation 29-25 𝑓𝑟𝑠,𝑎𝑝,𝑑 =
1.0 + 0.48 𝑅𝑟,𝑎𝑝,𝑑 + 0.39 𝑅𝑠,𝑎𝑝,𝑑
where
frs,ap,d = saturation flow adjustment factor for rainfall or snowfall during
analysis period ap and day d,
Rr,ap,d = rainfall rate during analysis period ap and day d (in./h), and
Rs,ap,d = precipitation rate when snow is falling during analysis period ap and
day d (in./h).
If Equation 29-25 is used for analysis periods with falling rain, the variable Rs
should equal 0.0. If it is used for analysis periods with falling snow, the variable
Rr should equal 0.0. The variable Rs equals the precipitation rate in terms of
equivalent inches of water per hour (i.e., it is not a snowfall rate).
The value obtained from Equation 29-25 applies when precipitation is falling.
If the pavement is wet and there is no rainfall, the adjustment factor frs,ap,d is 0.95.
If snow or ice is on the pavement and snow is not falling, the adjustment factor
frs,ap,d is 0.90.
Segments
The following equation is used to compute the free-flow speed adjustment
factor for analysis periods with poor weather conditions. It is used in Step 7 to
estimate the additional running time during weather events.
1.0
Equation 29-26 𝑓𝑠,𝑟𝑠,𝑎𝑝,𝑑 =
1.0 + 0.48 𝑅𝑟,𝑎𝑝,𝑑 + 1.4 𝑅𝑠,𝑎𝑝,𝑑

where
fs,rs,ap,d = free-flow speed adjustment factor for rainfall or snowfall during
Rr,ap,d = rainfall rate during analysis period ap and day d (in./h), and
Rs,ap,d = precipitation rate when snow is falling during analysis period ap and
day d (in./h).
If Equation 29-26 is used for analysis periods with falling rain, the variable Rs
should equal 0.0. If it is used for analysis periods with falling snow, the variable
Rr should equal 0.0. The variable Rs equals the precipitation rate in terms of
equivalent inches of water per hour (i.e., it is not a snowfall rate).
The value obtained from Equation 29-26 applies when precipitation is falling.
If the pavement is wet and there is no rainfall, the adjustment factor fs,rs,ap,d is 0.95.
If snow or ice is on the pavement and snow is not falling, the adjustment factor
fs,rs,ap,d is 0.90.
Step 3: Acquire Demand Adjustment Factors

During this step, the hour-of-day, day-of-week, and month-of-year demand
adjustment factors associated with each analysis period are acquired (as
identified previously in the traffic demand variation procedure). They are used
in Step 6 to estimate the analysis period volumes.
Step 4: Compute Incident Adjustment Factors for Intersections

The following equation is used to compute the saturation flow rate
adjustment factor for analysis periods associated with an incident. It is used in
Step 5 to estimate intersection saturation flow rate during incidents.
𝑁𝑖𝑐,𝑖𝑛𝑡(𝑖),𝑛,𝑚,𝑎𝑝,𝑑 𝑏𝑖𝑐,𝑖𝑛𝑡(𝑖),𝑛,𝑎𝑝,𝑑
𝑓𝑖𝑐,𝑖𝑛𝑡(𝑖),𝑛,𝑚,𝑎𝑝,𝑑 = (1 − ) (1 − ) ≥ 0.10 Equation 29-27
𝑁𝑛,𝑖𝑛𝑡(𝑖),𝑛,𝑚 ∑𝑚∈𝐿,𝑇,𝑅 𝑁𝑛,𝑖𝑛𝑡(𝑖),𝑛,𝑚
with
𝑏𝑖𝑐,𝑖𝑛𝑡(𝑖),𝑛,𝑎𝑝,𝑑 = 0.58 𝐼𝑓𝑖,𝑖𝑛𝑡(𝑖),𝑛,𝑎𝑝,𝑑 + 0.42 𝐼𝑝𝑑𝑜,𝑖𝑛𝑡(𝑖),𝑛,𝑎𝑝,𝑑 + 0.17 𝐼other,𝑖𝑛𝑡(𝑖),𝑛,𝑎𝑝,𝑑 Equation 29-28
where
fic,int(i),n,m,ap,d = saturation flow adjustment factor for incident presence for
movement m (m = L: left, T: through, R: right) on leg associated
with NEMA phase n (n = 2, 4, 6, 8) at intersection i during
Nn,int(i),n,m = number of lanes serving movement m under normal (i.e.,
nonincident) conditions on leg associated with NEMA phase n at
intersection i (ln),
Nic,int(i),n,m,ap,d = number of lanes serving movement m blocked by the incident on
leg associated with NEMA phase n at intersection i during
analysis period ap and day d (ln),

bic,int(i),n,ap,d = calibration coefficient based on incident severity on leg

associated with NEMA phase n at intersection i during analysis
period ap and day d,
Ipdo,int(i),n,ap,d = indicator variable for property-damage-only (PDO) crash on leg
associated with NEMA phase n at intersection i during analysis
period ap and day d (= 1.0 if PDO crash, 0.0 otherwise),
Ifi,int(i),n,ap,d = indicator variable for fatal-or-injury crash on leg associated with
NEMA phase n at intersection i during analysis period ap and
day d (= 1.0 if fatal-or-injury crash, 0.0 otherwise), and
Iother,int(i),n,ap,d = indicator variable for noncrash incident on leg associated with
NEMA phase n at intersection i during analysis period ap and
day d (= 1.0 if noncrash incident, 0.0 otherwise).
Equation 29-27 is applied to each approach traffic movement. For a given
movement, the first term of Equation 29-27 adjusts the saturation flow rate on the
basis of the number of lanes that are blocked by the incident. If the incident is
located on the shoulder or in the lanes associated with another movement m (i.e.,
Nic = 0), this term equals 1.0.
Equation 29-27 is used for each movement to estimate the saturation flow
rate adjustment factor for incidents. If all lanes associated with a movement are
closed because of the incident, an adjustment factor of 0.10 is used. This
approach effectively closes the lane but does not remove it from the intersection,
as described in the dataset.
Step 5: Compute Saturation Flow Rate for Intersections

During this step, the saturation flow rate for each intersection movement is
adjusted by using the factors computed in Steps 2 and 4. The weather adjustment
factor is applied to all movements at all intersections. The incident adjustment
factor is applied only to the movements affected by an incident.
The weather and incident factors are multiplied by the saturation flow rate in
the dataset to produce a revised estimate of the saturation flow rate.
Step 6: Compute Traffic Demand Volumes

Adjust Movement Volumes
During this step, the volume for each movement is adjusted by using the
appropriate hour-of-day, day-of-week, and month-of-year factors to estimate the
average hourly flow rate for the subject analysis period. The following equation
is used for this purpose:
𝑣input,𝑖𝑛𝑡(𝑖),𝑗
𝑣𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑 = 𝑓 𝑓 𝑓
Equation 29-29 𝑓ℎ𝑜𝑑,input 𝑓𝑑𝑜𝑤,input 𝑓𝑚𝑜𝑦,input ℎ𝑜𝑑,ℎ,𝑑 𝑑𝑜𝑤,𝑑 𝑚𝑜𝑦,𝑑
where
vint(i),j,h,d = adjusted hourly flow rate for movement j at intersection i during
hour h and day d (veh/h),

vinput,int(i),j = movement j volume at intersection i (from base dataset or alternative

dataset) (veh/h),
fhod,h,d = hour-of-day adjustment factor based on hour h and day d,
fdow,d = day-of-week adjustment factor based on day d,
fmoy,d = month-of-year adjustment factor based on day d,
fhod,input = hour-of-day adjustment factor for hour and day associated with vinput,
fdow,input = day-of-week adjustment factor for day associated with vinput, and
fmoy,input = month-of-year adjustment factor for day associated with vinput.
If a 15-min analysis period is used, the adjusted hourly flow rate is applied to
all four analysis periods coincident with the subject hour h. Equation 29-29 is also
used to adjust the volumes associated with each unsignalized access point on
each segment.
Random Variation Among 15-min Periods

If a 15-min analysis period is used, the analyst has the option of adding a
random element to the adjusted hourly volume for each movement and analysis
period. Doing so provides a more realistic estimate of performance measure
variability. However, it ensures that every analysis period is unique (thereby
lessening the likelihood that similar scenarios can be found for the purpose of
reducing the total number of scenarios to be evaluated). If this option is applied,
the turn movement volumes at each signalized intersection are adjusted by using
a random variability based on the peak hour factor. Similarly, the turn
movement volumes at each unsignalized access point are adjusted by using a
random variability based on a Poisson distribution.
If the analyst desires to add a random element to the adjusted hourly
volume, the first step is to estimate the demand flow rate variability adjustment
factor with the following equation:
1.0 − 𝑃𝐻𝐹𝑖𝑛𝑡(𝑖) −4
𝑓𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑 = √0.25𝑣𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑 × exp (−0.00679 + 0.004𝑃𝐻𝐹𝑖𝑛𝑡(𝑖) ) Equation 29-30
𝑃𝐻𝐹𝑖𝑛𝑡(𝑖)
where
fint(i),j,h,d = adjustment factor used to estimate the standard deviation of demand
flow rate for movement j at intersection i during hour h and day d,
PHFint(i) = peak hour factor for intersection i, and
vint(i),j,h,d = adjusted hourly flow rate for movement j at intersection i during
hour h and day d (veh/h).
The second step is to compute the randomized hourly flow rate for each
movement at each signalized intersection with the following equation:
𝑝 = 𝑅𝑓𝑎𝑝,𝑑 , 𝜇 = 0.25 𝑣𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑 ,
∗ −1
𝑣𝑖𝑛𝑡(𝑖),𝑗,𝑎𝑝,𝑑 = 4.0 × gamma ( ) Equation 29-31
𝜎 = 𝑓𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑 √0.25 𝑣𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑

where
v*int(i),j,ap,d = randomized hourly flow rate for movement j at intersection i
during analysis period ap and day d (veh/h),
gamma–1(p,μ,σ) = value associated with probability p for cumulative gamma
distribution with mean μ and standard deviation σ,
Rfap,d = random number for flow rate for analysis period ap and day d,
vint(i),j,h,d = adjusted hourly flow rate for movement j at intersection i
during hour h and day d (veh/h), and
fint(i),j,h,d = adjustment factor used to estimate the standard deviation of
demand flow rate for movement j at intersection i during hour
h and day d.
Similarly, the following equations are used to compute the randomized
hourly flow rates for each unsignalized access point. The first equation is used if
the adjusted hourly flow rate is 64 veh/h or less. The second equation is used if
the flow rate exceeds 64 veh/h.
If vint(i),j,h,d ≤ 64 veh/h,
Equation 29-32 ∗
𝑣𝑖𝑛𝑡(𝑖),𝑗,𝑎𝑝,𝑑 = 4.0 × Poisson−1(𝑝 = 𝑅𝑓𝑎𝑝,𝑑 , 𝜇 = 0.25 𝑣𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑 )
Otherwise,
𝑝 = 𝑅𝑓𝑎𝑝,𝑑 , 𝜇 = 0.25 𝑣𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑 ,
∗ −1
Equation 29-33 𝑣𝑖𝑛𝑡(𝑖),𝑗,𝑎𝑝,𝑑 = 4.0 × normal ( )
𝜎 = √0.25 𝑣𝑖𝑛𝑡(𝑖),𝑗,ℎ,𝑑
where
v*int(i),j,ap,d = randomized hourly flow rate for movement j at intersection i
during analysis period ap and day d (veh/h),
Poisson–1(p,μ) = value associated with probability p for the cumulative Poisson
distribution with mean μ,
Rfap,d = random number for flow rate for analysis period ap and day d,
vint(i),j,h,d = adjusted hourly flow rate for movement j at intersection i
during hour h and day d (veh/h), and
normal–1(p,μ,σ) = value associated with probability p for a cumulative normal
Step 7: Compute Speed for Segments

Additional Delay
During this step, the effect of incidents and weather on segment speed is
determined. This effect is added to the HCM dataset as an additional delay
incurred along the segment. The variable dother in Equation 18-7 is used with this
approach. This additional delay is computed with the following equations:
1.0 1.0
Equation 29-34 𝑑other,𝑠𝑒𝑔(𝑖),𝑛,𝑎𝑝,𝑑 = 𝐿𝑠𝑒𝑔(𝑖) ( ∗ − )
𝑆𝑓𝑜,𝑠𝑒𝑔(𝑖),𝑛,𝑎𝑝,𝑑 𝑆𝑓𝑜,𝑠𝑒𝑔(𝑖),𝑛

with
∗
𝑏𝑖𝑐,𝑠𝑒𝑔(𝑖),𝑛,𝑎𝑝,𝑑
𝑆𝑓𝑜,𝑠𝑒𝑔(𝑖),𝑛,𝑎𝑝,𝑑 = 𝑆𝑓𝑜,𝑠𝑒𝑔(𝑖),𝑛 × 𝑓𝑠,𝑟𝑠,𝑎𝑝,𝑑 × (1.0 − ) Equation 29-35
𝑁𝑜,𝑠𝑒𝑔(𝑖),𝑛
𝑏𝑖𝑐,𝑠𝑒𝑔(𝑖),𝑛,𝑎𝑝,𝑑 = 0.58 𝐼𝑓𝑖,𝑠𝑒𝑔(𝑖),𝑛,𝑎𝑝,𝑑 + 0.42 𝐼𝑝𝑑𝑜,𝑠𝑒𝑔(𝑖),𝑛,𝑎𝑝,𝑑 + 0.17 𝐼other,𝑠𝑒𝑔(𝑖),𝑛,𝑎𝑝,𝑑 Equation 29-36
where
dother,seg(i),n,ap,d = additional delay for the direction of travel served by NEMA
phase n (n = 2, 6) on segment i during analysis period ap and
day d (s/veh),
Lseg(i) = length of segment i (ft),
Sfo,seg(i),n = base free-flow speed for the direction of travel served by
NEMA phase n on segment i (ft/s),
S*fo,seg(i),n,ap,d = adjusted base free-flow speed for the direction of travel served
by NEMA phase n on segment i during analysis period ap and
day d (ft/s),
fs,rs,ap,d = free-flow speed adjustment factor for rainfall or snowfall
during analysis period ap and day d,
bic,seg(i),n,ap,d = calibration coefficient based on incident severity on leg
associated with NEMA phase n at intersection i during
No,seg(i),n = number of lanes serving direction of travel served by NEMA
phase n on segment i (ln),
Ipdo,seg(i),n,ap,d = indicator variable for property-damage-only (PDO) crash in
the direction of travel served by NEMA phase n on segment i
during analysis period ap and day d (= 1.0 if PDO crash, 0.0
otherwise),
Ifi,seg(i),n,ap,d = indicator variable for fatal-or-injury crash in the direction of
travel served by NEMA phase n on segment i during analysis
period ap and day d (= 1.0 if fatal-or-injury crash, 0.0
otherwise), and
Iother,seg(i),n,ap,d = indicator variable for noncrash incident in the direction of
travel served by NEMA phase n on segment i during analysis
period ap and day d (= 1.0 if noncrash incident, 0.0 otherwise).
The delay estimated from Equation 29-34 is added to the value of the “other
delay” variable in the dataset to produce a combined “other delay” value for
segment running speed estimation.
Segment Lane Closure

If an incident is determined to be located in one or more lanes, the variable
for the number of through lanes on the segment is reduced accordingly. This
adjustment is made for the specific segment and direction of travel associated
with the incident.

The variable indicating the number of major-street through lanes at each

unsignalized access point is reduced in a similar manner when the incident
occurs on a segment and closes one or more lanes. This adjustment is made for
each access point on the specific segment affected by the incident.
Step 8: Adjust Critical Left-Turn Headway

Research (1) indicates that the critical headway for left-turn drivers increases
by 0.7 to 1.2 s, depending on the type of weather event and the opposing lane
associated with the conflicting vehicle. The recommended increase in the critical
headway value for each weather condition is listed in Exhibit 29-5.
Exhibit 29-5 Weather Condition Additional Critical Left-Turn Headway (s)

Additional Critical Left-Turn Clear, snow on pavement 0.9
Headway due to Weather Clear, ice on pavement 0.9
Clear, water on pavement 0.7
Snowing 1.2
Raining 0.7
Step 9: Save Scenario Dataset

During this step, the dataset with the updated values is saved for evaluation
in the next stage of the reliability methodology. One dataset is saved for each
analysis period (i.e., scenario).

3. SUSTAINED SPILLBACK PROCEDURE
This section describes a procedure for using the methodologies described in

Chapter 16, Urban Street Facilities, and Chapter 18, Urban Street Segments, to
evaluate a facility with spillback in one or more travel directions on one or more
segments.
The discussion in this section addresses sustained spillback. Sustained
spillback occurs as a result of oversaturation (i.e., more vehicles discharging from
the upstream intersection than can be served at the subject downstream
intersection). The spillback can exist at the start of the study period, or it can
occur at some point during the study period. Spillback that first occurs after the
study period is not addressed.
OVERVIEW OF THE PROCEDURE

The effect of spillback on traffic flow is modeled through an iterative process
that applies the urban street segments methodology to each segment of the
subject urban street facility. If spillback occurs on a segment, the discharge rate of
each traffic movement entering the segment is reduced so that (a) the number of
vehicles entering the segment equals the number of vehicles exiting the segment
and (b) the residual queue length equals the available queue storage distance.
The approach used to model spillback effects is similar to the technique used
for multiple time period analysis, as described in the subsection Multiple Time
Period Analysis in Section 3 of Chapter 18. However, in this application, a single
analysis period is divided into subperiods for separate evaluation. Each
subperiod is defined by using the following rules:
• The first subperiod starts with the start of the analysis period.
• The current subperiod ends (and a new subperiod starts) with each new
occurrence of spillback on the facility.
• The total of all subperiod durations must equal the original analysis
period duration.
As with the multiple-time-period analysis technique, the residual queue
from one subperiod becomes the initial queue for the next subperiod. When all
subperiods have been evaluated by using the urban street segments
methodology, the performance measures for each subperiod are aggregated for
the analysis period with a weighted-average technique, where the weight is the
volume associated with the subperiod.
Section 3 of Chapter 30, Urban Street Segments: Supplemental, describes a
“spillback check” procedure for determining whether queue spillback occurs on
a segment during a given analysis period. That procedure also predicts the
controlling time until spillback. This time is used in the sustained spillback
procedure to determine when the current subperiod ends.
Section 3 of Chapter 30 also describes a procedure for predicting the effective
average vehicle spacing. This spacing is used in the sustained spillback procedure
to determine the maximum queue storage in a turn bay and along a segment.
Chapter 29/Urban Street Facilities: Supplemental Sustained Spillback Procedure

COMPUTATIONAL STEPS
This subsection describes the sequence of computational steps that culminate
in the calculation of facility performance for a specified analysis period. The
input data requirements for this procedure are the same as for the urban street
segments methodology (hereafter referred to as the “methodology”).
Step 1: Initialize Variables

Set the original analysis period variable To equal to the analysis period T that
is input by the analyst. Set the total time variable Ttotal,0 equal to zero and the
subperiod counter k to 0.
Step 2: Implement the Methodology

The methodology is used in this step to evaluate each segment on the facility.
The analysis period duration used in the methodology is computed as T = To –
Ttotal,k. Increase the value of the subperiod counter k by 1. Hence, for the first
subperiod (k = 0), the analysis period duration T equals To (i.e., T = To – 0.0).
Step 3: Check for Spillback

During this step, the results from Step 2 are examined to determine whether
there is a new occurrence of spillback. One direction of travel on one segment is
considered a “site.” Each site is checked in this step. Any site that has
experienced spillback during a previous subperiod is not considered in this step.
The predicted controlling time until spillback is recorded in this step. If
several sites experience spillback, the time of spillback that is recorded is based
on the site experiencing spillback first. The site that experiences spillback first is
flagged as having spilled back. The controlling time until spillback for the
subperiod Tcs,k is set equal to the time until spillback for this site. The total time
variable is computed with the following equation. It represents a cumulative
total time for the current and all previous subperiods.
Equation 29-37 𝑇total,𝑘 = 𝑇total,𝑘−1 + 𝑇𝑐𝑠,𝑘
where
Ttotal,k = total analysis time for subperiods 0 to k (h), and
Tcs,k = controlling time until spillback for the subperiod k (h).
If spillback does not occur, the performance measures from Step 2 are saved
by using the procedure described in a subsequent subsection. The analyst then
proceeds to Step 10 to determine the aggregate performance measures for the
analysis period.
Step 4: Implement the Methodology to Evaluate a Subperiod

At the start of this step, the analysis period is set equal to the controlling time
determined in Step 3 (i.e., T = Tcs,k). All other input variables remain unchanged.
Then, the methodology is implemented to evaluate the facility. The performance
measures from this evaluation are saved by using the procedure described in a
subsequent subsection.
Sustained Spillback Procedure Chapter 29/Urban Street Facilities: Supplemental

Step 5: Prepare for the Next Subperiod by Determining the Initial

Queue
During this step, the input data are modified by updating the initial queue
values for all movement groups at each intersection. This modification is
necessary to prepare for a new evaluation of the facility for the next subperiod.
The initial queue for each movement group is set to the estimated residual queue
from the previous evaluation.
The initial queue values for the movement groups at the downstream
intersection that exit each segment are checked by comparing them with the
available queue storage distance. The storage distance for the left-turn movement
group is computed with the following equation. The storage distance for the
right-turn movement group is computed with a variation of this equation.
𝐿𝑎,𝑡ℎ𝑟𝑢 + 𝐿𝑎,𝑙𝑡 (𝑁𝑙𝑡 − 1)
𝑁𝑞𝑥,𝑙𝑡,𝑛,𝑘 = Equation 29-38
𝐿∗ℎ,𝑘
where
Nqx,lt,n,k = maximum queue storage for left-turn movement group during
subperiod k (veh),
La,thru = available queue storage distance for the through movement (ft),
La,lt = available queue storage distance for the left-turn movement (ft),
Nlt = number of lanes in the left-turn bay (ln), and
L*h,k = effective average vehicle spacing in stationary queue during
subperiod k (ft/veh).
The available queue storage distance for the through movement equals the
segment length less the width of the upstream intersection. For turn movements
served from a turn bay, this length equals the length of the turn bay. For turn
movements served from a lane equal in length to that of the segment, the queue
storage length equals the segment length less the width of the upstream
intersection.
The maximum queue storage for the through movement group is computed
with the following equation:
𝐿𝑎,𝑡ℎ𝑟𝑢 𝑁𝑡ℎ
𝑁𝑞𝑥,𝑡ℎ𝑟𝑢,𝑛,𝑘 = Equation 29-39
𝐿∗ℎ,𝑘
where
Nqx,thru,n,k = maximum queue storage for through movement group during
subperiod k (veh), and
Nth = number of through lanes (shared or exclusive) (ln).
The initial queue for each movement group exiting a segment is compared
with the maximum queue storage values. Any initial queue that exceeds the
maximum value is set to equal the maximum value.

Step 6: Prepare for the Next Subperiod by Determining the Saturation

Flow Rate Adjustment
During this step, the saturation flow rate is recomputed for movement
groups entering the site identified in Step 3 as having spillback. This
modification is necessary to prepare for a new evaluation of the facility during
the next subperiod.
The process of recomputing this saturation flow rate uses an iterative loop.
The loop converges when the saturation flow rate computed for each upstream
movement is sufficiently small that the number of vehicles entering the spillback
segment just equals the number of vehicles that leave the segment. A “spillback”
saturation flow rate adjustment factor fsp is computed for each movement to
produce this result. Its value is set to 1.0 at the start of the first loop (i.e., fsp,0 = 1.0).
The process begins by setting the analysis time to equal the time remaining
in the original analysis period (i.e., T = To – Ttotal,k).
The next task is to compute the estimated volume arriving to each movement
exiting the segment at the downstream signalized intersection (i.e., the adjusted
destination volume). This calculation is based on the origin–destination matrix
and discharge volume for each movement entering the segment. These quantities
are obtained from the variables calculated by using the methodology, as
described in Section 2 of Chapter 30. The adjusted destination volume is
4
Equation 29-40 𝐷𝑎,𝑗,𝑘 = ∑ 𝑣𝑜𝑑,𝑖,𝑗,𝑘

𝑖=1
where
Da,j,k = adjusted volume for destination j ( j = 1, 2, 3, 4) for subperiod k
(veh/h), and
vod,i,j,k = volume entering from origin i and exiting at destination j for
subperiod k (veh/h).
The letters j and i in Equation 29-40 denote the following four movements:
1 = left turn, 2 = through, 3 = right turn, and 4 = combined midsegment access
points.
The next task is to compute the proportion of Da,j,k coming from upstream
origin i. These proportions are computed with the following equation:
𝑣𝑜𝑑,𝑖,𝑗,𝑘
Equation 29-41 𝑏𝑖,𝑗,𝑘 =
𝐷𝑎,𝑗,𝑘
where bi,j,k is the proportion of volume at destination j that came from origin i for
subperiod k (veh/h).
The next task is to estimate the maximum discharge rate for each upstream
movement. This estimate is based on consideration of the capacity of the
downstream movements exiting the segment and their volume. When the
segment has incurred spillback, the capacity of one or more of these exiting
movements is inadequate relative to the discharge rates of the upstream
movements entering the segment. The computed maximum discharge rate is

intended to indicate the amount by which each upstream movement’s discharge

needs to be limited so that there is a balance between the number of vehicles
entering and exiting the segment. The following equation is used for this
purpose. It is applied to each of the four upstream entry movements i.
𝑑𝑣𝑢,𝑖,𝑘 = 𝑏𝑖,2,𝑘 × 𝑐𝑑,2,𝑘 Equation 29-42
+min(𝑏𝑖,1,𝑘 × 𝑐𝑑,1,𝑘 , 𝑓𝑥𝑖,2,𝑘 × 𝑣𝑜𝑑,𝑖,1,𝑘 )
+min(𝑏𝑖,3,𝑘 × 𝑐𝑑,3,𝑘 , 𝑓𝑥𝑖,2,𝑘 × 𝑣𝑜𝑑,𝑖,3,𝑘 )
+𝑓𝑥𝑖,2,𝑘 × 𝑣𝑜𝑑,4,𝑘
with
𝑏𝑖,2,𝑘 × 𝑐𝑑,2,𝑘
𝑓𝑥𝑖,2,𝑘 = Equation 29-43
𝑣𝑜𝑑,𝑖,2,𝑘
where
dvu,i,k = maximum discharge rate for upstream movement i for subperiod k
(veh/h),
cd,j,k = capacity at the downstream intersection for movement j for
subperiod k (veh/h), and
fxi,2,k = volume adjustment factor for origin i for subperiod k.
The factor fx is the ratio of two quantities. The numerator is the downstream
through capacity that is available to the upstream through movement. The
denominator is the volume entering the segment as a through movement and
exiting as a through movement. The ratio is used to adjust the exiting turn
movement and access point volumes so that they are reduced by the same
proportion as is the volume for the exiting through movement.
The product bi,j,k × cd,j,k represents the maximum discharge rate for entry
movement i that can be destined for exit movement j such that the origin–
destination volume balance is maintained and the exit movement’s capacity is
not exceeded. It represents the allocation of a downstream movement’s capacity
to each of the upstream movements that use that capacity, where the allocation is
proportional to the upstream movement’s volume contribution to the
downstream movement volume.
The capacity for the combined set of access points is unknown and is
unlikely to be the source of spillback. Hence, this capacity is not considered in
Equation 29-42.
The next task is to estimate the saturation flow rate adjustment factor for the
movements at the upstream signalized intersection. The movements of interest
are those entering the subject segment. The following equation is used for this
purpose:
0.5
𝑑𝑣𝑢,𝑖,𝑘
𝑓𝑠𝑝,𝑖,𝑘,𝑙 = ( ) × 𝑓𝑚𝑠,𝑖,𝑘 × 𝑓𝑠𝑝,𝑖,𝑘,𝑙−1 Equation 29-44
𝑐𝑢,𝑖,𝑘

where
fsp,i,k,l = adjustment factor for spillback for upstream movement i for iteration
l in subperiod k,
cu,i,k = capacity at the upstream intersection for movement i for subperiod k
(veh/h), and
fms,i,k = adjustment factor for downstream lane blockage for movement i for
subperiod k.
The adjustment factor is shown to have a subscript l indicating that the factor
value is refined through an iterative process where the factor computed in a
previous iteration is updated by using Equation 29-44.
In theory, the exponent associated with the ratio in parentheses should be
1.0. However, an exponent of 0.5 was found to provide for a smoother
convergence to the correct factor value.
The procedure for calculating the adjustment factor for downstream lane
blockage fms is described in Section 3 of Chapter 30, Urban Street Segments:
Supplemental. This adjustment factor is incorporated into the spillback factor (as
shown in Equation 29-44) for segments with spillback.
The last task of this step is to adjust the access point entry volumes. The
following equation is used for this purpose. One factor is computed for each
access point movement that departs from the access point and enters the
direction of travel with spillback.
0.5
Equation 29-45 𝑓𝑎𝑝,𝑚,𝑛,𝑖,𝑘,𝑝 = (𝑓𝑥𝑖,4,𝑘 ) × 𝑓𝑎𝑝,𝑚,𝑛,𝑖,𝑘,𝑝−1
where fap,m,n,i,k,p is the access point volume adjustment factor for movement i at
access point n of site m for iteration p in subperiod k. The access point volume
adjustment factors are used to adjust the volume entering the segment at each
access point.
Step 7: Implement the Methodology to Evaluate the Remaining Time

The methodology is implemented in this step to evaluate each segment on
the facility. The analysis period was set in Step 6 to equal the time remaining in
the original analysis period. The saturation flow rate of each movement
influenced by spillback is adjusted by using the factors quantified in Step 6.
Step 8: Compute the Queue Prediction Error

During this step, the predicted residual queue for each movement group is
compared with the maximum queue storage. This distance is computed with the
equations described in Step 5. Any difference between the predicted and
maximum queues is considered a prediction error. If the sum of the absolute
errors for all movements is not equal to a small value, the analysis returns to
Step 6.
Step 9: Check the Total Time of Analysis

During this step, the total time of analysis Ttotal,k is compared with the original
analysis period To. If they are equal, the analysis continues with Step 10.

If the two times are not in agreement, the access point volumes are restored
to their original value and then multiplied by the most current access point
volume adjustment factor. The analysis then returns to Step 2.
Step 10: Compute the Performance Measure Summary

During this step, the average value of each performance measure is computed.
The value is a representation of the average condition for the analysis period. For
uniform delay at one intersection, it is computed with the following equation:
𝑑1,𝑎𝑔𝑔,𝑖,𝑗,all
𝑑1,𝑖,𝑗 = Equation 29-46
𝑇𝑜 × 𝑣𝑖,𝑗
where
d1,i,j = uniform delay for lane group j at intersection i (s/veh),
d1,agg,i,j,all = aggregated uniform delay for lane group j at intersection i for all
subperiods (s/veh),
To = analysis period duration for the first subperiod (h), and
vi,j = demand flow rate for lane group j at intersection i (veh/h).
A variation of Equation 29-46 is used to compute the average value for the
other intersection performance measures of interest. The equations for
computing the aggregated uniform delay are provided in the next subsection.
The following equation is used to compute the average running time for one
site, where a site is one direction of travel on one segment:
𝑡𝑅,𝑎𝑔𝑔,𝑚,all
𝑡𝑅,𝑚 = 𝑛 Equation 29-47
∑𝑘=0 𝑤𝑡ℎ𝑟𝑢,𝑚,𝑘
where
tR,m = segment running time for site m (s),
tR,agg,m,all = aggregated segment running time for site m for all n subperiods (s),
and
wthru,m,k = weighting factor for site m for subperiod k (veh).
A variation of Equation 29-47 is used to compute the average value for the
other intersection performance measures of interest. The term in the
denominator of Equation 29-47 equals the total through volume during the
analysis period. The equations for computing the aggregated segment running
time and weighting factor are provided in the next subsection.
PROCEDURE FOR SAVING PERFORMANCE MEASURES

The performance measures computed by using the methodology are saved at
selected points within the spillback procedure. These measures correspond to a
specific subperiod of the analysis period. Each measure is “saved” by
accumulating its value for each subperiod. This sum is then used to compute an
average performance measure value during the last step of the procedure.

The following equation is used to save the computed uniform delay for one
intersection lane group. The computed delay represents a cumulative total time
for the current and all previous subperiods.
Equation 29-48 𝑑1,𝑎𝑔𝑔,𝑖,𝑗,𝑘 = 𝑑1,𝑎𝑔𝑔,𝑖,𝑗,𝑘−1 + 𝑑1,𝑖,𝑗,𝑘 × 𝑤𝑖,𝑗,𝑘
with
Equation 29-49 𝑤𝑖,𝑗,𝑘 = 𝑇 × 𝑣𝑖,𝑗,𝑘
where
d1,agg,i,j,k = aggregated uniform delay for lane group j at intersection i for
subperiods 0 to k (s/veh),
d1,i,j,k = uniform delay for lane group j at intersection i for subperiod k
(s/veh),
wi,j,k = weighting factor for lane group j at intersection i for subperiod k
(veh), and
vi,j,k = demand flow rate for lane group j at intersection i for subperiod k
(veh/h).
The weighting factor represents the number of vehicles arriving during the
analysis period for the specified lane group.
A variation of Equation 29-48 is also used to compute the aggregated values
of the following performance measures at each intersection:
• Incremental delay,
• Initial queue delay,
• Uniform stop rate,
• Incremental stop rate based on second-term back-of-queue size, and
• Initial queue stop rate based on third-term back-of-queue size.
The following equation is used to save the computed running time for one
site, where a site is one direction of travel on one segment:
Equation 29-50 𝑡𝑅,𝑎𝑔𝑔,𝑚,𝑘 = 𝑡𝑅,𝑎𝑔𝑔,𝑚,𝑘−1 + 𝑡𝑅,𝑚,𝑘 × 𝑤𝑡ℎ𝑟𝑢,𝑚,𝑘
with
Equation 29-51 𝑤𝑡ℎ𝑟𝑢,𝑚,𝑘 = 𝑇 × [𝑣𝑡,𝑖,𝑗,𝑘 𝑁𝑡,𝑖,𝑗 + 𝑣𝑠𝑙,𝑖,𝑗,𝑘 (1 − 𝑃𝐿,𝑖,𝑗,𝑘 ) + 𝑣𝑠𝑟,𝑖,𝑗,𝑘 (1 − 𝑃𝑅,𝑖,𝑗,𝑘 )]
where
tR,agg,m,k = aggregated segment running time for site m for subperiods 0 to k (s),
tR,m,k = segment running time for site m for subperiod k (s),
wthru,m,k = weighting factor for site m for subperiod k (veh),
vt,i,j,k = demand flow rate in exclusive through lane group j at intersection i
for subperiod k (veh/h/ln),
Nt,i,j = number of lanes in exclusive through lane group j at intersection i
(ln),

vsl,i,j,k = demand flow rate in shared left-turn and through lane group j at
intersection i for subperiod k (veh/h),
vsr,i,j,k = demand flow rate in shared right-turn and through lane group j at
intersection i for subperiod k (veh/h),
PL,i,j,k = proportion of left-turning vehicles in the shared lane group j at
intersection i for subperiod k, and
PR,i,j,k = proportion of right-turning vehicles in the shared lane group j at
intersection i for subperiod k.
When Equation 29-50 and Equation 29-51 are applied, the lane groups j and
intersection i are located at the downstream end of the subject site m. The
weighting factor represents the number of through vehicles arriving at the
downstream intersection as a through movement during the analysis period.
A variation of Equation 29-50 is also used to compute the aggregated values
of the following performance measures at each intersection:
• Through movement delay,
• Through movement stop rate,
• Travel time at free-flow speed, and
• Travel time at base free-flow speed.
COMPUTATIONAL ENGINE DOCUMENTATION

This section describes the logic flow of the sustained spillback procedure.
The description uses a flowchart and linkage list to document the procedure’s
implementation in the computational engine.
The sequence of calculations in the spillback methodology is shown in
Exhibit 29-6. It consists of several routines and two loops, one of which is an
iterative loop with a convergence criterion.
The urban street segments methodology is implemented at three separate
points in the flowchart. Each point of implementation is indicated in the exhibit
with a box that references the phrase “HCM methodology.” The engine
documentation of this methodology is provided in Section 7 of Chapter 30,
Urban Street Segments: Supplemental.

Exhibit 29-6
Spillback Procedure Flowchart
A description of the logic flow is as follows. The urban street segments

methodology is initially implemented and the presence of spillback is checked. If
spillback does not occur, the results are reported and the process is concluded. If
spillback occurs on a segment, a subperiod is defined and the urban street
segments methodology is reimplemented by using an analysis period that is
shortened to equal the time until spillback.
The iterative loop shown on the right side of the exhibit is called to quantify
a saturation flow rate adjustment factor for each movement entering the segment
with spillback. The value of this factor is determined to be that needed to limit
the entry movement volume so that the residual queue on the segment does not
exceed the available queue storage distance.

The main routines identified in Exhibit 29-6 are listed in Exhibit 29-7. The list
provides more information about each routine’s function and the conditions for
its use.
Routine Description Conditions for Use Exhibit 29-7

SetupToSecondRun Find first segment to spill back (that None Sustained Spillback Module
has not previously spilled back) and Routines
reset the analysis time to equal the
controlling spillback time.
SavePerformanceMeasures Save results from current evaluation None
with those from all prior subperiods (if
any).
AdjustResidualQueue Set initial queue of next subperiod to Apply to all intersections
equal the residual queue from the subjected to spillback in
current subperiod. current subperiod.
ComputeAdjustedCapacity Compute a saturation flow rate Apply to all intersections
adjustment factor for all intersection subjected to spillback in
and driveway movements subjected to current subperiod.
spillback from a downstream
intersection.
ComputeQueueError Compare predicted queue length with Apply to all segments
available storage length for each experiencing spillback in
movement experiencing spillback. current subperiod.
Compute queue error as the absolute
value of the difference between the
predicted and available lengths.

4. USE OF ALTERNATIVE TOOLS
This section presents examples using alternative traffic analysis tools that
deal specifically with the limitations of the methodologies described in Chapters
16 to 22. Both deterministic and stochastic tools are used for this presentation.
The focus is on the motorized vehicle mode because alternative tools are applied
more frequently to deal with motorized vehicle traffic.
Several other chapters present examples covering the use of alternative tools
to deal with the limitations of specific methodologies. These chapters are
identified in the following list:
• Chapter 27, Freeway Weaving: Supplemental, presents a simulation
example that demonstrates the detrimental effect of queue backup from
an exit ramp signal on the operation of a freeway weaving section.
• Chapter 31, Signalized Intersections: Supplemental, presents simulation
examples that demonstrate the effect of storage bay overflow, right-turn-
on-red operation, short through lanes, and closely spaced intersections.
• Chapter 34, Interchange Ramp Terminals: Supplemental, presents a
simulation example that demonstrates the effect of ramp-metering signals
on the operation of a diamond interchange. Another simulation example
examines the effect of the diamond interchange on the operation of a
nearby intersection under two-way stop control.
• Chapter 36, Concepts: Supplemental, demonstrates the use of individual
vehicle trajectory analysis to examine cyclical queuing characteristics and
to assess queue spillover into an upstream segment.
The need to determine performance measures from an analysis of vehicle
trajectories was emphasized in Chapter 7, Interpreting HCM and Alternative
Tool Results, and Chapter 36, Concepts: Supplemental. Specific procedures for
defining measures in terms of vehicle trajectories were proposed to guide the
future development of alternative tools. Most of the examples presented in this
section have applied existing versions of alternative tools and, therefore, do not
reflect the proposed trajectory-based measures.
This section consists of three main subsections. The first describes the base
urban street facility used in the examples presented in the other two subsections.
The second describes the use of alternative tools for signal timing design and
evaluation. The third demonstrates the use of alternative tools in addressing
some of the limitations of the HCM methodologies.
BASIC EXAMPLE PROBLEM CONFIGURATION

The base configuration for the examples in this section is shown in Exhibit
29-8. Five signalized intersections are included with a spacing of 2,000 ft between
the upstream stop lines of each intersection. Each intersection has the same
layout, with two lanes for through and right-turn movements and one 150-ft-
long left-turn bay.
Use of Alternative Tools Chapter 29/Urban Street Facilities: Supplemental

Exhibit 29-8
Base Configuration for the
Examples
1 2 3 4 5
2,000 ft 2,000 ft 2,000 ft 2,000 ft
The phasing and demand flow rates for each intersection are shown in
Exhibit 29-9. Leading protected phases are provided for all protected left turns.
Intersections 1 and 5 have protected phases for all left turns. Intersections 2 and 4
have only permitted left turns. Intersection 3 has protected left turns on the
major street and permitted left turns on the minor street.
Peak 15-min Exhibit 29-9

Int. Adjusted Demand Demand Flow Rates and
No. Movement Left Through Right Phasing Plan Phasing Plan for Each
Intersection
Major st. 120 800 80
1
Minor st. 120 600 80
Major st. 80 800 120

2
Minor st. 80 600 120
Major st. 120 800 80

3
Minor st. 80 600 120
Major st. 80 800 120

4
Minor st. 80 600 120
Major st. 120 800 80

5
Minor st. 120 600 80
To simplify the discussion, the examples will focus on design and analysis
features that are beyond the stated limitations of the urban street analysis
procedures contained in Chapters 16 through 22. For example, pretimed control
will be assumed here because the ability to deal with traffic-actuated control is
not a limitation of the Chapter 19 signalized intersection analysis methodology.
For the same reason, the analysis of complex phasing schemes that fall within the
scope of the Chapter 19 procedures (e.g., protected-permitted phasing) will be
avoided. Parameters that influence the saturation flow rate (e.g., trucks, grade,
lane width, parking) will not be considered here because they are accommodated
in other chapters.
Chapter 29/Urban Street Facilities: Supplemental Use of Alternative Tools

A symmetrical demand volume pattern will be used to facilitate

interpretation of results. The demand volumes are assumed to be peak-hour
adjusted. Fixed yellow-change and red-clearance intervals of 4 s and 1 s,
respectively, will be assigned to all phases. Through-traffic phases and protected
left-turn phases will be assigned minimum green times of 10 s and 8 s,
respectively.
SIGNAL TIMING PLAN DESIGN

The methodologies in the HCM were developed to determine the
performance of a roadway segment under specific conditions. In simple cases,
the procedures may be applied in reverse for design purposes (e.g., determining
the number of required lanes). In more complex situations requiring
optimization of design parameters, the procedures must be applied iteratively
within an external software structure. Some alternative tools provide this type of
optimization structure and therefore offer a valuable extension of the HCM
methodologies. The extent of HCM compatibility varies among tools.
Two deterministic optimization tools are applied in this section. Each tool is
used to illustrate a different approach for producing the signal timing
parameters required by the procedures of Chapters 18 and 19. This discussion is
not intended as a comprehensive tutorial on signal timing plan design (STPD). A
more detailed treatment of this subject is available (3), which serves as a
comprehensive guide to traffic signal timing and includes a discussion of the use
of deterministic optimization tools. It represents a synthesis of traffic signal
timing concepts and their application and focuses on the use of detection, related
timing parameters, and effects on users at the intersection.
Deterministic STPD Tools

Several deterministic plan design tools are available commercially. Each tool
represents a comprehensive package with its own computational and interface
features. A typical tool configuration is illustrated in Exhibit 29-10. The following
elements are included in the configuration:
• The computational model, which performs the design, optimization, and
analysis functions. Two components are included in the computational
model. The first computes performance measures on the basis of specified
input data and operating parameters. The second contains the
optimization routines that seek a combination of operating parameters
that will produce the best performance.
• The data input editor, which organizes and facilitates the entry of traffic
data and operating parameters to be supplied to the computational
model. The data input editor establishes the “look and feel” of each tool.
The details vary considerably among tools. For example, some tools offer
the ability to compute saturation flow rates internally by using
procedures similar to those prescribed in Chapter 19, Signalized
Intersections.

• Import/export features, which facilitate communication of datasets

between other applications and devices. These features are intended to
enhance the productivity of each tool.
• Direct links to other applications, such as microscopic simulation tools
and fully HCM-compliant software.
• Graphic displays, which provide insight into time–space relationships,
queuing, and platoon propagation.
HCM Simulation Tools Exhibit 29-10

Procedures Elements of a Typical Signal
Timing Design Tool
Import/Export
Data Input
Editor Computational Graphic Displays
Model
The urban streets analysis procedures presented in the HCM deal with the
operation of an urban street facility as a set of interconnected segments. Most of
the commonly used STPD tools are configured to accommodate traffic control
networks involving multiple intersecting routes. To simplify the discussion, the
example presented here is limited to a single arterial route that will be analyzed
as a system.
Two widely used STPD tools will be applied to this example to illustrate
their features and to show how they can be used to supplement the urban street
facilities analysis procedures prescribed in this manual. Both tools are
commercially available software products. More information about these tools
can be found elsewhere (4, 5). The discussion in this section deals with the
combination of features available from both tools without reference to a specific
tool.
Both STPD tools deal with performance measures that are computed by the
procedures prescribed in this manual in addition to performance measures that
are beyond the scope of those procedures. The performance measures covered in
Chapters 16 and 18 include delay, stops, average speed, and queue length. The
discussion of those measures in this section will focus on their use in STPD and
not on comparison of the values computed by different methods.
Several other measures beyond the scope of the HCM methodologies are
commonly associated with signal timing plan design and evaluation. The
following measures are derived from analysis of travel characteristics, including
stops, delay, and queuing:

• Fuel consumption (gal/h), the amount of fuel consumed because of vehicle

miles traveled, stops, and delay, as computed by a model specific to each
tool;
• Operating cost ($/h), the total cost of operation of all vehicles as computed
by a model specific to each tool; and
• Time jammed, the percentage of time that the queue on a link has backed
up beyond the link limit.
STPD tools also deal with a set of performance measures related to the
quality of progression between intersections. These measures, all of which are
outside of the HCM scope, have been defined in the literature or by developers
of specific tools as follows:
• Bandwidth is defined by the number of seconds during which vehicles
traveling at the design speed will be able to progress through a set of
intersections. Link bandwidth is the width of the progression band (in
seconds) passing between adjacent intersections that define the link.
Arterial bandwidth is the width of the progression band that travels the
entire length of the arterial route.
• Progression efficiency is the ratio of the arterial bandwidth to the cycle
length. It thus represents the proportion of the cycle that contains the
arterial progression band. Suggested upper limits for “poor,” “fair,” and
“good” progression are 0.12, 0.24, and 0.36, respectively (5). Values above
0.36 are characterized as “great” progression.
• Progression attainability is the ratio of the arterial bandwidth to the shortest
green time for arterial through traffic on the route. By definition, the
arterial progression band cannot be greater than the shortest green time.
Therefore, an attainability of 100% indicates that further improvement is
only possible through the provision of additional green time. The need for
fine-tuning is suggested for attainability values between 70% and 99%,
with major changes needed for values below 70% (5).
• Progression opportunities (PROS) are a measure of arterial progression
quality that recognizes progression bands that are continuous between
two or more consecutive links but do not travel the full length of the
arterial. The number of PROS observed by a driver at any point in time
and space is defined by the number of intersections that lie ahead within
the progression band. The concept is based on the premise that driver
perception of progression quality increases with the number of
consecutive links that can be traversed within the progression band. The
measure is accumulated in a manner similar to the score in a game of
bowling, where success in one frame is passed on to the next frame to
increase the total score if the success continues. More detailed information
on the computation of PROS is available elsewhere (5).
• Interference is expressed as the percentage of time that an arterial through
vehicle entering a link on the green signal and traveling at the design
speed will be stopped at the next signal. This measure is arguably an
indication of poor perceived progression quality (5).

• Dilemma zone vehicles indicates the number of vehicles arriving on the

yellow interval. Thus, it offers a potential safety-related measure. The
computational details are described elsewhere (4).
• The coordinatability factor (CF), while it is not strictly a performance
measure as defined in this manual, is a measure of the desirability of
coordinating two intersections on the basis of several factors including
intersection spacing, speeds, and platoon formation. It is expressed as a
relative value between 0 and 100. This measure is described in more detail
elsewhere (4), where it is suggested that values above 80 indicate a
definite need for coordination.
Initial Timing Plan Design

An initial timing plan design will first be performed by using one of the
STPD tools. From the list of performance measures just discussed, fuel
consumption will be chosen in this example as the performance measure for
optimization. Other measures or combinations of measures could have been
selected. No recommendation is implied in the selection of this particular
measure. It serves this discussion because it supports an analysis of the trade-off
between other measures such as stops and delay.
A cycle length within a specified range must be selected first. Minimum and
maximum cycle lengths of 80 and 120 s, respectively, will be used. The cycle
optimization results are presented in Exhibit 29-11, which shows the effect of the
cycle length on delay, stops, and fuel consumption as computed by the STPD.
While delay and stops move in opposite directions, their combined effect
suggests that the minimum fuel consumption will be reached with an 80-s cycle.
This is not surprising because it is generally recognized that the optimal cycle
length for balanced progression is twice the link travel time at the design speed,
which is 2 × 34 = 68 s for a 2,000-ft link at 40 mi/h. However, 68 s is below the
minimum cycle length constraint. On the basis of these results, an 80-s cycle will
be selected for optimization of the other timing plan parameters.
81 35 Exhibit 29-11
80 30 Cycle Length Optimization
79 25 Results
Delay (s/veh)
Stops (%)
20
78
15
77
10
76 5
75 0
70 80 90 100 110 120 130 70 80 90 100 110 120 130
Cycle Length (s) Cycle Length (s)
(a) Stops Optimization (b) Delay Optimization
420
Fuel Consumption (gal/h)
415
410
405
400
395
390
385
70 80 90 100 110 120 130
Cycle Length (s)
(c) Fuel Consumption Optimization

The split and offset optimization was carried out next. The resulting timing
plan is shown in Exhibit 29-12. This table represents the initial timing plan to be
investigated and refined.
Exhibit 29-12 Intersection Offset Phase 1 Phase 2 Phase 3 Phase 4 Total

Timing Plan Developed by 1 0 13 29 13 25 80
Split and Offset Optimization 2 34 45 35 80
3 3 13 33 34 80
4 31 45 35 80
5 78 13 29 13 25 80
Notes: All times are in seconds.
Offsets are referenced to the first arterial through-traffic phase.
Initial Timing Plan Performance

A summary of the performance measures for the initial timing plan is
presented in Exhibit 29-13. Separate columns are included in this table for route
totals, which include only the segments that make up the urban street facility as
defined in Chapter 16, and system totals, which include the measures from the
cross-street segments. Note that some of the performance measures reported in
this table are also reported by the Chapter 16 methodology. While the STPD tool
definitions and model structures are similar to the HCM (e.g., uniform and
random components), no comparison of the values will be offered in this
discussion because the focus is on the STPD and not on modeling differences.
Exhibit 29-13 System Route

Performance Measures for the Performance Measure Units Totals Totals
Initial Timing Plan Total travel veh-mi/h 4,927 3,063
Total travel time veh-h/h 240 120
Uniform delay veh-h/h 95 34
Random delay veh-h/h 22 8
Total delay veh-h/h 116 43
Average delay s/veh 23.5 17.4
Passenger delay p-h/h 140 51
Uniform stops veh/h 12,893 5,576
Uniform stops % 72 63
Random stops veh/h 1,277 440
Random stops % 7 5
Total stops veh/h 14,171 6,016
Total stops % 79 68
Links with d/c >1 0 0
Links with queue overflow 0 0
Time jammed % 0 0
Period length s 900 900
System speed mi/h 20.5 25.6
Fuel consumption gal/h 387 195
Operating cost $/h 3,063 1,049
The initial timing plan design was based on minimizing fuel consumption as
a performance measure. The signal progression characteristics of this design are
also of interest. The progression characteristics will be examined in both
numerical and graphics representations. The numbers are presented in Exhibit
29-14 and are based on the progression performance measures that were defined
earlier. The interference values indicate the proportion of time that a vehicle
entering a link in the progression band would be stopped at the next signal. The
PROS are accumulated from progression bands that pass through some adjacent
signals along the route. The low progression efficiency and attainability and

PROS values suggest that this design, while optimal in some respects, would not
produce a very favorable motorist perception of progression quality.
Performance Measure Westbound Eastbound Average Exhibit 29-14

Bandwidth efficiency 10% 5% 8% Progression Quality Measures
Progression attainability 28% 14% 21% for the Initial Design
Interference 9% 10%
PROS 30% 28% 29%
Adjustments to Improve Progression Quality

Because of the low quality of progression, it is logical to revisit the initial
design with the objective of maximizing progression quality instead of
minimizing fuel consumption. The same cycle length range (80 to 120 s) was
used for this purpose, and the runs were repeated with the objective of
maximizing PROS. The maximum value of PROS was obtained with the same
cycle length and phase times as the initial design. The progression performance
measures associated with this timing plan are shown in Exhibit 29-15. These
measures do not differ substantially from the initial design, nor do the offsets.
The total PROS value increased from 29% to 30%, but the performance was
somewhat better balanced by direction. Thus, there is not a large trade-off
between the objectives of maximizing performance and maximizing progression
quality in this case.
A combination of factors peculiar to this example has led to the conclusion
that the signal timing parameters for optimizing performance and progression
are basically the same. The symmetry of the layout and phasing created a
situation in which fuel consumption could be minimized by favoring either
direction at the expense of the other. The balanced design was favored by the
PROS optimization because it offered a minimal numerical advantage (30%
versus 29%). One of the main reasons why both design approaches chose the
lowest acceptable cycle length is that, as pointed out previously, the theoretical
optimum cycle length was below the lowest acceptable cycle length.
Performance Measure Westbound Eastbound Average Exhibit 29-15

Bandwidth efficiency 8% 8% 8% Progression Quality Measures
Attainability 21% 21% 21% for the Improved Progression
Interference 9% 9% Design
PROS 30% 30% 30%
Time–Space Diagrams
STPD tools typically produce graphic displays depicting progression
characteristics. The most common display is the time–space diagram, which is
well documented in the literature and understood by all practitioners. The time–
space diagram reflecting the initial design is shown in Exhibit 29-16. Note that,
even though the traffic volumes are balanced in both directions, the design
appears to favor the westbound (right-to-left) direction. Because of the symmetry
of this example, a dual solution that yields the same performance but that favors
the eastbound direction is likely to exist.

Exhibit 29-16
Time–Space Diagram for the
Initial Design
The time–space diagram depicting the modified progression design is shown

in Exhibit 29-17. This design shows a better balance between the eastbound and
westbound directions. There is good progression into the system from both ends,
but the band in both directions is halted at the center intersection. The PROS
accumulation is evident in the bands that progress between some of the
intersections.
Exhibit 29-17
Modified Progression Design
The difference between the initial and modified designs appears to be

minimal. The modified design will be chosen for further investigation because it
offers a better balance between the two directions. The offset changes for this
design are presented in Exhibit 29-18.
Exhibit 29-18 Initial Revised

Offset Changes for the Intersection Offsets Offsets
Modified Progression Design 1 0 0
2 34 30
3 3 76
4 31 30
5 78 0
The time–space diagram for this operation from another STPD tool is shown
in Exhibit 29-19. The timing plan is the same as the plan that was depicted in
Exhibit 29-17, but the format of the display differs slightly. Both the link band
and the arterial band as defined previously are shown on this display. The
individual signal phases are also depicted. Both types of time–space diagrams
offer a manual adjustment feature whereby the offsets may be changed by
dragging the signal display back and forth on the monitor screen.

Exhibit 29-19
Alternative Time–Space
Diagram Format
Other Graphic Displays

Other graphics formats are not as ubiquitous as the time–space diagram but
can provide useful insights into the operation at and between intersections.
Flow Profile Diagrams

One example is the flow profile diagram, which is simply a plot of the flow
rate over one complete cycle. Flow profiles may be created to portray either the
arrival or the departure flows at a stop line.
An example illustrating the use of flow profiles is presented in Exhibit 29-20.
The eastbound segment between the first and second intersections is depicted in
this example. The traffic inputs to this segment come from three independent
movements at Intersection 1: southbound left, eastbound through, and
northbound right.
Four stages of the progress of traffic into and out of this segment are
depicted in the exhibit:
1. Uniform arrivals on external links: Each of the three movements entering the
segment will arrive with a flow profile that is constant throughout the cycle
because of the absence of platoon-forming phenomena on external links.
2. Departures on the green signal: Each movement proceeds on a different
phase and therefore enters the link at a different time.
3. Propagation on the segment with platoon dispersion: Each of the three
movements will be propagated downstream to the next signal by using a
model that applies the design speed and incorporates platoon dispersion.
Arrival of the platoons at the downstream end of the segment: The
composite arrival profile is illustrated in the figure. The profile represents
the sum of all of the movements entering the link.
4. Departure on the green signal: The platoons are regrouped at this point into
a new flow profile because of the effect of the signal. The extent of

regrouping will depend on the proportion of time that the signal is green.
If a continuous green signal were displayed, the output flow profile
would match the input flow profile exactly.
Exhibit 29-20
Southbound Left Eastbound Through Northbound Right
Example Illustrating the Use
of Flow Profiles
Stage 1: Uniform arrivals on

external links
Stage 2: Departures
on green signal
Stage 3: Propagation on the segment with platoon dispersion
Combined arrival profile at

the next downstream
signal
Stage 4: Departure profile

on the green signal
The departure profile for this movement forms one input to the next link and
is therefore equivalent to Stage 2 in the list above. The vehicles entering on
different phases from the cross street must be added to this movement to form
the input to the next segment as the process repeats itself throughout the facility.
The preceding description of the accumulation, discharge, and propagation
characteristics of flow profiles is of special interest to this discussion because the
same models used by the STPD tool have been adopted by the analysis

procedures given in Chapter 18, Urban Street Segments. These procedures are
described by Exhibit 30-3 through Exhibit 30-5 in Chapter 30, Urban Street
Segments: Supplemental. Therefore, the graphical representations given in
Exhibit 29-20 should be useful in facilitating understanding of the procedures
prescribed in Chapter 18.
Composite Flow Profiles

Another form of flow profile graphics is illustrated in Exhibit 29-21. This
text-based display offers a composite view of the flow profiles by showing the
arrival and departure graphics on the same figure represented by different
characters. The uniform arrival pattern from the external link is evident at the
upstream intersection, which corresponds to Stages 1 and 2 of Exhibit 29-20. The
effect of the platooned arrivals is also evident at the downstream intersection,
corresponding to Stages 4 and 5. More details on interpreting the composite flow
profiles are given elsewhere (5).
Exhibit 29-21
Composite Flow Profiles for
the First Eastbound Segment
(a) Upstream Intersection (Uniform Arrivals) (b) Downstream Intersection (Platooned Arrivals)
Queue Length Graphics

The accumulation and discharge of queues can also be represented
graphically in a manner that is consistent with the analysis procedures of
Chapters 16 through 19. An example of graphics depicting the queue length
throughout the cycle is presented in Exhibit 29-22. The upstream signal shows
the familiar triangular shape that is the basis of the uniform delay equation. The
downstream signal shows the effect of platooned arrivals on the length of the
queue.
Exhibit 29-22
Variation of Queue Length
Throughout the Signal Cycle
for the First Eastbound
Segment
(a) Upstream Intersection Queue Length (b) Downstream Intersection

(Uniform Arrivals) (Platooned Arrivals)

Adding Flows and Queues to the Time–Space Diagram

One useful display superimposes the flow profiles and queuing
characteristics on the time–space diagram to give a complete picture of the
operation of the facility. An example of this display representing the improved
progression design is presented in Exhibit 29-23. The flow rates are represented
by the density of the lines progressing between intersections at the design speed.
The queues are represented by horizontal lines upstream of each intersection.
From this diagram, the effect of the design on queue accumulation and discharge
and on the propagation of flows between intersections can be visualized.
Exhibit 29-23
Time–Space Diagram with
Flows and Queues
Potential Improvements from Phasing Optimization

The quality of progression in this example was improved from the initial
design, but the results leave room for further improvement. For example, there
are minimal arterial through bands. The current design was based on leading
phases for all protected left turns. The operation might be improved by the
application of lagging left-turn phases on some approaches. The procedures
given in Chapter 18 are sensitive to the phase order. These procedures could be
applied manually to seek a better operation. The use of STPD tools for this
purpose will be demonstrated here because phasing optimization is internalized
in the tools as a computational feature.

The phasing optimization process recommended changes at two of the five

intersections. The phasing modifications are shown in Exhibit 29-24. Lead-lag
phasing was applied at both intersections. As a result of the optimization, the
arterial bandwidth increased from 6 to 16 s in both directions. The total signal
delay decreased from 220 to 200 s/veh. The arterial speed increased from 22.1 to
23.0 mi/h. Thus, the phasing optimization would improve both the progression
quality and the operational performance of the route. The progression quality
improvement is evident in the time–space diagram presented in Exhibit 29-25.
Original Phasing Optimized Phasing Exhibit 29-24

Optimized Phasing
Modifications
Intersection 1
Intersection 5
Exhibit 29-25
Optimized Phasing Plan
The decision to implement lead-lag phasing involves many factors including

safety and local preferences. This discussion has been limited to a demonstration
of how STPD tools can be used in the assessment of the operational effects of
phasing optimization as one input to the decision process. The suggested
modifications will not be implemented in the balance of the examples.

DEMONSTRATION OF ALTERNATIVE TOOL APPLICATIONS

Effect of Midsegment Parking Activities
The HCM methodology in Chapter 18 recognizes midsegment activities such
as cross-street entry between signals and access point density. A procedure is
provided in the methodology for estimating the delay due to vehicles turning left
or right into an access point approach. However, no procedures are included for
estimating the delay or stops due to other causes such as pedestrian interference
and parking maneuvers. Alternative tools must be used to assess these effects.
This section will demonstrate the use of a typical microscopic simulation tool
(6) to assess the effects of midsegment parking maneuvers on the performance of
an urban street facility. The signal timing plan example from the previous section
will be used for this purpose. The offsets will be modified first to create “ideal”
progression in the eastbound direction at the expense of the westbound flow.
The investigation will focus on the eastbound flow. The offsets and time–space
diagram depicting this operation are shown in Exhibit 29-26. Offset 1 is
referenced to the first phase for arterial through movements. Offset 2 is
referenced to Phase 1. Their values will differ because of leading left-turn phases
at some intersections. Different tools require different offset references.
Exhibit 29-26
Time–Space Diagram Showing
Ideal Eastbound Progression
Offset Offset
Signal 1 2
1 0 0
2 35 47
3 63 68
4 23 35
5 57 56
The treatment of parking maneuvers by the selected simulation tool is

described in the tool’s user guide (6). The following parameters must be supplied
for each segment that contains on-street parking spaces:
• Beginning of the parking area with respect to the downstream end of the
segment,

• Length of the parking area,

• Mean duration of a parking maneuver, and
• Mean frequency of parking maneuvers.
The occurrence and duration of parking maneuvers are randomized around
their specified mean values. The parameters that will be used in this example are
Parameter Value Exhibit 29-27

Beginning of the parking area 200 ft from the downstream intersection Parameters for the Parking
Example
Length of the parking area 1,600 ft (leaving 200 ft to the upstream
intersection)
Mean duration of a parking maneuver 30 s
Mean frequency of parking maneuvers 0 veh/h (no parking maneuvers)
60 veh/h
120 veh/h
180 veh/h
240 veh/h
Represents a range of approximately 15 min to
60 min average parking duration
The simulation runs covered 80 cycles of operation. Separate runs were made
for each level of parking frequency. The default simulation parameters of the
selected tool were used.
The effect of the parking activity on travel time and delay is presented in
Exhibit 29-28, which shows the total travel time for the facility as well as the two
delay components of travel time (total delay and control delay). Each of the
values represents the sum of the individual segment values. The graphs
demonstrate that all of the relationships were more or less linear with respect to
the parking activity level.
350 Exhibit 29-28

Effect of Parking Activity Level
on Travel Time and Delay
300
250
Delay Time (s/veh)
200
150
100
50
0
0 50 100 150 200 250
Parking Maneuvers per Hour
Total Travel Time Total Delay Control Delay

The effect of the parking activity on stops is presented in Exhibit 29-29. For
this example, the average percentage of stops for all eastbound vehicles increased
from slightly more than 40% to slightly less than 60% throughout the range of
parking activity levels. Both of these exhibits indicate that the simulation tool
was able to extend the capability for analysis of urban street facilities beyond the
stated limitations of the methodology presented in Chapter 16.
Exhibit 29-29 70
Effect of Parking Activity Level
on the Percentage of Stops
60
50
40
Stops (%)
30
20
10
0
0 50 100 150 200 250
Parking Maneuvers per Hour
Effect of Platooned Arrivals at a Roundabout

Chapter 22, Roundabouts, describes a methodology for analyzing the
operation of an isolated roundabout. Section 9 of Chapter 30, Urban Street
Segments: Supplemental, describes a methodology for analyzing the operation of
street segments bounded by roundabouts. Neither methodology explicitly
accounts for the effect that platooned arrivals from a signal may have on
roundabout operational performance. Therefore, the analysis of a roundabout as
a part of a coordinated traffic control system is likely better accomplished with
alternative tools. The alternative deterministic tools described earlier in this
section do not deal explicitly with roundabouts in coordinated systems. Most
simulation tools offer some roundabout modeling capability, although the level
of modeling detail varies among tools.
This subsection describes the use of a typical simulation tool (7) in analyzing
a roundabout within the arterial configuration of the previous example in this
section. For this purpose, Intersection 3 at the center of the system will be
converted to a roundabout with two lanes on each approach. To simplify the
discussion, a basic symmetrical configuration will be used, because the
discussion will be limited to the effect of platooned arrivals on the operation. The
design aspects of roundabouts are covered in Chapter 22, Roundabouts, with
more details provided in Chapter 33, Roundabouts: Supplemental, and
elsewhere (8). The default traffic modeling parameters of the simulation tool will
be applied. The roundabout configuration is shown schematically in Exhibit 29-30.

Exhibit 29-30
Roundabout Configuration for
Intersection 3
This example will examine two STPDs that create substantially different
platoon arrival characteristics on the arterial approaches to the roundabout. The
time–space diagrams representing the two designs are shown in Exhibit 29-31.
The first design provides simultaneous arrival of the arterial platoons from both
directions. The second creates a situation in which one platoon will arrive in the
first half of the cycle and the other will arrive during the second half. The two
cases will be described as “simultaneous” and “alternating” platoon arrivals.
Exhibit 29-31
Time–Space Diagrams
Showing Simultaneous and
Alternating Platoon Arrivals at
the Roundabout
(a) Simultaneous Platoon Arrivals (b) Alternating Platoon Arrivals
The platoon arrival characteristics can only be expected to influence the

operation of a roundabout with relatively free-flowing traffic. While a two-lane
roundabout could accommodate the demand volumes used in the previous
examples in which the intersection was signalized, the initial simulation runs
indicated enough queuing on all approaches to obscure the effect of the
progression design. Since the focus of this example is on the effect of the adjacent
signal timing plan, the demand volumes on the cross-street approaches to the
roundabout will be reduced by 100 veh/h (approximately 17%) to provide a
better demonstration of that effect.

Ten simulation runs were performed for both progression designs, and the
average values of the performance measures were used to compare the two
designs. The performance measures illustrated in Exhibit 29-32 include delay and
stops on all approaches to the roundabout and travel times on individual link
segments and on the route as a whole.
Exhibit 29-32 Movement Alternating Simultaneous Difference Percent

Performance Comparison for Delay
Simultaneous and Alternating
Major-street approaches 15.81 14.18 1.64 10.34
Platoon Arrivals at a
Minor-street approaches 19.36 19.88 –0.52 –2.69
Roundabout
Stops
Major-street approaches 0.59 0.52 0.08 12.71
Minor-street approaches 0.88 0.89 –0.01 –0.57
Average Travel Times
Through vehicles traveling
250.60 237.74 12.86 5.13
the full route
Approach links 58.06 56.30 1.76 3.03
Exit links 50.76 45.66 5.10 10.05
As a general observation, the simultaneous design performed noticeably

better than did the alternating design on the major street, with a slight
degradation to the cross-street performance. Travel times for vehicles traveling
the full length of the facility were improved by about 5%. Travel times on the
arterial segments entering and leaving the roundabout were improved by 3%
and 10%, respectively.
This example has demonstrated that the simulation tool was able to describe
the effect of two signal progression schemes on the performance of a roundabout
within a coordinated arterial signal system. The next example will deal with the
same basic arterial layout except that the roundabout will be replaced by a two-
way STOP-controlled (TWSC) intersection. The platoon arrival types can be
expected to have a greater influence on the TWSC operation than the roundabout
because the effect is much more direct. Major-street vehicles always have the
right-of-way over minor-street vehicles. Simultaneous platoons arriving from
both directions will provide more opportunity for gaps in the major-street flow.
Alternating platoons will keep major-street vehicles in the intersection for a
greater proportion of time, thereby restricting cross-street access.
The effect at a roundabout is much more subtle because minor-street vehicles
have the right-of-way over major-street vehicles once they have entered the
roundabout. With simultaneous arrivals, platoons from opposite directions assist
each other by keeping the minor-street vehicles from entering and seizing control
of the roadway. When there is no traffic from the opposite direction, as in the
case of alternating arrivals, a major-street movement is more likely to encounter
minor-street vehicles within the roundabout. This phenomenon explains the 10%
improvement in performance for simultaneous arrivals in the roundabout
example as indicated in Exhibit 29-32.

Queue Length Analysis Based on Vehicle Trajectories

The HCM’s segment-based chapters provide deterministic procedures for
estimating the extent of queue backup on either signalized or unsignalized
approaches. Most of the procedures are sensitive to some degree to platoon
formation from adjacent signals. Most provide estimates of the average back of
queue (BOQ) and the expected BOQ at some level of probability.
One additional queuing measure that can be derived from simulation is the
proportion of time that the BOQ might be expected to extend beyond a specified
point. This measure can be obtained directly from the analysis of individual
vehicle trajectories by using the procedures set forth in Chapter 7, Interpreting
HCM and Alternative Tool Results, and Chapter 36, Concepts: Supplemental.
Those procedures will be applied in this example to examine the queuing
characteristics on the minor-street approach to a TWSC intersection operating
within a signalized arterial system. The criteria and procedures prescribed in
Chapter 36 for identifying the onset and release from the queued state will be used.
The same urban street configuration will be used for this purpose. The center
intersection that was converted to a roundabout in the previous example will
now be converted to TWSC. Because of the unique characteristics of TWSC, a few
changes will have to be made to the configuration. TWSC capacities are lower
than those of signals or roundabouts, so the minor-street demand volumes will
have to be reduced. The two-lane approaches will be preserved, but the
additional left-turn bay will be eliminated. The same two platoon arrival
configurations (simultaneous and alternating) will be examined to determine
their effect on the minor-street queuing characteristics. The signal timing plans
from the roundabout example, as illustrated in Exhibit 29-31, will also be used
here. Twelve cycles covering 960 s will be simulated for each case to be
examined, and the individual vehicle trajectories will be recorded.
Queuing Characteristics
The first part of this example will demonstrate TWSC operation with an
idealized scenario to provide a starting point for more practical examples. Two
intersecting streams of through movements with completely uniform
characteristics will be simulated. As many of the stochastic features of the
simulation model as possible will be disabled. This is a highly theoretical
situation with no real practical applications in the field. Its purpose is to provide
a baseline for comparison.
The formation of queues under these conditions is illustrated in Exhibit 29-
33, which shows the instantaneous BOQ for all time steps in the simulation. The
cross-street entry volume was 600 veh/h in each direction, representing
approximately the capacity of the approach. The cyclical operation is evident
here, with 12 discernible cycles observed. Each cycle has a similar appearance.
The differences among cycles are due to embedded stochastic features that could
not be disabled.

Exhibit 29-33 240
Queuing Results for the 220

Theoretical Example
200
180
160
Back of Queue (ft)

140
120
100
80
60
40
20
0
1 61 121 181 241 301 361 421 481 541 601 661 721 781 841 901 961
Time Step (s)
The signal timing plan with simultaneous platoon arrivals should produce
the most cyclical operation that could actually be observed in the field. This
configuration was simulated by loading the minor street to near capacity levels
as determined experimentally. The entry volume was 350 veh/h.
The queuing results are shown in Exhibit 29-34. Some cyclical characteristics
are still evident here, but they are considerably diminished from the idealized
case. The loss of cyclical characteristics results from cross-street turning
movements entering the segments at their upstream intersections and from the
general stochastic nature of simulation modeling.
Exhibit 29-34 240

Queuing Results for
Simultaneous Platoons 220
200
180
160
Back of Queue (ft)
140
120
100
80
60
40
20
0
1 61 121 181 241 301 361 421 481 541 601 661 721 781 841 901 961
Time Step (s)

The operation was simulated next with alternating platoon arrivals. Again
the demand volumes were set to the experimentally determined approach
capacity, which was 270 veh/h, or about 25% lower than the capacity with
simultaneous platoons. The results are presented in Exhibit 29-35. Some further
loss of cyclical properties due to the spreading of entry opportunities across a
greater proportion of the cycle is observed here.
240 Exhibit 29-35

Queuing Results for
220 Alternating Platoons
200
180
160
Back of Queue (ft)
140
120
100
80
60
40
20
0
1 61 121 181 241 301 361 421 481 541 601 661 721 781 841 901 961
Time Step (s)
The least cyclical characteristics would be expected from simulation of a

completely isolated operation. The 2,000-ft link lengths were retained for this
case, but no adjacent intersections existed. All other parameters remained the
same, including the entry volume because the entry capacity for isolated
operation was found to be the same as the case with alternating platoons.
The results are presented in Exhibit 29-36. There are no cyclical
characteristics here because there is no underlying cycle in the operation. Also,
even with the same entry volume as the alternating platoon case, the peak BOQs
are much lower. This is because the entry opportunities are distributed randomly
in time instead of being concentrated at specific points in the cycle.

Exhibit 29-36 240

Queuing Results for Isolated
220
TWSC Operation
200
180
160
Back of Queue (ft)

140
120
100
80
60
40
20
0
1 61 121 181 241 301 361 421 481 541 601 661 721 781 841 901
Time Step (s)
Back-of-Queue Assessment
The discussion to this point has focused on instantaneous BOQs in an effort
to understand the general nature of queuing under the conditions that were
examined. With knowledge of the instantaneous BOQ values available from
simulation, useful performance measures related to queuing can be produced
from simulation. One such measure is the proportion of time that a queue would
be expected to back up beyond a specified point. This concept is different from
the probability of backup to that point normally associated with deterministic
tools. The balance of the discussion will deal with the proportion of time with
queue backup (PTQB) beyond a specified point.
The three cases examined in this example were simulated with cross-street
demand volumes of 80, 160, 240, 320, and 400 veh/h, and the PTQB
characteristics were determined by simulation for each case. The results were
plotted for a specified distance of 100 ft from the stop line as shown in Exhibit 29-
37. Each case is represented by a separate line that shows the percentage of time
that the queue would be expected to back up beyond 100 ft from the stop line for
each cross-street entry volume level. The simultaneous platoon case showed the
lowest BOQ levels, starting with no time with BOQ beyond 100 ft below 240
veh/h, and reached a value of nearly 90% of the time at the maximum volume of
400 veh/h. Predictably, the isolated case was the most susceptible to queue
backup, and the alternating platoon case fell somewhere in between.

100 Exhibit 29-37

Effect of Cross-Street Demand
90 Volume on Queue Backup
Beyond 100 ft from the Stop
80
Line
Time with BOQ > 100 ft (%)
70
60
50
40
30
20
10
0
0 50 100 150 200 250 300 350 400
Cross-Street Demand Volume (veh/h)
Simultaneous Alternating Isolated
This example has demonstrated the use of simulation to produce potentially

useful queuing measures based on the analysis of individual vehicle trajectories.
It has also demonstrated how simulation can be used to assess the queuing
characteristics of a minor-street approach to a TWSC intersection operating in a
coordinated signal environment.

5. EXAMPLE PROBLEMS
This section describes the application of the motorized vehicle, pedestrian,

bicycle, and transit methodologies through a series of example problems. Exhibit
29-38 provides an overview of these problems. The focus of the examples is to
illustrate the multimodal facility evaluation process. An operational analysis
level is used for all examples. The planning and preliminary engineering analysis
level is identical to the operational analysis level in terms of the calculations
except that default values are used when field-measured values are not available.
Exhibit 29-38 Problem Analysis

Example Problems Number Description Level
1 Automobile-oriented urban street Operational
2 Widen the sidewalks and add bicycle lanes on both sides of facility Operational
3 Widen the sidewalks and add parking on both sides of facility Operational
4 Urban street reliability under existing conditions Operational
5 Urban street reliability strategy evaluation Planning
EXAMPLE PROBLEM 1: AUTOMOBILE-ORIENTED URBAN STREET

The Urban Street Facility
A 1-mi urban street facility is shown in Exhibit 29-39. It is located in a
downtown area and oriented in an east–west travel direction. The facility
consists of five segments with a signalized boundary intersection for each
segment. Segments 1, 2, and 3 are 1,320 ft long and have a speed limit of 35 mi/h.
Segments 4 and 5 are 660 ft long and have a speed limit of 30 mi/h. Each segment
has two active access point intersections.
Exhibit 29-39
Second Ave.
Fourth Ave.
Example Problem 1: Urban
Sixth Ave.
Fifth Ave.
Third Ave
First Ave.
Street Schematic
Segment 1 Segment 2 Segment 3 Seg. 4 Seg. 5

35 mi/h 35 mi/h 35 mi/h 30 mi/h 30 mi/h N
1,320 1,320 1,320 660 ft 660 ft
1 2 3 4 5 6
Access points
Signal Signal Signal Signal Signal Signal
Offset 0 s Offset 50 s Offset 50 s Offset 0 s Offset 0 s Offset 0 s
Segments 1, 2, and 3 pass through a mixture of office and strip commercial.

Segments 4 and 5 are in a built-up shopping area.
The geometry of the typical street segment is shown in Exhibit 29-40. It is the
same for each segment. The street has a curbed, four-lane cross section with two
lanes in each direction. There is a 1.5-ft curb-and-gutter section on each side of
the street. There are 200-ft left-turn bays on each approach to each signalized
intersection. Right-turn vehicles share the outside lane with through vehicles on
each intersection approach. A 6-ft sidewalk is provided on each side of the street
Example Problems Chapter 29/Urban Street Facilities: Supplemental

adjacent to the curb. No fixed objects are located along the outside of the
sidewalk. Midsegment pedestrian crossings are legal. No bicycle lanes are
provided on the facility or its cross streets. No parking is allowed along the
street.
Pavement condition rating: 3.5 Exhibit 29-40

N Curbed cross section Example Problem 1: Segment
Cross-street lane width: 12 ft Geometry
Corner radius: 6.0 ft
11 ft 13 ft
11 ft 12 ft
Signal 9 ft Signal
11 ft 12 ft
11 ft 13 ft
Crosswalk width: 12 ft
Total walkway width: 6 ft
Not to scale Buffer: 0 ft
The Question
What are the travel speed and level of service (LOS) of the motorized vehicle,
pedestrian, bicycle, and transit modes in both directions of travel along the
facility?
The Facts
The traffic counts for one segment are shown in Exhibit 29-41. The counts are
the same for all of the other segments. The counts were taken during the 15-min
analysis period of interest. However, they have been converted to hourly flow
rates.
60 480 60 49 48 48 49 60 480 60 Exhibit 29-41

Example Problem 1:
80 39 38 80
640 702 684 640
Intersection Turn Movement
80 38 39 80
640
Signal 80
684
Access Point 39
702
Access Point 38
640
Signal 80 Counts
Intersection Intersection
80 38 39 80
60 480 60 49 48 48 49 60 480 60
The signalization conditions are shown in Exhibit 29-42. The conditions

shown are identified as belonging to Signalized Intersection 1; however, they are
the same for the other signalized intersections (with the exception of offset). The
signals operate with coordinated–actuated control. The left-turn movements on
the northbound and southbound approaches operate under permitted control.
The left-turn movements on the major street operate as protected–permitted in a
lead–lead sequence.
Exhibit 29-42 indicates that the passage time for each phase is 2.0 s. The
minimum green setting is 5 s for the major-street left-turn phases and 18 s for the
cross-street phases. The offset to Phase 2 (the reference phase) end-of-green
interval is 0.0 s. The offset for each of the other intersections is shown in Exhibit
29-39. A fixed-force mode is used to ensure that coordination is maintained. The
cycle length is 100 s.
Chapter 29/Urban Street Facilities: Supplemental Example Problems

Exhibit 29-42
Example Problem 1: Signal
Conditions for Intersection 1
Geometric conditions and traffic characteristics for Signalized Intersection 1

are shown in Exhibit 29-43. They are the same for the other signalized
intersections. The movement numbers follow the numbering convention shown
in Exhibit 19-1 of Chapter 19.
All intersection movements include 3% heavy vehicles. The segment and
intersection approaches are effectively level. No parking is allowed along the
facility or its cross-street approaches. With a few exceptions (discussed below),
local buses stop on the eastbound and westbound approaches to each signalized
intersection at a rate of 3 buses/h.
Arrivals for all cross-street movements are effectively random, so a platoon
ratio of 1.00 is used. The through movement arriving to the eastbound approach
at Intersection 1 exhibits favorable progression from an upstream signal, so a
platoon ratio of 1.33 is used. For similar reasons, a ratio of 1.33 is also used for
the through movement arriving to the westbound approach at Intersection 6.
Right-turn-on-red volume is estimated at 5.0% of the right-turn volume.

Approach Eastbound
Intersection Data Worksheet
Westbound Northbound Southbound
Exhibit 29-43
Movement L T R L T R L T R L T R Example Problem 1:
Movement number
Intersection Geometry
5 2 12 1 6 16 3 8 18 7 4 14
Geometric Conditions and
Number of lanes 1 2 0 1 2 0 1 2 0 1 2 0 Traffic Characteristics for
Lane assignment
Average lane width, ft
L
9.0
TR
11.0
n.a.
0.0
L
9.0
TR
11.0
n.a.
0.0
L
12.0
TR
12.0
n.a. L
12.0
TR
12.0
n.a.
Signalized Intersection 1
Number of receiving lanes 2 2 2 2
Turn bay or segment length, ft 200 0 200 200 1320 200 200 999 200 200 999
Traffic Characteristics
Volume, veh/h 80 640 80 80 640 80 60 480 60 60 480 60
Right-turn-on-red volume, veh/h 4 4 3 3
Percent heavy vehicles, % 3 3 3 3 3 3 3 3 3 3 3 3
Lane utilization adjustment factor 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Peak hour factor 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Start-up lost time, s 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
Extension of eff. green time, s 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
Platoon ratio 1.000 1.333 1.000 1.333 1.000 1.000 1.000 1.000 1.000
Upstream filtering factor 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Pedestrian volume, p/h 100 100 100 100
Bicycle volume, bicycles/h 1 1 1 1
Opposing right-turn lane influence Yes Yes Yes Yes
Initial queue, veh 0 0 0 0 0 0 0 0 0 0 0
Speed limit, mi/h 35 35 35 35 35 35 35 35 35 35 35 35
Unsignalized movement volume, veh/h 0 0 0 0 0 0 0 0 0 0 0 0
Unsignalized movement delay, s/veh 0 0 0 0 0 0 0 0 0 0 0 0
Unsignalized mvmt. stop rate, stops/veh 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Approach Data Left Side Right Side Left Side Right Side Left Side Right Side Left Side Right Side
Parking present? No No No No No No No No
Parking maneuvers, maneuvers/h 0 0 0 0 0 0 0 0
Bus stopping rate, buses/h 3 3 0 0
Approach grade, % 0 0 0 0 0 0 0 0 0 0 0 0
Detection Data
Stop line detector presence Presence Presence No det. Presence Presence No det. Presence Presence Presence Presence Presence No det.
Stop line detector length, ft 40 40 40 40 40 40 40 40 40 40 40 40
Each segment has a barrier curb along the outside of the street in each
direction of travel. With allowance for the upstream signal width, the percentage
of the segment length with curb is estimated at 94% for Segments 1, 2, and 3. It is
estimated as 88% for Segments 4 and 5.
The traffic and lane assignment data for the two access point intersections for
Segment 1 are shown in Exhibit 29-44. These data are the same for the other
segments; however, the access point locations (shown in the first column) are
reduced by one-half for Segments 4 and 5. The movement numbers follow the
numbering convention shown in Exhibit 20-1 of Chapter 20, Two-Way STOP-
Controlled Intersections. There are no turn bays on the segment at the two access
point intersections.
Access Point Input Data
Access Approach Eastbound Westbound Northbound Southbound Exhibit 29-44
Point Movement
Location,ft Movement number
L T R L T R L T R L T R Example Problem 1: Access
1 2 3 4 5 6 7 8 9 10 11 12
440 Volume, veh/h 38 684 38 39 702 39 49 0 48 48 0 49 Point Data
West end Lanes 0 2 0 0 2 0 1 0 1 1 0 1
880 Volume, veh/h 39 702 39 38 684 38 48 0 49 49 0 48
Lanes 0 2 0 0 2 0 1 0 1 1 0 1
A low wall is located along about 25% of the sidewalk in Segments 1, 2, and
3. In contrast, 10% of the sidewalk along Segments 4 and 5 is adjacent to a low
wall, 35% to a building face, and 15% to a window display.
Office and strip commercial activity along Segments 1, 2, and 3 generates a
pedestrian volume of 100 p/h on the adjacent sidewalks and crosswalks.
Shopping activity along Segments 4 and 5 generates a pedestrian volume of
300 p/h on the adjacent sidewalks and crosswalks. A lack of bicycle lanes has
discouraged bicycle traffic on the facility and its cross streets; however, a bicycle
volume of 1.0 bicycle/h is entered for each intersection approach.
Local buses stop on the eastbound and westbound approaches to each
signalized intersection, with the exception of Intersection 5. There are no stops on
either approach to Intersection 5. However, transit stops are provided along the
facility at 0.25-mi intervals, so the service is considered to be local. As a result,

the westbound transit frequency on Segment 5 and the eastbound transit

frequency on Segment 4 are considered to be the same as for the adjacent
segments (i.e., 3 buses/h). The bus dwell time at each stop averages 20 s. Buses
arrive within 5 min of their scheduled time about 75% of the time and have a
load factor of 0.80 passengers/seat. Each bus stop has a bench but no shelter.
Outline of Solution
This section outlines the results of the facility evaluation. To complete this
evaluation, the motorized vehicle, pedestrian, and bicycle methodologies in
Chapter 19 were used to evaluate each of the signalized intersections on the
facility. The procedure in Chapter 20 was used to estimate delay for pedestrians
crossing at a midsegment location. The motorized vehicle, pedestrian, bicycle,
and transit methodologies in Chapter 18 were then used to evaluate both
directions of travel on each segment. Finally, the methodologies described in
Chapter 16 were used to evaluate all four travel modes in both directions of
travel on the facility. The findings from each evaluation are summarized in the
following three subparts.
Intersection Evaluation
The results of the evaluation of Intersection 1 (i.e., First Avenue) are shown
in Exhibit 29-45. The results for Intersections 2, 3, and eastbound Intersection 4
are similar. In contrast, Intersections 5 and 6 are associated with a shorter
segment length, lower speed limit, and higher pedestrian volume, so their
operation is different from that of the other intersections. The results for
Intersection 5 (i.e., Fifth Avenue) are shown in Exhibit 29-46. Intersection 6 and
westbound Intersection 4 have similar results.
Exhibit 29-45 Intersection Approach Eastbound

Intersection Evaluation Summary
Example Problem 1: First Avenue Basic Description
Applicable lane assignments L TR n.a. L TR n.a. L TR n.a. L TR n.a.
Intersection 1 Evaluation Primary movement number 5 2 12 1 6 16 3 8 18 7 4 14
Vehicle volume, veh/h 80 640 80 80 640 80 60 480 60 60 480 60
Conflicting crosswalk volume, p/h 100 100 100 100
Bicycle volume, bicycle/h 1 1 1 1
Approach lanes, ln 1 2 0 1 2 0 1 2 0 1 2 0
Vehicle Level of Service
Int. delay, s/veh Volume-to-capacity ratio 0.17 0.33 0.33 0.15 0.33 0.33 0.37 0.62 0.62 0.37 0.62 0.62
20.4 Control delay, s/veh 7.78 5.75 5.78 7.06 13.43 13.78 43.24 34.18 34.26 43.24 34.18 34.26
Int. level of service Stop rate, stops/veh 0.27 0.20 0.20 0.26 0.48 0.50 0.85 0.77 0.77 0.85 0.77 0.77
C Level of service A A A A B B D C C D C C
Pedestrian Level of Service
Corner location Adjacent to Eastbound Adjacent to Westbound Adjacent to Northbound Adjacent to Southbound
Corner circulation area, ft2/p 93.7 93.7 93.7 93.7
Crosswalk location Crossing major Crossing major Crossing minor Crossing minor
Crosswalk circulation area, ft2/p 75.9 75.9 82.4 82.4
Pedestrian delay, s/p 42.3 42.3 42.3 42.3
Pedestrian LOS score 2.75 2.75 2.66 2.66
Level of service C C B B
Bicycle Level of Service
Bicycle delay, s/bicycle 5.8 13.8 34.3 34.3
Bicycle LOS score 3.72 3.72 2.87 2.87
Level of service D D C C

Intersection Approach Eastbound

Exhibit 29-46
Fifth Avenue Basic Description Example Problem 1:
Primary movement number 5 2 12 1 6 16 3 8 18 7 4 14 Intersection 5 Evaluation
B Level of service A A A A A A D C C D C C
Level of service B B B B
Level of service D D C C
Both exhibits indicate that the major-street vehicular through movements

(i.e., eastbound Movement 2 and westbound Movement 6) operate with very low
delay and few stops. The LOS is A and B for the eastbound and westbound
through movements, respectively.
Pedestrian circulation area on the corners of Intersection 1 is generous, with
pedestrians able to move in their desired path without conflict. Corner
circulation area at Intersection 5 is restricted, with pedestrians having limited
ability to pass slower pedestrians.
At Intersection 1, the low pedestrian volume results in generous crosswalk
circulation area. Pedestrians rarely need to adjust their path to avoid conflicts. In
contrast, the high pedestrian volume at Intersection 5 results in a constrained
crosswalk circulation area. Pedestrians frequently adjust their path to avoid
conflict. At each intersection, pedestrians experience an average wait of about
42 s at the corner to cross the street in any direction. This delay is lengthy, and
some pedestrians may not comply with the signal indications. At Intersection 1,
the pedestrian LOS is C for the major-street crossing and B for the minor-street
crossing. At Intersection 5, the pedestrian LOS is B for the major-street and
minor-street crossings.
The lack of a bicycle lane combined with a moderately high traffic volume
results in a bicycle LOS D on the eastbound and westbound approaches of
Intersection 1 and Intersection 5.
Segment Evaluation
The results of the evaluation of Segment 1 (i.e., First Avenue to Second
Avenue) are shown in Exhibit 29-47. The results for Segments 2 and 3 are similar.
In contrast, Segments 4 and 5 are associated with a shorter segment length, lower
speed limit, and higher pedestrian volume, so their operation is different from
that of the other intersections. The results for Segment 5 (i.e., Fifth Avenue to
Sixth Avenue) are shown in Exhibit 29-48. Segment 4 has similar results.

Exhibit 29-47 Segment Evaluation Summary

Example Problem 1: Segment Travel Direction Eastbound Westbound
Segment 1 Evaluation First Avenue Basic Description
to Speed limit, mi/h 35 35
Second Avenue Vehicle volume, veh/h 800 800
Through lanes, ln 2 2
Segment length, ft Vehicle Level of Service
1,320 Base free-flow speed, mi/h 40.9 40.9
Travel speed, mi/h 24.2 23.4
Spatial stop rate, stops/mi 1.72 1.93
Level of service C C
Pedestrian space, ft2/p 593.9 593.9
Pedestrian travel speed, ft/s 3.54 3.54
Pedestrian LOS score 3.48 3.48
Bicycle travel speed, mi/h 12.44 12.20
Bicycle LOS score 3.67 3.67
Level of service D D
Transit Level of Service
Transit travel speed, mi/h 12.8 12.4
Transit LOS score 3.17 3.20

Segment 5 Evaluation Fifth Avenue Basic Description
Sixth Avenue Vehicle volume, veh/h 800 800
660 Base free-flow speed, mi/h 37.9 37.9
Level of service D C
Exhibit 29-47 indicates that the vehicular through movements on Segment 1

in the eastbound and westbound travel directions have a travel speed of 24 and
23 mi/h, respectively (i.e., about 58% of the base free-flow speed). The LOS of
each movement is C. In contrast, Exhibit 29-48 indicates that the through
movements have a travel speed of only about 17 mi/h on Segment 5 (or 46% of
the base free-flow speed), which is LOS D. Vehicles stop at a rate of about
1.8 stops/mi on Segment 1 and about 2.7 stops/mi on Segment 5.
Pedestrian space on the sidewalk along the segment is generous on Segment 1
and adequate on Segment 5. These characterizations are based on Exhibit 16-9
and an assumed dominance of platoon flow for Segments 4 and 5. Pedestrians on
these sidewalks can walk freely without having to alter their path to accommodate
other pedestrians. The segment travel speed (3.54 ft/s for Segment 1 and 3.18 ft/s
for Segment 5) is adequate but would desirably exceed 4.0 ft/s. Nevertheless, the
sidewalk is near the traffic lanes, and crossing the street at a midsegment location
can be difficult. As a result, the pedestrian LOS is C on all segments.

The lack of a bicycle lane, combined with a moderately high traffic volume,
results in a bicycle LOS D for both directions of travel on all segments.
Transit travel speed is about 12 mi/h on Segment 1 and corresponds to
LOS C. On Segment 5, the travel speed is about 8 mi/h and 17 mi/h in the
eastbound and westbound directions, respectively. The low speed for the
eastbound direction results in LOS D. The higher speed for the westbound
direction is due to the lack of a westbound transit stop on Segment 5. It results in
LOS C for this direction.
Facility Evaluation
The methodologies described in Chapter 16 were used to compute the
aggregate performance measures for each travel direction along the facility. The
results are shown in Exhibit 29-49. This exhibit indicates that the vehicle travel
speed is about 22 mi/h in each travel direction (or 56% of the base free-flow
speed). An overall LOS C applies to both vehicular movements on the facility;
however, it is noted that LOS D applies to Segments 4 and 5. Vehicles incur stops
along the facility at a rate of about 1.9 stops/mi.
Facility Evaluation Summary Exhibit 29-49

Travel Direction Eastbound Westbound Example Problem 1:
Vehicle Level of Service 2 6 Facility Evaluation
Base free-flow speed, mi/h 40.1 40.1
Facility length, ft Level of service C C
5,280 Poorest perf. segment LOS D D
Poorest perf. segment LOS C C
Poorest perf. segment LOS D D
Pedestrian space on the sidewalk along the facility is generous. Pedestrians

on the sidewalks can walk freely without having to alter their path to
accommodate other pedestrians. The facility travel speed of about 3.4 ft/s is
adequate but would desirably exceed 4.0 ft/s. Nevertheless, the sidewalk is near
the traffic lanes, and crossing the street at a midsegment location can be difficult.
As a result, the pedestrian LOS is C for both directions of travel.
The lack of a bicycle lane, combined with a moderately high traffic volume,
results in an overall bicycle LOS D for both directions of travel.
Transit travel speed is about 12 mi/h on the facility in each direction of travel.
An overall LOS C is assigned to each direction. The lower speed on westbound
Segment 4 and eastbound Segment 5 is noted to result in LOS D for those
segments.

EXAMPLE PROBLEM 2: PEDESTRIAN AND BICYCLE IMPROVEMENTS

The 1-mi urban street facility shown in Exhibit 29-39 is being considered for
geometric design modifications to improve pedestrian and bicycle service. The
following changes to the facility are proposed:
• Eliminate one vehicle lane in each direction,
• Add a 12-ft raised-curb median,
• Add a 4-ft bicycle lane in each direction,
• Increase the total walkway width to 9 ft,
• Add a 3-ft buffer between the sidewalk and the curb, and
• Add bushes to the buffer with a 10-ft spacing.
No fixed objects are located along the outside of the sidewalk. The analysis
for Example Problem 1 represents the existing condition, against which this
alternative will be evaluated.
same for each segment. Additional segment details are provided in the
discussion for Example Problem 1.
Exhibit 29-50 Pavement condition rating: 3.5

Example Problem 2: N Curbed cross section
Segment Geometry Cross-street lane width: 12 ft
4 ft
12 ft
Signal 12 ft Raised-curb median Signal
12 ft
4 ft
The Question
What are the travel speed and LOS of the motorized vehicle, pedestrian,
bicycle, and transit modes in both directions of travel along the facility?
The Facts
The traffic counts, signalization, and intersection geometry are listed in
Exhibit 29-41 to Exhibit 29-44. They are unchanged from Example Problem 1.
Outline of Solution
This section outlines the results of the facility evaluation. The motorized
vehicle, pedestrian, and bicycle methodologies in Chapter 19 were used to
evaluate each of the signalized intersections on the facility. The procedure in
Chapter 20 was used to estimate delay for pedestrians crossing at a midsegment
location. The motorized vehicle, pedestrian, bicycle, and transit methodologies in
Chapter 18 were then used to evaluate both directions of travel on each segment.
Finally, the methodologies described in Chapter 16 were used to evaluate all four

travel modes in both directions of travel on the facility. The findings from each
evaluation are summarized in the following three subparts.

Exhibit 29-51
First Avenue Basic Description Example Problem 2:
C Level of service B A A A B B D C C D C C
Level of service C C C C

Exhibit 29-52
C Level of service B B B B B B D C C D C C
Level of service C C C C
Both exhibits indicate that the vehicular through movements on the facility
(i.e., eastbound Movement 2 and westbound Movement 6) operate with low
delay and few stops. For the eastbound through movement, the LOS is A at
Intersection 1 and B at Intersection 5. The LOS is B for the westbound through
movement at both intersections. Relative to Example Problem 1, the delay for the
through movements has increased by 1 to 3 s at Intersection 1 and by 6 to 8 s at
Intersection 5. This increase is sufficient to lower the LOS designation for the
through movements at Intersection 5 (i.e., from A to B).
Pedestrian circulation area on the corners of Intersections 1 and 5 is
generous, with few instances of conflict. This condition is improved from
Example Problem 1 and reflects the provision of wider sidewalks.

Relative to Example Problem 1, the reduction in through lanes has reduced

the time provided to pedestrians to cross the major street. This reduction resulted
in larger pedestrian groups using the crosswalk and a small reduction in
crosswalk pedestrian space. At Intersection 1, pedestrian space is still generous,
with few instances of conflict. At Intersection 5, the problem is amplified by a
higher pedestrian demand. Pedestrian space in the crosswalks is constrained,
and pedestrians are likely to find that their ability to pass slower pedestrians is
limited.
At each intersection, pedestrians experience an average wait of about 42 s at
the corner to cross the street in any direction. This condition has not changed
from Example Problem 1.
At both intersections, the pedestrian LOS is B for the major-street and minor-
street crossings. Relative to Example Problem 1, the pedestrian LOS score for the
major-street crossings has improved a small amount at all intersections. At
Intersection 1, this change is sufficient to result in a change in service level (i.e.,
from C to B) for the major-street crossings.
Bicyclists using the bicycle lanes are expected to be delayed about 8 s/bicycle
on the eastbound and westbound approaches at each intersection. This level of
delay is desirably low. However, the bicycle lane is relatively narrow at 4 ft,
which leads to LOS C on the eastbound and westbound approaches of both
intersections. This LOS is an improvement over the LOS D identified in Example
Problem 1.
Segment Evaluation
the other intersections. The results for Segment 5 (i.e., Fifth Avenue to Sixth
Avenue) are shown in Exhibit 29-54. Segment 4 has similar results.
The results reported in this section reflect the segment geometry shown in
Exhibit 29-50. These results are compared with those from Example Problem 1.
The differences in performance are a result of the changes identified in the bullet
list that precedes Exhibit 29-50. Most notable in this list is the reduction in lanes
for motorized vehicles, which results in a doubling of vehicles in the remaining
lanes. The vehicle volume in these lanes has a significant influence on bicycle and
pedestrian performance.

Segment Evaluation Summary Exhibit 29-53

Segment Travel Direction Eastbound Westbound Example Problem 2:
First Avenue Basic Description Segment 1 Evaluation
Segment Evaluation Summary Exhibit 29-54

Segment Travel Direction Eastbound Westbound Example Problem 2:
Fifth Avenue Basic Description Segment 5 Evaluation
Level of service E E

in the eastbound and westbound travel directions have a travel speed of about
22 mi/h (i.e., about 56% of the base free-flow speed). LOS C applies to both
movements. In contrast, Exhibit 29-54 indicates that the through movements
have a travel speed of only about 13 mi/h on Segment 5 (or 37% of the base free-
flow speed), which is LOS E. Vehicles stop at a rate of about 1.8 stops/mi on
Segment 1 and about 4.6 stops/mi on Segment 5. Relative to Example Problem 1,
the quality of service has been degraded for vehicles traveling along Segment 5.
Pedestrian space on the sidewalk along the segment is generous on
Segment 1. Pedestrians can walk freely without having to alter their path to
accommodate other pedestrians. Pedestrian space is adequate on Segment 5,
with pedestrians in platoons occasionally needing to adjust their path to avoid
conflict. These characterizations are based on Exhibit 16-9 and on an assumed
dominance of platoon flow for Segments 4 and 5. Relative to Example Problem 1,

the sidewalks are more distant from the traffic lanes and crossing the street at a
midsegment location is easier because of the raised curb median. The LOS score
indicates improved pedestrian service; however, the pedestrian LOS remains at
C on all segments.
Bicyclists using the bicycle lanes experience a travel speed of 13 mi/h on
Segment 1 and 12 mi/h on Segment 5. This travel speed is considered desirable.
However, the bicycle lane is relatively narrow at 4 ft, so a bicycle LOS C results
for both directions of travel on each segment. The bicycle LOS scores, while still
poor, indicate that bicycle service has improved on both segments relative to that
found in Example Problem 1. In fact, the bicycle LOS for each segment has
improved by one letter designation.
Transit travel speed is 10 mi/h on Segment 1 and corresponds to LOS C. On
Segment 5, the travel speed is about 5 mi/h and 13 mi/h in the eastbound and
westbound directions, respectively. The low speed for the eastbound direction
results in LOS D. The higher speed for the westbound direction is due to the lack
of a westbound transit stop on Segment 5. It results in LOS C. Relative to
Example Problem 1, the slower vehicular travel speed has increased the transit
LOS scores, which indicates a lower quality of service.
Facility Evaluation
The methodologies described in Chapter 16 were used to compute the
aggregate performance measures for each travel direction along the facility. The
results are shown in Exhibit 29-55. This exhibit indicates that the vehicle travel
speed is about 18 mi/h in each travel direction (or 48% of the base free-flow
speed). An overall LOS D applies to vehicle travel in each direction on the
facility. It is noted that LOS E applies to Segments 4 and 5. Vehicles incur stops
along the facility at a rate of about 2.6 stops/mi. Relative to Example Problem 1,
vehicular travel speed has dropped about 4 mi/h, and motorized vehicle LOS has
degraded one level for this scenario.
Exhibit 29-55 Facility Evaluation Summary

Example Problem 2: Travel Direction Eastbound Westbound
Facility Evaluation Vehicle Level of Service 2 6
Facility length, ft Level of service D D
5,280 Poorest perf. segment LOS E E


accommodate other pedestrians. Increasing the separation between the sidewalk
and traffic lanes and improving pedestrians’ ability to cross the street at
midsegment locations (by adding a raised-curb median) have resulted in a lower
LOS score, which indicates improved service relative to Example Problem 1.
However, the pedestrian LOS letter (C) is unchanged.
Bicyclists in the bicycle lanes are estimated to experience an average travel
speed of about 13 mi/h. This travel speed is considered desirable. However, the
4-ft bicycle lane is relatively narrow and produces LOS C. This level is one level
improved over that found for Example Problem 1.
An overall LOS C is assigned to each direction. Relative to Example Problem 1,
the LOS designation is unchanged; however, the transit speed is slower, and the
transit LOS score higher, which indicates a reduction in the quality of service.
EXAMPLE PROBLEM 3: PEDESTRIAN AND PARKING IMPROVEMENTS

The 1-mi urban street facility shown in Exhibit 29-39 is being considered for
geometric design modifications to improve parking and pedestrian service. The
following changes to the facility are proposed:
• Eliminate one vehicle lane in each direction,
• Add a 12-ft raised-curb median,
• Add a 9.5-ft parking lane in each direction, and
• Increase the total walkway width to 7 ft.
No fixed objects will be located along the outside of the sidewalk. The on-
street parking is expected to be occupied 50% of the time. Parking maneuvers are
estimated to cause 1.8 s/veh additional delay on Segments 1, 2, and 3. On
Segments 4 and 5, these maneuvers are estimated to cause 0.3 s/veh additional
delay. The analysis for Example Problem 1 represents the existing condition,
against which this alternative will be evaluated.
same for each segment. Additional segment details are provided in the
discussion for Example Problem 1.
The Question
What are the travel speed and LOS of the motorized vehicle, pedestrian,
bicycle, and transit modes in both directions of travel along the facility?
The Facts
The traffic counts, signalization, and intersection geometry are listed in
Exhibit 29-41 to Exhibit 29-44. They are unchanged from Example Problem 1.

Exhibit 29-56 Pavement condition rating: 3.5

Example Problem 3: N Curbed cross section
Segment Geometry Cross-street lane width: 12 ft
9.5 ft
10 ft
Signal 12 ft Raised-curb median Signal
10 ft
9.5 ft
Outline of Solution
This section outlines the results of the facility evaluation. To complete this
evaluation, the motorized vehicle, pedestrian, and bicycle methodologies in
Chapter 19 were used to evaluate each of the signalized intersections on the
facility. The procedure in Chapter 20 was used to estimate pedestrian delay
when crossing at a midsegment location. The motorized vehicle, pedestrian,
bicycle, and transit methodologies in Chapter 18 were then used to evaluate both
directions of travel on each segment. Finally, the methodologies described in
Chapter 16 were used to evaluate all four travel modes in both directions of
travel on the facility. The findings from each evaluation are summarized in the
following three subparts.
Exhibit 29-57 Intersection Approach Eastbound

Example Problem 3: First Avenue Basic Description
Applicable lane assignments L T R L T R L TR n.a. L TR n.a.
Intersection 1 Evaluation Primary movement number 5 2 12 1 6 16 3 8 18 7 4 14
C Level of service B A A A B B D C C D C C
Level of service E E C C


Exhibit 29-58
Applicable lane assignments L T R L T R L TR n.a. L TR n.a.
C Level of service A B A A B A D C C D C C
Level of service E E C C
Both exhibits indicate that the vehicular through movements on the facility
(i.e., eastbound Movement 2 and westbound Movement 6) operate with very low
delay and few stops. For the eastbound through movement, the LOS is A at
Intersection 1 and B at Intersection 5. The LOS is B for the westbound through
movement at both intersections. Relative to Example Problem 1, the delay for the
through movements has increased by a few seconds at both intersections.
However, this increase is sufficient to lower the LOS designation for only the
eastbound through movement at Intersection 5.
Pedestrian circulation area on the corners of Intersection 1 is generous.
However, corner circulation area at Intersection 5 is constrained, with
pedestrians frequently needing to adjust their path to avoid slower pedestrians.
Regardless, this condition is improved from Example Problem 1 and reflects the
provision of wider sidewalks.
Relative to Example Problem 1, the reduction in lanes has reduced the time
provided to pedestrians to cross the major street. This reduction resulted in
larger pedestrian groups using the crosswalk and a slight reduction in crosswalk
pedestrian space. At Intersection 1, pedestrian space is generous. However,
pedestrian space is constrained at Intersection 5, with pedestrians having limited
ability to pass slower pedestrians as they cross the street.
At each intersection, pedestrians experience an average wait of about 42 s at
the corner to cross the street in any direction. At both intersections, the
pedestrian LOS is B for the major-street crossing and the minor-street crossing.
The LOS designation has improved for the major-street crossing at Intersection 1
by one letter, relative to Example Problem 1, and remains unchanged at
Intersection 5.
The lack of a bicycle lane combined with a high traffic volume results in a
bicycle LOS E on the eastbound and westbound approaches of Intersection 1 and
Intersection 5. This level is worse than the LOS D identified in Example Problem
1 because the traffic volume per lane has doubled.
Segment Evaluation

that of the other intersections. The results for Segment 5 (i.e., Fifth Avenue to
Sixth Avenue) are shown in Exhibit 29-60. Segment 4 has similar results.
The results reported in this section reflect the segment geometry shown in
Exhibit 29-56. These results are compared with those from Example Problem 1.
The differences in performance are a result of the changes identified in the bullet
list that precedes Exhibit 29-56. Most notable in this list is the reduction in lanes
for motorized vehicles, which results in a doubling of vehicles in the remaining
lanes. The vehicle volume in these lanes has a significant influence on bicycle and
pedestrian performance.

Segment 1 Evaluation First Avenue Basic Description

Segment 5 Evaluation Fifth Avenue Basic Description

in the eastbound and westbound travel directions have a travel speed of about
19 mi/h (i.e., about 53% of the base free-flow speed). LOS C applies to both

movements. In contrast, Exhibit 29-60 indicates that the through movements

have a travel speed of only about 14 mi/h on Segment 5 (or 42% of the base free-
flow speed), which is LOS D. Vehicles stop at a rate of about 2.1 stops/mi on
Segment 1 and about 3.5 stops/mi on Segment 5. Relative to Example Problem 1,
conditions have degraded for vehicles traveling along these segments, but not
enough to drop the LOS designation.
Pedestrian space on the sidewalk along the segment is generous on
Segment 1 and adequate on Segment 5. These characterizations are based on
Exhibit 16-9 and on an assumed dominance of platoon flow for Segments 4 and 5.
Pedestrians on these sidewalks can walk freely without having to alter their path
to accommodate other pedestrians. Relative to Example Problem 1, the sidewalks
are more distant from the traffic lanes, and crossing the street at a midsegment
location is easier because of the raised-curb median. The LOS score indicates
improved pedestrian service; however, the pedestrian LOS remains at C on all
segments.
The lack of a bicycle lane combined with a high traffic volume results in a
bicycle LOS D for both directions of travel on Segment 1 and Segment 5. Relative
to Example Problem 1, the quality of service has degraded for bicyclists on all
segments. This reduction in service is due largely to the increased density of
vehicles in the mixed traffic lanes.
Transit travel speed is about 10 mi/h on Segment 1 and corresponds to
LOS C. On Segment 5, the travel speed is about 6 mi/h and 14 mi/h in the
eastbound and westbound directions, respectively. The low speed for the
eastbound direction results in LOS D. The higher speed for the westbound
direction is due to the lack of a westbound transit stop on Segment 5. It results in
LOS C. Relative to Example Problem 1, the slower vehicular travel speed has
increased the transit LOS scores, which indicates a lower quality of service.
Facility Evaluation
The methodology described in Section 2 is used to compute the aggregate
performance measures for each travel direction along the facility. The results are
shown in Exhibit 29-61. This exhibit indicates that the vehicle travel speed is
about 18 mi/h in each travel direction (or 51% of the base free-flow speed). An
overall LOS C applies to both vehicular movements on the facility; however, it is
noted that LOS D applies to Segments 4 and 5. Vehicles incur stops along the
facility at a rate of about 2.3 stops/mi. Relative to Example Problem 1, the quality
of vehicular service has degraded, but not enough to drop the LOS designation.
accommodate other pedestrians. Increasing the separation between the sidewalk
and traffic lanes and improving pedestrians’ ability to cross the street at
midsegment locations (by adding a raised-curb median) have resulted in a lower
LOS score, which indicates improved service relative to Example Problem 1.
However, the pedestrian LOS letter (C) is unchanged.

Exhibit 29-61 Facility Evaluation Summary

Example Problem 3: Travel Direction Eastbound Westbound
Facility Evaluation Vehicle Level of Service 2 6
Facility length, ft Level of service C C
5,280 Poorest perf. segment LOS D D
The lack of a bicycle lane combined with a high traffic volume results in an
overall bicycle LOS D for both directions of travel. The quality of service has
degraded slightly, relative to Example Problem 1, but not enough to drop the
LOS designation.
An overall LOS C is assigned to each direction. Conditions have degraded
slightly, relative to Example Problem 1, but not enough to drop the transit LOS
designation.
EXAMPLE PROBLEM 4: EXISTING URBAN STREET RELIABILITY

Objective
This example problem illustrates
• The steps involved in calculating reliability statistics for an urban street
facility using the minimum required data for the analysis,
• Identification of the key reliability problems on the facility, and
• Diagnosis of the causes (e.g., demand, weather, incidents) of reliability
problems on the facility.
Site
The selected site for this example problem is an idealized 3-mi-long principal
arterial street located in Lincoln, Nebraska. The street is a two-way, four-lane,
divided roadway with shoulders. There are seven signalized intersections that
are spaced uniformly at 0.5-mi intervals along the street. The posted speed limit
on the major street and the minor streets is 35 mi/h. A portion of this street is
shown in Exhibit 29-62. The distances shown are the same for the other segments
of the facility.

50 500 100 100 80 Exhibit 29-62

Example Problem 4:
10 100 Urban Street Facility
1,000 1,050
200 Signal 80 Access Point
200 80
1,000 1,050
10 100
100 500 50 80 100
2,640 ft 2,640 ft N
600 ft 600 ft 600 ft 600 ft
1 2 3
AP1 AP2 AP3 AP4
Signal Segment 1 Signal Segment 2 Signal
Also shown in Exhibit 29-62 are the traffic movement volumes for each
intersection and access point on the facility. Each intersection has the same
volume, and each access point has the same volume. Intersection geometry and
signal timing are described in a subsequent section.
Required Input Data

This section describes the input data needed for both the reliability
methodology and the core HCM urban streets methodology. The dataset that
describes conditions under which no work zones or special events are present is
known as the base dataset. Other datasets used to describe work zones or special
events are called alternative datasets.
Reliability Methodology Input Data

Exhibit 29-63 lists the input data needed for an urban street reliability
evaluation. The agency does not collect traffic volume data on a continual basis,
so the factors and ratios that describe demand patterns will be defaulted. Traffic
counts for one representative day are provided by the analysis and used as the
basis for estimating volume during other hours of the year. Lincoln, Nebraska, is
one of the communities for which a 10-year summary of weather data is
provided, so the default weather data will be used. Incident data are available
locally as annual crash frequencies by intersection and street segment. It was
determined that the effect of work zones or special events on reliability would
not be considered in the evaluation.
HCM Urban Street Methodology Input Data

This subsection describes the data gathered to develop the base dataset. The
base dataset contains all of the input data required to conduct an urban street
facility analysis with the methodologies described in HCM Chapters 16 through
19. Alternative datasets are not needed because the effects of work zones and
special events are not being considered in the evaluation.

Exhibit 29-63 Data Category Input Data Need Data Value

Example Problem 4: Input Functional class Urban street functional class Urban principal arterial
Data Needs and Sources
Nearest city Required when defaulted weather Lincoln, Nebraska
data used
Geometrics Presence of shoulder Yes
Time periods Analysis period 15 min
Study period 7–10 a.m.
Reliability reporting period Weekdays for 1 year
Demand patterns Hour-of-day factors Will be defaulted
Day-of-week demand ratio
Month-of year demand ratio
Demand change due to rain, snow
Weather Rain, snow, and temperature data Will be defaulted
by month
Pavement runoff duration
Incidents Segment and intersection crash Available locally (See Step 5)
frequencies
Crash frequency adjustment factors Not required (no work zones)
for work zones or special events
Factors influencing incident duration Will be defaulted
Work zones and Changes to base conditions Not required (no work zones)
special events (alternative dataset) and schedule
Traffic counts Day and time of traffic counts used Tuesday, January 4, 7–8 a.m.
in base and alternative datasets No alternative datasets required
(no work zones)
Traffic count data for the hour beginning at 7:00 a.m. are available from a
recent traffic count taken on a Tuesday, January 4. Weather conditions were clear
and the pavement was dry. The traffic volumes are shown in Exhibit 29-62. They
are the same at all seven intersections for this idealized example.
Exhibit 29-64 provides the signal timing data for Intersection 1. The other
signalized intersections have the same signal timing.
Exhibit 29-64 Approach Eastbound Westbound Northbound Southbound

Example Problem 4: Movement L T R L T R L T R L T R
Intersection 1 Signal Timing NEMA movement no. 5 2 12 1 6 16 3 8 18 7 4 14
Data Volume (veh/h) 200 1000 10 200 1000 10 100 500 50 100 500 50
Lanes 1 2 1 1 2 1 1 2 0 1 2 0
Turn bay length (ft) 200 0 200 200 0 200 200 0 0 200 0 0
Saturation flow rate (veh/h/ln) 1,800 1,800 1,800 1,800 1,800 1,800 1,800 1,800 1,800 1,800 1,800 1,800
Platoon ratio 1.000 1.333 1.000 1.000 1.333 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Initial queue (veh) 0 0 0 0 0 0 0 0 0 0 0 0
Speed limit (mi/h) -- 35 -- -- 35 -- -- 35 -- -- 35 --
Detector length (ft) 40 40 -- -- 40 40 -- 40 40 --
Lead/lag left-turn phase Lead -- Lead -- Lead -- Lead --
Left-turn mode Prot. -- Prot. -- Pr/Pm -- Pr/Pm --
Passage time (s) 2.0 -- 2.0 -- 2.0 2.0 2.0 2.0
Minimum green (s) 5 -- 5 -- 5 5 5 5
Change period (Y+Rc) (s) 3.0 4.0 3.0 4.0 3.0 4.0 3.0 4.0
Phase splits (s) 20.0 35.0 20.0 35.0 20.0 25.0 20.0 25.0
Max. recall Off -- Off -- Off Off Off Off
Min. recall Off -- Off -- Off Off Off Off
Dual entry No Yes No Yes No Yes No Yes
Simultaneous gap out Yes Yes Yes Yes Yes Yes Yes Yes
Dallas phasing No No No No No No No No
Reference phase 2
Offset (s) 0 or 50
Notes: L = left turn, T = through, R = right turn, Prot. = protected, Pr/Pm = protected-permitted.
See Chapter 18 for definitions of signal timing variables.

At each signalized intersection, there are left- and right-turn bays on each of
the two major-street approaches, left-turn bays on each of the minor-street
approaches, and two through lanes on each approach. Two unsignalized access
points exist between each signal.
The posted speed limit for the major street and the minor streets is 35 mi/h.
The traffic signals operate in coordinated-actuated mode at a 100-s cycle. The
offset for the eastbound through phase alternates between 0 and 50 s at
successive intersections to provide good two-way progression.
The peak hour factor is 0.99, 0.92, 0.93, 0.94, 0.95, 0.96, and 0.97 at
Intersections 1 through 7, respectively.
Analysis Replications
The urban street reliability method uses a Monte Carlo approach to generate A Monte Carlo approach is
used when there is some
variables describing weather events, incidents, and random demand fluctuations randomness in the value of a
for each scenario in the reliability reporting period. One variation of this variable due to unknown
influences (and known
approach is to use an initial random number seed. The use of a seed number influences by other variables
ensures that the same random number sequence is used each time a set of that also have some
randomness) such that it is
scenarios is generated for a given reliability reporting period. Any positive difficult to determine the
integer can be used as a seed value. Each set of scenarios is called a replication. frequency (or probability) of
the subject variable’s value
Because events (e.g., a storm, a crash) are generated randomly in the urban accurately.
street method, the possibility exists that highly unlikely events could be
overrepresented or underrepresented in a given set of scenarios. To minimize
any bias these rare events may cause, the set of scenarios should be replicated Multiple analysis replications
are needed to determine the
and evaluated two or more times. Each time the set of scenarios is created, the confidence interval for the final
inputs should be identical, except that a different set of random number seeds is performance results.
used. Then, the performance measures of interest from the evaluation of each set
of scenarios are averaged to produce the final performance results.
Five replications were found to provide sufficient precision in the predicted
reliability measures for this example problem. The seed numbers in the following
list were selected by the analyst for this example problem. The first replication
used seed numbers 82, 11, and 63. The second replication used numbers 83, 12,
and 64. This pattern continues for the other three replications.
• Weather event generator: 82, 83, 85, 87, 89
• Demand event generator: 11, 12, 14, 16, 18
• Incident event generator: 63, 64, 66, 68, 70
The random number sequence created by a specific seed number may be
specific to the software implementation and computer platform used in the
analysis. As a result, evaluating the same dataset and seed number in different
software or on a different platform may produce results different from those
shown here. Each result, though different, will be equally valid.
Computational Steps
This example problem proceeds through the following steps:
1. Establish the purpose, scope, and approach.
2. Code datasets.

3. Estimate weather events.

4. Estimate demand volumes.
5. Estimate incident events.
6. Generate scenarios.
7. Apply the Chapter 16 motorized vehicle methodology.
8. Conduct quality control and error checking.
9. Interpret results.
Step 1: Establish the Purpose, Scope, and Approach

Define the Purpose
The agency responsible for this urban street wishes to perform a reliability
analysis of existing conditions to determine whether the facility is experiencing
significant reliability problems. It also wants to diagnose the primary causes of
any identified reliability problems on the facility so that an improvement
strategy can be developed.
Define the Reliability Analysis Box

The results from a preliminary evaluation of the facility were used to define
the general spatial and temporal boundaries of congestion on the facility under
fair weather, nonincident conditions. A study period consisting of the weekday
morning peak period (7–10 a.m.) and a study area consisting of the 3-mi length of
facility between Intersections 1 and 7 encompass all of the recurring congestion.
The reliability reporting period is to include all weekdays during the course
of a year. The duration of the analysis period will be 15 min.
Select Reliability Performance Measures

Reliability will be reported by using the following performance measures:
mean travel time index (TTI), 80th percentile TTI, 95th percentile TTI (PTI),
reliability rating, and total delay (in vehicle hours) for the reliability reporting
period.
Step 2: Code Datasets

Select Reliability Factors for Evaluation
The major causes of travel time reliability problems are demand surges,
weather, and incidents. Reliability problems associated with work zones and
special events were determined not to be key elements of the evaluation of this
specific facility.
Code the Base Dataset

The base dataset was developed for the selected study section and study
period. This dataset describes the traffic demand, geometry, and signal timing
conditions for the intersections and segments on the subject urban street facility
during the study period when no work zones are present and no special events
occur. The data included in this dataset are described in Chapters 16 through 19.

Code the Alternative Datasets

Only the base dataset will be required because no work zones are planned in
the next year and no special events affect the facility on weekdays.
Step 3: Estimate Weather Events

This step predicts weather event date, time, type (i.e., rain or snow), and
duration for each study period day in the reliability reporting period.
Identify Input Data

The default weather data for Lincoln, Nebraska, are a compilation of 10 years
of historical data from the National Climatic Data Center (2, 9) and include the
following statistics:
• Total normal precipitation,
• Total normal snowfall,
• Number of days with precipitation of 0.01 in. or more,
• Normal daily mean temperature, and
• Precipitation rate.
One inch of snowfall is estimated to have the water content of 0.1 in. of rain.
Exhibit 29-65 shows the historical weather data for 2 months of the year.
Weather Data January April Exhibit 29-65

Normal precipitationa (in.) 0.67 2.90 Example Problem 4: Sample
Normal snowfall (in.) 6.60 1.50 Weather Data for Lincoln,
Days with precipitation (days) 5 9 Nebraska
Daily mean temperature (˚F) 22.40 51.20
Precipitation rate (in./h) 0.030 0.062
a
Note: Rainfall plus water content of snow.
Determine Weather Events for Each Day

At this point in the analysis, weather is estimated for all days during a 2-year
period. The analysis is not yet confined to the days within the reliability
reporting period or the hours within the study period. The purpose of the extra
calculations is to define the expected weather pattern for the study facility, which
will be used in a later step to estimate incident frequencies.
A Monte Carlo approach is used to decide whether precipitation will occur
in a given day. If it does, a Monte Carlo approach is also used to determine the
type of precipitation (i.e., rain or snow), precipitation rate, total precipitation,
and start time for the current day. The details of the scenario generation process
are described in Section 2.
Exhibit 29-66 illustrates the results of the calculations for 2 weeks in January
and 2 weeks in April. These results are based on the historical weather data for
Lincoln, Nebraska, as shown in Exhibit 29-65. The random number values shown
in the exhibit are intended to illustrate the computations within this specific
table. Different values are obtained if the random number seed is changed. Only
dates falling within the reliability reporting period are shown.
For reliability evaluation, total precipitation is assumed to be perfectly
correlated with the precipitation rate such that storms producing a large total

precipitation are associated with a high precipitation rate. This relationship is

replicated by estimating both values by using the same random number.
As can be seen from Exhibit 29-66, the computed event durations may exceed
24 h, but when the end times are set for the event, any event that ends beyond
24:00 is truncated to 24:00.
Exhibit 29-66
Precipitation Start RN RSd,m

Example Problem 4: Sample
Total Precipitation RN RTPd
Start of Precipitation Event
Time Wet After Precip. (h)

Precipitation Duration (h)
Precipitation Rate RN RPd
Precipitation Rate (in./h)

Generated Weather Events
Total Event Duration (h)

Precipitation? (Yes/No)
Mean Temperature (˚F)
Total Precipitation (in.)
End of Wet Pavement

Temperature RN RTd
End of Precipitation
Precipitation RN RD
Snow/Rain?
Day/Night?
Date
Jan 10 0.03 Yes 0.94 30 Snow 0.83 0.54 0.83 2.08 0.23 4:30 3.88 1.22 Night 5.10 8:23 9:36
Jan 12 0.30 No
Jan 13 0.90 No
Jan 14 0.20 No
Jan 25 0.53 No
Jan 26 0.45 No
Jan 27 0.21 No
Jan 28 0.60 No
Apr 4 0.64 No
Apr 5 0.24 Yes 0.11 45 Rain 0.40 0.03 0.40 0.02 1.00 23:15 0.68 0.07 Night 0.75 23:56 24:00
Apr 6 0.22 Yes 0.19 47 Rain 0.31 0.02 0.31 0.01 0.08 1:45 0.34 0.92 Night 1.26 2:05 3:00
Apr 7 0.78 No
Apr 8 0.39 No
Apr 11 0.55 No
Apr 12 0.37 No
Apr 13 0.10 Yes 0.28 48 Rain 0.82 0.11 0.82 0.54 0.39 7:15 5.05 0.72 Day 5.76 12:18 13:01
Apr 14 0.78 No
Apr 15 0.27 Yes 0.98 61 Rain 0.73 0.08 0.73 0.30 0.57 11:30 3.62 0.66 Day 4.28 15:07 15:47
Note: RN = random number.
Determine Weather Events for Each Analysis Period

The days that have weather events are subsequently examined to determine
whether the event occurs during the study period. Specifically, each analysis
period is examined to determine whether it is associated with a weather event.
An examination of the start and end times in Exhibit 29-66 indicates that the
snow on January 10 and the rain on April 13 occur during the 7:00 to 10:00 a.m.
study period.
Step 4: Estimate Demand Volumes

This step identifies the appropriate traffic volume adjustment factors
(demand ratios) for each date and time during the reliability reporting period.
These factors are used during the scenario file generation procedure to estimate
the volume associated with each analysis period. If the analyst does not provide
demand ratios based on local data, the default ratios provided in Section 5,
Applications, of Chapter 17 are used.
Identify Input Data

The input data needed for this step are identified in the following list:
• Hour-of-day demand ratio,
• Day-of-week demand ratio,

• Month-of-year demand ratio,

• Demand change factor for rain event, and
• Demand change factor for snow event.
The default values for these factors are obtained from Exhibit 17-5 to Exhibit
17-8. Their selection is based on the functional class of the subject facility, which
is “urban principal arterial.”
Determine Base Demand Ratio

First, the demand ratios for the day of the traffic count are determined. The
count was taken on Tuesday, January 4, during the 7:00 a.m. hour. By using the
default demand ratio data from Exhibit 17-5 through Exhibit 17-7, the following
can be seen:
• The hour-of-day ratio for the 7:00 a.m. hour for principal arterials is 0.071,
• The day-of-week ratio for Tuesdays is 0.98, and
• The month-of-year ratio for principal arterials in January is 0.831.
Multiplying these three factors together yields the base demand ratio of
0.0578. This ratio indicates that counted traffic volumes represent 5.78% of
annual average daily traffic (AADT), if this urban street’s demand pattern is
similar to that of the default demand data.
Determine Analysis Period Demand Ratio

A similar process is used to determine the demand ratio represented by each
analysis period, except that an additional adjustment is made for weather. From
Exhibit 17-8, a default 1.00 demand adjustment factor is applied to analysis
periods with rain and a 0.80 adjustment factor is applied to analysis periods with
snow.
As an example, the weather generator produced snow conditions for
Monday, January 10, at 7:00 a.m. Default demand ratio data are obtained again
from Exhibit 17-5 through Exhibit 17-7. The text accompanying Exhibit 17-8
states that a demand change factor of 0.80 is appropriate for snowing conditions.
Therefore, the factor values in the following list are established for the
evaluation:
• The hour-of-day ratio for the 7:00 a.m. hour for principal arterials is 0.071,
• The day-of-week ratio for Mondays is 0.98,
• The month-of-year ratio for principal arterials in January is 0.831, and
• The demand change factor is 0.80.
Multiplying these factors together yields the demand ratio of 0.0463. This
ratio indicates that the analysis period volumes represent 4.63% of AADT.
Therefore, the traffic counts are multiplied by (0.0463 / 0.0578) = 0.800 to produce
equivalent volumes for the hour starting at 7:00 a.m. on Monday, January 10.
Exhibit 29-67 shows a selection of demand profile computations for different
hours, days, months, and weather events. Each row in this exhibit corresponds to
one analysis period (i.e., scenario). The ratio shown in the last column of this

exhibit is multiplied by the traffic counts for each signalized intersection to

estimate the equivalent hourly flow rate for the associated analysis period.
Exhibit 29-67 Weather Hour Day Month Total Total/

Example Problem 4: Sample Date Weekday Time Weather Factor Factor Factor Factor Factor Base
Demand Profile Calculations Jan 10 Mon 7:00 Snow 0.80 0.071 0.980 0.831 0.0463 0.800
Jan 10 Mon 7:15 Snow 0.80 0.071 0.980 0.831 0.0463 0.800
Jan 10 Mon 7:30 Snow 0.80 0.071 0.980 0.831 0.0463 0.800
Jan 10 Mon 7:45 Snow 0.80 0.071 0.980 0.831 0.0463 0.800
Jan 10 Mon 8:00 Snow 0.80 0.058 0.980 0.831 0.0378 0.654
Jan 10 Mon 8:15 Snow 0.80 0.058 0.980 0.831 0.0378 0.654
Jan 10 Mon 8:30 Dry 1.00 0.058 0.980 0.831 0.0472 0.817
Jan 10 Mon 8:45 Dry 1.00 0.058 0.980 0.831 0.0472 0.817
Jan 10 Mon 9:00 Dry 1.00 0.047 0.980 0.831 0.0383 0.662
Jan 10 Mon 9:15 Dry 1.00 0.047 0.980 0.831 0.0383 0.662
Jan 10 Mon 9:30 Dry 1.00 0.047 0.980 0.831 0.0383 0.662
Jan 10 Mon 9:45 Dry 1.00 0.047 0.980 0.831 0.0383 0.662
Apr 6 Wed 7:00 Dry 1.00 0.071 1.000 0.987 0.0701 1.212
Apr 6 Wed 7:15 Dry 1.00 0.071 1.000 0.987 0.0701 1.212
Apr 6 Wed 7:30 Dry 1.00 0.071 1.000 0.987 0.0701 1.212
Apr 6 Wed 7:45 Dry 1.00 0.071 1.000 0.987 0.0701 1.212
Apr 6 Wed 8:00 Dry 1.00 0.058 1.000 0.987 0.0572 0.990
Apr 6 Wed 8:15 Dry 1.00 0.058 1.000 0.987 0.0572 0.990
Apr 6 Wed 8:30 Dry 1.00 0.058 1.000 0.987 0.0572 0.990
Apr 6 Wed 8:45 Dry 1.00 0.058 1.000 0.987 0.0572 0.990
Apr 6 Wed 9:00 Dry 1.00 0.047 1.000 0.987 0.0464 0.802
Apr 6 Wed 9:15 Dry 1.00 0.047 1.000 0.987 0.0464 0.802
Apr 6 Wed 9:30 Dry 1.00 0.047 1.000 0.987 0.0464 0.802
Apr 6 Wed 9:45 Dry 1.00 0.047 1.000 0.987 0.0464 0.802
Step 5: Estimate Incident Events

The procedure described in this step is used to predict incident event dates,
times, and durations. It also determines each incident event’s type (i.e., crash or
noncrash), severity level, and location on the facility. The procedure uses
weather event and demand variation information from the two previous steps as
part of the incident prediction process. Crash frequency data are used to estimate
the frequency of both crash-related incidents and non-crash-related incidents.
For an urban street reliability evaluation, incidents are categorized as being
• Segment-related or
• Intersection-related.
These two categories are mutually exclusive.
Identify Input Data

Incident frequency data. Three-year average crash frequencies are
determined from locally available crash records for each segment and
intersection along the facility. These averages are shown in Exhibit 29-68. The
frequency of noncrash incidents is estimated from the crash frequency data in a
subsequent step. Noncrash incident frequency is not an input quantity due to the
difficulty agencies have in acquiring noncrash incident data.

Crash Frequency Exhibit 29-68

Location (crashes/year) Example Problem 4: Locally
Segment 1-2 (Intersections 1 to 2) 15 Available Crash Frequency
Segment 2-3 (Intersections 2 to 3) 16 Data
Segment 3-4 (Intersections 3 to 4) 17
Intersection 1 32
Intersection 2 33
Intersection 3 34
Intersection 4 35
Intersection 5 36
Intersection 6 37
Intersection 7 38
Work zone/special event crash frequency adjustment factors. Work zones

and special events are not being considered in this example; therefore, these
crash frequency adjustment factors do not need to be provided.
Weather event crash frequency adjustment factors. The default crash
frequency adjustment factors given in Exhibit 17-9 are used.
Incident duration factors. The default incident detection and response times
given in Exhibit 17-9 and the default clearance times given in Exhibit 17-10 are
used.
Incident distribution. The default incident distribution given in Exhibit 17-
11 for urban street facilities with shoulders is used.
Compute Equivalent Crash Frequency for Weather

This step converts the average crash frequencies (supplied as input data) into
the equivalent crash frequencies for each weather type.
First, the input crash frequency data for segments and intersections are
converted into an equivalent crash frequency for each of the following weather
conditions: clear and dry, rainfall, wet pavement (not raining), and snow or ice
on pavement (not snowing). This conversion is based on the number of hours
during a 2-year period that a particular weather condition occurs and the crash
adjustment factor corresponding to each weather condition. For this example
problem, the number of hours in a year with a particular weather condition is
determined from the default weather data for Lincoln, Nebraska.
The equivalent crash frequency when every day is dry for street location i is
computed with Equation 29-13 and Equation 29-14.
Exhibit 29-69 illustrates the computations of the equivalent crash frequencies
by weather type for two segments and three intersections. The calculations are
similar for the other segments and intersections.

Exhibit 29-69 Segments Intersections

Example Problem 4: Variable Definition 1-2 2-3 1 2 3
Computation of Crash Fcstr(i) Observed average crash frequency 15 16 65 66 67
Frequency by Weather Type Ny Number of years 2 2 2 2 2
Nhdry Hours of dry weather 17026.98 17026.98 17026.98 17026.98 17026.98
Nhrf Hours of rainfall 278.22 278.22 278.22 278.22 278.22
Nhwp Hours of wet pavement 104.33 104.33 104.33 104.33 104.33
Nhsf Hours of snowfall 64.61 64.61 64.61 64.61 64.61
Nhsp Hours of snow or ice on pavement 45.86 45.86 45.86 45.86 45.86
Crash frequency adjustment factors
for…
CFAFrf Rainfall 2.0 2.0 2.0 2.0 2.0
CFAFwp Wet pavement 3.0 3.0 3.0 3.0 3.0
CFAFsf Snowfall 1.5 1.5 1.5 1.5 1.5
CFAFsp Snow or ice on pavement 2.75 2.75 2.75 2.75 2.75
Calculated crash frequencies for…
Fcstr(i),dry Dry weather 14.50 15.47 30.94 31.91 32.88
Fcstr(i),rf Rainfall 29.01 30.94 61.89 63.82 65.75
Fcstr(i),wp Wet pavement 43.51 46.41 92.83 95.73 98.63
Fcstr(i),sf Snowfall 21.76 23.21 46.41 47.86 49.32
Fcstr(i),sp Snow or ice on pavement 39.89 42.54 85.09 87.75 90.41
Note: Total hours of dry, rainfall, wet pavement, snowfall, and snow or ice on pavement = 17,520 h (2 years).
Establish Crash Frequency Adjustment Factors for Work Zones or Special

Events
This step is skipped because work zones and special events are not being
considered for this evaluation.
Determine Whether an Incident Occurs

This step goes through each of the 24 h of each day that is represented in the
reliability reporting period. For each hour, whether an incident occurs is
determined. If an incident occurs, its duration is also determined. Finally, for
each incident identified in this manner, whether some portion (or all) of the
incident occurs during a portion of the study period is determined.
Weather-adjusted incident frequencies. First, for a given hour in a given
day, the weather event data are checked to see which weather condition (dry,
rainfall, snowfall, wet pavement and not raining, or snow or ice on pavement
and not snowing) was generated for that hour. The expected incident frequencies
for street locations (i.e., segments and intersections) Fistr(i),wea(h,d) are determined
from (a) the corresponding crash frequency for the given weather condition
Fcstr(i),wea (from a previous step) and (b) a factor pcstr,wea relating total crashes to total
incidents for the given weather condition (from the default values in the third
column of Exhibit 17-11). If a special event or work zone was present on the
given hour and day, the expected incident frequency is multiplied by the
segment or intersection (as appropriate) crash frequency adjustment factor
CFAFstr specified by the analyst for special events and work zones. Equation 29-
15 is used to compute the expected incident frequency:
𝐹𝑐𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎
𝐹𝑖𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑) = 𝐶𝐹𝐴𝐹𝑠𝑡𝑟
𝑝𝑐𝑠𝑡𝑟,𝑤𝑒𝑎
For example, weather was dry on Wednesday, April 6, at 9:00 a.m. For
Segment 1-2, the equivalent crash frequency for dry weather is 14.50 crashes/year
(from Exhibit 29-69). The ratio of crashes to incidents for segments in dry
weather is 0.358. There is no work zone or special event, so the crash adjustment
factor is 1.0. Then

14.50
𝐹𝑖𝑠𝑒𝑔1−2,dry = (1.0) = 40.5 incidents/year
0.358
Similarly, snow was falling on Monday, January 10, at 7:00 a.m. The
equivalent crash frequency for snowfall on Segment 1-2 is 21.76 crashes/year. The
ratio of crashes to incidents for segments in snowy weather is 0.358. Therefore,
21.76
𝐹𝑖𝑠𝑒𝑔1−2,𝑠𝑓 = (1.0) = 60.8 incidents/year
0.358
Conversion to hourly frequencies. Next, the incident frequency Fistr(i),wea(h,d) is
converted to an hourly frequency fistr(i),wea(h,d),h,d by multiplying it by the percent of
annual demand represented by the hour and by dividing by the number of days
in a year (expressed as a ratio of hours). The same hour-of-day fhod,h,d, day-of-week
fdow,d, and month-of-year fmoy,d demand ratios used in Step 4 are used here.
Equation 29-16 is used, where “8,760” represents the number of hours in a year
and “24” represents the number of hours in a day.
𝐹𝑖𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑)
𝑓𝑖𝑠𝑡𝑟(𝑖),𝑤𝑒𝑎(ℎ,𝑑),ℎ,𝑑 = (24𝑓ℎ𝑜𝑑,ℎ,𝑑 )𝑓𝑑𝑜𝑤,𝑑 𝑓𝑚𝑜𝑦,𝑑
8,760
The month-of-year demand ratio for April is 0.987, the day-of-week demand
ratio for Wednesday is 1.00, and the hour-of-day demand ratio for 9:00 a.m. is
0.047. The incident frequency for this day and time is calculated above as 40.5
incidents per year. Therefore, the equivalent hourly incident frequency for
Segment 1-2 on Wednesday, April 6, at 9:00 a.m. is
40.5
𝑓𝑖𝑠𝑒𝑔1−2,dry,0900,Apr06 = (24 × 0.047)(1.00)(0.987) = 0.00515 incidents/h
8,760
Similarly, the equivalent hourly incident frequency for Segment 1-2 on
Monday, January 10, at 7:00 a.m. is
60.8
𝑓𝑖𝑠𝑒𝑔1−2,𝑠𝑓,0700,Jan10 = (24 × 0.071)(0.98)(0.831) = 0.00963 incidents/h
8,760
Probability of no incidents. Incidents for a given day, street location,
incident type, and hour of day are assumed to follow a Poisson distribution, as
given in Equation 29-17.
Exhibit 29-70 demonstrates the determination of incidents for Segment 1-2 on
April 6 for the 9:00 a.m. hour. Exhibit 29-71 does the same for January 10 for the
7:00 a.m. hour.
If more than one incident occurs at the same time and location, the more
serious incident is considered in the methodology. During an incident, the
methodology requires that at least one lane remain open in each direction of
travel on a segment and on each intersection approach. If the number of lanes
blocked by an incident is predicted to equal the number of lanes available on the
segment or intersection approach, one lane is maintained open and the
remaining lanes are blocked. For example, if the segment has two lanes in the
subject travel direction and an incident occurs and is predicted to block two
lanes, the incident is modeled as blocking only one lane.

Exhibit 29-70 Hourly

Example Problem 4: Incident Incident Incident exp Random Incident
Determination for April 6, Incident Type Proportion Frequency (-fi × pi) Number ?
9:00 a.m., for Segment 1-2 Crash 1 lane Fatal/injury 0.036 0.00515 0.99981 0.90019 No
Crash 1 lane PDO 0.083 0.00515 0.99957 0.38078 No
Crash 2 lane Fatal/injury 0.028 0.00515 0.99986 0.90860 No
Crash 2 lane PDO 0.030 0.00515 0.99984 0.06081 No
Crash Shoulder Fatal/injury 0.021 0.00515 0.99990 0.82183 No
Crash Shoulder PDO 0.016 0.00515 0.99918 0.34916 No
Noncrash 1 lane Breakdown 0.456 0.00515 0.99766 0.99900 Yes
Noncrash 1 lane Other 0.089 0.00515 0.99954 0.59842 No
Noncrash 2 lane Breakdown 0.059 0.00515 0.99970 0.69323 No
Noncrash Shoulder Breakdown 0.014 0.00515 0.99993 0.13012 No
Noncrash Shoulder Other 0.007 0.00515 0.99996 0.44620 No
Notes: Incident proportions total 100%. PDO = property damage only.
Random numbers have been selected to illustrate this particular step of the computations. They are not
necessarily the same results that would be achieved in a full run of the procedure.
Exhibit 29-71 Hourly

Example Problem 4: Incident Incident Incident exp Random Incident
Determination for January 10, Incident Type Proportion Frequency (-fi × pi) Number ?
7:00 a.m., for Segment 1-2 Crash 1 lane Fatal/injury 0.036 0.00963 0.99965 0.21041 No
Crash 1 lane PDO 0.083 0.00963 0.99920 0.83017 No
Crash 2 lane Fatal/injury 0.028 0.00963 0.99973 0.58437 No
Crash 2 lane PDO 0.030 0.00963 0.99971 0.80487 No
Crash Shoulder Fatal/injury 0.021 0.00963 0.99981 0.35441 No
Crash Shoulder PDO 0.016 0.00963 0.99846 0.64888 No
Noncrash Shoulder Breakdown 0.014 0.00963 0.99987 0.30502 No
Noncrash Shoulder Other 0.007 0.00963 0.99993 0.33279 No
Note: Incident proportions total 100%. PDO = property damage only.
Random numbers have been selected to illustrate this particular step of the computations. They are not
necessarily the same results that would be achieved in a full run of the procedure.
Determine Incident Duration

If the result of the previous step indicates that an incident occurs in a given
segment or intersection during a given hour and day, the incident duration is
then determined randomly from a gamma distribution by using the average
incident duration and the standard deviation of incident duration as inputs.
These values are supplied as input data.
The duration is used in a subsequent step to determine which analysis
periods are associated with an incident. The incident duration is rounded to the
nearest quarter hour for 15-min analysis periods. This rounding is performed to
ensure the most representative match between event duration and analysis
period start and end times. This approach causes events that are shorter than
one-half the analysis period duration to be ignored (i.e., they are not recognized
in the scenario generation process).
Exhibit 29-70 shows that a noncrash, one-lane, breakdown incident was
generated for Segment 1-2 on April 6 starting at the 9:00 a.m. hour. Exhibit 29-72
shows the inputs into the incident duration calculation and the result. As with
other computations in this example problem involving random numbers,
different values are obtained if the random number seed is changed.

Variable Value Exhibit 29-72

Location Segment 1-2 Example Problem 4: Sample
Incident type Noncrash Calculation of Incident
Number of lanes involved 1-lane Duration
Incident severity Breakdown
Weather Dry
Incident detection time (min) 2.0
Incident response time, dry weather (min) 15.0
Incident clearance time (min) 10.8
Average incident duration (min) 27.8
Standard deviation of incident duration (min) 22.2
Average incident duration (h) 0.463
Standard deviation of incident duration (h) 0.371
Random number 0.57455
Gamma function alpha parameter (mean2/variance) 1.5625
Gamma function beta parameter (variance/mean) 0.2965
Duration (h) 0.433
Rounded duration (nearest 15 min) (h) 0.50
Incident start time 9:00
Incident end time 9:30
Determine Incident Location

If an incident occurs at a segment or intersection during a given hour and
day, its location is determined in this step. For intersections, the location is one of
the intersection legs. For segments, the location is one of the two segment travel
directions.
In the case of the incident identified on Segment 1-2 at 9:00 a.m. on April 6,
the two directions of the segment have equal traffic volumes (see Exhibit 29-62)
and therefore have equal probability of having the incident occur. This time, the
scenario generator randomly assigned the incident to the westbound direction
(identified as being associated with NEMA Phase 6 at the intersection).
Identify Analysis Period Incidents

The preceding steps of the incident estimation procedure are repeated for
each hour of each day in the reliability reporting period. During this step, the
analysis periods associated with an incident are identified. Specifically, each
hour of the study period is examined to determine whether it coincides with an
incident. If an incident occurs, its event type, lane location, severity, and street
location are identified and recorded. Each subsequent analysis period coincident
with the incident is also recorded.
Step 6: Generate Scenarios

This step uses the results from Steps 3 to 5 to create one scenario for each
analysis period in the reliability reporting period. The base dataset coded in
Step 2 represents the “seed” file from which the new scenarios are created.
As discussed previously, each analysis period is considered to be one
scenario. There are 3,120 analysis periods in the reliability reporting period
(= 4 analysis periods/hour × 3 hours/day × 5 days/week × 52 weeks/year ×
1 year/reporting period). Thus, there are 3,120 scenarios.
Each scenario created in this step includes the appropriate adjustments to
segment running speed and intersection saturation flow rate associated with the
weather events or incidents that are predicted to occur during the corresponding

analysis period. If an analysis period has an incident, the number of lanes is

reduced, the saturation flow rate is adjusted for affected intersection lanes, and a
free-flow speed adjustment factor is applied to the affected lanes in the segment.
If an analysis period has rainfall, snowfall, wet pavement, or snow or ice on the
pavement, the saturation flow rate is adjusted for all intersections, the free-flow
speed is adjusted for all segments, and the left-turn critical headways are
adjusted for all intersections.
The traffic demand volumes in each dataset are adjusted for monthly,
weekly, and hourly variations.
Step 7: Apply the Chapter 16 Motorized Vehicle Methodology

The analysis methodology for urban street facility evaluation is applied to
each scenario generated in the previous step. At the conclusion of this step, the
delay and queue length for each intersection, as well as the speed and travel time
for each segment, are computed for each scenario.
Step 8: Conduct Quality Control and Error Checking

The quality control of thousands of scenarios is difficult, so it is
recommended that the analyst focus on error checking and quality control on the
base dataset. The results should be error-checked to the analyst’s satisfaction to
ensure that they accurately represent real-world congestion on the facility under
recurring demand conditions with no incidents and under dry weather
conditions. The same criteria for error checking should be used as for a
conventional HCM analysis, but with the recognition that any error in the base
dataset will be crucial, because it will be reproduced thousands of times by the
scenario generator.
The total delay for each scenario should be scanned to identify the study
periods likely to be associated with exceptionally long queues. For a given study
period, the final queue on each entry intersection approach for the last analysis
period should not be longer than the corresponding initial queue for the first
analysis period. The study period duration should be increased (i.e., started
earlier, ended later) such that this condition is satisfied. Ideally, the study period
is sufficiently long that these reference initial and final queues both equal zero.
An efficient approach for making this check is to start by evaluating the scenario
with the largest total delay.
Step 9: Interpret Results

This step examines the reliability results for the existing facility. These results
are listed in Exhibit 29-73. Although both travel directions have the same volume
and capacity, several of the values in this exhibit vary slightly by travel direction
because of the use of Monte Carlo methods.
The vehicle miles traveled (VMT) is computed for each scenario and added
for all scenarios in the reliability reporting period. This statistic describes overall
facility utilization for the reliability reporting period.

Measure Eastbound Westbound Exhibit 29-73

Vehicle miles traveleda 2,260 2,257 Example Problem 4: Reliability
Number of scenariosa 3,120 3,120 Performance Measure Results
Base free-flow travel timeb (s) 262.9 262.9
Mean TTIb 1.69 1.64
80th percentile TTI 1.57 1.56
95th percentile TTI (PTI) 2.98 2.61
Reliability rating 93.2 94.1
Total delayb (veh-h) 72.0
a
Notes: This statistic represents a total for the reliability reporting period.
b
This statistic represents an average of the value for each scenario (i.e., an average value for all
scenarios).
The travel time indices shown in Exhibit 29-73 were computed by finding the
average (i.e., mean), 80th, and 95th percentile travel times for a given direction of
travel across all scenarios and dividing by the facility’s base free-flow speed.
Since hourly demands, geometry, weather, and signal timings are identical in
both directions, the differences between the indices illustrate the effects of
random variation in incidents and 15-min demands for the two directions.
The reliability rating describes the percent of VMT on the facility associated
with a TTI less than 2.5. A facility that satisfies this criterion during a given
scenario is likely to provide LOS D or better for that scenario. The reliability
ratings shown in the exhibit indicate that more than 90% of the vehicle miles of
travel on the facility are associated with LOS D or better.
The total delay (in vehicle hours) combines the delay per vehicle and volume
of all intersection lane groups at each intersection during a scenario. This statistic
increases with an increase in volume or delay. It is the only statistic of those
listed in Exhibit 29-73 that considers the performance of all traffic movements
(i.e., the other measures consider just the major-street through movement).
Hence, it is useful for quantifying the overall change in operation associated with
a strategy. When considered on a scenario-by-scenario basis, this statistic can be
used to identify those scenarios with extensive queuing on one or more “entry”
approaches (i.e., the cross-street intersection approaches and the major-street
approaches that are external to the facility).
Exhibit 29-74 shows the travel time distribution for the facility’s eastbound
travel direction. That for the westbound direction has a similar shape. The longer
travel times tend to be associated with poor weather. The longest travel times
coincide with one or more incidents and poor weather.
The reliability methodology was repeated several times to examine the
variability in the reliability performance measures. Each replication used the
same input data, with the exception that the three random numbers were
changed for each replication. Exhibit 29-75 shows the predicted average and 95th
percentile travel times for the eastbound travel direction based on five replications.

Exhibit 29-74
Example Problem 4:
Eastbound Travel Time
Distribution
Exhibit 29-75 Replication Average Travel Time (s) 95th Percentile Travel Time (s)
Example Problem 4: 1 443.7 783.8
Confidence Interval 2 441.4 787.5
Calculation for Eastbound 3 432.8 758.4
Direction 4 439.3 740.0
5 433.7 772.9
Average 438.2 768.5
Standard deviation 4.79 19.6
95th% confidence interval 432.2–444.1 744.4–792.8
(±1.36%) (±3.16%)
The last three rows of Exhibit 29-75 show the statistics for the sample of five
observations. The 95th percentile confidence interval was computed by using
Equation 17-3. The confidence interval for the average travel time is 432.2 to
441.1 s, which equates to ±1.36% of the overall average travel time. Similarly, the
confidence interval for the 95th percentile travel time is ±3.16% of the average of
the 95th percentile travel times. This confidence interval is larger than that of the
average travel time because the 95th percentile travel time tends to be influenced
more by the occurrence of incidents and poor weather. As suggested by the
formulation of Equation 17-3, the confidence interval can be reduced in width by
increasing the number of replications.
The contribution of demand, incidents, and weather to total vehicle hours of
delay (VHD) during the reliability reporting period is used to determine the
relative contributions of each factor to the facility’s reliability. The annual VHD
takes into account both the severity of the event and its likelihood of occurrence.
VHD is computed by identifying the appropriate category for each scenario and
adding the estimated VHD for each scenario in this category. The results are
summed for all scenarios in each category in the reliability reporting period.
They are presented in Exhibit 29-76 and Exhibit 29-77. The categories have been
condensed to facilitate the diagnosis of the primary causes of reliability problems
on the urban street. Demand has been grouped into two levels. All foul weather
and incident scenarios have been grouped into a single category each.
An examination of the cell values in Exhibit 29-77 yields the conclusion that
the single most significant cause of annual delay on the urban street example is
high demand, which accounts for 53.6% of annual delay during fair weather with

no incidents. Incidents or bad weather collectively account for 22.9% of annual

delay on the facility (17.8% + 7.3% + 2.8% – 5.1% – 0.0%).
Total Delay by Demand and Weather (veh-h) Exhibit 29-76

Low Demand High Demand Example Problem 4: Annual
Fair Weather Foul Weather Fair Weather Foul Weather Total VHD by Cause
No incidents 52,957 6,337 120,393 5,025 184,712
Incidents 5,865 23 22,714 11,437 40,039
Total 58,822 6,360 143,107 16,462 224,751
Low Demand High Demand Exhibit 29-77

Fair Weather Foul Weather Fair Weather Foul Weather Total Example Problem 4:
No incidents 23.6% 2.8% 53.6% 2.2% 82.2% Percentage of Annual VHD by
Incidents 2.6% 0.0% 10.1% 5.1% 17.8% Cause
Total 26.2% 2.8% 63.7% 7.3% 100.0%
EXAMPLE PROBLEM 5: URBAN STREET STRATEGY EVALUATION

Objective
This example problem illustrates an application of the reliability
methodology for alternatives analysis. The objective is to demonstrate the utility
of reliability information in evaluating improvement strategies. The strategies
considered in this example involve changes in the urban street’s geometric
design or its signal operation. These changes are shown to affect traffic operation
and safety, both of which can influence reliability.
Site
The same urban street described in Example Problem 4 is used in this
example problem.
Required Input Data

The same types of required input data described in Example Problem 4 are
used here. The conditions described in Example Problem 4 are used as the
starting point for evaluating each of three strategies that have been identified as
having the potential to improve facility reliability. One base dataset is used to
describe the “existing” facility of Example Problem 4, while one base dataset is
associated with each strategy, resulting in a total of four base datasets. Specific
changes to the Example Problem 4 base dataset required to represent each
strategy are described later. The three strategies are as follows:
1. Shift 5 s from the cross-street left-turn phase to the major-street through
phase.
2. Change the major-street left-turn mode from protected-only to protected-
permitted.
3. Eliminate major-street right-turn bays and add a second lane to major-
street left-turn bays.
These strategies were formulated to address a capacity deficiency for the
major-street through movements at each intersection. This deficiency was noted
as part of the analysis described in Example Problem 4. The change associated with
each strategy was implemented at each of the seven intersections on the street.

For this example problem, the changes needed to implement the strategies
require changes only to the base datasets. However, some strategies may require
changes to the reliability methodology input data, the base datasets, or the
alternative datasets.
Computational Steps
This example problem proceeds through the following steps:
1. Establish the purpose, scope, and approach.
2. Code datasets.
3. Generate scenarios.
4. Apply the Chapter 16 motorized vehicle methodology.
5. Interpret results.
Step 1: Establish the Purpose, Scope, and Approach

Define the Purpose
The agency responsible for this urban street wishes to perform a reliability
analysis of existing conditions to determine which of the three strategies offers
the greatest potential for improvement in facility reliability.
Define the Reliability Analysis Box

The results from a preliminary evaluation of the facility were used to define
the general spatial and temporal boundaries of congestion on the facility under
fair weather, nonincident conditions. A study period consisting of the weekday
morning peak period (7–10 a.m.) and a study area consisting of the 3-mi length of
facility between Intersections 1 and 7 encompass all of the recurring congestion.
The reliability reporting period is desired to include all weekdays during the
course of a year. The duration of the analysis period will be 15 min.
Select Reliability Performance Measures

Reliability will be reported by using the following performance measures:
mean TTI, 80th percentile TTI, 95th percentile TTI (PTI), reliability rating, and
total delay (in vehicle hours) for the reliability reporting period.
Step 2: Code Datasets

Code the Base Dataset
The first base dataset represents existing conditions and is identical to the
base dataset described in Example Problem 4. This base dataset was modified as
follows to create a new base dataset (three in all) for each strategy being evaluated:
• The signal timing parameters for the Strategy 1 base dataset were
modified at each intersection to reduce the phase splits for the minor-
street left-turn movements by 5 s and to increase the phase splits for the
major-street through movements by 5 s.
• The signal timing parameters for the Strategy 2 base dataset were
modified at each intersection to change the major-street left-turn mode

from protected-only to protected-permitted. Furthermore, Chapter 12 of

the Highway Safety Manual (10) indicates that intersection crash frequency
increases by 11% on average when this change is made. Therefore, the
crash frequency input data for each intersection were increased to reflect
this change.
• The geometric parameters for the Strategy 3 base dataset were modified at
each intersection to eliminate the major-street right-turn bays and to add a
second lane to the major-street left-turn bays. Furthermore, Chapter 12 of
the Highway Safety Manual (10) indicates that intersection crash frequency
increases by 9% for this change. Therefore, the crash frequency input data
for each intersection were increased to reflect this change.
Code the Alternative Datasets

Since no work zones are planned in the next year and no special events affect
the facility on weekdays, only the base datasets will be required.
Step 3: Generate Scenarios

During this step, the reliability methodology is used to create one scenario
for each analysis period in the reliability reporting period. The base datasets
coded in Step 2 represent the “seed” files from which the scenarios associated
with each strategy are created. As in Example Problem 4, one set of 3,120
scenarios is created for the existing facility. Additional sets of 3,120 scenarios are
created for each of the three strategies.
Step 4: Apply the Chapter 16 Motorized Vehicle Methodology

The analysis methodology for urban street facility evaluation is applied to
each scenario generated in the previous step, as described in Example Problem 4.
Step 5: Interpret Results

This step examines the reliability results for the facility. Initially, the results
for the existing facility are described. Then, the results for each of the three
strategies are summarized and compared with those of the existing facility. The
formulation of these strategies was motivated by an examination of the results
for the existing facility. The examination indicated that the major-street through
movements had inadequate capacity during the morning peak traffic hour for
several high-volume months of the year.
Results for the Existing Facility

The results for the existing facility are the same as for Example Problem 4,
given previously in Exhibit 29-73 through Exhibit 29-77.
Results for Strategy 1

In Strategy 1, 5 s are taken from the cross-street left-turn phase split. This
change increases the time available to the major-street through (i.e., coordinated)
phase and increases the through movement capacity. The results for this strategy
are listed in Exhibit 29-78. The first two rows list the average values obtained

from five replications. The third row lists the change in the performance measure
value. The last row indicates whether the change is statistically significant.
Exhibit 29-78 Travel Time (s) Total Delay Reliability

Example Problem 5: Results Case Average 95th Percentile (veh-h) Rating
for Strategy 1
Existing 438.2 768.5 70.7 93.2
Strategy 1 400.7 542.2 66.2 96.8
Change -37.5 -226.3 -4.5 3.6
Significant? Yes Yes Yes Yes
Note: Results based on five replications.
The statistics in Exhibit 29-78 indicate that the strategy produces a relatively
large improvement in travel time, particularly in the 95th percentile travel time.
The strategy improves reliability during the peak hour for the high-volume
months, which is reflected by the increase in the reliability rating. It forecasts an
increase of 3.6% in the VMT for which LOS D or better is provided. On the other
hand, delay to the cross-street left-turn movements increases. This increase
partially offsets the decrease in delay to the major-street through movements.
This trade-off is reflected by a small reduction of 4.5 veh-h total delay.

In Strategy 2, the major-street left-turn mode is changed from protected-only
to protected-permitted. This change reduces the time required by the major-
street left-turn phase, which increases the time available to the coordinated phase
and increases the through movement capacity. The results of the evaluation of
this strategy are given in Exhibit 29-79.
Exhibit 29-79 Travel Time (s) Total Delay Reliability

Example Problem 5: Results Case Average 95th Percentile (veh-h) Rating
for Strategy 2
Existing 438.2 768.5 70.7 93.2
Strategy 2 382.9 473.5 49.6 97.3
Change -55.3 -295.0 -21.1 4.1
Significant? Yes Yes Yes Yes
The statistics in Exhibit 29-79 indicate that Strategy 2 produces a relatively

large improvement in travel time, particularly in the average travel time, relative
to Strategy 1. The strategy improves reliability during the peak hour for the high-
volume months, reflected by the increase in the reliability rating. It forecasts an
increase of 4.1 percent in the VMT for which LOS D or better is provided. The
delay to the major-street through movements decreases without a significant
increase in the delay to the other movements. This trend is reflected by the
notable reduction of 21.1 veh-h total delay.

In Strategy 3, the major-street right-turn bays are eliminated and second
lanes are added to the major-street left-turn bays. This change reduced the time
required by the major-street left-turn phase, which increased the time available
to the coordinated phase and increased the through movement capacity. The
results for this strategy are listed in Exhibit 29-80.

Travel Time (s) Total Delay Reliability Exhibit 29-80

Case Average 95th Percentile (veh-h) Rating Example Problem 5: Results
for Strategy 3
Existing 438.2 768.5 70.7 93.2
Strategy 3 410.0 460.2 59.0 98.5
Change -28.2 -308.3 -11.7 5.3
Significant? No Yes Yes Yes
The statistics in Exhibit 29-80 indicate that the strategy produces a relatively
large improvement in travel time, particularly in the 95th percentile travel time.
The strategy improves reliability during the peak hour for the high-volume
months, reflected by the increase in the reliability rating. It forecasts an increase
of 5.3% in the VMT for which LOS D or better is provided. Delay to the major-
street through movements decreases, as reflected by the reduction of 11.7 veh-h
total delay. The change in average travel time is not statistically significant
because the loss of the right-turn bays shifts the location of many incidents from
the bays to the through lanes. This shift causes the average travel time for
Strategy 3 to vary more widely among scenarios.
Summary of Findings
All three strategies improved the facility’s reliability and overall operation.
Strategy 1 (shift 5 s to the coordinated phase) provides some improvement in
reliability of travel through the facility and some reduction in total delay in the
system.
Strategy 2 (protected-only to protected-permitted) provides the lowest average
travel time and the lowest total delay. It also provides a notable improvement in
travel reliability.
Strategy 3 (eliminate right-turn lanes, increase left-turn lanes) provides the
biggest improvement in reliability of travel. It also provides some overall benefit in
terms of lower travel time and total delay.
The selection of the best strategy should include consideration of the change
in road user costs, as measured in terms of reliability, total delay, and crash
frequency. Viable strategies are those for which the reduction in road user costs
exceeds the construction costs associated with strategy installation and
maintenance.

6. REFERENCES
Some of these references can 1. Zegeer, J., J. Bonneson, R. Dowling, P. Ryus, M. Vandehey, W. Kittelson, N.
Reference Library in Volume 4. Rouphail, B. Schroeder, A. Hajbabaie, B. Aghdashi, T. Chase, S. Sajjadi, R.
Margiotta, and L. Elefteriadou. Incorporating Travel Time Reliability into the
2. Comparative Climatic Data for the United States Through 2010. National Climatic
Data Center, National Oceanic and Atmospheric Administration, Asheville,
N.C., 2011. http://www.ncdc.noaa.gov. Accessed Sept. 21, 2011.
3. Urbanik, T., A. Tanaka, B. Lozner, E. Lindstrom, K. Lee, S. Quayle, S. Beaird,
S. Tsoi, P. Ryus, D. Gettman, S. Sunkari, K. Balke, and D. Bullock. NCHRP
Report 812: Signal Timing Manual, 2nd ed. Transportation Research Board,
4. Husch, D., and J. Albeck. Synchro Studio 7 User’s Guide. Trafficware, Ltd.,
2006.
5. Wallace, C., K. Courage, M. Hadi, and A. Gan. TRANSYT-7F User’s Guide,
Vol. 4 in a Series: Methodology for Optimizing Signal Timing. University of
Florida, Gainesville, March 1998.
6. Corridor-Microscopic Simulation Program (CORSIM) Version 6.1 User's Guide.
University of Florida, Gainesville, 2008.
7. VISSIM 5.10 User Manual. PTV Vision, Karlsruhe, Germany, 2008.
8. Rodegerdts, L., J. Bansen, C. Tiesler, J. Knudsen, E. Myers, M. Johnson, M.
Moule, B. Persaud, C. Lyon, S. Hallmark, H. Isebrands, R. B. Crown, B.
Guichet, and A. O’Brien. NCHRP Report 672: Roundabouts: An Informational
Guide, 2nd ed. Transportation Research Board of the National Academies,
9. Rainfall Frequency Atlas of the U.S.: Rainfall Event Statistics. National Climatic
Data Center, National Oceanic and Atmospheric Administration, Asheville,
N.C., 2011. http://www.ncdc.noaa.gov/oa/documentlibrary/rainfall.html.
Accessed Sept. 21, 2011.
10. Highway Safety Manual, 1st ed. American Association of State Highway and
Transportation Officials, Washington, D.C., 2010.
References Chapter 29/Urban Street Facilities: Supplemental

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 30
URBAN STREET SEGMENTS: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION .............................................................................................. 30-1
2. TRAFFIC DEMAND ADJUSTMENTS .......................................................... 30-2

Capacity Constraint and Volume Balance ...................................................... 30-2
Origin–Destination Distribution....................................................................... 30-4
3. SIGNALIZED SEGMENT ANALYSIS .......................................................... 30-7

Discharge Flow Profile ....................................................................................... 30-7
Running Time ..................................................................................................... 30-8
Projected Arrival Flow Profile .......................................................................... 30-8
Proportion of Time Blocked .............................................................................30-11
Sustained Spillback............................................................................................30-12
Midsegment Lane Restriction ..........................................................................30-19
4. DELAY DUE TO TURNS ................................................................................ 30-21

Delay due to Left Turns ....................................................................................30-21
Delay due to Right Turns .................................................................................30-26
5. PLANNING-LEVEL ANALYSIS APPLICATION...................................... 30-29

Overview of the Application............................................................................30-29
Required Data and Sources ..............................................................................30-29
Methodology ......................................................................................................30-29
Example Problem...............................................................................................30-36
6. FIELD MEASUREMENT TECHNIQUES .................................................... 30-38

Free-Flow Speed ................................................................................................30-38
Average Travel Speed .......................................................................................30-39
7. COMPUTATIONAL ENGINE DOCUMENTATION ............................... 30-42

Flowcharts ..........................................................................................................30-42
Linkage Lists ......................................................................................................30-45
Chapter 30/Urban Street Segments: Supplemental Contents

8. EXAMPLE PROBLEMS ...................................................................................30-48

Example Problem 1: Motorized Vehicle LOS ................................................ 30-48
Example Problem 2: Pedestrian LOS ............................................................. 30-56
Example Problem 3: Bicycle LOS .................................................................... 30-62
Example Problem 4: Transit LOS .................................................................... 30-66
9. ROUNDABOUT SEGMENT METHODOLOGY .......................................30-71

Scope of the Methodology ............................................................................... 30-71
Limitations of the Methodology ..................................................................... 30-71
Required Input Data and Sources .................................................................. 30-71
Geometric Design Data .................................................................................... 30-72
Computational Steps ........................................................................................ 30-73
10. REFERENCES ....................................................................................................30-82
Contents Chapter 30/Urban Street Segments: Supplemental

LIST OF EXHIBITS
Exhibit 30-1 Entry and Exit Movements on the Typical Street Segment ............. 30-2
Exhibit 30-2 Default Seed Proportions for Origin–Destination Matrix ............... 30-5
Exhibit 30-3 Platoon Dispersion Model ................................................................... 30-9
Exhibit 30-4 Arrival Flow Profile Estimation Procedure ..................................... 30-10
Exhibit 30-5 Estimation of Blocked Period Duration ........................................... 30-12
Exhibit 30-6 Vehicle Trajectories During Spillback Conditions ......................... 30-13
Exhibit 30-7 Required Input Data for the Planning-Level Analysis
Application ......................................................................................................... 30-29
Exhibit 30-8 Planning-Level Analysis Application for Urban Street
Segments ............................................................................................................. 30-30
Exhibit 30-9 Planning-Level Analysis: Running Time Worksheet ..................... 30-31
Exhibit 30-10 Planning-Level Analysis: Proportion Arriving During
Green Worksheet ............................................................................................... 30-32
Exhibit 30-11 Planning-Level Analysis: Control Delay Worksheet ................... 30-33
Exhibit 30-12 Planning-Level Analysis: Stop Rate Worksheet ........................... 30-34
Exhibit 30-13 Planning-Level Analysis: Travel Speed and Spatial Stop
Rate Worksheet .................................................................................................. 30-35
Exhibit 30-14 Planning-Level Analysis: Example Problem ................................. 30-36
Exhibit 30-15 Travel Time Field Worksheet .......................................................... 30-41
Exhibit 30-16 Methodology Flowchart ................................................................... 30-42
Exhibit 30-17 Setup Module .................................................................................... 30-43
Exhibit 30-18 Segment Evaluation Module ........................................................... 30-43
Exhibit 30-19 Segment Analysis Module ............................................................... 30-44
Exhibit 30-20 Delay due to Turns Module ............................................................ 30-44
Exhibit 30-21 Performance Measures Module ...................................................... 30-45
Exhibit 30-22 Segment Evaluation Module Routines........................................... 30-46
Exhibit 30-23 Segment Analysis Module Routines .............................................. 30-46
Exhibit 30-24 Delay due to Turns Module Routines ............................................ 30-47
Exhibit 30-25 Example Problems ............................................................................ 30-48
Exhibit 30-26 Example Problem 1: Urban Street Segment Schematic ................ 30-48
Exhibit 30-27 Example Problem 1: Intersection Turn Movement Counts ......... 30-49
Exhibit 30-28 Example Problem 1: Signal Conditions for Intersection 1 ........... 30-49
Exhibit 30-29 Example Problem 1: Geometric Conditions and Traffic
Characteristics for Signalized Intersection 1 .................................................. 30-50
Exhibit 30-30 Example Problem 1: Segment Data ................................................ 30-51
Exhibit 30-31 Example Problem 1: Access Point Data ......................................... 30-51
Chapter 30/Urban Street Segments: Supplemental Contents

Exhibit 30-32 Example Problem 1: Movement-Based Output Data ...................30-51

Exhibit 30-33 Example Problem 1: Timer-Based Phase Output Data ................30-53
Exhibit 30-34 Example Problem 1: Timer-Based Movement Output Data .......30-54
Exhibit 30-35 Example Problem 1: Movement-Based Access Point Output
Data ......................................................................................................................30-54
Exhibit 30-36 Example Problem 1: Performance Measure Summary ................30-56
Exhibit 30-37 Example Problem 2: Segment Geometry .......................................30-57
Exhibit 30-40 Validity Range of Inputs and Calculated Values for
Analysis of Motor Vehicles on an Urban Street Roundabout Segment .....30-71
Exhibit 30-41 Additional Required Input Data, Potential Data Sources,
and Default Values for Analysis of Motor Vehicles on an Urban
Street Roundabout Segment .............................................................................30-72
Exhibit 30-42 Illustration of Geometric Design Data ...........................................30-72
Exhibit 30-43 Base Free-Flow Speed Adjustment Factors ...................................30-74
Exhibit 30-44 Illustration of Subsegment Dimensions .........................................30-76
Contents Chapter 30/Urban Street Segments: Supplemental

1. INTRODUCTION
Chapter 30 is the supplemental chapter for Chapter 18, Urban Street VOLUME 4: APPLICATIONS
GUIDE
Segments, which is found in Volume 3 of the Highway Capacity Manual (HCM). 25. Freeway Facilities:
This chapter presents detailed information about the following aspects of the Supplemental
Chapter 18 motorized vehicle methodology: Segments: Supplemental
• The adjustments made to the input vehicular demand flow rates at Supplemental
signalized boundary intersections so that they reasonably reflect actual 28. Freeway Merges and
operating conditions during the analysis period, 29. Urban Street Facilities:
Supplemental
• The process for analyzing vehicular traffic flow on a segment bounded by 30. Urban Street Segments:
signalized intersections, and Supplemental
• The process for estimating through-vehicle delay due to vehicle turning Supplemental
32. STOP-Controlled
movements at unsignalized midsegment access points. Intersections:
Supplemental
This chapter provides a simplified version of the Chapter 18 motorized 33. Roundabouts:
vehicle methodology that is suitable for planning applications. It describes Supplemental
techniques for measuring free-flow speed and average travel speed in the field Terminals: Supplemental
and provides details about the computational engine that implements the 35. Pedestrians and Bicycles:
Supplemental
Chapter 18 motorized vehicle methodology. Chapter 30 provides four example
problems that demonstrate the application of the motorized vehicle, pedestrian, 37. ATDM: Supplemental
bicycle, and transit methodologies to an urban street segment. Finally, the
chapter provides an overview of the methodology for evaluating the
performance of the motor vehicle mode on an urban street segment bounded by
one or more roundabouts.
Chapter 30/Urban Street Segments: Supplemental Introduction

2. TRAFFIC DEMAND ADJUSTMENTS
This section describes adjustments made to the input vehicular demand flow
rates at signalized boundary intersections so that they reasonably reflect actual
operating conditions during the analysis period. These adjustments have no
effect if existing vehicular flow rates are accurately quantified for the subject
segment and all movements operate below their capacity. However, if the
demand flow rate for any movement exceeds its capacity or if there is
disagreement between the count of vehicles entering and the count exiting the
segment, some movement flow rates will need to be adjusted for accurate
evaluation of segment operation.
This section describes two procedures that check the input flow rates and
make adjustments if necessary. These procedures are
• Capacity constraint and volume balance and
• Origin–destination distribution.
These procedures can be extended to the analysis of unsignalized boundary
intersections; however, the mechanics of this extension are not described.
CAPACITY CONSTRAINT AND VOLUME BALANCE

This subsection describes the procedure for determining the turn movement
flow rates at each intersection along the subject urban street segment. The
analysis is separately applied to each travel direction and proceeds in the
direction of travel. The procedure consists of a series of steps that are completed
in sequence for the entry and exit movements associated with each segment.
These movements are shown in Exhibit 30-1.
Exhibit 30-1 Upstream Intersection Access Points Subject Intersection N

Entry and Exit Movements on
the Typical Street Segment
Signal Signal
- Subject segment Entry and exit volume for all

- Entry movements access points combined
- Exit movements
As indicated in Exhibit 30-1, three entry movements are associated with the
upstream signalized intersection and three exit movements are associated with
the downstream signalized intersection. Entry and exit movements also exist at
each access point intersection. However, these movements are aggregated into
one entry and one exit movement for simplicity.
The analysis procedure is described in the following steps. Frequent
reference is made to “volume” in these steps. In this application, volume is
considered to be equivalent to average flow rate for the analysis period and to
have units of vehicles per hour (veh/h).
Traffic Demand Adjustments Chapter 30/Urban Street Segments: Supplemental

Step 1: Identify Entry and Exit Volumes

The volume for each entry and exit movement is identified during this step.
The volume entering the segment from each access point intersection should be
identified and added to obtain a total for the segment. Similarly, the volume
exiting the segment from each access point intersection should be identified and
added for the segment.
A maximum of eight entry volumes are identified in this step. The seven
volumes at the upstream boundary intersection include signalized left-turn
volume, signalized through volume, signalized right-turn volume, unsignalized
left-turn volume, unsignalized through volume, unsignalized right-turn volume,
and right-turn-on-red (RTOR) volume. The eighth entry volume is the total
access point entry volume.
A maximum of eight exit volumes are identified in this step. The seven
volumes at the downstream boundary intersection include signalized left-turn
volume, signalized through volume, signalized right-turn volume, unsignalized
left-turn volume, unsignalized through volume, unsignalized right-turn volume,
and RTOR volume. The eighth exit volume is the total access point exit volume.
Step 2: Estimate Movement Capacity

During this step, the capacity of each signalized entry movement is
estimated. This estimate should be a reasonable approximation based on
estimates of the saturation flow rate for the corresponding movement and the
phase splits established for signal coordination. The capacity of the RTOR
movements is not calculated during this step.
If the right-turn movement at the upstream intersection shares a lane with its
adjacent through movement, the discharge flow rate for the turn movement can
be estimated by using Equation 30-1.
𝑠𝑞|𝑟 = 𝑠𝑠𝑟 𝑃𝑅 Equation 30-1
where
sq|r = shared lane discharge flow rate for upstream right-turn traffic
movement in vehicles per hour per lane (veh/h/ln),
ssr = saturation flow rate in shared right-turn and through-lane group with
permitted operation (veh/h/ln), and
PR = proportion of right-turning vehicles in the shared lane (decimal).
The procedure described in Section 2 of Chapter 31, Signalized Intersections:
Supplemental, is used to estimate the two variables shown in Equation 30-1. A
similar equation can be constructed to estimate the shared lane discharge flow
rate for an upstream left-turn movement in a shared lane.
The capacity for the right-turn movement in the shared-lane lane group is
then computed with Equation 30-2.
𝑐𝑞|𝑟 = 𝑠𝑞|𝑟 𝑔/𝐶 Equation 30-2
where
cq|r = shared lane capacity for upstream right-turn traffic movement (veh/h),
Chapter 30/Urban Street Segments: Supplemental Traffic Demand Adjustments

sq|r = shared lane discharge flow rate for upstream right-turn traffic
movement (veh/h/ln),
g = effective green time (s), and
The procedure described in Section 2 of Chapter 31 is used to estimate the
signal timing variables shown in Equation 30-2. A similar equation can be
constructed for an upstream left-turn movement in a shared lane.
Step 3: Compute Volume-to-Capacity Ratio

During this step, the volume-to-capacity ratio is computed for each
signalized entry movement. This ratio is computed by dividing the arrival
volume from Step 1 by the capacity estimated in Step 2. Any movements with a
volume-to-capacity ratio in excess of 1.0 will meter the volume arriving to the
downstream intersection. This ratio is not computed for the RTOR movements.
Step 4: Compute Discharge Volume

The discharge volume from each of the three signalized entry movements is
equal to the smaller of its entry volume or its associated movement capacity. The
total discharge volume for the combined access point approach is assumed to be
equal to the total access point entry volume. Similarly, the discharge volume for
each unsignalized and RTOR movement is assumed to equal its corresponding
entry volume. As a last calculation, the eight discharge volumes are added to
obtain the total discharge volume.
Step 5: Compute Adjusted Exit Volume

The total discharge volume from Step 4 should be compared with the total
exit volume. The total exit volume is the sum of the eight exit volumes identified
in Step 1. If the two totals do not agree, the eight exit volumes must be adjusted
so that their sum equals the total discharge volume. The adjusted exit volume for
a movement equals its exit volume multiplied by the “volume ratio.” The volume
ratio equals the total discharge volume divided by the total exit volume.
Step 6: Repeat Steps 1 Through 5 for Each Segment

The preceding steps should be completed for each segment in the facility in
the subject direction of travel. The procedure should then be repeated for the
opposing direction of travel.
ORIGIN–DESTINATION DISTRIBUTION
The volume of traffic that arrives at a downstream intersection for a given
downstream movement represents the combined volume from each upstream
point of entry weighted by its percentage contribution to the downstream exit
movement. The distribution of these contribution percentages between each
upstream and downstream pair is represented as an origin–destination
distribution matrix.
The origin–destination matrix is important for estimating the arrival pattern
of vehicles at the downstream intersection. Hence, the focus here is on upstream

entry movements that are signalized, because (a) they are typically the higher-
volume movements and (b) the signal timing influences their time of arrival
downstream. For these reasons, the origin–destination distribution is focused on
the three upstream signalized movements. All other movements (i.e.,
unsignalized movements at the boundary intersections, access point movements,
RTOR movements) are combined into one equivalent movement—referred to
hereafter as the “access point” movement—that is assumed to arrive uniformly
throughout the signal cycle.
Ideally, an origin–destination survey would be conducted for an existing
segment, or the origin–destination data would be available from traffic forecasts
by planning models. One matrix would be available for each direction of travel
on the segment. In the absence of such information, origin–destination volumes
can be estimated from the entry and exit volumes for a segment, where the exit
volumes equal the adjusted arrival volumes from the procedure described in the
previous subsection, Capacity Constraint and Volume Balance.
Each of the four entry movements to the segment shown in Exhibit 30-1 is
considered an origin. Each of the four exit movements is a destination. The
problem then becomes one of estimating the origin–destination table given the
entering and exiting volumes.
This procedure is derived from research (1). It is based on the principle that
total entry volume is equal to total exit volume. It uses seed proportions to
represent the best estimate of the volume distribution. These proportions are
refined through implementation of the procedure. It is derived to estimate the
most probable origin–destination volumes by minimizing the deviation from the
seed percentages while ensuring the equivalence of entry and exit volumes.
The use of seed percentages allows the procedure to adapt the origin–
destination volume estimates to factors or geometric situations that induce
greater preference for some entry–exit combinations than is suggested by simple
volume proportion (e.g., a downstream freeway on-ramp). The default seed
proportions are listed in Exhibit 30-2.
Seed Proportion by Origin Movement Destination Exhibit 30-2

Left Through Right Access Point Movement Default Seed Proportions for
Origin–Destination Matrix
0.02 0.10 0.05 0.02 Left
0.91 0.78 0.92 0.97 Through
0.05 0.10 0.02 0.01 Right
0.02 0.02 0.01 0.00 Access point
1.00 1.00 1.00 1.00
Step 1: Set Adjusted Origin Volume

𝑂𝑎,𝑖 = 𝑂𝑖 Equation 30-3
where
Oa,i = adjusted volume for origin i (i = 1, 2, 3, 4) (veh/h), and
Oi = volume for origin i (i = 1, 2, 3, 4) (veh/h).
The letter i denotes the four movements entering the segment. This volume is
computed for each of the four origins.
Chapter 30/Urban Street Segments: Supplemental Traffic Demand Adjustments

Step 2: Compute Adjusted Destination Volume

4
Equation 30-4 𝐷𝑎,𝑗 = ∑ 𝑂𝑎,𝑖 𝑝𝑖,𝑗

𝑖=1
where
Da,j = adjusted volume for destination j (j = 1, 2, 3, 4) (veh/h),
Oa,i = adjusted volume for origin i (i = 1, 2, 3, 4) (veh/h), and
pi,j = seed proportion of volume from origin i to destination j (decimal).
The letter j denotes the four movements exiting the segment. This volume is
computed for each of the four destinations.
Step 3: Compute Destination Adjustment Factor

Equation 30-5
𝐷𝑗
𝑏𝑑,𝑗 =
𝐷𝑎,𝑗
where
bd,j = destination adjustment factor j (j = 1, 2, 3, 4),
Dj = volume for destination j (j = 1, 2, 3, 4) (veh/h), and
Da,j = adjusted volume for destination j (j = 1, 2, 3, 4) (veh/h).
This factor is computed for each of the four destinations.
Step 4: Compute Origin Adjustment Factor

4
Equation 30-6 𝑏𝑜,𝑖 = ∑ 𝑏𝑑,𝑗 𝑝𝑖,𝑗

𝑗=1
where bo,i is the origin adjustment factor i (i = 1, 2, 3, 4). This factor is computed
for each of the four origins.
Step 5: Compute Adjusted Origin Volume

𝑂𝑖
Equation 30-7 𝑂𝑎,𝑖 =
𝑏𝑜,𝑖
where Oa,i is the adjusted volume for origin i (i = 1, 2, 3, 4) (veh/h). This volume is
computed for each of the four origins. It replaces the value previously
determined for this variable.
For each origin, compute the absolute difference between the adjusted origin
volume from Equation 30-7 and the previous estimate of the adjusted origin
volume. If the sum of these four differences is less than 0.01, proceed to Step 6;
otherwise, set the adjusted origin volume for each origin equal to the value from
Equation 30-7, go to Step 2, and repeat the calculation sequence.
Step 6: Compute Origin–Destination Volume

Equation 30-8 𝑣𝑖,𝑗 = 𝑂𝑎,𝑖 𝑏𝑑,𝑗 𝑝𝑖,𝑗
where vi,j is the volume entering from origin i and exiting at destination j (veh/h).
This volume is computed for all 16 origin–destination pairs.

3. SIGNALIZED SEGMENT ANALYSIS
This section describes the process for analyzing vehicular traffic flow on a
segment bounded by signalized intersections. Initially, this process computes the
flow profile of discharging vehicles at the upstream intersection as influenced by
the signal timing and phase sequence. It uses this profile to compute the arrival
flow profile at a downstream junction. The arrival flow profile is then compared
with the downstream signal timing and phase sequence to compute the
proportion of vehicles arriving during green. The arrival flow profile is also used
to compute the proportion of time that a platoon blocks one or more traffic
movements at a downstream access point intersection. These two platoon
descriptors are used in subsequent procedures to compute delay and other
performance measures.
This section describes six procedures that are used to define the arrival flow
profile and compute the related platoon descriptors. These procedures are
• Discharge flow profile,
• Running time,
• Projected arrival flow profile,
• Proportion of time blocked,
• Sustained spillback, and
• Midsegment lane restriction.
Each procedure is described in the following subsections.
DISCHARGE FLOW PROFILE

A flow profile is a macroscopic representation of steady traffic flow
conditions for the average signal cycle during the specified analysis period. The
cycle is represented as a series of 1-s time intervals (hereafter referred to as “time
steps”). The start time of the cycle is 0.0 s, relative to the system reference time.
The time steps are numbered from 1 to C’, where C’ is the cycle length in units of
time steps. The flow rate for step i represents an average of the flows that occur
during the time period corresponding to step i for all cycles in the analysis
period. This approach is conceptually the same as that used in the TRANSYT-7F
model (2).
A discharge flow profile is computed for each of the upstream signalized
left-turn, through, and right-turn movements. Each profile is defined by the time
that the signal is effectively green and by the time that the queue service time
ends. During the queue service time, the discharge flow rate is equal to the
saturation flow rate. After the queue service time is reached, the discharge rate is
set equal to the “adjusted discharge volume.” The adjusted discharge volume is
equal to the discharge volume computed by using the procedures described in
Section 2, but it is adjusted to reflect the “proportion of arrivals during green.”
The latter adjustment adapts the discharge flow pattern to reflect platoon arrivals
on the upstream segment.
Chapter 30/Urban Street Segments: Supplemental Signalized Segment Analysis

The discharge flow profile is dependent on movement saturation flow rate,

queue service time, phase duration, and proportion of arrivals during green for
the discharging movements. The movement saturation flow rate is computed by
using the procedure described in Section 3 of Chapter 19, Signalized Intersections.
Procedures for calculating the remaining variables are described in subsequent
subsections. This relationship introduces a circularity in the computations that
requires an iterative sequence of calculations to converge on the steady-state
solution.
RUNNING TIME
The running time procedure describes the calculation of running time
between the upstream intersection and a downstream intersection. This
procedure is described as Step 2 of the motorized vehicle methodology in
Chapter 18, Urban Street Segments.
One component of running time is the delay due to various midsegment
sources. One notable source of delay is left or right turns from the segment at an
access point intersection. This delay is computed by using the procedure
described in Section 4. Other sources of delay include on-street parking
maneuvers and pedestrian crosswalks. Delay from these sources represents an
input variable to the methodology.
PROJECTED ARRIVAL FLOW PROFILE

This subsection describes the procedure for predicting the arrival flow
profile at a downstream intersection (i.e., access point or boundary intersection).
This flow profile is based on the discharge flow profile and running time
computed previously. The discharge flow profile is used with a platoon
dispersion model to compute the arrival flow profile. The platoon dispersion
model is summarized in the next part of this subsection. The procedure for using
this model to estimate the arrival flow profile is described in the second part.
Platoon Dispersion Model

The platoon dispersion model was originally developed for use in the
TRANSYT model (3). Input to the model is the discharge flow profile for a
specified traffic movement. Output statistics from the model include (a) the
arrival time of the leading vehicles in the platoon to a specified downstream
intersection and (b) the flow rate during each subsequent time step.
In general, the arrival flow profile has a lower peak flow rate than the
discharge flow profile owing to the dispersion of the platoon as it travels down
the street. For similar reasons, the arrival flow profile is spread out over a longer
period of time than the discharge flow profile. The rate of dispersion increases
with increasing segment running time, which may be caused by access point
activity, on-street parking maneuvers, and other midsegment delay sources.
The platoon dispersion model is described by Equation 30-9.
Equation 30-9 ′
𝑞𝑎|𝑢,𝑗 ′
= 𝐹 𝑞𝑢,𝑖 ′
+ (1 − 𝐹) 𝑞𝑎|𝑢,𝑗−1
with
Equation 30-10 𝑗 = 𝑖 + 𝑡′
Signalized Segment Analysis Chapter 30/Urban Street Segments: Supplemental

where
q’a|u,j = arrival flow rate in time step j at a downstream intersection from
upstream source u (veh/step),
q’u,i = departure flow rate in time step i at upstream source u (veh/step),
F = smoothing factor,
j = time step associated with platoon arrival time t’, and
t’ = platoon arrival time (steps).
The upstream flow source u can be the left-turn, through, or right-turn
movement at the upstream boundary intersection. It can also be the collective set
of left-turn or right-turn movements at access point intersections between the
upstream boundary intersection and the subject intersection.
Exhibit 30-3 illustrates an arrival flow profile obtained from Equation 30-9. In
this figure, the discharge flow profile is input to the model as variable q’u,i. The
dashed rectangles that form the discharge flow profile indicate the flow rate
during each of nine time steps (i = 1, 2, 3, . . . , 9) that are each dt seconds in
duration. The vehicles that depart in the first time step (i = 1) arrive at the
downstream intersection after traveling an amount of time equal to t’ steps. The
arrival flow at any time step j (= i + t’) is computed with Equation 30-9.
Street Segment Discharge Flow Profile Exhibit 30-3

Flow, veh/step
saturation flow rate, s Platoon Dispersion Model

Distance from
Upstream Upstream
Intersection
Intersection
0 ft dt
Direction of
Traffic Flow
Flow, veh/step
Arrival Flow Profile
t'
Downstream
Intersection
1,000 ft red green
0
Time (s)
Research (4) indicates that Equation 30-11 describes the relationship between
the smoothing factor and running time.
1
𝐹= Equation 30-11
1 + 0.138 𝑡𝑅′ + 0.315/𝑑𝑡
where
t’R = segment running time = tR/dt (steps),
tR = segment running time (s), and
dt = time step duration (s/step).

The recommended time step duration for this procedure is 1.0 s/step. Shorter
values can be rationalized to provide a more accurate representation of the
profile, but they also increase the time required for the computations. Experience
indicates that 1.0 s/step provides a good balance between accuracy and
computation time.
Equation 30-12 is used to compute platoon arrival time to the subject
downstream intersection.
1
Equation 30-12 𝑡 ′ = 𝑡𝑅′ − + 1.25
𝐹

This subsection describes the procedure for computing the arrival flow
profile. Typically, there are three upstream signalized traffic movements that
depart at different times during the signal cycle; they are the minor-street right
turn, major-street through, and minor-street left turn. Traffic may also enter the
segment at various midblock access points or as an unsignalized movement at
the boundary intersection. Exhibit 30-4 illustrates how these movements join to
form the arrival flow profile for the subject downstream intersection.
Exhibit 30-4 Upstream Intersection Downstream Intersection

Arrival Flow Profile Estimation
Procedure
Discharge Flow Profile

Combined Arrival Flow Profile
Cross-street right turn
Flow (veh/step)
Flow (veh/step)
2.30 2.3
Primarily through
Major-street through
1.10 1.1 Primarily left
Primarily
right
Cross-street left turn
-0.10 -0.10
0 0
Time (steps) Time (steps)

Flow (veh/step)
Cross-street right turn

2.30
Major-street through
1.10
Cross-street left turn

-0.10
0
Time (steps)
In application, the discharge flow profile for each of the departing

movements is obtained from the discharge flow profile procedure described
previously. These profiles are shown in the first of the three x-y plots in Exhibit

30-4. The platoon dispersion model is then used to estimate the arrival flows for
each movement at a downstream intersection. These arrival flow profiles are
shown in the second x-y plot in the exhibit. Arrivals from midsegment access
points, which are not shown, are assumed to have a uniform arrival flow profile
(i.e., a constant flow rate for all time steps).
Finally, the origin–destination distribution procedure is used to distribute
each arrival flow profile to each of the downstream exit movements. The four
arrival flow profiles associated with the subject exit movement are added
together to produce the combined arrival flow profile. This profile is shown in
the third x-y plot. The upstream movement contributions to this profile are
indicated by arrows.
Comparison of the profiles in the first and second x-y plots of Exhibit 30-4
illustrates the platoon dispersion process. In the first x-y plot, the major-street
through movement has formed a dense platoon as it departs the upstream
intersection. However, by the time this platoon reaches the downstream
intersection it has spread out and has a lower peak flow rate. In general, the
amount of platoon dispersion increases with increasing segment length. For very
long segments, the platoon structure degrades and arrivals become uniform
throughout the cycle.
Platoon structure can also degrade as a result of significant access point
activity along the segment. Streets with frequent active access point intersections
tend to have more vehicles leave the platoon (i.e., turn from the segment at an
access point) and enter the segment after the platoon passes (i.e., turn in to the
segment at an access point). Both activities result in significant platoon decay.
The effect of platoon decay is modeled by using the origin–destination
matrix, in which the combined access point activity is represented as one volume
assigned to midsegment origins and destinations. A large access point volume
corresponds to a smaller volume that enters at the upstream boundary
intersection as a defined platoon. This results in a larger portion of the combined
arrival flow profile defined by uniform (rather than platoon) arrivals. When a
street has busy access points, platoon decay tends to be a more dominant cause
of platoon degradation than platoon dispersion.
PROPORTION OF TIME BLOCKED

The combined arrival flow profile can be used to estimate the time that a
platoon passes through a downstream access point intersection. During this time
period, the platoon can be dense enough to preclude a minor movement driver
from finding an acceptable gap.
The use of the arrival flow profile to estimate the blocked period duration is
shown in Exhibit 30-5. The profile shown represents the combined arrival flow
profile for the through-lane group at a downstream access point intersection. The
dashed line represents the critical platoon flow rate. Flow rates in excess of this
threshold are rationalized to be associated with platoon headways that are too
short to be entered (or crossed) by minor movements. The critical platoon flow
rate qc is equal to the inverse of the critical headway tc associated with the minor

movement (i.e., qc = 3,600/tc). The appropriate critical headway values for various
movements are identified in Chapter 20, Two-Way STOP-Controlled Intersections.
Exhibit 30-5
Estimation of Blocked Period
Duration
In the situation of a driver desiring to complete a left turn from the major
street across the traffic stream represented by Exhibit 30-5, the proportion of time
blocked is computed by using Equation 30-13. For this maneuver, the blocked
period duration is based on the flow profile of the opposing through-lane group.
𝑡𝑝′ 𝑑𝑡
Equation 30-13 𝑝𝑏 =
𝐶
where
pb = proportion of time blocked (decimal),
t’p = blocked period duration (steps),
dt = time step duration (s/step), and
Equation 30-13 is also used for the minor-street right-turn movement.
However, in this situation, the blocked period duration is computed for the
through-lane group approaching from the left. For the minor-street left-turn and
through movements, the arrival flow profiles from both directions are evaluated.
In this instance, the blocked period duration represents the time when a platoon
from either direction is present in the intersection.
SUSTAINED SPILLBACK
This subsection describes two procedures that were developed for the
evaluation of segments that experience sustained spillback. Sustained spillback
occurs as a result of oversaturation (i.e., more vehicles discharging from the
upstream intersection than can be served at the subject downstream intersection).
The spillback can exist at the start of the study period, or it can occur at some
point during the study period. Spillback that first occurs after the study period is
not addressed.

Effective Average Vehicle Spacing

One piece of information needed to evaluate segments experiencing sustained
spillback is the effective average vehicle spacing (5). A simple estimate of this
spacing is computed as the sum of the average vehicle length and the average
distance between two queued vehicles (as measured from the back bumper of the
lead vehicle to the front bumper of the trailing vehicle).
Presumably, this estimate of average spacing could be divided into the
segment length to determine the maximum number of queued vehicles on the
segment during spillback. However, this result is biased because it is based on
the assumption that all vehicles on the segment will always be stationary during
spillback. This is a weak assumption because the downstream signal operation
creates backward-traveling waves of starting and stopping. Between the starting
wave and the stopping wave, vehicles are moving at the saturation headway and
its associated speed. Their spacing exceeds that of the aforementioned “simple”
estimate.
The procedure described in this subsection is used to estimate the effective
average vehicle spacing L*h on a segment with spillback. The derivation of this new
variable is based on the vehicle trajectories shown in Exhibit 30-6. The segment of
interest is shown on the left side of the figure. Spillback is present for all of the
cycles shown; however, trajectories are shown only for two cycles. The solid
trajectories coincide with vehicles that enter the segment as a through movement
at the upstream intersection. The dashed lines coincide with vehicles that enter
the segment as a turn movement. A vehicle that enters the segment traveling
north as a through vehicle is shown to experience four cycles before exiting the
segment. The trajectories show that the vehicles move forward at a saturation
headway of 3,600/s seconds per vehicle and a speed of Va feet per second.
C Vehicle Source Exhibit 30-6

Through movement at upstream intersection
r g Turn movement at upstream intersection Vehicle Trajectories During
Spillback Conditions
tpr
Lh Lh
1
Va
L
3,600/s
0 t
The lines that slope downward from the upper left to lower right represent
the waves of reaction time. They have a slope of tpr seconds per vehicle. The
starting wave originates at the onset of the green indication, and the stopping
wave originates at the onset of the red indication. The average vehicle spacing
when vehicles are stopped is Lh feet per vehicle.

On the basis of the relationships shown in Exhibit 30-6, the following

procedure can be used to estimate the effective average vehicle spacing.
Step 1. Compute Wave Travel Time

The time required for the driver reaction wave to propagate backward to the
upstream intersection is computed with the following equation:
Equation 30-14 (𝐿 − 𝑊𝑖 ) 𝑡𝑝𝑟
𝑡𝑚𝑎𝑥 =
𝐿ℎ
with
Equation 30-15 𝐿ℎ = 𝐿𝑝𝑐 (1 − 0.01 𝑃𝐻𝑉 ) + 0.01 𝐿𝐻𝑉 𝑃𝐻𝑉
where
tmax = wave travel time (s);
L = segment length (ft);
Wi = width of upstream signalized intersection, as measured along the
segment centerline (ft);
tpr = driver starting response time (= 1.3) (s/veh);
Lh = average vehicle spacing in stationary queue (ft/veh);
Lpc = stored passenger car lane length = 25 (ft);
LHV = stored heavy vehicle lane length = 45 (ft); and
PHV = percent heavy vehicles in the corresponding movement group (%).
Step 2. Compute Speed of Moving Queue

The average speed of the moving queue is computed with Equation 30-16:
Equation 30-16 𝐿ℎ
𝑉𝑎 =
2.0 − 𝑡𝑝𝑟
where Va is the average speed of moving queue (ft/s).
Step 3. Compute Effective Average Vehicle Spacing

The relationship between the trajectories of the moving vehicles defines the
following association between speed, saturation flow rate, signal timing, and
vehicle spacing.
Equation 30-17 If 0.0 ≤ 𝑡𝑚𝑎𝑥 < 𝑟, then 𝐿∗ℎ = 𝐿ℎ
𝑟 1 −1
If 𝑟 ≤ 𝑡𝑚𝑎𝑥 < 𝐶, then 𝐿∗ℎ = 2.0 ( + ) ≥ 𝐿ℎ
𝐿−𝑊 𝑉 𝑖 𝑎
𝐿ℎ
If 𝐶 ≤ 𝑡𝑚𝑎𝑥 , then 𝐿∗ℎ =
1.0 − 0.5 𝑡𝑝𝑟 𝑔/𝐶
where
Lh* = effective average vehicle spacing in stationary queue (ft/veh),
r = effective red time (= C – g) (s),
g = effective green time (s), and

Equation 30-17 has three component equations. The component equation

used for a given segment and analysis period will be based on the value of tmax, r,
and C. The value of average vehicle spacing from the first component equation is
the smallest that can be obtained from Equation 30-17. The value from the last
equation is the largest that can be obtained. The value obtained from the
equation in the middle varies between these two extreme values, depending on
the value of tmax.
Spillback Check
This subsection describes the procedure for determining whether queue
spillback occurs on a segment during a given analysis period (4). The analysis is
applied separately to each travel direction and proceeds in the direction of travel.
The procedure consists of a series of steps that are completed in sequence for the
signalized exit movements associated with each segment. These movements
were shown in Exhibit 30-1. Spillback due to the movements associated with the
access points is not specifically addressed.
Step 1: Identify Initial Queue

During this step, the initial queue for each signalized exit movement is
identified. This value represents the queue present at the start of the analysis
period (the total of all vehicles in all lanes serving the movement). The initial
queue estimate would likely be available for the evaluation of an existing
condition for which field observations indicate the presence of a queue at the
start of the analysis period. For planning or preliminary design applications, it
can be assumed to equal 0.0 vehicles.
Step 2: Identify Queue Storage Length

The length of queue storage for each exit movement is identified during this
step. For turn movements served from a turn bay, this length equals the length of
the turn bay. For through movements, this length equals the segment length less
the width of the upstream intersection. For turn movements served from a lane
equal in length to that of the segment, the queue storage length equals the
segment length less the width of the upstream intersection.
Step 3: Compute Maximum Queue Storage

The maximum queue storage for the exiting through movement is computed
with Equation 30-18:
(𝑁𝑡ℎ − 𝑃𝐿 − 𝑃𝑅 ) 𝐿𝑎,𝑡ℎ𝑟𝑢
𝑁𝑞𝑥,𝑡ℎ𝑟𝑢 = Equation 30-18
𝐿∗ℎ
where
Nqx,thru = maximum queue storage for the through movement (veh),
Nth = number of through lanes (shared or exclusive) (ln),
PL = proportion of left-turning vehicles in the shared lane (decimal),
PR = proportion of right-turning vehicles in the shared lane (decimal),

La,thru = available queue storage distance for the through movement (ft/ln), and
Lh* = effective average vehicle spacing in stationary queue (ft/veh).
The procedure described in Section 2 of Chapter 31, Signalized Intersections:
Supplemental, is used to estimate PL and PR. If there are no shared lanes, PL = 0.0
and PR = 0.0.
The maximum queue storage for a turn movement is computed with
Equation 30-19:
𝑁turn 𝐿𝑎,turn + 𝑃turn 𝐿𝑎,𝑡ℎ𝑟𝑢
Equation 30-19 𝑁𝑞𝑥,turn =
𝐿ℎ
where
Nqx,turn = maximum queue storage for a turn movement (veh),
Nturn = number of lanes in the turn bay (ln),
La,turn = available queue storage distance for the turn movement (ft/ln),
Pturn = proportion of turning vehicles in the shared lane = PL or PR (decimal),
and
Lh = average vehicle spacing in stationary queue (ft/veh).
This equation is applicable to turn movements in exclusive lanes (i.e., Pturn
= 0.0) and to turn movements that share a through lane.
Step 4: Compute Available Storage Length

The available storage length is computed for each signalized exit movement
Equation 30-20 𝑁𝑞𝑎 = 𝑁𝑞𝑥 − 𝑄𝑏 ≥ 0.0
where
Nqa = available queue storage (veh),
Nqx = maximum queue storage for the movement (veh), and
Qb = initial queue at the start of the analysis period (veh).
The analysis thus far has treated the three signalized exit movements as if
they were independent. At this point, the analysis must be extended to include
the combined through and left-turn movement when the left-turn movement has
a bay (i.e., it does not have a lane that extends the length of the segment). The
analysis must also be extended to include the combined through and right-turn
movement when the right-turn movement has a bay (but not a full-length lane).
The analysis of these newly formed “combined movements” is separated
into two parts. The first part is the analysis of just the bay. This analysis is a
continuation of the exit movement analysis using the subsequent steps of this
procedure. The second part is the analysis of the length of the segment shared by
the turn movement and the adjacent through movement. The following rules are
used to evaluate the combined movements for the shared segment length:
1. The volume for each combined movement equals the sum of the adjusted
arrival volumes for the two contributing movements. These volumes are

obtained from the procedure described in a previous subsection, Origin–

Destination Distribution.
2. The initial queue for each combined movement is computed with
Equation 30-21.
𝐿𝑎,turn 𝑁turn 𝐿𝑎,turn 𝑁𝑡ℎ
𝑄𝑏,𝑐𝑜𝑚𝑏 = max (0.0, 𝑄𝑏,turn − ∗ , 𝑄𝑏,𝑡ℎ𝑟𝑢 − ) Equation 30-21
𝐿ℎ 𝐿∗ℎ
where Qb,comb is the initial queue for the combined movement (veh). The
other variables were defined previously and are evaluated for the
movement indicated by the variable subscript.
3. The queue storage length for a combined movement La,comb equals the
queue storage length for the through movement less the queue storage
length of the turn movement (i.e., La,comb = La,thru – La,turn).
4. The number of lanes available to the combined movement Ncomb equals the
number of lanes available to the through movement.
5. The maximum queue storage for the combined movement Nqx,comb is
𝑁𝑡ℎ 𝐿𝑎,𝑡ℎ𝑟𝑢
𝑁𝑞𝑥,𝑐𝑜𝑚𝑏 = Equation 30-22
𝐿∗ℎ
6. The available storage length for the combined movement Nqa,comb is
𝑁𝑞𝑎,𝑐𝑜𝑚𝑏 = 𝑁𝑞𝑥,𝑐𝑜𝑚𝑏 − 𝑄𝑏,𝑐𝑜𝑚𝑏 ≥ 0.0 Equation 30-23
Step 5: Compute Capacity

The capacity for both the exit movements and the combined movements is
established in this step. The capacity for each exit movement was computed in
Step 2 in the subsection titled Capacity Constraint and Volume Balance. The
capacity of the combined movements is computed by using Equation 30-24.
𝑣𝑎,1 𝑐𝑡ℎ𝑟𝑢 (𝑁𝑡ℎ − 1)
𝑐= + Equation 30-24
𝑋1 𝑁𝑡ℎ
with
𝑣𝑎,turn + 𝑣𝑎,𝑡ℎ𝑟𝑢
𝑣𝑎,1 = max (𝑣𝑎,turn , ) Equation 30-25
𝑁𝑡ℎ
𝑣𝑎,turn 𝑣𝑎,1 − 𝑣𝑎,turn
𝑋1 = + Equation 30-26
𝑐turn 𝑐𝑡ℎ𝑟𝑢 /𝑁𝑡ℎ
where
c = capacity of the combined movements (veh/h),
va,1 = adjusted arrival volume in the shared lane (veh/h),
X1 = volume-to-capacity ratio in the shared lane,
cthru = capacity for the exiting through movement (veh/h),
cturn = capacity for the exiting turn movement (veh/h),
va,turn = adjusted arrival volume for the subject turn movement (veh/h),

va,thru = adjusted arrival volume for the subject through movement (veh/h), and
The two adjusted arrival volumes va,turn and va,thru are obtained from the
procedure described in the Origin–Destination Distribution subsection.
Step 6: Compute Queue Growth Rate

During this step, the queue growth rate is computed for each signalized exit
movement for which the storage extends the length of the segment. Typically,
the through movement satisfies this requirement. A turn movement may also
satisfy this requirement if it is served by an exclusive lane that extends the length
of the segment. The queue growth rate is computed as the difference between the
adjusted arrival volume va and the capacity c for the subject exit movement.
Equation 30-27 is used to compute this rate.
Equation 30-27 𝑟𝑞𝑔 = 𝑣𝑎 − 𝑐 ≥ 0.0
where rqg is the queue growth rate (veh/h).
The queue growth rate is also computed for the combined movements
formulated in Step 4. The adjusted volume used in Equation 30-27 represents the
sum of the through and turn movement volumes in the combined group. The
capacity for the group was computed in Step 5.
Step 7: Compute Time Until Spillback

During this step, the time until spillback is computed for each signalized exit
movement for which the storage extends the length of the segment. This time is
computed with Equation 30-28 for any movement with a nonzero queue growth
rate.
𝑁𝑞𝑎
Equation 30-28 𝑇𝑐 =
𝑟𝑞𝑔
where Tc is the time until spillback (h).
For turn movements served by a bay, the computed spillback time is the time
required for the bay to overflow. It does not represent the time at which the turn-
related queue reaches the upstream intersection.
Equation 30-28 is also used to compute the spillback time for the combined
movements formulated in Step 4. However, this spillback time is the additional
time required for the queue to grow along the length of segment shared by the
turn movement and the adjacent through movement. This time must be added to
the time required for the corresponding turn movement to overflow its bay to
obtain the actual spillback time for the combined movement.
Step 8: Repeat Steps 1 Through 7 for Each Segment

The preceding steps should be completed for each segment in the facility in
the subject direction of travel. The procedure should then be repeated for the
opposing direction of travel.

Step 9: Determine Controlling Spillback Time

During this step, the shortest time until spillback for each of the exit
movements (or movement groups) for each segment and direction of travel is
identified. If the segment supports two travel directions, two values are
identified (one value for each direction). The smaller of the two values is the
controlling spillback time for the segment. If a movement (or movement group)
does not spill back, it is not considered in this process for determining the
controlling spillback time.
Next, the controlling segment times are compared for all segments that make
up the facility. The shortest time found is the controlling spillback time for the
facility.
If the controlling spillback time exceeds the analysis period, the results from
the motorized vehicle methodology are considered to reflect the operation of the
facility accurately. If spillback occurs before the end of the desired analysis
period, the analyst should consider either (a) reducing the analysis period so that
it ends before spillback occurs or (b) using the sustained spillback evaluation
procedure in Chapter 29, Urban Street Facilities: Supplemental.
MIDSEGMENT LANE RESTRICTION

When one or more lanes on an urban street segment are temporarily closed,
the flow in the lanes that remain open can be adversely affected. The closure can
be due to a work zone, an incident, or a similar event. Occasionally, the lane
closure can adversely affect the performance of traffic movements that are
entering or exiting the segment at the boundary signalized intersection.
Logically, the magnitude of the effect will increase as the distance between the
intersection and lane closure decreases. The impact on the intersection that has a
downstream lane closure is the subject of discussion in this subsection.
The procedure described in this subsection is used to adjust the saturation
flow rate of the movements entering a segment when one or more downstream
lanes are blocked. The procedure is developed for incorporation within the
motorized vehicle methodology described in Chapters 18 and 19 (5). Specifically,
the procedure is inserted into the motorized vehicle methodology in Chapter 18,
Urban Street Segments, and used to compute a saturation flow rate adjustment
factor for the movements entering the segment at the intersection. This
adjustment factor is then implemented in the motorized vehicle methodology in
Chapter 19, Signalized Intersections, to compute the adjusted saturation flow rate
of the affected movements.
This procedure is added to the end of Step 4 of the motorized vehicle
methodology described in Chapter 18. It occurs after the saturation flow rate and
phase duration have been determined. It is implemented as part of the iterative
convergence loop identified in the motorized vehicle methodology framework
The calculation sequence begins with an estimate of the capacity for each
traffic movement discharged to the downstream segment. This estimate is
obtained by using the motorized vehicle methodology in Chapter 19. The next
step is to compute the capacity of the downstream segment as influenced by the

midsegment lane restriction. The estimate of movement capacity is then compared

with the downstream segment capacity. If the movement capacity exceeds the
downstream segment capacity, the movement saturation flow rate is reduced
proportionally by using an adjustment factor for downstream lane blockage.
The lane blockage saturation flow rate adjustment factor is computed for
each movement entering the subject segment. The following equations are used
to compute the factor value.
Equation 30-29 𝑐𝑚𝑠
If 𝑐𝑚𝑠 < 𝑐𝑖 or 𝑓𝑚𝑠,𝑖−1 < 1.0, then 𝑓𝑚𝑠,𝑖 = 𝑓𝑚𝑠,𝑖−1 ≥ 0.1
𝑐𝑖
Otherwise, 𝑓𝑚𝑠,𝑖 = 1.0
with
Equation 30-30 𝑐𝑚𝑠 = 0.25 𝑘𝑗 𝑁𝑢𝑛𝑏𝑙𝑘 𝑆𝑓 ≤ 1,800 𝑁𝑢𝑛𝑏𝑙𝑘
where
fms, i = adjustment factor for downstream lane blockage during iteration i,
cms = midsegment capacity (veh/h),
ci = movement capacity during iteration i (veh/h),
kj = jam density (= 5,280 / Lh) (veh/mi/ln),
Lh = average vehicle spacing in stationary queue (ft/veh),
Sf = free-flow speed (mi/h), and
Nunblk = number of open lanes when blockage is present (ln).
The number of lanes used in Equation 30-30 equals the number of unblocked
lanes (i.e., the open lanes) while the blockage is present.
The variable i in the adjustment factor subscript indicates that the factor’s
value is incrementally revised during each iteration of the convergence loop
associated with the motorized vehicle methodology. Ultimately, the factor
converges to a value that results in a movement capacity matching the available
midsegment capacity. For the first iteration, the factor value is set to 1.0 for all
movements. The factor value is also set to 1.0 if the segment is experiencing
spillback. In this situation, a saturation flow rate adjustment factor for spillback
(which incorporates the downstream lane blockage effect) is computed for the
movement. The calculation of the factor for spillback is described in Chapter 29,
Urban Street Facilities: Supplemental.
Equation 30-29 indicates that the factor is less than 1.0 when the midsegment
capacity is smaller than the movement capacity. If the factor has been set to a
value less than 1.0 in a previous iteration, it continues to be adjusted during each
subsequent iteration until convergence is achieved. A minimum factor value of
0.1 is imposed as a practical lower limit.

4. DELAY DUE TO TURNS
This section describes a process for estimating the delay to through vehicles
that follow vehicles turning from the major street into an unsignalized access point
intersection. This delay can be incurred at any access point intersection along the
street. For right-turn vehicles, the delay results when the following vehicles’ speed
is reduced to accommodate the turning vehicle. For left-turn vehicles, the delay
results when the following vehicles must wait in queue while a vehicle ahead
executes a left-turn maneuver at the access point. This delay occurs primarily on
undivided streets; however, it can occur on divided streets when the left-turn
queue exceeds the available storage and spills back into the inside through lane.
The delay estimation process consists of the following two procedures:
• Delay due to left turns and
• Delay due to right turns.
Each procedure is described in the following subsections. These procedures
are based on the assumption that the segment traffic flows are random. While
this assumption may not be strictly correct for urban streets, it is conservative in
that it will yield slightly larger estimates of delay. Moreover, expansion of the
models to accommodate platooned flows would not likely be cost-effective given
the small amount of delay caused by turning vehicles.
DELAY DUE TO LEFT TURNS

Through vehicles on the major-street approach to an unsignalized
intersection can incur delay when the left-turn queue exceeds the available
storage and blocks the adjacent through lane (in this context, the undivided cross
section is considered a major-street approach having no left-turn storage). The
through vehicles that follow are delayed when they stop behind the queue of
turning vehicles. This delay ends when the left-turn vehicle departs or the
through vehicle merges into the adjacent through lane. By merging into the
adjacent lane, drivers reduce their delay relative to the delay they would have
incurred had they waited for the left-turn queue to clear. This delay is computed
1 𝑃𝑙𝑡
𝑑𝑎𝑝,𝑙 = 𝑝𝑜𝑣 𝑑𝑡,1 ( − 1) Equation 30-31
𝑃𝐿 1 − 𝑃𝑙𝑡 − 𝑃𝑟𝑡
where
dap,l = through-vehicle delay due to left turns (s/veh),
pov = probability of left-turn bay overflow (decimal),
dt,1 = average delay to through vehicles in the inside lane (s/veh),
PL = proportion of left-turning vehicles in the shared lane (decimal),
Plt = proportion of left-turning vehicles on the subject approach (decimal), and
Prt = proportion of right-turning vehicles on the subject approach (decimal).
Chapter 30/Urban Street Segments: Supplemental Delay due to Turns

As indicated by Equation 30-31, the delay due to left turns is based on the
value of several variables. The following sequence of computations can be used
to estimate these values (6).
Step 1: Compute the Probability of a Lane Change

𝑣𝑎𝑝𝑝 2
Equation 30-32 𝑃𝑙𝑐 = 1 − [(2 ) − 1] ≥ 0.0
𝑠𝑙𝑐
with
𝑣𝑙𝑡 + 𝑣𝑡ℎ + 𝑣𝑟𝑡
Equation 30-33 𝑣𝑎𝑝𝑝 =
𝑁𝑠𝑙 + 𝑁𝑡 + 𝑁𝑠𝑟
where
Plc = probability of a lane change among the approach through lanes,
vapp = average demand flow rate per through lane (upstream of any turn
bays on the approach) (veh/h/ln),
slc = maximum flow rate in which a lane change can occur = 3,600/tlc
(veh/h/ln),
tlc = critical merge headway = 3.7 (s),
vlt = left-turn demand flow rate (veh/h),
vth = through demand flow rate (veh/h),
vrt = right-turn demand flow rate (veh/h),
Nsl = number of lanes in shared left-turn and through-lane group (ln),
Nt = number of lanes in exclusive through-lane group (ln), and
Nsr = number of lanes in shared right-turn and through-lane group (ln).
If the ratio vapp/slc in Equation 30-32 exceeds 1.0, then it should be set to 1.0.
Step 2: Compute Through-Vehicle Equivalent for Left-Turn Vehicle

If there is a left-turn bay on the major street at the access point, the through-
vehicle equivalent EL1 is 1.0. However, if there is no left-turn bay, the following
equation is used to compute the through-vehicle equivalent.
1,800
Equation 30-34 𝐸𝐿1 =
𝑐𝑙
with
𝑣𝑜 𝑒 −𝑣𝑜 𝑡𝑐𝑔 /3,600
Equation 30-35 𝑐𝑙 =
1 − 𝑒 −𝑣𝑜 𝑡𝑓ℎ/3,600
where
EL1 = equivalent number of through cars for a permitted left-turning vehicle,
cl = capacity of a left-turn movement with permitted left-turn operation
(veh/h),
vo = opposing demand flow rate (veh/h),
Delay due to Turns Chapter 30/Urban Street Segments: Supplemental

tfh = follow-up headway = 2.2 (s), and

tcg = critical headway = 4.1 (s).
Step 3: Compute Modified Through-Vehicle Equivalent

𝐸𝐿1,𝑚 = (𝐸𝐿1 − 1)𝑃𝑙𝑐 + 1 Equation 30-36
𝐸𝑅,𝑚 = (𝐸𝑅,𝑎𝑝 − 1)𝑃𝑙𝑐 + 1 Equation 30-37
where
EL1,m = modified through-car equivalent for a permitted left-turning vehicle,
ER,m = modified through-car equivalent for a protected right-turning vehicle,
and
ER,ap = equivalent number of through cars for a protected right-turning
vehicle at an access point (2.20 if there is no right-turn bay on the
major street at the access point; 1.0 if there is a right-turn bay).
Step 4: Compute Proportion of Left Turns in Inside Through Lane

−𝑏 + √𝑏2 − 4 𝐼𝑡 𝑅 𝑐
𝑃𝐿 = ≤ 1.0 Equation 30-38
2 𝐼𝑡 𝑅
with
𝑏 = 𝑅 − 𝐼𝑙𝑡 𝑃𝑙𝑡 {𝐼𝑡 + (𝑁𝑠𝑙 + 𝑁𝑡 + 𝑁𝑠𝑟 − 1)[(1 + 𝐼𝑡 )𝐸𝐿1,𝑚 − 1]} Equation 30-39
𝑐 = −𝐼𝑙𝑡 𝑃𝑙𝑡 (𝑁𝑠𝑙 + 𝑁𝑡 + 𝑁𝑠𝑟 ) Equation 30-40
𝑅 = 1 + 𝐼𝑟𝑡 𝑃𝑟𝑡 (𝐸𝑅,𝑚 − 1) Equation 30-41
where
R, b, c = intermediate calculation variables;
Ilt = indicator variable (1.0 when there is no left-turn bay on the major
street at the access point, 0.0 when there is a left-turn bay);
Irt = indicator variable (1.0 when there is no right-turn bay on the major
street at the access point, 0.0 when there is a right-turn bay); and
It = indicator variable (1.0 when equations are used to evaluate delay due
to left turns, 0.00001 when equations are used to evaluate delay due to
right turns).
If the number of through lanes on the subject intersection approach (= Nsl + Nt
+ Nsr) is equal to 1.0, then PL = Plt.
The indicator variable It is used to adapt the equations to the analysis of lane
volume for both left-turn- and right-turn-related delays. The variable has a value
of 1.0 in the evaluation of left-turn-related delays. In this situation, it models the
condition in which one or more left-turning vehicles are blocking the inside lane.
In contrast, the variable has a negligibly small value when it is applied to right-
turn-related delays. It models flow conditions in which all lanes are unblocked.

Step 5: Compute Proportion of Right Turns in Outside Through Lane

𝑠1
+ 𝑁𝑠𝑙 + 𝑁𝑡 + 𝑁𝑠𝑟 − 1
1,800
Equation 30-42 𝑃𝑅 = 𝐼𝑟𝑡 𝑃𝑟𝑡 𝑠1 ≤ 1.0
1 − 𝐼𝑟𝑡 𝑃𝑟𝑡 (1,800 + 𝑁𝑠𝑙 + 𝑁𝑡 + 𝑁𝑠𝑟 − 2) (𝐸𝑅,𝑚 − 1)
with
1,800 (1 + 𝑃𝐿 𝐼𝑡 )
Equation 30-43 𝑠1 =
1 + 𝑃𝐿 (𝐸𝐿1,𝑚 − 1) + (𝑃𝐿 𝐸𝐿1,𝑚 𝐼𝑡 )
where s1 is the saturation flow rate for the inside lane (veh/h/ln). If the number of
through lanes on the subject intersection approach (= Nsl + Nt + Nsr) is equal to 1.0,
then PR = Prt.
Step 6: Compute Inside Lane and Outside Lane Flow Rates

𝑣𝑙𝑡
Equation 30-44
𝑣1 =
𝑃𝐿
𝑣𝑟𝑡
if 𝑃𝑅 > 0.0
𝑃𝑅
𝑣𝑛 =
Equation 30-45 𝑣𝑙𝑡 + 𝑣𝑡ℎ + 𝑣𝑟𝑡 − 𝑣1
if 𝑃𝑅 = 0.0
{ 𝑁𝑠𝑙 + 𝑁𝑡 + 𝑁𝑠𝑟 − 1
where
v1 = flow rate for the inside lane (veh/h/ln) and
vn = flow rate for the outside lane (veh/h/ln).
Step 7: Compute Intermediate Lane Flow Rate

If there are more than two lanes on the subject intersection approach,
Equation 30-46 can be used to estimate the flow rate in the intermediate lanes.
𝑣𝑙𝑡 + 𝑣𝑡ℎ + 𝑣𝑟𝑡 − 𝑣1 − 𝑣𝑛
Equation 30-46 𝑣𝑖 =
𝑁𝑠𝑙 + 𝑁𝑡 + 𝑁𝑠𝑟 − 2
where vi is the flow rate for lane i (veh/h/ln). The flow rates in lanes 2, 3, . . . ,
n – 1 are identical and equal to the value obtained from Equation 30-46.
Step 8: Compute Merge Capacity

Equation 30-47 is used to compute the merge capacity available to through
drivers waiting in the inside lane of a multilane approach.
𝑣2 𝑒 −𝑣2 𝑡𝑙𝑐 /3,600
Equation 30-47 𝑐𝑚𝑔 =
1 − 𝑒 −𝑣2 𝑡𝑙𝑐/3,600
where
cmg = merge capacity (veh/h),
v2 = flow rate in the adjacent through lane (veh/h/ln), and
tlc = critical merge headway = 3.7 (s).

Step 9: Compute Delay to Through Vehicles That Merge

2
1 1 𝑣𝑚𝑔 𝑣𝑚𝑔 8 𝑣𝑚𝑔
𝑑𝑚𝑔 = 3,600 ( − ) + 900 𝑇 [ − 1 + √( − 1) + 2 ] Equation 30-48
𝑐𝑚𝑔 1,800 𝑐𝑚𝑔 𝑐𝑚𝑔 𝑐𝑚𝑔 𝑇
with
𝑣𝑚𝑔 = 𝑣1 − 𝑣𝑙𝑡 ≥ 0.0 Equation 30-49
where
dmg = merge delay (s/veh),
vmg = merge flow rate (veh/h/ln), and
T = analysis period duration (h).
This delay is incurred by through vehicles that stop in the inside lane and
eventually merge into the adjacent through lane. The “1/1,800” term included in
Equation 30-48 extracts the service time for the through vehicle from the delay
estimate, so that the delay estimate represents the increase in travel time
resulting from the left-turn queue.
Step 10: Compute Inside Lane Capacity

Equation 30-50 is used to compute the capacity of the inside lane for vehicles
that do not merge.
1,800(1 + 𝑃𝐿 )
𝑐𝑛𝑚 = Equation 30-50
1 + 𝑃𝐿 (𝐸𝐿1 − 1) + (𝑃𝐿 𝐸𝐿1 )
where cnm is the nonmerge capacity for the inside lane (veh/h). The unadjusted
through-vehicle equivalent for a left-turn vehicle EL1 is used in this equation to
estimate the nonmerge capacity.
Step 11: Compute Delay to Through Vehicles That Do Not Merge
1 1 𝑣1 𝑣1 2 8 𝑣1
𝑑𝑛𝑚 = 3,600 ( − ) + 900 𝑇 [ − 1 + √( − 1) + 2 ] Equation 30-51
𝑐𝑛𝑚 1,800 𝑐𝑛𝑚 𝑐𝑛𝑚 𝑐𝑛𝑚 𝑇
where dnm is the nonmerge delay for the inside lane (s/veh). This delay is incurred
by through vehicles that stop in the inside lane and wait for the queue to clear.
These vehicles do not merge into the adjacent lane.
Step 12: Compute Delay to Through Vehicles in the Inside Lane

This delay is estimated as the smaller of the delay relating to the merge and
nonmerge maneuvers. It is computed with Equation 30-52.
𝑑𝑡,1 = min(𝑑𝑛𝑚 , 𝑑𝑚𝑔 ) Equation 30-52
Step 13: Compute the Probability of Left-Turn Bay Overflow

The probability of left-turn bay overflow is computed by using the following
equation:
𝑣𝑙𝑡 𝑁𝑞𝑥,𝑙𝑡+1
𝑝𝑜𝑣 = ( ) Equation 30-53
𝑐𝑙

with
𝑁𝑙𝑡 𝐿𝑎,𝑙𝑡
Equation 30-54 𝑁𝑞𝑥,𝑙𝑡 =
𝐿ℎ
where
pov = probability of left-turn bay overflow (decimal),
Nqx,lt = maximum queue storage for the left-turn movement (veh),
Nlt = number of lanes in the left-turn bay (ln),
La,lt = available queue storage distance for the left-turn movement (ft/ln), and
Lh = average vehicle spacing in the stationary queue (see Equation 30-15)
(ft/veh).
For an undivided cross section, the number of left-turn vehicles that can be
stored, Nqx,lt, is equal to 0.0.
Step 14: Compute Through-Vehicle Delay due to Left Turns

The through-vehicle delay due to left turns dap,l is computed with Equation
30-31.
DELAY DUE TO RIGHT TURNS

A vehicle turning right from the major street into an access point often delays
the through vehicles that follow it. Through vehicles are delayed because they
have to reduce speed to avoid a collision with the vehicle ahead, the first of
which has reduced speed to avoid a collision with the right-turning vehicle. This
delay can be several seconds in duration for the first few through vehicles but
will always decrease to negligible values for subsequent vehicles as the need to
reduce speed diminishes. For purposes of running time calculation, this delay
must be averaged over all through vehicles traveling in the subject direction. The
resulting average delay is computed with Equation 30-55.
𝑃𝑟𝑡
Equation 30-55
𝑑𝑎𝑝,𝑟 = 0.67 𝑑𝑡|𝑟
1 − 𝑃𝑙𝑡 − 𝑃𝑟𝑡
where
dap,r = through-vehicle delay due to right turns (s/veh) and
dt|r = through-vehicle delay per right-turn maneuver (s/veh).
The variable dt|r in Equation 30-55 converges to 0.0 as the proportion of
turning vehicles approaches 1.0. The constant 0.67 is a calibration factor based on
field data. The steps undertaken to quantify this factor are described in the
remainder of this subsection. Equation 30-55 can also be used to estimate the
delay due to left-turn vehicles on a one-way street. In this case, variables
associated with the right-turn movement would be redefined as applicable to the
left-turn movement and vice versa.
As indicated by Equation 30-55, the delay due to right turns is based on the
value of several variables. The following sequence of computations can be used
to estimate these values (7).

Step 1: Compute Minimum Speed for the First Through Vehicle

𝑢𝑚 = 1.47 𝑆𝑓 − 𝑟𝑑 (𝐻1 − ℎ|𝛥<ℎ<𝐻1 ) ≥ 𝑢𝑟𝑡 Equation 30-56
with
1 Δ − 𝐻1 𝑒 −𝜆(𝐻1 −Δ)
ℎ|Δ<ℎ<𝐻1 = + Equation 30-57
𝜆 1 − 𝑒 −𝜆(𝐻1 −Δ)
1.47 𝑆𝑓 − 𝑢𝑟𝑡 𝐿ℎ
𝐻1 = + 𝑡𝑐𝑙 + ≥Δ Equation 30-58
𝑟𝑑 1.47 𝑆𝑓
1
𝜆= Equation 30-59
1
−Δ
𝑞𝑛
where
um = minimum speed of the first through vehicle given that it is delayed (ft/s),
urt = right-turn speed = 20 (ft/s),
Sf = free-flow speed (mi/h),
h|∆<h<H1 = average headway of those headways between ∆ and H1 (s/veh),
Δ = headway of bunched vehicle stream = 1.5 (s/veh),

H1 = maximum headway that the first through vehicle can have and still
incur delay (s/veh),
rd = deceleration rate = 6.7 (ft/s2),
tcl = clearance time of the right-turn vehicle = 0.6 (s),
Lh = average vehicle spacing in stationary queue (see Equation 30-15)
(ft/veh),
λ = flow rate parameter (veh/s),
qn = outside lane flow rate = vn/3,600 (veh/s), and
vn = flow rate for the outside lane (veh/h/ln).
The right-turn speed urt used in Equation 30-56 and Equation 30-58 is likely
to be sensitive to access point design, including the approach profile, throat
width, and curb radius. For level profiles and nominal throat widths, the speed
can vary from 15 to 25 ft/s for radii varying from 20 to 60 ft, respectively. A
default turn speed of 20 ft/s is recommended when information is not available
to make a more accurate estimate.
The flow rate for the outside lane vn is computed by using Steps 3, 4, 5, and 6
from the procedure described in the previous subsection, Delay due to Left
Turns. However, the probability of a lane change Plc is set equal to 1.0 when the
calculations in Step 3 are made. In Steps 4 and 5, the variable It is set equal to
0.00001. The proportion of right-turning vehicles in the shared lane PR is also
computed at this point and used in a later step.

Step 2: Compute Delay to the First Through Vehicle

2
(1.47 𝑆𝑓 − 𝑢𝑚 ) 1 1
Equation 30-60 𝑑1 = ( + )
2 (1.47 𝑆𝑓 ) 𝑟𝑑 𝑟𝑎
where d1 is the conditional delay to the first through vehicle (s/veh), and ra is the
acceleration rate = 3.5 (ft/s2).
Step 3: Compute Delay to the Second Through Vehicle

Equation 30-61 𝑑2 = 𝑑1 − (ℎ|𝛥<ℎ<𝐻2 − 𝛥)
with
1 𝛥 − 𝐻2 𝑒 −𝜆(𝐻2 −𝛥)
Equation 30-62 ℎ|𝛥<ℎ<𝐻2 = +
𝜆 1 − 𝑒 −𝜆(𝐻2 −𝛥)
Equation 30-63 𝐻2 = 𝑑1 + 𝛥
where d2 is the conditional delay to Vehicle 2 (s/veh).
Step 4: Compute Delay to the Third and Subsequent Through Vehicles

Equation 30-64 𝑑𝑖 = 𝑑𝑖−1 − (ℎ|𝛥<ℎ<𝐻𝑖 − 𝛥)
with
1 𝛥 − 𝐻𝑖 𝑒 −𝜆(𝐻𝑖 −𝛥)
Equation 30-65 ℎ|𝛥<ℎ<𝐻𝑖 = +
𝜆 1 − 𝑒 −𝜆(𝐻𝑖 −𝛥)
Equation 30-66 𝐻𝑖 = 𝑑𝑖−1 + 𝛥
where di is the conditional delay to vehicle i (i = 3, 4, . . . , ) (s/veh). As shown by
Equation 30-61 and Equation 30-64, the delay to each subsequent through vehicle
is less than or equal to that of the preceding vehicle. In fact, the sequence of
delays always converges to zero when the average flow rate in the outside lane is
less than 1/Δ.
Step 4 should be repeated for the third and subsequent through vehicles
until the delay computed for vehicle i is less than 0.1 s. In general, this criterion
results in delay being computed for only the first two or three vehicles.
Step 5: Compute Through-Vehicle Delay per Right-Turn Maneuver

The through-vehicle delay for the first two vehicles is computed with
Equation 30-67.
Equation 30-67 𝑑𝑡|𝑟 = 𝑑1 (1 − 𝑒 −𝜆(𝐻1 −𝛥) )(1 − 𝑃𝑅 ) + 𝑑2 (1 − 𝑒 −𝜆(𝐻1 −𝛥) )(1 − 𝑒 −𝜆(𝐻2 −𝛥) )(1 − 𝑃𝑅 )2
where dt|r is the through-vehicle delay per right-turn maneuver (s/veh). If three
or more vehicles are delayed, an additional term needs to be added to Equation
30-67 for each subsequent vehicle. In this situation, Equation 30-68 can be used to
compute the delay for any number of vehicles.
∞ 𝑖
Equation 30-68 𝑑𝑡|𝑟 = ∑ [𝑑𝑖 × ∏(1 − 𝑒 −𝜆(𝐻𝑗−𝛥) ) × (1 − 𝑃𝑅 )𝑖 ]

𝑖=1 𝑗=1
Step 6: Compute Through-Vehicle Delay due to Right Turns

The through-vehicle delay due to right turns dap,r is computed with Equation
30-55.

5. PLANNING-LEVEL ANALYSIS APPLICATION
OVERVIEW OF THE APPLICATION

This section describes a simplified method for evaluating the operation of a
coordinated street segment with signalized boundary intersections. The
application addresses motorized vehicle operation. It is focused on the analysis
of the through movement at the boundary intersections. This method can be
used when minimal data are available for the analysis and only approximate
results are desired.
REQUIRED DATA AND SOURCES

The overall data requirements are summarized in Exhibit 30-7. Some of the
input requirements may be met by assumed values or default values. Other data
items are site-specific and must be obtained in the field. The objective of using
the planning-level analysis application is to minimize the need for the collection
of detailed field data.
Exhibit 30-7
Data Category Location Input Data Element
Required Input Data for the
Traffic characteristics Boundary intersection Through-demand flow rate Planning-Level Analysis
Through-saturation flow rate Application
Volume-to-capacity ratio of the upstream
movements
Segment Platoon ratio
Midsegment flow rate
Midsegment delay
Geometric design Boundary intersection Number of through lanes
Upstream intersection width
Segment Number of through lanes
Segment length
Restrictive median length
Nonrestrictive median length
Proportion of segment with curb
Number of access point approaches
Proportion of segment with on-street parking
Signal control Boundary intersection Effective green-to-cycle-length ratio
Cycle length
Other Segment Analysis period duration
Speed limit
At a minimum, the analyst must provide traffic volumes and the approach-
lane configuration for the subject intersection. Default values for several
variables are specifically identified in the methodology and integrated into the
method. These values have been selected to be generally representative of typical
conditions. Additional default values are identified in Section 3 of Chapter 18,
Urban Street Segments.
METHODOLOGY
The methodology consists of five computational steps. These steps are
• Determine running time;
• Determine proportion arriving during green;
• Determine through control delay;
Chapter 30/Urban Street Segments: Supplemental Planning-Level Analysis Application

• Determine through stop rate; and

• Determine travel speed, spatial stop rate, and level of service (LOS).
Each step is executed in the sequence presented in the preceding list. This
sequence is illustrated by the flowchart in Exhibit 30-8. The rectangles with
rounded corners indicate the computational steps. The parallelograms indicate
where input data are needed.
Exhibit 30-8 Input volume and

Planning-Level Analysis geometry
Application for Urban Street
Segments
Segment Analysis Module
Input delays due to

Compute segment running time
turns
Compute proportion arriving

Input platoon ratio
during green
Signalized Intersection Module

Compute control delay
Input g/C ratio

Compute stop rate
Performance Measures Module

Compute segment travel speed
Report segment delay, travel time,
stop rate, and travel speed
Compute spatial stop rate
The computations associated with each step identified in Exhibit 30-8 are
described in Section 3 of Chapter 18. These computations are conveniently
illustrated here in a series of worksheets; each worksheet corresponds to one or
two of the calculation steps.
The first of the computational worksheets is the Running Time worksheet. It
is shown as Exhibit 30-9 (values shown apply to the Example Problem, as
discussed in a subsequent section).
Planning-Level Analysis Application Chapter 30/Urban Street Segments: Supplemental

RUNNING TIME WORKSHEET Exhibit 30-9

Planning-Level Analysis:
General Information Site Information Running Time Worksheet
Analyst JME Street Texas Avenue
Agency or Company ACME Engr. Jurisdiction
Date Performed 9/30/15 Analysis Year 2015
Analysis Period 5:30 p.m. to 5:45 p.m. Analysis Level planning
Base free-flow speed calibration factor (Scalib), mi/h: 0.0
Input Data
Segment 1 Segment 2
Direction of travel EB/NB WB/SB EB/NB WB/SB
Segment Data
Number of through lanes for length of segment Nth (ln) 2 2
Speed limit Spl (mi/h) 35 35
Midsegment volume vm (veh/h) 1,150 1,150
Total delay due to turns into access points Σdap (s/veh) 0.52 0.52
Delay due to other midsegment sources dother (s/veh) 0 0
Length of segment L (ft) 1,800 1,800
Width of upstream boundary intersection Wi (ft) 50 50
Length of segment with restrictive median Lrm (ft) 0 0
Length of segment with nonrestrictive median Lnr (ft) 0 0
Start-up lost time l1 (s) 2.0 2.0
Access Data
Proportion of segment with curb on right-hand side pcurb 0.70 0.70
Number of access points on right-hand side Nap 4 4
Proportion of segment with on-street parking ppk 0.00 0.00
Running Time Computation
Adjusted segment length Ladj (ft) Ladj = L – Wi 1,750 1,750
Proportion of segment length with restrictive median prm, 0.0 0.0
prm = Lrm /Ladj
Speed constant S0 (mi/h), S0 = 25.6 + 0.47 Spl 42.1 42.1
Adjustment for cross section fCS (mi/h), -0.3 -0.3
fCS = 1.5 prm – 0.47 pcurb – 3.7 pcurb prm
Access point density Da (access points/mi), 24.1 24.1
Da = 5,280 (Nap,EB/NB + Nap,WB/SB) /Ladj
Adjustment for access points fA (mi/h), -0.9 -0.9
fA = –0.078 Da /Nth
Adjustment for on-street parking fpk (mi/h), fpk = –3 ppk 0.0 0.0
Base free-flow speed Sfo (mi/h), Sfo =Scalib +S0 +fCS +fA +fpk 40.8 40.8
Segment length adjustment factor fL, 0.96 0.96
fL = 1.02 – 4.7 (Sfo – 19.5)/max(L, 400) ≤ 1.0
Free-flow speed Sf (mi/h), Sf = Sfo fL ≥ Spl 39.3 39.3
fv = 2 1.03 1.03
Proximity 0.21
 vm 
adjustment 1 + 1 − 
 52.8 N th S f 
factor fv 
Running 6.0 −l1 3,600 L 33.7 33.7

tR = + f +  d ap +d other
time tR (s) 0.0025 L 5,280 S f v
Note: The first term in the running time equation is only applicable to segments with signal-controlled, STOP-
controlled, or YIELD-controlled through movement at the boundary intersection.

The Running Time worksheet combines input data describing the segment
geometric design, speed limit, volume, and access point frequency to estimate
the base free-flow speed. This speed is then adjusted for segment length effects to
obtain the expected free-flow speed. The free-flow speed is then used to estimate
a free-flow travel time, which is adjusted for the proximity of other vehicles.
Delay that is caused by turns into access points or other sources is added to the
adjusted travel time. Default values for the delay due to turns at midsegment
access points are listed in Exhibit 18-13 in Chapter 18. These defaults can be used
when more accurate estimates of this delay are not available. The result of these
adjustments is an estimate of the expected segment running time.
The second of the computational worksheets is the Proportion Arriving
During Green worksheet. It is shown as Exhibit 30-10. This worksheet is
designed for the analysis of the segment through-lane group. It documents the
calculation of the proportion of vehicles that arrive during the green indication.
Input data include the effective green-to-cycle-length ratio and platoon ratio.
Exhibit 30-10 PROPORTION ARRIVING DURING GREEN WORKSHEET

Proportion Arriving During General Information
Green Worksheet
Project Description Texas Avenue, 5:30 p.m. to 5:45 p.m.
Input Data
Segment 1 Segment 2
Signal Timing Data
Effective green-to-cycle-length ratio g/C 0.47 0.47
Traffic Data
Platoon ratio Rp 1.43 0.67
Proportion Arriving During Green Computation
Proportion arriving during green P, P = Rp (g/C) 0.67 0.31
The third computational worksheet is the Control Delay worksheet. It is

shown as Exhibit 30-11. This worksheet is designed for the analysis of the
segment through-lane group. Input variables include the analysis period
duration, cycle length, effective green-to-cycle-length ratio, volume, saturation
flow rate, and lanes. The proportion of arrivals during green is obtained from the
previous worksheet.
The equation for computing the progression adjustment factor PF* that is
provided in Exhibit 30-11 is a simplified version of the exact equation (as
provided in Section 3 of Chapter 19). The simplified equation, in combination
with the supplemental adjustment factor fPA, is sufficiently accurate for purposes
of the planning-level analysis application.
The control delay is computed as the sum of two components. The first
component to be computed is the uniform delay. The notation “min(1, X)” is
shown in the equation used to compute this delay. It means that the value to be
substituted for this text is the smaller of 1.0 and the volume-to-capacity ratio.

CONTROL DELAY WORKSHEET Exhibit 30-11

General Information Control Delay Worksheet
Input Data
Analysis period T (h): 0.25 Segment 1 Segment 2
Signal Timing Data
Cycle length C (s) 100 100
Traffic Data
Through-lane group volume vth (veh/h) 968 950
Lane group saturation flow rate s (veh/h/ln) 1,800 1,800
Proportion of arrivals during green P 0.67 0.31
Volume-to-capacity ratio Xu of the upstream movements 0.57 0.57
Geometric Design Data
Number of through lanes Nth (ln) 2 2
Delay Computation
Capacity c (veh/h), c = Nth s g/C 1,692 1,692
Volume-to-capacity ratio X, X = vth/c 0.57 0.56
Supplemental adjustment factor for platoons arriving 1.15 0.93
during green fPA, fPA = 1.00 except as noted below:
If 0.50 < Rp ≤ 0.85, then fPA = 0.93
If 1.15 < Rp ≤ 1.50, then fPA = 1.15
Progression adjustment factor PF*, 0.71 1.20
PF* = fPA (1 – P)/(1 – g/C)
( )10−.5min(
2 13.6 23.0
* C (1 − g / C )
Uniform delay d1 (s/veh), d1 = PF
1, X )g / C 
Upstream filtering adjustment factor I, 0.80 0.80
I = 1.0 − 0.91 X u2.68  0.090
Incremental delay d2 (s/veh), 1.13 1.08
 4I X 
d 2 = 900 T ( X − 1) + ( X − 1)2 + 
 c T 
Control delay d (s/veh), d = d1 + d2 14.7 24.1
The second delay component is the incremental delay, which is based on the
upstream filtering adjustment factor. This factor requires the variable Xu, which
can be estimated as the volume-to-capacity ratio of the segment through-lane
group at the upstream signalized intersection. Additional detail on the calculation
of this ratio is provided in Section 3 of Chapter 19, Signalized Intersections.
The fourth computational worksheet is the Stop Rate worksheet. It is shown
as Exhibit 30-12. This worksheet is designed for the analysis of the segment
through-lane group. The input variables are the same as those needed for the
Control Delay worksheet with the addition of speed limit. The average speed

during the analysis period is estimated by using the equation provided. If the
average speed is known, it should be substituted for the estimated value.
Exhibit 30-12 STOP RATE WORKSHEET

Stop Rate Worksheet General Information
Input Data
Analysis period T (h): 0.25 Segment 1 Segment 2
Signal Timing Data
Cycle length C (s) 100 100
Traffic Data
Through-lane group volume vth (veh/h) 968 950
Lane group saturation flow rate s (veh/h/ln) 1,800 1,800
Proportion of arrivals during green P 0.67 0.31
Speed limit Spl (mi/h) 35 35
Incremental delay d2 (s/veh) 1.13 1.08
Number of through lanes Nth (ln) 2 2
Stop Rate Computation
Effective green time g (s), g = C (g/C) 47 47
Effective red time r (s), r = C – g 53 53
Capacity c (veh/h), c = Nth s g/C 1,692 1,692
Average speed Sa (mi/h), Sa = 0.90 (25.6 + 0.47 Spl) 37.8 37.8
Threshold acceleration–deceleration delay (s), (1 – P) g X 8.8 18.1
Acceleration–deceleration delay da (s), 11.2 11.2
da = 0.393 (Sa – 5.0)2/Sa
Deterministic stop rate h1 (stops/veh), 0.31 0.74
1 − P (1 + d a / g )
h1 = if d a  (1 − P ) g X
1−P X
(1 − P )(r − d a )
h1 = if d a  (1 − P ) g X
r − (1 − P ) g X
Second-term back-of-queue size Q2 (veh/ln), 0.26 0.25
Q2 = c d2 /(3,600 Nth)
Full stop rate h (stops/veh), h = h1 + 3,600 Nth Q2 /(vth C ) 0.33 0.76
The stop rate is computed as the sum of two components. The first
component to be computed is the deterministic stop rate. Two equations are
available for this computation. The correct equation to use is based on a check of
the acceleration–deceleration delay relative to the computed threshold value.
The second stop rate component is based on the second-term back-of-queue
size. This queue represents the average number of vehicles that are unserved at
the end of the green interval. It is based on the incremental delay computed for
the Control Delay worksheet.

The fifth computational worksheet is the Travel Speed and Spatial Stop Rate
worksheet. It is shown as Exhibit 30-13. This worksheet is designed for the
analysis of the segment through-lane group. The input values include segment
length and the full stop rate associated with other midsegment events (e.g., turns
at access points). The other input data listed represent computed values and are
obtained from the previous worksheets.
TRAVEL SPEED AND SPATIAL STOP RATE WORKSHEET Exhibit 30-13

General Information Travel Speed and Spatial Stop
Rate Worksheet
Input Data
Segment 1 Segment 2
Length of segment L (ft) 1,800 1,800
Base free-flow speed Sfo (mi/h) 40.8 40.8
Running time tR (s) 33.7 33.7
Control delay d (s/veh) 14.7 24.1
Full stop rate h (stops/veh) 0.33 0.76
Full stop rate due to other midsegment 0 0
sources hother (stops/veh)
Travel Speed Computation
Travel time TT (s), TT = tR + d 48.4 57.7
Travel speed ST,seg (mi/h), 25.4 21.3
3,600 L
ST , seg =
5,280TT
Spatial Stop Rate Computation

Total stop rate hT (stops/veh), 0.33 0.76
hT = h + hother
Spatial stop rate Hseg (stops/mi), 0.96 2.23
5,280 hT
H seg =
L
Level-of-Service Computation
Travel speed thresholds for base free-flow A: >32.6 A: >32.6
speed (Sfo) by interpolation of values in B: >27.3 B: >27.3
Exhibit 18-1 (mi/h) C: >20.4 C: >20.4
D: >16.3 D: >16.3
E: >12.2 E: >12.2

EXAMPLE PROBLEM
The Urban Street Segment
The total length of an undivided urban street segment is 1,800 ft. It is shown
in Exhibit 30-14. Both of the boundary intersections are signalized. The street has
a four-lane cross section with two lanes in each direction. There are left-turn bays
on the subject segment at each signalized intersection.
Exhibit 30-14 1,800 ft N

Example Problem 950 veh/h
Platoon Ratio = 0.67
1 2
968 veh/h
Platoon Ratio = 1.43
Signal Segment 1 Signal
The segment has two access point intersections. Each intersection has two
STOP-controlled side-street approaches, and each approach has sufficient traffic
volume during the analysis period to be considered active. The segment also has
two driveways on each side of the street; however, their turn movement volumes
are too low for them to be considered active.
The Question
What are the travel speed, spatial stop rate, and LOS during the analysis
hour for through-vehicle traffic in both directions of travel along the segment?
The Facts
Some details of the segment are shown in Exhibit 30-14. Both boundary
intersections are signalized. The following additional information is known
about the street segment:
Through saturation flow rate: 1,800 veh/h/ln
Midsegment volume: 1,150 veh/h
Midsegment delay: 0.52 s/veh
Number of through lanes at boundary intersection: 2
Upstream intersection width: 50 ft
Number of through lanes on segment: 2
Proportion of street with curb: 0.70
Proportion of street with on-street parking: 0.0
g/C ratio: 0.47
Cycle length: 100 s
Analysis period: 0.25 h
Speed limit: 35 mi/h
Percent left turns at active access points: 6%
Percent right turns at active access points: 8%

Selected Calculations
1. Compute total delay due to turns into access Midsegment lanes = 2 lanes
points Midsegment lane volume = 575 veh/h/ln
Interpolate in Exhibit 18-13 to obtain
0.37 s/veh/pt through-vehicle delay.
Number of active access points = 2

Percent turns = 7% [= (6 + 8)/2]
Total delay per access pt. = 7/10 × 0.37
= 0.26 s/veh/pt
Total delay per segment = 2 × 0.26
= 0.52 s/veh
2. Compute upstream filtering factor No information was available about the
volume-to-capacity ratio for the upstream
movements, so this ratio was estimated to
equal the volume-to-capacity ratio for the
subject movement.
Results
The calculations are shown in Exhibit 30-9 to Exhibit 30-13. The travel speed
for the eastbound direction is 25.4 mi/h. The travel speed for the westbound
direction is 21.3 mi/h. The eastbound and westbound spatial stop rates are 0.96
and 2.23 stops/mi, respectively.
The base free-flow speed is 40.8 mi/h. By interpolating this value between
those in Exhibit 18-1, the threshold travel speeds for LOS A, B, C, D, and E are
>32.6, >27.3, >20.4, >16.3, and >12.2 mi/h, respectively. Thus, the travel speed for
the eastbound direction of 26.3 mi/h corresponds to LOS C. The westbound LOS
is similarly determined to be C.

6. FIELD MEASUREMENT TECHNIQUES
This section describes two techniques for estimating key vehicular traffic
characteristics by using field data. The first technique is used to estimate free-
flow speed. The second technique is used to estimate average travel speed.
The field measurements for both techniques should occur during a time
period that is representative of the analysis period. This approach recognizes a
possible difference in driver speed choice during different times of day (and,
possibly, days of week and months of year).
FREE-FLOW SPEED
The following steps can be used to determine the free-flow speed for
vehicular traffic on an urban street segment. The definition of “urban street
segment” is provided in Section 2 of Chapter 18.
The speed measured with the technique described in this section describes
the free-flow speed for the subject segment. It is not necessarily an accurate
measurement of the free-flow speed on an adjacent segment because of possible
differences in geometry, access point spacing, or speed limit.
Some urban streets have characteristics that can influence free-flow speed
but that are not considered in the predictive procedure. If free-flow speed is
measured for these segments, the results should be qualified to acknowledge the
possible influence of these characteristics on the measured speed. These
characteristics include a change in the posted speed limit along the segment, the
display of an advisory speed sign that has an advisory speed lower than the
speed limit, a change in the number of through lanes along the segment,
significant grade, or a midsegment capacity constraint (e.g., narrow bridge).
Step 1. Conduct a spot-speed study at a midsegment location during low-
volume conditions. Record the speed of 100 or more free-flowing passenger cars.
A car is free-flowing when it has a headway of 8 s or more to the vehicle ahead
and 5 s or more to the vehicle behind in the same traffic lane. In addition, a free-
flow vehicle is not influenced (i.e., slowed) by the following factors: (a) vehicles
turning onto (or off of) the subject segment at the boundary intersection or at a
midsegment access point, (b) traffic control devices at the boundary intersections,
or (c) traffic control devices deployed along the segment.
In view of the aforementioned definition of “free-flow vehicle,” vehicles
turning into (or out of) an access point should not be included in the database.
Vehicles that are accelerating or decelerating as a result of driver response to a
traffic control signal should not be included in the database. Vehicles should not
be included if they are influenced by signs that require a lower speed limit
during school hours or signs that identify a railroad crossing.
Step 2. Compute the average of the spot speeds Sspot and their standard
deviation σspot.
Step 3. Compute the segment free-flow speed Sf as a space mean speed by
Field Measurement Techniques Chapter 30/Urban Street Segments: Supplemental

2
𝜎spot
𝑆𝑓 = 𝑆spot − Equation 30-69
𝑆spot
where
Sspot = average spot speed (mi/h), and
σspot = standard deviation of spot speeds (mi/h).
Step 4. If the base free-flow speed Sfo is also desired, it can be computed by
𝑆𝑓
𝑆𝑓𝑜 = Equation 30-70
𝑓𝐿
with
𝑆𝑓 − 19.5
𝑓𝐿 = 1.02 − 4.7 ≤ 1.0 Equation 30-71
max(𝐿𝑠 , 400)
where
Sfo = base free-flow speed (mi/h),
Ls = distance between adjacent signalized intersections (ft), and
fL = signal spacing adjustment factor.
Equation 30-71 was originally derived with the intent of using the base free-
flow speed Sfo in the numerator of the second term. However, use of the free-flow
speed Sf in its place is sufficient for this application.
Equation 30-71 was derived by using signalized boundary intersections. For
more general applications, the definition of distance Ls is broadened so that it
equals the distance between the two intersections that (a) bracket the subject
segment and (b) each have a type of control that can impose on the subject
through movement a legal requirement to stop or yield.
AVERAGE TRAVEL SPEED

The following steps can be used to determine the average travel speed for
vehicular traffic on an urban street segment.
Step 1. Identify the time of the day (e.g., morning peak, evening peak, off-
peak) during which the study will be conducted. Identify the segments to be
evaluated.
Step 2. Conduct the test car travel time study for the identified segments
during the identified study period. The following factors should be considered
before or during the field study:
• The number of travel time runs will depend on the range of speeds found
on the street. Six to 12 runs for each traffic volume condition are typically
adequate. The analyst should determine the minimum number of runs on
the basis of guidance provided elsewhere (8).
Chapter 30/Urban Street Segments: Supplemental Field Measurement Techniques

• The objective of the data collection is to obtain the information identified

in the Travel Time Field Worksheet (i.e., vehicle location and arrival and
departure times at each boundary intersection). This worksheet is shown
in Exhibit 30-15. In general, each row of this worksheet represents the
data for one direction of travel on one segment. If the street serves traffic
in two travel directions, separate worksheets are typically used to record
the data for each direction of travel.
• The equipment used to record the data may include a Global Positioning
System–equipped laptop computer or simply a pair of stopwatches. If
available, an instrumented test car should be used to reduce labor
requirements and to facilitate recording and analysis.
• During the test run, the average-car technique is typically used and
requires that the test car travel at the average speed of the traffic stream,
as judged by its driver (8).
• The cumulative travel time is recorded as the vehicle passes the center of
each boundary intersection. Whenever the test car stops or slows (i.e.,
5 mi/h or less), the observer uses a second stopwatch to measure the
duration of time the vehicle is stopped or slowed. This duration (and the
cause of the delay) is recorded on the worksheet on the same row that is
associated with the next boundary intersection to be reached. The rows
are intentionally tall so that a midsegment delay and the signal delay can
both be recorded in the same cell.
• Test car runs should begin at different time points in the signal cycle to
avoid having all runs start from a “first in platoon” position.
• Some midsegment speedometer readings should also be recorded to
check on unimpeded travel speeds and to see how they relate to the
estimated free-flow speed.
Step 3. The cumulative travel time observations between adjacent boundary
intersections are subtracted to obtain the travel time for the corresponding
segment. This travel time can be averaged for all test runs to obtain an average
segment travel time. The average is then divided into the segment length to
obtain an estimate of the average travel speed. This speed should be computed
for each direction of travel for the segment.
The data should be summarized to provide the following statistics for each
segment travel direction: average travel speed, average delay time for the
boundary intersection, and average delay time for other sources (pedestrian,
parking maneuver, etc.).
The average segment travel time for each of several consecutive segments in
a common direction of travel can be added to obtain the total travel time for the
facility. This total travel time can then be divided into the facility length (i.e., the
total length of all segments) to obtain the average travel speed for the facility.
This calculation should be repeated to obtain the average travel speed for the
other direction of travel.
Field Measurement Techniques Chapter 30/Urban Street Segments: Supplemental

TRAVEL TIME FIELD WORKSHEET Exhibit 30-15

Travel Time Field Worksheet
General Information Site Information
Analyst Street
Agency or Company Jurisdiction
Date Performed Analysis Year
Analysis Period Direction of
Travel
Field Data
Run Number: __________________ Run Number:
_________________
Delays due to Slow Delays due to Slow
Location Cumulative
or Stop or Stop
(typically a Cumulative Travel Time
boundary Travel Time at Delay at Location Delay
intersection) Location (s) Cause a Time (s) (s) Cause a Time (s)
a
Notes: Cause of delay: Ts = signal; Lt = left turn; Pd = pedestrian; Pk = parking; Ss = STOP sign; Ys = YIELD sign.
Chapter 30/Urban Street Segments: Supplemental Field Measurement Techniques

7. COMPUTATIONAL ENGINE DOCUMENTATION
This section uses a series of flowcharts and linkage lists to document the
logic flow for the computational engine.
FLOWCHARTS
The methodology flowchart is shown in Exhibit 30-16. The methodology
consists of five main modules:
• Setup Module,
• Segment Evaluation Module,
• Segment Analysis Module,
• Delay due to Turns Module, and
• Performance Measures Module.
This subsection provides a separate flowchart for each of these modules.
Exhibit 30-16
Methodology Flowchart
Computational Engine Documentation Chapter 30/Urban Street Segments: Supplemental

The Setup Module is shown in Exhibit 30-17. This module consists of five
main routines, as shown in the large rectangles of the exhibit. The main function
of each routine, as well as the name given to it in the computational engine, is
also shown in the exhibit. These routines and the Initial Queue Delay Module are
described in Chapter 31, Signalized Intersections: Supplemental.
Start Exhibit 30-17

Setup Module
Initial estimate of cycle length

(InitialSetupRoutine) Establish lane groups; estimate
initial group sat. flow rate, group
volume, and phase duration
(InitialCapacityEstimate)
Set demand flow = input flow rate
for current analysis period
(PeriodVolumeSetup)
Convert input movement initial
queue to lane group initial queue
If initial queue exists, obtain (InitialQueueSetup)
saturated delay estimate
(AnalysisSetup)
Finish
The Segment Evaluation Module is shown in Exhibit 30-18. This module

consists of eight main routines. The main function of each routine, as well as the
name given to it in the computational engine, is also shown in the exhibit. The
Segment Analysis Module and the Delay due to Turns Module are outlined in
the next two exhibits. The Signalized Intersection Module and the Compute
Average Phase Duration routine are described in Chapter 31. The Volume Check,
Define Origin–Destination Matrix, Spillback Check, and Midsegment Capacity
routines are described further in the next subsection.
Exhibit 30-18
Segment Evaluation Module
Chapter 30/Urban Street Segments: Supplemental Computational Engine Documentation

The Segment Analysis Module is shown in Exhibit 30-19. This module

consists of seven main routines, six of which are implemented for both segment
travel directions. The main function of each routine, as well as the name given to
it in the computational engine, is also shown in the exhibit. These routines are
described further in the next subsection.
Exhibit 30-19 Start

Compute initial proportion of

arrivals during green
(InitialPortionOnGreen) Compute arrival flow profile to
downstream intersection
(ComputeProjectedProfile)
Compute discharge flow profile Compute arrival flow profile for

for entry movements exit movements
(ComputeDischargeProfile) (ComputeConflictFlowRate)
Compute running time to Compute proportion of arrivals

downstream intersection during green
(GetRunningTime) (ComputePortionOnGreen)
No Evaluated both travel Yes

directions?
Compute proportion of time

Finish blocked
(ComputeBlockTime)
The Delay due to Turns Module is shown in Exhibit 30-20. This module
consists of two main routines, each of which is implemented for both segment
travel directions. The main function of each routine, as well as the name given to
it in the computational engine, is also shown in the exhibit. These routines are
described further in the next subsection.
Exhibit 30-20 Start

Delay due to Turns Module
Compute lane volume distribution Compute lane volume distribution

when inside lane blocked when inside lane not blocked
(ComputeAcPtApproachVolumeDist) (ComputeAcPtApproachVolumeDist)
Compute delay due to left-turning Compute delay due to left-turning

vehicles in the inside lane vehicles in the inside lane
(ComputeThruDelayAtAcPT) (ComputeThruDelayAtAcPT)
No Evaluated both travel Yes

Finish
directions?

The Performance Measures Module is shown in Exhibit 30-21. This module

consists of four routines. The main function of each routine is also shown in the
exhibit. One of the routines (i.e., EstimateIncrementalDelay) is complicated
enough to justify its development as a separate entity in the computational
engine. This routine is described in Chapter 31, Signalized Intersections:
Supplemental.
Exhibit 30-21
Module
LINKAGE LISTS
This subsection uses linkage lists to describe the main routines that make up
the computational engine. Each list is provided in a table that identifies the
routine and the various subroutines that it references. Conditions for which the
subroutine is used are also provided.
The lists are organized by module, as described in the previous subsection. A
total of three tables are provided to address the following three modules:
• Segment Evaluation Module,
• Segment Analysis Module, and
• Delay due to Turns Module.
The linkage list for the Segment Evaluation Module is provided in Exhibit
30-22. The main routines are listed in Column 1 and were previously identified in
Exhibit 30-18.
The linkage list for the Segment Analysis Module is provided in Exhibit 30-
23. The main routines are listed in Column 1 and were previously identified in
Exhibit 30-19.
Finally, the linkage list for the Delay due to Turns Module is provided in
Exhibit 30-24. The main routines are listed in Column 1 and were previously
identified in Exhibit 30-20.

Exhibit 30-22 Routine Subroutine Conditions for Use

Segment Evaluation Module
VolumeCheck Ensure that discharge volume for Apply for both segment
Routines
each entry movement does not travel directions.
exceed its capacity.
DefineODMatrix ComputeODs Apply to all intersections on
(compute origin–destination volume segment and for both
for movements that enter and exit segment travel directions.
segment)
SpillbackCheck ComputeSpillbackTime Apply for both segment
(compute spillback time for each travel directions.
exit movement at the downstream
boundary intersection)
SegmentAnalysisModule See Exhibit 30-23.
SignalizedIntersectionModule See Chapter 31.
ComputeMidSegmentCapacity Compute midsegment capacity Apply to each upstream
when restricted and reduce signalized intersection traffic
saturation flow rate of upstream movement that enters
movements so upstream discharge segment.
is less than or equal to the
midsegment capacity.
DelayDueToTurnsModule See Exhibit 30-24.
ComputeAveragePhaseDuration See Chapter 31.

InitialPortionOnGreen Compute proportion of arrivals during None
Routines
green (P) based on current signal
timing.
ComputeDischargeProfile Compute discharge flow rate for each Apply to each upstream
1-s interval of signal cycle at upstream boundary intersection
boundary intersection. movement that enters
segment.
GetRunningTime Compute running time on length of Apply to all intersections on the
street between upstream boundary segment and for both segment
intersection and subject downstream travel directions.
intersection.
ComputeProjectedProfile Compute arrival flow profile reflecting Apply to each upstream
dispersion of platoons formed at boundary intersection
upstream boundary intersection. movement that enters
segment.
ComputeConflictFlowRate Use arrival flow profile and origin– Apply to all intersections on the
destination matrix to compute arrival segment and for both segment
flow rate for movements at subject travel directions.
intersection.
Compute conflicting flow rate at access Apply to all access point
point intersections on basis of the intersections and for both
projected arrivals at each intersection. segment travel directions.
ComputePortionOnGreen For each exit movement, compute Apply to each downstream
count of vehicles arriving at boundary intersection.
downstream boundary intersection
during green.
ComputeBlockTime Use computed conflicting flow rates at Apply to all access point
each access point intersection to intersections and for both travel
compute the proportion of time segment travel directions.
blocked for each nonpriority
movement.

Routine Subroutine Conditions for Use Exhibit 30-24

Delay due to Turns Module
ComputeAcPtApproach- Compute the volume for each lane Apply lane volume routine for
Routines
VolumeDist on the approach to the access case in which inside lane is
point intersection when blocked blocked by a turning vehicle.
by a left-turning vehicle. Apply to all access point
intersections and for both
segment travel directions.
Compute the volume for each lane Apply lane volume routine for
on the approach to the access case in which inside lane is not
point intersection when not blocked by a turning vehicle.
blocked by a left-turning vehicle. Apply to all access point
intersections and for both
segment travel directions.
ComputeThruDelayAtAcPT Compute the probability of left- If segment is undivided, the
turn bay overflow at access point probability of bay overflow is
intersection. 1.0.
Compute the delay to through Apply to all access point
movements due to a left turn at intersections and for both
an access point. segment travel directions.
Based on lane volume estimate for
case in which inside lane is
blocked by a turning vehicle.
Compute the delay to through Apply to all access point
movements due to a right turn at intersections and for both
an access point. segment travel directions.
Based on lane volume estimate for
case in which inside lane is not
blocked by a turning vehicle.

8. EXAMPLE PROBLEMS
This section describes the application of each of the motorized vehicle,

pedestrian, bicycle, and transit methodologies through the use of example
problems. Exhibit 30-25 provides an overview of these problems. The focus of the
examples is on an operational analysis. A planning and preliminary engineering
analysis is identical to the operational analysis in terms of the calculations, except
that default values are used when field-measured values are not available.
Exhibit 30-25 Problem

Example Problems Number Description Analysis Type
1 Motorized Vehicle LOS Operational
2 Pedestrian LOS Operational
3 Bicycle LOS Operational
4 Transit LOS Operational
EXAMPLE PROBLEM 1: MOTORIZED VEHICLE LOS

The Urban Street Segment
The total length of an undivided urban street segment is 1,800 ft. The segment
is shown in Exhibit 30-26. Both of the boundary intersections are signalized. The
street has a four-lane cross section with two lanes in each direction. There are
left-turn bays on the subject segment at each signalized intersection.
Exhibit 30-26 1,800 ft N

Example Problem 1: Urban
600 ft 600 ft
Street Segment Schematic
1 2
AP1 AP2
The segment has two active access point intersections, shown in the exhibit
as AP1 and AP2. Each intersection has two STOP-controlled side-street
approaches. The segment has some additional driveways on each side of the
street; however, their turn movement volumes are too low during the analysis
period for them to be considered active. The few vehicles that do turn at these
locations during the analysis period have been added to the corresponding
volumes at the two active access point intersections.
The Question
What are the travel speed, spatial stop rate, and LOS during the analysis
period for the segment through movement in both directions of travel?
Example Problems Chapter 30/Urban Street Segments: Supplemental

The Facts
The segment’s traffic counts are listed in Exhibit 30-27. The counts were
taken during the 15-min analysis period of interest. However, they have been
converted to hourly flow rates. Note that the volumes leaving the signalized
intersections do not add up to the volume arriving at the downstream access
point intersection.
50 500 100 100 80 100 80 50 500 100
Exhibit 30-27
10 100 100 10
Example Problem 1:
200
1,000
80 Access Point 1,050
80 Access Point 1,050
200
1,000 Intersection Turn Movement
Signal 1 Signal 2
1,000
200
1,050
Intersection 1 80
1,050
Intersection 2 80
1,000
200 Counts
10 100 100 10
100 500 50 80 100 80 100 100 500 50
The signalization conditions are shown in Exhibit 30-28. The conditions

shown are identified as belonging to Signalized Intersection 1; however, they are
the same for Signalized Intersection 2. The signals operate with coordinated–
actuated control. The left-turn movements on the northbound and southbound
approaches operate under protected–permitted control and lead the opposing
through movements (i.e., a lead–lead phase sequence). The left-turn movements
on the major street operate as protected-only in a lead–lead sequence.
Controller Data Worksheet Exhibit 30-28

General Information Example Problem 1: Signal
Cross street: First Avenue Analysis period: 7:15 am to 7:30 am Conditions for Intersection 1
Phase Sequence
Phases 1 and 2 Phases 3 and 8
1. WB left (1) with WB thru (6) 1. NB left (3) with NB thru (8)
Enter choice 2 2. WB left (1) before EB thru (2) Enter choice 2 2. NB left (3) before SB thru (4)
3. EB thru (2) before WB left (1) 3. SB thru (4) before NB left (3)
Phases 5 and 6 Phases 4 and 7
1. EB left (5) with EB thru (2) 1. SB left (7) with SB thru (4)
Enter choice 2 2. EB left (5) before WB thru (6) Enter choice 2 2. SB left (7) before NB thru (8)
3. WB thru (6) before EB left (5) 3. NB thru (8) before SB left (7)
Left-Turn Mode
Phase 1 or 2 Phase 3 or 8
Enter choice 3 2. WB left (1) prot-perm Enter choice 2 2. NB left (3) prot-perm
3. WB left (1) protected 3. NB left (3) protected
Phase 5 or 6 Phase 4 or 7
Enter choice 3 2. EB left (5) prot-perm Enter choice 2 2. SB left (7) prot-perm
3. EB left (5) protected 3. SB left (7) protected
Phase Settings
Approach Eastbound Westbound Northbound Southbound
Phase number 5 2 1 6 3 8 7 4
Movement L T+R L T+R L T+R L T+R
Lead/lag left-turn phase Lead -- Lead -- Lead -- Lead --
Left-turn mode Prot. -- Prot. -- Pr/Pm -- Pr/Pm --
Passage time, s 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
Phase split, s 20 35 20 35 20 25 20 25
Minimum green, s 5 8 5 8 5 5 5 5
Yellow change, s 3.0 4.0 3.0 4.0 3.0 4.0 3.0 4.0
Red clearance, s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Walk+ ped. clear, s 0 0 0 0
Recall? No No No No No No No No
Dual entry ? No Yes No Yes No Yes No Yes
Simultaneous gap-out? Yes Yes
Dallas left-turn phasing? No No
Coordination settings Offset, s: 0 Offset Ref.: End of Green Force Mode: Fixed
Cycle, s: 100 Reference phase: 2
Chapter 30/Urban Street Segments: Supplemental Example Problems

Exhibit 30-28 indicates that the passage time for each actuated phase is 2.0 s.
The minimum green setting for each actuated phase is 5 s. The offset to Phase 2
(the reference phase) end-of-green interval is 0.0 s. A fixed-force mode is used to
ensure that good coordination is maintained. The cycle length is 100 s.
Geometric conditions and traffic characteristics for Signalized Intersection 1
are shown in Exhibit 30-29. They are the same for Signalized Intersection 2. The
movement numbers follow the numbering convention shown in Exhibit 19-1 of
Chapter 19.
Exhibit 30-29 Approach Eastbound

Intersection Data Worksheet
Example Problem 1: Movement L T R L T R L T R L T R
Movement number 5 2 12 1 6 16 3 8 18 7 4 14
Geometric Conditions and Intersection Geometry
Traffic Characteristics for Number of lanes
Lane assignment
1
L
2
T
1
R
1
L T
2 1
R
1
L
2
TR
0
n.a.
1
L
2
TR
0
n.a.
Signalized Intersection 1 Average lane width, ft 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0
Number of receiving lanes 2 2 2 2
Turn bay or segment length, ft 200 999 200 200 1800 200 200 999 200 200 999
Traffic Characteristics
Volume, veh/h 200 1000 10 200 1000 10 100 500 50 100 500 50
Right-turn-on-red volume, veh/h 0 0 0 0
Percent heavy vehicles, % 0 0 0 0 0 0 0 0 0 0 0 0
Lane utilization adjustment factor 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Peak hour factor 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Start-up lost time, s 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
Extension of eff. green time, s 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
Platoon ratio 1.000 1.333 1.000 1.000 1.333 1.000 1.000 1.000 1.000 1.000 1.000
Upstream filtering factor 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Pedestrian volume, p/h 0 0 0 0
Bicycle volume, bicycles/h 0 0 0 0
Opposing right-turn lane influence Yes Yes Yes Yes
Initial queue, veh 0 0 0 0 0 0 0 0 0 0 0
Speed limit, mi/h 35 35 35 35 35 35 35 35 35 35 35 35
Unsignalized movement volume, veh/h 0 0 0 0 0 0 0 0 0 0 0 0
Unsignalized movement delay, s/veh 0 0 0 0 0 0 0 0 0 0 0 0
Unsignalized mvmt. stop rate, stops/veh 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Approach Data Left Side Right Side Left Side Right Side Left Side Right Side Left Side Right Side
Parking present? No No No No No No No No
Parking maneuvers, maneuvers/h 0 0 0 0 0 0 0 0
Bus stopping rate, buses/h 0 0 0 0
Approach grade, % 0 0 0 0 0 0 0 0 0 0 0 0
Detection Data
Stop line detector presence Presence Presence Presence Presence Presence Presence Presence Presence Presence Presence Presence No det.
Stop line detector length, ft 40 40 40 40 40 40 40 40 40 40 40 40
All signalized intersection approaches have a 200-ft left-turn bay and two
through lanes. The east–west approaches have a 200-ft right-turn lane. The
north–south approaches have a shared through and right-turn lane. Many of the
geometric and traffic characteristics shown in the exhibit are needed to compute
the saturation flow rate with the procedure described in Section 3 of Chapter 19.
The platoon ratio is entered for all movements associated with an external
approach to the segment. The eastbound through movement at Signalized
Intersection 1 is known to be coordinated with the upstream intersection so that
favorable progression occurs, as described by a platoon ratio of 1.333. The
westbound through movement at Signalized Intersection 2 is also coordinated
with its upstream intersection, and arrivals are described by a platoon ratio of
1.33. Arrivals to all other movements are characterized as “random” and are
described with a platoon ratio of 1.00. The movements for the westbound
approach at Signalized Intersection 1 (and eastbound approach at Signalized
Intersection 2) are internal movements, so a platoon ratio (and upstream filtering
factor) is not entered for them. More accurate values are computed during
subsequent iterations by using a procedure provided in the methodology.
The speed limit on the segment and on the cross-street approaches is 35 mi/h.
With a couple of exceptions, detection is located just upstream of the stop line in
each traffic lane at the two signalized intersections. A 40-ft detection zone is used
in each instance. The exceptions are the traffic lanes serving the major-street

through movement at each intersection. There is no detection for these

movements because they are not actuated.
The geometric conditions that describe the segment are shown in Exhibit 30-
30. These data are used to compute the free-flow speed for the segment.
Segment Data Worksheet Exhibit 30-30

Input Data Example Problem 1: Segment
EB WB Data
Basic Segment Data
Number of through lanes that extend the length of the segment: 2 2
Speed limit, mph 35 35
Segment Length Data
Length of segment (measured stopline to stopline), ft 1800 1800
Width of upstream signalized intersection, ft 50 50
Adjusted segment length, ft 1750 1750
Length of segment with a restrictive median (e.g, raised-curb), ft 0 0
Length of segment with a non-restrictive median (e.g, two-way left-turn lane), ft 0 0
Length of segment with no median, ft 1750 1750
Percentage of segment length with restrictive median, % 0 0
Access Data
Percentage of street with curb on right-hand side (in direction of travel), % 70 70
Number of access points on right-hand side of street (in direction of travel) 4 4
Percentage of street with on-street parking on right-hand side (in direction of travel),% 0 0
Other Delay Data
Mid-segment delay, s/veh 0 0
The traffic and lane assignment data for the two access point intersections
are shown in Exhibit 30-31. The movement numbers follow the numbering
convention shown in Exhibit 20-1 of Chapter 20, Two-Way STOP-Controlled
Intersections. There are no turn bays on the segment at the two access point
intersections.
Access Point Input Data
Access Approach Eastbound Westbound Northbound Southbound
Exhibit 30-31
Point
Location,ft
Movement L T R L T R L T R L T R Example Problem 1: Access
Movement number 1 2 3 4 5 6 7 8 9 10 11 12
600 Volume, veh/h 80 1,050 100 80 1,050 100 80 0 100 80 0 100 Point Data
West end Lanes 0 2 0 0 2 0 1 0 1 1 0 1
1200 Volume, veh/h 80 1,050 100 80 1,050 100 80 0 100 80 0 100
Lanes 0 2 0 0 2 0 1 0 1 1 0 1
Outline of Solution
Movement-Based Data
Exhibit 30-32 provides a summary of the analysis of the individual traffic
movements at Signalized Intersection 1.
INTERSECTION 1 EB EB EB WB WB WB NB NB NB SB SB SB
L T R L T R L T R L T R Exhibit 30-32
Movement:
Volume, veh/h
5
200
2
1,000
12
10
1
194
6
968
16
10
3
100
8
500
18
50
7
100
4
500
14
50
Example Problem 1:
Initial Queue, veh 0 0 0 0 0 0 0 0 0 0 0 0 Movement-Based Output Data
Ped-Bike Adj. Factor (A_pbT) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Parking, Bus Adj. Factors (f_bb x f_p) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Downstream Lane Blockage Factor (f_ms) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Spillback Factor (f_sp) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Adjusted Sat. Flow Rate, veh/h/ln 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900
Lanes 1 2 1 1 2 1 1 2 0 1 2 0
Lane Assignment L T R L T R L TR n.a. L TR n.a.
Capacity, veh/h 236 1,856 789 233 1,848 785 217 617 61 217 617 61
Discharge Volume, veh/h 0 1,000 0 0 0 0 0 0 50 100 0 0
Proportion Arriving On Green 0.131 0.651 0.488 0.045 0.493 0.501 0.061 0.181 0.181 0.061 0.181 0.181
Approach Volume, veh/h 1,210 1,172 650 650
Approach Delay, s/veh 18.0 23.4 39.7 39.7
Approach Stop Rate, stops/veh 0.442 0.617 0.831 0.831
With the exception of Initial Queue, Lanes, and Lane Assignment, the
variables listed in Exhibit 30-32 have computed values. The volumes shown for
the eastbound (EB), northbound (NB), and southbound (SB) movements are
identical to the input volumes. The westbound (WB) volumes were computed
from the input volumes during Step 1: Determine Traffic Demand Adjustments.

Specifically, they were reduced because the input westbound volume for this
intersection exceeded the volume departing the upstream access point
intersection (i.e., AP1).
Four factors are listed in the top half of Exhibit 30-32. These factors represent
saturation flow rate adjustment factors. Their values are dependent on signal
timing or lane volume, quantities that are computed during the iterative
convergence loop (identified in the motorized vehicle methodology framework
shown in Exhibit 18-8). As a result, the value of each factor also converges within
this loop. The procedure for calculating the pedestrian–bicycle adjustment factor
is described in Section 2 of Chapter 31. The procedure for calculating the
parking–bus adjustment factor is described in Section 3 of Chapter 19. The
procedure for calculating the downstream lane blockage (due to midsegment
lane restriction) factor is described in Section 3 of this chapter. The methodology
for calculating the spillback factor is described in Chapter 29.
Capacity for a movement is computed by using the movement volume
proportion in each approach lane group, lane group saturation flow rate, and
corresponding phase duration. This variable represents the capacity of the
movement, regardless of whether it is served in an exclusive lane or a shared
lane. If the movement is served in a shared lane, the movement capacity
represents the portion of the lane group capacity available to the movement, as
distributed in proportion to the volume of the movements served by the
associated lane group.
Discharge volume is computed for movements that enter a segment during
Step 1: Determine Traffic Demand Adjustments. At Signalized Intersection 1, the
movements entering the segment are the eastbound through movement, the
northbound right-turn movement, and the southbound left-turn movement. A
value of 0.0 veh/h is shown for all other movements, which indicates that they
are not relevant to this calculation. If volume exceeds capacity for any given
movement, the discharge volume is set equal to the capacity. Otherwise, the
discharge volume is equal to the movement volume.
The proportion arriving during green P is computed for internal movements
during Step 3: Determine the Proportion Arriving During Green. In contrast, it is
computed from the input platoon ratio for external movements.
The last three rows in Exhibit 30-32 represent summary statistics for the
approach. The approach volume is the sum of the three movement volumes.
Approach delay and approach stop rate are computed as volume-weighted
averages for the lane groups served on an intersection approach.
Timer-Based Phase Data

Exhibit 30-33 provides a summary of the output data for Signalized
Intersection 1 from a signal controller perspective. The controller has eight
timing functions (or timers), with Timers 1 to 4 representing Ring 1 and Timers 5
to 8 representing Ring 2. The ring structure and phase assignments are described
in Section 2 of Chapter 19. Timers 1, 2, 5, and 6 are used to control the east–west
traffic movements on the segment. Timers 3, 4, 7, and 8 are used to control the
north–south movements that cross the segment.

Timer Data Exhibit 30-33

Timer: 1 2 3 4 5 6 7 8
WB EB NB SB EB WB SB NB Example Problem 1: Timer-
L T.R L T.T+R L T.R L T.T+R Based Phase Output Data
Assigned Phase 1 2 3 4 5 6 7 8
Phase Duration (G+Y+Rc), s 15.90 52.84 9.13 22.13 16.10 52.63 9.13 22.13
Change Period (Y+Rc), s 3.00 4.00 3.00 4.00 3.00 4.00 3.00 4.00
Phase Start Time, s 35.27 51.16 4.00 13.14 35.27 51.37 4.00 13.14
Phase End Time, s 51.16 4.00 13.13 35.27 51.37 4.00 13.13 35.27
Max. Allowable Headway (MAH), s 3.13 0.00 3.13 3.06 3.13 0.00 3.13 3.06
Equivalent Maximum Green (Gmax), s 30.73 0.00 17.00 31.87 30.73 0.00 17.00 31.87
Max. Queue Clearance Time (g_c+l1), s 12.646 0.000 6.442 16.165 12.829 0.000 6.442 16.165
Green Extension Time (g_e), s 0.311 0.000 0.099 1.968 0.322 0.000 0.099 1.968
Probability of Phase Call (p_c) 0.995 0.000 0.938 1.000 0.996 0.000 0.938 1.000
Probability of Max Out (p_x) 0.000 0.000 0.000 0.016 0.000 0.000 0.000 0.016
Cycle Length, s: 100
The timing function construct is essential to the modeling of a ring-based

signal controller. Timers always occur in the same numeric sequence (i.e., 1 then 2
then 3 then 4 in Ring 1; 5 then 6 then 7 then 8 in Ring 2). The practice of
associating movements with phases (e.g., the major-street through movement
with Phase 2), coupled with the occasional need for lagging left-turn phases and
split phasing, creates the situation in which phases do not always time in
sequence. For example, with a lagging left-turn phase sequence, major-street
through Phase 2 times first and then major-street left-turn Phase 1 times second.
The modern controller accommodates the assignment of phases to timing
functions by allowing the ring structure to be redefined manually or by time-of-
day settings. Specification of this structure is automated in the computational
engine by the assignment of phases to timers.
The methodology is based on modeling timers, not on directly modeling
movements or phases. The methodology converts movement and phase input
data into timer input data. It then models controller response to these inputs and
computes timer duration and related performance measures.
The two signalized intersections in this example problem have lead–lead left-
turn sequences. Hence, the timer number is equal to the phase number (e.g., the
westbound movement is associated with Phase 1, which is assigned to Timer 1).
The phase duration shown in Exhibit 30-33 is the estimated average phase
duration during the analysis period. It represents the sum of the green, yellow
change, and red clearance intervals. For Timer 2 (i.e., Phase 2), the average green
interval duration can be computed as 48.84 s (= 52.84 – 4.00).
The phase start time is the time the timer (and phase) starts, relative to
system time 0.0. For Phase 2, the start time is 51.16 s. The end of the green
interval associated with this phase is 100.0 s (= 51.16 + 48.84). This time is equal to
the cycle length, so the end of green actually occurs at 0.0 s. This result is
expected because Phase 2 is the coordinated phase and the offset to the end of
Phase 2 (relative to system time 0.0) was input as 0.0 s.
The phase end time is the time the timer (and phase) ends relative to system
time 0.0. For Phase 2, the end of the green interval occurs at 0.0 s and the end of
the phase occurs 4.0 s later (i.e., the change period duration).
The remaining variables in Exhibit 30-33 apply to the noncoordinated phases
(i.e., the actuated phases). These variables describe the phase timing and
operation. They are described in more detail in Section 2 of Chapter 19 and
Section 2 of Chapter 31.

Timer-Based Movement Data

Exhibit 30-34 summarizes the output for Signalized Intersection 1 as it relates
to the movements assigned to each timer. Separate sections of output are shown
in the exhibit for the left-turn, through, and right-turn movements. The assigned
movement row identifies the movement (previously identified in Exhibit 30-32)
assigned to each timer.
The saturation flow rate shown in Exhibit 30-34 is the saturation flow rate for
the movement. The procedure for calculating these rates is described in Section 3
of Chapter 19 and Section 3 of Chapter 31. In general, the rate for a movement is
the same as for a lane group when the lane group serves one movement. The rate
is split between the movements when the lane group is shared by two or more
movements.
Exhibit 30-34 Timer Data

Timer: 1 2 3 4 5 6 7 8
Example Problem 1: Timer- WB EB NB SB EB WB SB NB
Based Movement Output Data L T.R L T.T+R L T.R L T.T+R
Left-Turn Movement Data
Assigned Movement 1 3 5 7
Mvmt. Sat Flow, veh/h 1,805.00 1,805.00 1,805.00 1,805.00
Through Movement Data
Mvmt. Sat Flow, veh/h 3,800.00 3,401.19 3,800.00 3,401.19
Right-Turn Movement Data
Mvmt. Sat Flow, veh/h 1,615.00 338.99 1,615.00 338.99
Timer-Based Lane Group Data

The motorized vehicle methodology described in Chapter 19 computes a
variety of output statistics that portray the operation of each intersection lane
group. The example problem in Chapter 19 illustrates these statistics and discusses
their interpretation. The output data for the individual lane groups are not
repeated in this chapter. Instead, the focus of the remaining discussion is on the
access point output and the performance measures computed for the two through
movements on the segment (i.e., eastbound through and westbound through).
Access Point Data

Exhibit 30-35 illustrates the output statistics for the two access point
intersections located on the segment. The first six rows listed in the exhibit
correspond to Access Point Intersection 1 (AP1), and the second six rows
correspond to Access Point Intersection 2 (AP2). Additional sets of six rows
would be provided in this table if additional access point intersections were
evaluated.
Exhibit 30-35 Access Point Data

Segment 1
EB
L
EB
T
EB
R
WB
L
WB
T
WB
R
NB
L
NB
T
NB
R
SB
L
SB
T
SB
R
Example Problem 1: Movement:
Access Point Intersection No. 1
1 2 3 4 5 6 7 8 9 10 11 12
Movement-Based Access Point 1: Volume, veh/h 74.80 981.71 93.50 75.56 991.70 94.45 80.00 0.00 100.00 80.00 0.00 100.00
1: Lanes 0 2 0 0 2 0 1 0 1 1 0 1
Output Data 1: Proportion time blocked 0.150 0.160 0.250 0.250 0.160 0.250 0.250 0.150
1: Delay to through vehicles, s/veh 0.193 0.194
1: Prob. inside lane blocked by left 0.115 0.115
1: Dist. from West/South signal, ft 600
2: Volume, veh/h 75.56 991.70 94.45 74.80 981.71 93.50 80.00 0.00 100.00 80.00 0.00 100.00
2: Lanes 0 2 0 0 2 0 1 0 1 1 0 1
2: Proportion time blocked 0.160 0.150 0.250 0.250 0.150 0.250 0.250 0.160
2: Dist. from West/South signal, ft 1,200

The eastbound and westbound volumes listed in Exhibit 30-35 are not equal
to the input volumes. These volumes were adjusted during Step 1: Determine
Traffic Demand Adjustments so that they equal the volume discharging from the
upstream intersection. This routine achieves balance between all junction pairs
(e.g., between Signalized Intersection 1 and Access Point Intersection 1, between
Access Point Intersection 1 and Access Point Intersection 2, and so forth).
The “proportion of time blocked” is computed during Step 3: Determine the
Proportion Arriving During Green. It represents the proportion of time during
the cycle that the associated access point movement is blocked by the presence of
a platoon passing through the intersection. For major-street left turns, the
platoon of concern approaches from the opposing direction. For the minor-street
left turn, platoons can approach from either direction and can combine to block
this left turn for extended time periods. This trend can be seen by comparing the
proportion of time blocked for the eastbound (major-street) left turn (i.e., 0.15)
with that for the northbound (minor-street) left turn (i.e., 0.25) at Access Point
Intersection 1.
The “delay to through vehicles” is computed during Step 2: Determine
Running Time. It represents the sum of the delay due to vehicles turning left from
the major street and the delay due to vehicles turning right from the major street.
This delay tends to be small compared with typical signalized intersection delay
values. But it can reduce overall travel speed if there are several high-volume
access points on a street and only one or two through lanes in each direction of
travel.
The “probability of the inside through lane being blocked” is also computed
during Step 2: Determine Running Time as part of the delay-to-through-vehicles
procedure. This variable indicates the probability that the left-turn bay at an
access point will overflow into the inside through lane on the street segment.
Hence, it indicates the potential for a through vehicle to be delayed by a left-turn
maneuver. The segment being evaluated has an undivided cross section, and no
left-turn bays are provided at the access point intersections. In this situation, the
probability of overflow is 0.115, indicating that the inside lane is blocked about
11.5% of the time.
Results
Exhibit 30-36 summarizes the performance measures for the segment. Also
shown are the results from the spillback check conducted during Step 1:
Determine Traffic Demand Adjustments. The movements indicated in the
column heading are those exiting the segment at a boundary intersection. Thus,
the westbound movements on Segment 1 are those occurring at Signalized
Intersection 1. Similarly, the eastbound movements on Segment 1 are those
occurring at Signalized Intersection 2.

Exhibit 30-36 Segment Summary EB EB EB WB WB WB

Example Problem 1: L T R L T R
Performance Measure Seg.No. Movement: 5 2 12 1 6 16
Summary 1 Bay/Lane Spillback Time, h never never never never never never
1 ShrdLane Spillback Time, h never never never never never never
1 Base Free-Flow Speed, mph 40.78 40.78
1 Running Time, s 33.54 33.54
1 Running Speed, mph 36.59 36.59
1 Through Delay, s/veh 18.310 18.310
1 Travel Speed, mph 23.67 23.67
1 Stop Rate, stops/veh 0.547 0.547
1 Spatial Stop Rate, stops/mi 1.61 1.61
1 Through vol/cap ratio 0.52 0.52
1 Level of Service C C
1 Proportion Left Lanes 0.33 0.33
1 Auto. Traveler Perception Score 2.53 2.53
SPILLBACK TIME, h: never
The spillback check procedure computes the time of spillback for each of the
internal movements. For turn movements, the bay/lane spillback time is the time
before the turn bay overflows. For through movements, the bay/lane spillback
time is the time before the through lane overflows due only to through demand.
If a turn bay exists and it overflows, the turn volume will queue in the adjacent
through lane. For this scenario, the shared lane spillback time is computed and
used instead of the bay/lane spillback time. If several movements experience
spillback, the time of first spillback is reported at the bottom of Exhibit 30-36.
The output data for the two through movements are listed in Exhibit 30-36,
starting with the third row. The base free-flow speed (FFS) and running time
statistics are computed during Step 2: Determine Running Time. The through
delay listed is computed during Step 5: Determine Through Control Delay. It is a
weighted average delay for the lane groups serving through movements at the
downstream boundary intersection. The weight used in this average is the
volume of through vehicles served by the lane group.
The base free-flow speed is 40.78 mi/h. By interpolating this value between
those in Exhibit 18-1, the threshold travel speeds for LOS A, B, C, D, and E are as
follows: >32.6, >27.5, >20.5, >16.3, and >12.3 mi/h, respectively. Thus, the travel
speed for the eastbound direction of 23.67 mi/h corresponds to LOS C. The same
conclusion is reached for the westbound travel direction.
Each travel direction has one left-turn bay and three intersections. Thus, the
proportion of intersections with left-turn lanes is 0.33. This proportion is used in
Step 10: Determine Automobile Traveler Perception Score to compute the score
of 2.53, which suggests that most automobile travelers would find segment
service to be very good.
EXAMPLE PROBLEM 2: PEDESTRIAN LOS

The Segment
The sidewalk of interest is located along a 1,320-ft urban street segment. The
segment is part of a collector street located near a community college. It is shown
in Exhibit 30-37. Sidewalk is only shown for the south side of the segment for the
convenience of illustration. It also exists on the north side of the segment.

Exhibit 30-37
Geometry
The Question
What is the pedestrian LOS for the sidewalk on the south side of the segment?
The Facts
The geometric details of the sidewalk and street cross section are shown in
Exhibit 30-37. Both boundary intersections are signalized. Crossing the segment
at uncontrolled midsegment locations is legal. The following additional
information is known about the sidewalk and street segment:
Traffic characteristics:
Midsegment flow rate in eastbound direction: 940 veh/h
Pedestrian flow rate in south sidewalk (walking in both directions): 2,000 p/h
Proportion of on-street parking occupied during analysis period: 0.20
Geometric characteristics:
Outside shoulder width: none
Parking lane width: 9.5 ft
Cross section has raised curb along outside edge of roadway
Effective width of fixed objects on sidewalk: 0.0 ft (no objects present)
Presence of trees, bushes, or other vertical objects in buffer: No
Other data:
Pedestrians can cross the segment legally and do so somewhat uniformly
along its length
Proportion of sidewalk adjacent to window display: 0.0
Proportion of sidewalk adjacent to building face: 0.0
Proportion of sidewalk adjacent to fence: 0.50
Performance measures obtained from supporting methodologies:
Motorized vehicle running speed: 33 mi/h
Pedestrian delay when walking parallel to the segment: 40 s/p
Pedestrian delay when crossing the segment at the nearest signal-controlled
crossing: 80 s/p
Pedestrian delay crossing the segment at an uncontrolled midsegment
location: 740 s/p
Pedestrian LOS score for the downstream intersection: 3.60

Outline of Solution
First, the pedestrian space will be calculated for the sidewalk. This measure
will then be compared with the qualitative descriptions of pedestrian space listed
in Exhibit 18-15. Next, the pedestrian travel speed along the sidewalk will be
calculated. Finally, LOS for the segment will be determined by using the
computed pedestrian LOS score and the pedestrian space variables.
Computational Steps
Step 1: Determine Free-Flow Walking Speed
The average free-flow walking speed is estimated to be 4.4 ft/s on the basis of
the guidance provided.
Step 2: Determine Average Pedestrian Space

The shy distance on the inside of the sidewalk is computed with Equation 18-24.
𝑊𝑠,𝑖 = max(𝑊𝑏𝑢𝑓 , 1.5)
𝑊𝑠,𝑖 = max (5.0, 1.5)
𝑊𝑠,𝑖 = 5.0 ft
The shy distance on the outside of the sidewalk is computed with Equation 18-25.
𝑊𝑠,𝑜 = 3.0 𝑝window + 2.0 𝑝building + 1.5 𝑝fence
𝑊𝑠,𝑜 = 3.0(0.0) + 2.0(0.0) + 1.5(0.50)
𝑊𝑠,𝑜 = 0.75 ft
There are no fixed objects present on the sidewalk, so the adjusted fixed-
object effective widths for the inside and outside of the sidewalk are both equal
to 0.0 ft. The effective sidewalk width is computed with Equation 18-23.
𝑊𝐸 = 𝑊𝑇 − 𝑊𝑂,𝑖 − 𝑊𝑂,𝑜 − 𝑊𝑠,𝑖 − 𝑊𝑠,𝑜 ≥ 0.0
𝑊𝐸 = 10 − 0.0 − 0.0 − 5.0 − 0.75
𝑊𝐸 = 4.25 ft
The pedestrian flow per unit width of sidewalk is computed with Equation
18-28 for the subject sidewalk.
𝑣𝑝𝑒𝑑
𝑣𝑝 =
60 𝑊𝐸
2,000
𝑣𝑝 =
60(4.25)
𝑣𝑝 = 7.84 p/ft/min
The average walking speed Sp is computed with Equation 18-29.
𝑆𝑝 = (1 − 0.00078 𝑣𝑝2 ) 𝑆𝑝𝑓 ≥ 0.5 𝑆𝑝𝑓
𝑆𝑝 = [1 − 0.00078(7.84)2 ](4.4)
𝑆𝑝 = 4.19 ft/s

Finally, Equation 18-30 is used to compute average pedestrian space.

𝑆𝑝
𝐴𝑝 = 60
𝑣𝑝
4.19
𝐴𝑝 = 60
7.84
𝐴𝑝 = 32.0 ft 2 /p
The pedestrian space can be compared with the ranges provided in Exhibit
18-15 to make some judgments about the performance of the subject intersection
corner. The criteria for platoon flow are considered applicable given the
influence of the signalized intersections. According to the qualitative
descriptions provided in this exhibit, walking speed will be restricted, as will the
ability to pass slower pedestrians.
Step 3: Determine Pedestrian Delay at Intersection

The pedestrian methodology in Chapter 19, Signalized Intersections, was
used to estimate two pedestrian delay values. One is the delay at the boundary
intersection experienced by a pedestrian walking parallel to segment dpp. This
delay was computed to be 40 s/p. The second is the delay experienced by a
pedestrian crossing the segment at the nearest signal-controlled crossing dpc. This
delay was computed to be 80 s/p.
The pedestrian methodology in Chapter 20, Two-Way STOP-Controlled
Intersections, was used to estimate the delay incurred while waiting for an
acceptable gap in traffic dpw at a midsegment location. As given in The Facts, this
delay was computed to be 740 s/p.
Step 4: Determine Pedestrian Travel Speed

The pedestrian travel speed is computed with Equation 18-31.
𝐿
𝑆𝑇𝑝,𝑠𝑒𝑔 =
𝐿
+ 𝑑𝑝𝑝
𝑆𝑝
1,320
𝑆𝑇𝑝,𝑠𝑒𝑔 =
1,320
4.19 + 40
𝑆𝑇𝑝,𝑠𝑒𝑔 = 3.72 ft/s
This walking speed is slightly less than 4.0 ft/s and is considered acceptable,
but a higher speed is desirable.
Step 5: Determine Pedestrian LOS Score for Intersection

The pedestrian methodology in Chapter 19 was used to determine the
pedestrian LOS score for the downstream boundary intersection Ip,int. As given in
The Facts, it was computed to be 3.60.

Step 6: Determine Pedestrian LOS Score for Link

The pedestrian LOS score for the link is computed from three factors.
However, before these factors can be calculated, several cross-section variables
need to be adjusted and several coefficients need to be calculated. These
variables and coefficients are calculated first. Then, the three factors are
computed. Finally, they are combined to determine the desired score.
The midsegment demand flow rate is greater than 160 veh/h. The street cross
section is curbed but there is no shoulder, so the adjusted width of paved outside
shoulder Wos* is 0.0 ft. Therefore, the effective total width of the outside through
lane, bicycle lane, and shoulder Wv is computed as
𝑊𝑣 = 𝑊𝑜𝑙 + 𝑊𝑏𝑙 + 𝑊𝑜𝑠∗ + 𝑊𝑝𝑘
𝑊𝑣 = 12 + 5 + 0 + 9.5
𝑊𝑣 = 26.5 ft
Because the proportion of occupied on-street parking is less than 0.25 and the
sum of the bicycle lane and parking lane widths exceeds 10.0 ft, the effective
width of the combined bicycle lane and parking lane Wl is set to 10.0 ft.
The adjusted available sidewalk width WaA is computed as
𝑊𝑎𝐴 = min (𝑊𝑇 − 𝑊𝑏𝑢𝑓 , 10)
𝑊𝑎𝐴 = min (10 − 5, 10)
𝑊𝑎𝐴 = 5 ft
The sidewalk width coefficient fsw is computed as
𝑓𝑠𝑤 = 6.0 − 0.3 𝑊𝑎𝐴
𝑓𝑠𝑤 = 6.0 − 0.3(5.0)
𝑓𝑠𝑤 = 4.5 ft
The buffer area coefficient fb is equal to 1.0 because there is no continuous
barrier at least 3.0 ft high located in the buffer area.
The motorized vehicle methodology described in Section 3 of Chapter 18 was
used to determine the motorized vehicle running speed SR for the subject
segment. This speed was computed to be 33.0 mi/h.
The cross-section adjustment factor is computed with Equation 18-33.
𝐹𝑤 = −1.2276 ln (𝑊𝑣 + 0.5 𝑊𝑙 + 50 𝑝𝑝𝑘 + 𝑊𝑏𝑢𝑓 𝑓𝑏 + 𝑊𝑎𝐴 𝑓𝑠𝑤 )
𝐹𝑤 = −1.2276 ln [26.5 + 0.5(10) + 50(0.20) + 5.0(1.0) + 5.0(4.5)]
𝐹𝑤 = −5.20
The motorized vehicle volume adjustment factor is computed with Equation
18-34.
𝑣𝑚
𝐹𝑣 = 0.0091
4 𝑁𝑡ℎ
940
𝐹𝑣 = 0.0091
4(2)

𝐹𝑣 = 1.07
The motorized vehicle speed adjustment factor is computed with Equation
18-35.
𝑆𝑅 2
𝐹𝑠 = 4 ( )
100
33.0 2
𝐹𝑠 = 4 ( )
100
𝐹𝑠 = 0.44
Finally, the pedestrian LOS score for the link Ip,link is calculated with Equation
18-32.
𝐼𝑝,link = 6.0468 + 𝐹𝑤 + 𝐹𝑣 + 𝐹𝑠
𝐼𝑝,link = 6.0468 + (−5.20) + 1.07 + 0.44
𝐼𝑝,link = 2.35
Step 7: Determine Link LOS

The pedestrian LOS for the link is determined by using the pedestrian LOS
score from Step 6. This score is compared with the link-based pedestrian LOS
thresholds on the right side of Exhibit 18-2 to determine that the LOS for the
specified direction of travel along the subject link is B.
Step 8: Determine Roadway Crossing Difficulty Factor

Crossings occur somewhat uniformly along the length of the segment, and
the segment is bounded by two signalized intersections. Thus, the distance Dc is
assumed to equal one-third of the segment length, or 440 ft (= 1,320/3), and the
diversion distance Dd is computed as 880 ft (= 2 × 440 ft).
The delay incurred due to diversion is calculated by using Equation 18-37.
𝐷𝑑
𝑑𝑝𝑑,𝐿𝑂𝑆 = 0.084 + 𝑑𝑝𝑐
𝑆𝑝
880
𝑑𝑝𝑑,𝐿𝑂𝑆 = 0.084 + 80
4.19
𝑑𝑝𝑑,𝐿𝑂𝑆 = 98 s/p
Both the perceived diversion delay dpd,LOS of 98 s/p and the delay waiting for
an adequate gap dpw of 740 s/p are greater than 70 s and therefore, from Exhibit
18-20A, the LOS scores associated with each delay equal 6. Therefore, based on
Equation 18-38, the midsegment crossing LOS score is 6:
𝐼𝑝,𝑚𝑥 = min[𝐼𝑝𝑤 , 𝐼𝑝𝑑 , 6] = min[6, 6, 6] = 6
Step 9: Determine Pedestrian LOS Score for Segment

Equation 18-39 is used to determine the pedestrian LOS score for the
segment. The proportion of pedestrian demand desiring to cross midblock pmx is
assumed to be the default value of 0.35.

3 3 1/3
( 𝐼𝑝,𝑙𝑖𝑛𝑘 [1 − 𝑝𝑚𝑥 ] + 𝐼𝑝,𝑚𝑥 𝑝𝑚𝑥 ) 𝐿/𝑆𝑝 + (𝐼𝑝,𝑖𝑛𝑡 ) 𝑑𝑝𝑝
𝐼𝑝,𝑠𝑒𝑔 =[ ]
𝐿/𝑆𝑝 + 𝑑𝑝𝑝
1/3
1,320
( 2.35 [1 − 0.35] + 6.0 × 0.35)3 ( (3.60)3 × 40
𝐼𝑝,𝑠𝑒𝑔 = [ 4.19 ) + ]
1,320
( 4.19 ) + 40
𝐼𝑝,𝑠𝑒𝑔 = 3.62
Step 10: Determine Segment LOS

The pedestrian LOS for the segment is determined by using the pedestrian
LOS score from Step 9 and the average pedestrian space from Step 2. These two
performance measures are compared with their respective thresholds on the left
side of Exhibit 18-2 to determine that the LOS for the specified direction of travel
along the subject segment is D.
EXAMPLE PROBLEM 3: BICYCLE LOS

The Segment
The bicycle lane of interest is located along a 1,320-ft urban street segment. The
segment is part of a collector street located near a community college. The bicycle
lane is provided for the eastbound direction of travel, as shown in Exhibit 30-38.
Exhibit 30-38
Geometry
The Question
What is the bicycle LOS for the eastbound bicycle lane?
The Facts
The geometric details of the street cross section are shown in Exhibit 30-38.
Both boundary intersections are signalized. The following additional information
is known about the street segment:
Traffic characteristics:
Midsegment flow rate in eastbound direction: 940 veh/h
Percent heavy vehicles: 8.0%
Proportion of on-street parking occupied during analysis period: 0.20
Geometric characteristics:
Outside shoulder width: none

Parking lane width: 9.5 ft

Median type: undivided
Cross section has raised curb along the outside edge of the roadway
Number of access point approaches on right side of segment in subject travel
direction: 3
Other data:
Pavement condition rating: 2.0
Bicycle control delay: 40 s/bicycle
Bicycle LOS score for the downstream intersection: 0.08
Outline of Solution
First, the bicycle delay at the boundary intersection will be computed. This
delay will then be used to compute the bicycle travel speed. Next, a bicycle LOS
score will be computed for the link. It will then be combined with a similar score
for the boundary intersection and used to compute the bicycle LOS score for the
segment. Finally, LOS for the segment will be determined by using the computed
score and the thresholds in Exhibit 18-3.
Computational Steps
Step 1: Determine Bicycle Running Speed
The average bicycle running speed Sb could not be determined from field
data. Therefore, it was estimated to be 15 mi/h on the basis of the guidance provided.
Step 2: Determine Bicycle Delay at Intersection

The motorized vehicle methodology in Chapter 19, Signalized Intersections,
was used to estimate the bicycle delay at the boundary intersection db. This delay
was computed to be 40.0 s/bicycle.
Step 3: Determine Bicycle Travel Speed

The segment running time of through bicycles is computed as
3,600 𝐿
𝑡𝑅𝑏 =
5,280 𝑆𝑏
3,600(1,320)
𝑡𝑅𝑏 =
5,280(15)
𝑡𝑅𝑏 = 60.0 s
The average bicycle travel speed is computed with Equation 18-40.
3,600 𝐿
𝑆𝑇𝑏,𝑠𝑒𝑔 =
5,280 (𝑡𝑅𝑏 + 𝑑𝑏 )
3,600(1,320)
𝑆𝑇𝑏,𝑠𝑒𝑔 =
5,280 (60.0 + 40.0)
𝑆𝑇𝑏,𝑠𝑒𝑔 = 9.0 mi/h

This travel speed is adequate, but a speed of 10 mi/h or more is considered

desirable.
Step 4: Determine Bicycle LOS Score for Intersection

The bicycle methodology in Chapter 19 was used to determine the bicycle
LOS score for the boundary intersection Ib,int. It was computed to be 0.08.
Step 5: Determine Bicycle LOS Score for Link

The bicycle LOS score is computed from four factors. However, before these
factors can be calculated, several cross-section variables need to be adjusted.
These variables are calculated first, and then the four factors are computed.
Finally, they are combined to determine the desired score.
The street cross section is curbed but there is no shoulder, so the adjusted
width of the paved outside shoulder Wos* is 0.0 ft. Therefore, the total width of the
outside through lane, bicycle lane, and paved shoulder Wt is computed as
𝑊𝑡 = 𝑊𝑜𝑙 + 𝑊𝑏𝑙 + 𝑊𝑜𝑠∗
𝑊𝑡 = 12 + 5 + 0
𝑊𝑡 = 17 ft
The variable Wt does not include the width of the parking lane in this
instance because the proportion of occupied on-street parking exceeds 0.0.
The total width of shoulder, bicycle lane, and parking lane Wl is computed as
𝑊𝑙 = 𝑊𝑏𝑙 + 𝑊𝑜𝑠∗ + 𝑊𝑝𝑘
𝑊𝑙 = 5 + 0 + 9.5
𝑊𝑙 = 14.5 ft
The midsegment demand flow rate is greater than 160 veh/h. Therefore, the
effective total width of the outside through lane, bicycle lane, and shoulder as a
function of traffic volume Wv is equal to Wt.
The total width of shoulder, bicycle lane, and parking lane Wl exceeds 4.0 ft.
Therefore, the effective width of the outside through lane is computed as
𝑊𝑒 = 𝑊𝑣 + 𝑊𝑙 − 20 𝑝𝑝𝑘 ≥ 0.0
𝑊𝑒 = 17 + 14.5 − 20(0.20) ≥ 0.0
𝑊𝑒 = 27.5 ft
The percent heavy vehicles is less than 50%, so the adjusted percent heavy
vehicles PHVa is equal to the input percent heavy vehicles PHV of 8.0%.
The motorized vehicle methodology described in Section 3 of Chapter 18 was
used to determine the motorized vehicle running speed SR for the subject
segment. This speed was computed to be 33.0 mi/h, which exceeds 21 mi/h.
Therefore, the adjusted motorized vehicle speed SRa is also equal to 33.0 mi/h.
The midsegment demand flow rate is greater than 8 veh/h (= 4 Nth), so the
adjusted midsegment demand flow rate vma is equal to the input demand flow
rate of 940 veh/h.

The cross-section adjustment factor is computed with Equation 18-42.

𝐹𝑤 = −0.005 𝑊𝑒2
𝐹𝑤 = −0.005(27.5)2
𝐹𝑤 = −3.78
The motorized vehicle volume adjustment factor comes from Equation 18-43.
𝑣𝑚𝑎
𝐹𝑣 = 0.507 ln ( )
4 𝑁𝑡ℎ
940
𝐹𝑣 = 0.507 ln ( )
4(2)
𝐹𝑣 = 2.42
The motorized vehicle speed adjustment factor is computed with Equation
18-44.
𝐹𝑆 = 0.199[1.1199 ln(𝑆𝑅𝑎 − 20) + 0.8103](1 + 0.1038𝑃𝐻𝑉𝑎 )2
𝐹𝑆 = 0.199[1.1199 ln(33.0 − 20) + 0.8103][1 + 0.1038(8.0)]2
𝐹𝑆 = 2.46
The pavement condition adjustment factor is computed with Equation 18-45.
7.066
𝐹𝑝 =
𝑃𝑐2
7.066
𝐹𝑝 =
(2.0)2
𝐹𝑝 = 1.77
Finally, the bicycle LOS score for the link Ib,link is calculated with Equation
18-41.
𝐼𝑏,link = 0.760 + 𝐹𝑤 + 𝐹𝑣 + 𝐹𝑆 + 𝐹𝑝
𝐼𝑏,link = 0.760 − 3.78 + 2.42 + 2.46 + 1.77
𝐼𝑏,link = 3.62
Step 6: Determine Link LOS

The bicycle LOS for the link is determined by using the bicycle LOS score
from Step 5. This score is compared with the link-based bicycle LOS thresholds
in Exhibit 18-3 to determine that the LOS for the specified direction of travel
along the subject link is D.
Step 7: Determine Bicycle LOS Score for Segment

The unsignalized conflicts factor is computed with Equation 18-47.
5,280 𝑁𝑎𝑝,𝑠
𝐹𝑐 = 0.035 ( − 20)
𝐿
5,280 (3)
𝐹𝑐 = 0.035 [ − 20]
1,320
𝐹𝑐 = −0.28

The bicycle LOS score for the segment is computed with Equation 18-46.
1
3 3 3
(𝐹𝑐 + 𝐼𝑏,link + 1) 𝑡𝑅,𝑏 + ( 𝐼𝑏,𝑖𝑛𝑡 + 1) 𝑑𝑏
𝐼𝑏,𝑠𝑒𝑔 = 0.75 [ ] + 0.125
𝑡𝑅,𝑏 + 𝑑𝑏
1
[(−0.28) + 3.62 + 1]3 (60) + (0.08 + 1)3(40) 3
𝐼𝑏,𝑠𝑒𝑔 = 0.75 [ ] + 0.125
60 + 40
𝐼𝑏,𝑠𝑒𝑔 = 2.88
Step 8: Determine Segment LOS

The bicycle LOS for the segment is determined by using the bicycle LOS
score from Step 7. This score is compared with the segment-based bicycle LOS
thresholds in Exhibit 18-3 to determine that the LOS for the specified direction of
travel along the subject segment is C.
EXAMPLE PROBLEM 4: TRANSIT LOS

The Segment
The transit route of interest travels east along a 1,320-ft urban street segment.
The segment is part of a collector street located near a community college. It is
shown in Exhibit 30-39. A bus stop is provided on the south side of the segment
for the subject route.
Exhibit 30-39
Geometry
The Question
What is the transit LOS for the eastbound bus route on the subject segment?
The Facts
The geometric details of the segment are shown in Exhibit 30-39. Both
boundary intersections are signalized. There is one stop in the segment for the
eastbound route. The following additional information is known about the bus
stop and street segment:
Transit characteristics:
Dwell time: 20.0 s
Transit frequency: 4 veh/h

Excess wait time data are not available for the stop, but the on-time
performance of the route (based on a standard of up to 5 min late being
considered “on time”) at the previous time point is known (92%)
Passenger load factor: 0.83 passengers/seat
Other data:
Area type: not in a central business district
g/C ratio at downstream boundary intersection: 0.4729
Cycle length: 140 s
The bus stop in the segment has a bench, but no shelter
Number of routes serving the segment: 1
The bus stop is accessed from the right-turn lane (i.e., the stop is off-line).
Buses are exempt from the requirement to turn right but have no other
traffic priority
Pedestrian LOS score for the link: 3.53
Through vehicle control delay at the downstream boundary intersection:
19.4 s/veh
Reentry delay: 16.17 s
Outline of Solution
First, the transit vehicle segment running time will be computed. Next, the
control delay at the boundary intersection will be obtained and used to compute
the transit vehicle segment travel speed. Then the transit wait–ride score will be
computed. This score will be combined with the pedestrian LOS score for the
link to compute the transit LOS score for the segment. Finally, LOS for the
segment will be determined by comparing the computed score with the
thresholds identified in Exhibit 18-3.
Computational Steps
Step 1: Determine Transit Vehicle Running Time
The transit vehicle running time is based on the segment running speed and
delay due to a transit vehicle stop. These components are calculated first, and
then running time is calculated.
Transit vehicle segment running speed can be computed with Equation 18-48.
61
𝑆𝑅𝑡 = min (𝑆𝑅 , )
1 + 𝑒 −1.00+(1,185 𝑁𝑡𝑠 /𝐿)
61
𝑆𝑅𝑡 = min (33.0, )
1+ 𝑒 −1.00+(1,185(1)/1,320)
𝑆𝑅𝑡 = 32.1 mi/h

The acceleration and deceleration rates are unknown, so they are assumed to
be 3.3 ft/s2 and 4.0 ft/s2, respectively, on the basis of data given in the Transit
Capacity and Quality of Service Manual (9).
The bus stop is located on the near side of a signalized intersection. From
Equation 18-50, the average proportion of bus stop acceleration–deceleration
delay not due to the intersection’s traffic control fad is equal to the g/C ratio for the
through movement in the bus’s direction of travel (in this case, eastbound). The
effective green time g is 66.21 s (calculated as the phase duration minus the
change period), and the cycle length is 140 s. Therefore, fad is 0.4729.
Equation 18-49 can now be used to compute the portion of bus stop delay
due to acceleration and deceleration.
5,280 𝑆𝑅𝑡 1 1
𝑑𝑎𝑑 = ( ) ( + ) 𝑓𝑎𝑑
3,600 2 𝑟𝑎𝑡 𝑟𝑑𝑡
5,280 32.1 1 1
𝑑𝑎𝑑 = ( )( + ) (0.4729)
3,600 2 3.3 4.0
𝑑𝑎𝑑 = 6.15 s
Equation 18-51 is used to compute the portion of bus stop delay due to
serving passengers. The input average dwell time of 20.0 s and an fdt value of
0.4729 are used in the equation, on the basis of the stop’s near-side location at a
traffic signal and the g/C ratio computed in a previous step. The fdt factor is used
to avoid double-counting the portion of passenger service time that occurs during
the signal’s red indication and is therefore included as part of control delay.
𝑑𝑝𝑠 = 𝑡𝑑 𝑓𝑑𝑡
𝑑𝑝𝑠 = (20.0)(0.4729)
𝑑𝑝𝑠 = 9.46 s
The bus stop is located in the right-turn lane; therefore, the bus is subject to
reentry delay on leaving the stop. On the basis of the guidance for reentry delay
for a near-side stop at a traffic signal, the reentry delay dre is equal to the queue
service time gs. This time is calculated to be 16.17 s by following the procedures
in Section 3 of Chapter 31, Signalized Intersections: Supplemental.
Equation 18-52 is used to compute the total delay due to the transit stop.
𝑑𝑡𝑠 = 𝑑𝑎𝑑 + 𝑑𝑝𝑠 + 𝑑𝑟𝑒
𝑑𝑡𝑠 = 6.15 + 9.46 + 16.17
𝑑𝑡𝑠 = 31.78 s
Equation 18-53 is used to compute transit vehicle running time on the basis
of the previously computed components.
𝑁𝑡𝑠
3,600 𝐿
𝑡𝑅𝑡 = + ∑ 𝑑𝑡𝑠,𝑖
5,280 𝑆𝑅𝑡
𝑖=1
3,600(1,320)
𝑡𝑅𝑡 = + 31.78
5,280(32.1)
𝑡𝑅𝑡 = 59.9 s

Step 2: Determine Delay at Intersection

The through delay dt at the boundary intersection is set equal to the through
vehicle control delay exiting the segment at this intersection. The latter delay is
19.4 s/veh. Thus, the through delay dt is equal to 19.4 s/veh.
Step 3: Determine Travel Speed

The average transit travel speed is computed with Equation 18-55.
3,600 𝐿
𝑆𝑇𝑡,𝑠𝑒𝑔 =
5,280 (𝑡𝑅𝑡 + 𝑑)
3,600(1,320)
𝑆𝑇𝑡,𝑠𝑒𝑔 =
5,280(59.9 + 19.4)
𝑆𝑇𝑡,𝑠𝑒𝑔 = 11.3 mi/h
Step 4: Determine Transit Wait–Ride Score

The wait–ride score is based on the headway factor and the perceived travel
time factor. Each of these components is calculated separately. The wait–ride
score is then calculated.
The input data indicate that there is one route on the segment, and its
frequency is 4 veh/h. The headway factor is computed with Equation 18-56.
𝐹ℎ = 4.00𝑒 −1.434/(𝑣𝑠 +0.001)
𝐹ℎ = 4.00𝑒 −1.434/(4+0.001)
𝐹ℎ = 2.80
The perceived travel time factor is based on several intermediate variables
that need to be calculated first. The first of these calculations is the amenity time
rate. It is calculated by using Equation 18-60. A default passenger trip length of
3.7 mi is used in the absence of other information.
1.3 𝑝𝑠ℎ + 0.2 𝑝𝑏𝑒
𝑇𝑎𝑡 =
𝐿𝑝𝑡
1.3(0.0) + 0.2(1.0)
𝑇𝑎𝑡 =
3.7
𝑇𝑎𝑡 = 0.054 min/mi
Since no information is available for actual excess wait time but on-time
performance information is available for the route, Equation 18-61 is used to
estimate excess wait time.
𝑡𝑒𝑥 = [𝑡𝑙𝑎𝑡𝑒 (1 − 𝑝𝑜𝑡 )]2
𝑡𝑒𝑥 = [5.0(1 − 0.92)]2
𝑡𝑒𝑥 = 0.16 min
The excess wait time rate Tex is then the excess wait time tex divided by the
average passenger trip length Lpt: 0.16/3.7 = 0.043 min/mi.

The passenger load waiting factor is computed with Equation 18-59.

4 (𝐹𝑙 − 0.80)
𝑎1 = 1 +
4.2
4 (0.83 − 0.80)
𝑎1 = 1 +
4.2
𝑎1 = 1.03
The perceived travel time rate is computed with Equation 18-58.
60
𝑇𝑝𝑡𝑡 = (𝑎1 ) + (2 𝑇𝑒𝑥 ) − 𝑇𝑎𝑡
𝑆𝑇𝑡,𝑠𝑒𝑔
60
𝑇𝑝𝑡𝑡 = (1.03 ) + [2(0.043)] − 0.054
11.3
𝑇𝑝𝑡𝑡 = 5.50 min/mi
The segment is not located in a central business district of a metropolitan
area with a population of 5 million or more, so the base travel time rate Tbtt is equal
to 4.0 min/mi. The perceived travel time factor is computed with Equation 18-57.
(𝑒 − 1) 𝑇𝑏𝑡𝑡 − (𝑒 + 1) 𝑇𝑝𝑡𝑡
𝐹𝑡𝑡 =
(𝑒 − 1) 𝑇𝑝𝑡𝑡 − (𝑒 + 1) 𝑇𝑏𝑡𝑡
(−0.40 − 1)(4.0) − (−0.40 + 1)(5.50)
𝐹𝑡𝑡 =
(−0.40 − 1)(5.50) − (−0.40 + 1)(4.0)
𝐹𝑡𝑡 = 0.881
Finally, the transit wait–ride score is computed with Equation 18-62.
𝑠𝑤-𝑟 = 𝐹ℎ 𝐹𝑡𝑡
𝑠𝑤-𝑟 = (2.80)(0.883)
𝑠𝑤-𝑟 = 2.47
Step 5: Determine Pedestrian LOS Score for Link

The pedestrian methodology described in Chapter 18 was used to determine
the pedestrian LOS score for the link Ip,link. This score was computed to be 3.53.
Step 6: Determine Transit LOS Score for Segment

The transit LOS score for the segment is computed with Equation 18-63.
𝐼𝑡,𝑠𝑒𝑔 = 6.0 − 1.50 𝑠𝑤-𝑟 + 0.15 𝐼𝑝,link
𝐼𝑡,𝑠𝑒𝑔 = 6.0 − 1.50(2.47) + 0.15(3.53)
𝐼𝑡,𝑠𝑒𝑔 = 2.83

The transit LOS is determined by using the transit LOS score from Step 6.
This performance measure is compared with the thresholds in Exhibit 18-3 to
determine that the LOS for the specified bus route is C.

9. ROUNDABOUT SEGMENT METHODOLOGY
SCOPE OF THE METHODOLOGY

This subsection provides an overview of the methodology for evaluating the
performance of the motor vehicle mode on an urban street segment bounded by
one or more roundabouts. The methodology is based on national research that
measured the travel time performance of nine facilities containing three or more
roundabouts in series (10). The methodology is designed to be integrated into the
general motorized vehicle methodology for urban street segments described in
Chapter 18. Only the relevant deviations from the general methodology are
provided in this subsection.
LIMITATIONS OF THE METHODOLOGY

The methodologies in this subsection are based on regression analyses of
field-measured data. The limits of these field data are provided in Exhibit 30-40.
The analyst is cautioned with regard to the validity of the results when an input
or intermediate calculated value is outside the range of the research data. In
addition, the methodology does not account for capacity constraint caused by
oversaturated conditions or the possible effects of an upstream signal on a
downstream roundabout.
Input or Calculated Value Minimum Maximum Exhibit 30-40

Input Data Validity Range of Inputs and
Calculated Values for Analysis
Inscribed circle diameter (ft) 84 245
of Motor Vehicles on an Urban
Number of circulating lanes 1 2
Street Roundabout Segment
Segment length (ft) 540 7,900
Posted speed limit (mi/h) 25 50
Intermediate Calculations
Central island diameter (ft) 48 187
Length of first portion of segment (ft) 270 3,953
Length of second portion of segment (ft) 244 3,993
Free-flow speed (mi/h) 26 53
Roundabout influence area for first portion of segment (ft) 235 1,446
Roundabout influence area for second portion of segment (ft) 73 897
Geometric delay for first portion of segment (s) 0.1 9.5
Geometric delay for second portion of segment (s) 0.1 6.6
REQUIRED INPUT DATA AND SOURCES

Exhibit 30-41 lists the additional required input data, potential data sources,
and suggested default values for applying the methodology in this subsection.
The reader should refer to Chapter 18 for a complete list of required input data.
Guidance on selecting values for inscribed circle diameter and width of
circulating lanes can be obtained elsewhere (11).
Chapter 30/Urban Street Segments: Supplemental Roundabout Segment Methodology

Exhibit 30-41 Required Data and Units Potential Data Source(s) Suggested Default Value
Additional Required Input Geometric Design Data
Data, Potential Data Sources,
130 ft for one-lane
and Default Values for
Inscribed circle diameter of upstream Field data, aerial photo, roundabout
Analysis of Motor Vehicles on
and downstream roundabout (ft) preliminary design 180 ft for two-lane
an Urban Street Roundabout
roundabout
Segment
Number of circulating lanes of upstream Field data, aerial photo,
Must be provided
and downstream roundabout (ft) preliminary design
Average width of circulating lanes of
Field data, aerial photo, 20 ft for one-lane roundabout
upstream and downstream roundabout
preliminary design 15 ft for two-lane roundabout
(ft)
Performance Measure Data
Control delay by lane at boundary
HCM method output Must be provided
roundabout (s/veh)
Capacity by lane at boundary
HCM method output Must be provided
roundabout (veh/h)
GEOMETRIC DESIGN DATA

This subsection describes the geometric design data listed in Exhibit 30-41.
These data describe the additional geometric elements of the roundabouts
beyond the geometric elements of the intersections and segments described in
Exhibit 18-5.
Inscribed Circle Diameter

The inscribed circle diameter, ICD, is the diameter of the largest circle that
can be inscribed within the outer edges of the circulatory roadway. The ICD
serves as the width of the roundabout. This is illustrated in Exhibit 30-42.
Exhibit 30-42
Illustration of Geometric
Design Data
ICD
wc
For the purposes of this methodology, if the ICD is variable throughout the
roundabout (e.g., to accommodate a variable number of circulating lanes, as
illustrated in Exhibit 30-42), the larger dimension should be used.
Roundabout Segment Methodology Chapter 30/Urban Street Segments: Supplemental

Number of Circulating Lanes

The number of circulating lanes Nc is the count of circulating lanes
immediately downstream of the entry that forms the end of the segment under
study.
Average Width of Circulating Lanes

The average width of circulating lanes wc is measured in the section of
circulatory roadway immediately downstream of the entry, that is, the same
location where the number of circulating lanes is counted. This is illustrated in
Exhibit 30-42.
COMPUTATIONAL STEPS
The computational steps described below are illustrated in the flowchart
provided in Exhibit 18-8. The path followed is that of a noncoordinated system
with YIELD control.
Step 1: Determine Traffic Demand Adjustments

The models developed for estimating travel speed through a series of
roundabouts were calibrated by using roundabouts that were operating below
capacity. Neither the capacity estimation procedures for roundabouts in
Chapter 22 nor the procedures in this subsection explicitly account for capacity
constraint that restricts (or meters) discharge volume from the intersection when
the demand volume for an intersection traffic movement exceeds its capacity.
Similarly, the methodology does not account for the effect on roundabout
operations or travel time that may be created by queue spillback between two
roundabouts. The occurrence of any of these conditions should be flagged, and
an alternative tool should be considered.
Step 2: Determine Running Time

A procedure for determining running time for a segment bounded by one or
more roundabouts is described in this step. It builds on the procedure described
in Chapter 18. Each calculation is discussed in the following subparts, which
culminate with the calculation of segment running time.
A. Determine Free-Flow Speed

Free-flow speed represents the average running speed of through vehicles
traveling along a segment under low-volume conditions and not delayed by
traffic control devices or other vehicles. It reflects the effect of the street
environment on driver speed choice. Elements of the street environment that
influence this choice under free-flow conditions include speed limit, access point
density, median type, curb presence, and segment length. Further discussion on
free-flow speed can be found in Section 3 of Chapter 18.
Free-flow speed (when the influence of roundabouts at one or both ends of
the segment is considered) is calculated by separately determining the free-flow
speed influenced by the roundabout at each end of the segment and then
comparing these two free-flow speed estimates with the free-flow speed that
would be estimated without the presence of roundabouts.

Base Free-Flow Speed

The base free-flow speed is defined to be the free-flow speed on longer
segments and is computed the same for segments bounded by roundabouts as
for segments bounded by signals. It includes the influence of speed limit, access
point density, median type, curb presence, and on-street parking presence. It is
Equation 30-72 𝑆𝑓𝑜 = 𝑆𝑐𝑎𝑙𝑖𝑏 + 𝑆0 + 𝑓𝑐𝑠 + 𝑓𝐴 + 𝑓𝑝𝑘
where
Sfo = base free-flow speed (mi/h),
Scalib = base free-flow speed calibration factor (mi/h),
S0 = speed constant (mi/h),
fCS = adjustment for cross section (mi/h),
fA = adjustment for access points (mi/h), and
fpk = adjustment for on-street parking (mi/h).
The speed constant and adjustment factors used in Equation 30-72 are listed
in Exhibit 30-43. The exhibit is the same as Exhibit 18-11, except that the width of
the signalized intersection used in the calculation for the adjustment for access
points fA has been replaced with the inscribed circle diameter of the roundabout,
and the range of speed limits is restricted to the validity range for this method.
Equations provided in the table footnote can also be used to compute these
adjustment factors for conditions not shown in the exhibit. Further discussion of
this equation and adjustment factors can be found in Chapter 18.
Exhibit 30-43 Speed Percent with Adjustment for Cross

Base Free-Flow Speed Speed Limit Constant S0 Restrictive Section fCS (mi/h)b
Adjustment Factors (mi/h) (mi/h)a Median Type Median (%) No Curb Curb
25 37.4 Restrictive 20 0.3 -0.9
30 39.7 40 0.6 -1.4
35 42.1 60 0.9 -1.8
40 44.4 80 1.2 -2.2
45 46.8 100 1.5 -2.7
50 49.1 Nonrestrictive Not applicable 0.0 -0.5
No median Not applicable 0.0 -0.5
Access Adjustment for Access Points fA by Lanes Percent with Adjustment
Density Da Nth (mi/h)c On-Street for Parking
(points/mi) 1 Lane 2 Lanes 3 Lanes Parking (%) (mi/h)d
0 0.0 0.0 0.0 0 0.0
2 -0.2 -0.1 -0.1 20 -0.6
4 -0.3 -0.2 -0.1 40 -1.2
10 -0.8 -0.4 -0.3 60 -1.8
20 -1.6 -0.8 -0.5 80 -2.4
40 -3.1 -1.6 -1.0 100 -3.0
60 -4.7 -2.3 -1.6
a
Notes: S0 = 25.6 + 0.47Spl, where Spl = posted speed limit (mi/h).
b
fCS = 1.5 prm – 0.47 pcurb – 3.7 pcurb prm, where prm = proportion of link length with restrictive median
(decimal) and pcurb = proportion of segment with curb on the right-hand side (decimal).
c
fA = –0.078 Da /Nth with Da = 5,280 (Nap,s + Nap,o)/(L – ICDi), where Da = access point density on segment
(points/mi); Nth = number of through lanes on the segment in the subject direction of travel (ln); Nap,s =
number of access point approaches on the right side in the subject direction of travel (points); Nap,o =
number of access point approaches on the right side in the opposing direction of travel (points); L =
segment length (ft); and ICDi = inscribed circle diameter of roundabout (ft).
d
fpk = –3.0 × proportion of link length with on-street parking available on the right-hand side (decimal).

Equation 30-72 has been calibrated by using data for many urban street
segments collectively located throughout the United States, so the default value
of 0.0 mi/h for Scalib is believed to yield results that are reasonably representative
of driver behavior in most urban areas. However, if desired, a locally
representative value can be determined from field-measured estimates of the
base free-flow speed for several street segments. The local default value can be
established for typical street segments or for specific street types. This calibration
factor is determined as the one value that provides a statistically based best fit
between the prediction from Equation 30-72 and the field-measured estimates. A
procedure for estimating the base free-flow speed from field data is described in
Section 6.
Roundabout Geometry and Speed Parameters

The computation of free-flow speed, roundabout influence area, and
geometric delay requires measurement or estimation of a series of geometric
parameters associated with the roundabout at one or both ends of the segment.
These computations are performed separately for each roundabout.
The central island diameter is equal to the inscribed circle diameter minus
the width of the circulatory roadway on each side of the central island. The
circulatory roadway width is equal to the average width of each circulating lane
times the number of circulating lanes. These calculations are combined into a
single equation as given in Equation 30-73.
𝐶𝐼𝐷 = 𝐼𝐶𝐷 − 2𝑁𝑐 𝑤𝑐 Equation 30-73
where
CID = central island diameter (ft),
ICD = inscribed circle diameter (ft),
Nc = number of circulating lane(s), and
wc = average width of circulating lane(s) (ft).
The circulating speed, Sc, can be approximated by assuming that the
circulating path occupies the centerline of the circulatory roadway with a radius
equal to half the central island diameter plus half the total width of the
circulatory roadway. This radius can be computed with Equation 30-74.
𝐼𝐶𝐷 𝑁𝑐 𝑤𝑐
𝑟𝑐,𝑡ℎ = + Equation 30-74
2 2
where
rc,th = average radius of circulating path of through movement (ft),
ICD = inscribed circle diameter (ft),
Nc = number of circulating lane(s), and
wc = average width of circulating lane(s) (ft).
The speed associated with this radius can be estimated with Equation 30-75
(12), which assumes a negative cross slope of the circulatory roadway of –0.02,
typical of many roundabouts.

Equation 30-75
𝑆𝑐 = 3.4614𝑟𝑐,𝑡ℎ 0.3673
where
Sc = circulating speed (mi/h), and
rc,th = average radius of circulating path of through movement (ft).
For the purposes of calculating free-flow speed, roundabout influence area,
and geometric delay, the segment length is divided into two subsegments.
Subsegment 1 consists of the portion of the segment from the yield line of the
upstream roundabout to the midpoint between the two roundabouts, defined as
halfway between the cross-street centerlines of the two roundabouts.
Subsegment 2 consists of the portion of the segment from this midpoint to the
yield line of the downstream roundabout. The lengths of these subsegments are
calculated with Equation 30-76 and Equation 30-77. These dimensions are
illustrated in Exhibit 30-44.
1 𝐼𝐶𝐷1 𝐼𝐶𝐷2 𝐼𝐶𝐷1
Equation 30-76 𝐿1 = (𝐿 − + )+
2 2 2 2
Equation 30-77 𝐿2 = 𝐿 − 𝐿1
where
L1 = length of Subsegment 1 (ft),
L = length of segment (ft),
ICD1 = inscribed circle diameter of Roundabout 1 (ft), and
ICD2 = inscribed circle diameter of Roundabout 2 (ft).
Exhibit 30-44
Illustration of Subsegment
Dimensions
Free-Flow Speed for Upstream Subsegment (Subsegment 1)

Free-slow speed for Subsegment 1 (the upstream subsegment) is computed
in a three-step process by first determining an initial free-flow speed. A
roundabout influence area is then computed as the distance over which the
geometric features of the roundabout influence travel speed. The initial free-flow
speed is then adjusted downward if the roundabout influence area meets or
exceeds the length of the subsegment.

The initial free-flow speed for Subsegment 1 is estimated from the

subsegment length, posted speed limit, and central island diameter of the
roundabout at the upstream end of the segment by using Equation 30-78.
𝑆𝑓,1,initial = 14.6 + 0.0039𝐿1 + 0.48𝑆𝑃𝐿 + 0.02𝐶𝐼𝐷1 Equation 30-78
where
Sf,1,initial = initial free-flow speed for Subsegment 1 (mi/h),
SPL = posted speed limit (mi/h), and
CID1 = central island diameter for roundabout at upstream end of
Subsegment 1 (ft).
The roundabout influence area for Subsegment 1, RIA1, is estimated from the
free-flow speed and circulating speed with Equation 30-79. This equation yields
positive values for inputs within the range limits.
𝑅𝐼𝐴1 = −149.8 + 31.4𝑆𝑓,1,initial − 22.5𝑆𝑐,1 Equation 30-79
where
RIA1 = roundabout influence area for Subsegment 1 (ft),
Sf,1,initial = initial free-flow speed for Subsegment 1 (mi/h), and
Sc,1 = through movement circulating speed for roundabout at upstream end
of segment (mi/h).
The roundabout influence area is then compared with the length of the
subsegment, as shown in Equation 30-80. If the roundabout influence area is
equal to or exceeds the length of the subsegment, the subsegment free-flow
speed is reduced.
𝑆𝑓,1 = 𝑆𝑓,1,𝑖nitial − 4.43 if 𝑅𝐼𝐴1 ≥ 𝐿1 , else Equation 30-80
𝑆𝑓,1 = 𝑆𝑓,1,initial
where Sf,1 is the free-flow speed for Subsegment 1 (mi/h).
Free-Flow Speed for Downstream Subsegment (Subsegment 2)

The initial free-flow speed for Subsegment 2, Sf,2,initial, is estimated with
Equation 30-81.
𝑆𝑓,2,initial = 15.1 + 0.0037𝐿2 + 0.43𝑆𝑃𝐿 + 0.05𝐶𝐼𝐷2 Equation 30-81
where
Sf,2,initial = initial free-flow speed for Subsegment 2 (mi/h),
SPL = posted speed limit (mi/h), and
CID2 = central island diameter for roundabout at downstream end of
Subsegment 2 (ft).
The roundabout influence area for the subsegment RIA2 is estimated from
the free-flow speed and downstream circulating speed with Equation 30-82.

Equation 30-82
𝑅𝐼𝐴2 = 165.9 + 13.8𝑆𝑓,2,initial − 21.1𝑆𝑐,2
where
RIA2 = roundabout influence area for Subsegment 2 (ft),
Sf,2,initial = initial free-flow speed for Subsegment 2 (mi/h), and
Sc,2 = through movement circulating speed for roundabout at downstream
end of subsegment (mi/h).
The roundabout influence area is then compared with the length of the
subsegment, as shown in Equation 30-83. If the roundabout influence area is
equal to or exceeds the length of the subsegment, the subsegment free-flow
speed is reduced to account for the overlap.
Equation 30-83
𝑆𝑓,2 = 𝑆𝑓,2,initial − 4.73 if 𝑅𝐼𝐴2 ≥ 𝐿2 , else
𝑆𝑓,2 = 𝑆𝑓,2,initial
where Sf,2 is the free-flow speed for Subsegment 2 (mi/h).
Free-Flow Speed Without Influence of Roundabouts

The calculation for free-flow speed without the geometric influence of
roundabouts is the same as for segments bounded by signalized intersections, as
provided in Chapter 18. Equation 30-84 is used to compute the value of an
adjustment factor that accounts for the influence of short spacing of boundary
intersections.
𝑆𝑓𝑜 − 19.5
Equation 30-84 𝑓𝐿 = 1.02 − 4.7 ≤ 1.0
max(𝐿𝑠 , 400)
where
fL = boundary intersection spacing adjustment factor;
Sfo = base free-flow speed (mi/h); and
Ls = distance between adjacent boundary intersections that (a) bracket the
subject segment and (b) each have a type of control that can impose on
the subject through movement a legal requirement to stop or yield,
such as a roundabout (ft).
The predicted free-flow speed without the geometric influence of
roundabouts is computed with Equation 30-85 on the basis of estimates of base
free-flow speed and the signal spacing adjustment factor.
Equation 30-85 𝑆𝑓,𝑛𝑜𝑛-𝑟𝑏𝑡 = 𝑆𝑓𝑜 𝑓𝐿 ≥ 𝑆𝑝𝑙
where Sf,non-rbt is the free-flow speed for nonroundabout segments (mi/h) and Spl is
the posted speed limit. If the speed obtained from Equation 30-85 is less than the
speed limit, the speed limit is used.
Free-Flow Speed
The free-flow speeds for each subsegment are then compared with each
other and with the nonroundabout free-flow speed with Equation 30-86. The
lowest of these speeds is the governing free-flow speed for the segment. The
analyst is cautioned that if the result of this calculation is outside the validity

range presented in Exhibit 30-40, the calculation is an extrapolation of the model.

Note that the resulting free-flow speed for a segment bounded by one or more
roundabouts may be lower than the posted speed, even though the
nonroundabout free-flow speed is constrained by the posted speed in accordance
with the motorized vehicle methodology in Chapter 18.
𝑆𝑓 = min(𝑆𝑓,1 , 𝑆𝑓,2 , 𝑆𝑓,𝑛𝑜𝑛-𝑟𝑏𝑡 ) Equation 30-86
B. Compute Adjustment for Vehicle Proximity

This step is the same as in Chapter 18.
C. Compute Delay due to Turning Vehicles

D. Estimate Delay due to Other Sources

E. Compute Segment Running Time

Equation 30-87 is used to compute the segment running time, which is based
on Equation 18-7. It incorporates the conditions specified in Chapter 18 for a
yield-controlled boundary exiting the segment: a start-up lost time of 2.5 s and
the influence of the volume-to-capacity ratio of the roundabout entry.
𝑁𝑎𝑝
3.5 𝑣𝑡ℎ 3,600 𝐿
𝑡𝑅 = × min ( , 1.00) + 𝑓 + ∑ 𝑑𝑎𝑝,𝑖 + 𝑑other Equation 30-87
0.0025 𝐿 𝑐𝑡ℎ 5,280 𝑆𝑓 𝑣
𝑖=1
where
tR = segment running time (s),
L = segment length (ft),
vth = through-demand flow rate (veh/h),
cth = through-movement capacity (veh/h),
fv = proximity adjustment factor,
dap,i = delay due to left and right turns from the street into access point
intersection i (s/veh),
Nap = number of influential access point approaches along the segment = Nap,s
+ pap,lt Nap,o (points),
Nap,s = number of access point approaches on the right side in the subject
direction of travel (points),
Nap,o = number of access point approaches on the right side in the opposing
direction of travel (points),
pap,lt = proportion of Nap,o that can be accessed by a left turn from the subject
direction of travel, and
dother = delay due to other sources along the segment (e.g., curb parking or
pedestrians) (s/veh).

The variables vth and cth used in Equation 30-87 apply to the through
movement exiting the segment at the boundary roundabout.
Step 3: Determine the Proportion Arriving During Green

This step does not apply to a segment with a downstream roundabout. The
methodology does not account for the possible effects of an upstream signal on a
downstream roundabout.
Step 4: Determine Signal Phase Duration

This step does not apply to a segment with a downstream roundabout.
Step 5: Determine Through Delay

The through delay for a segment with a roundabout at one or both ends is
computed as a combination of control delay and geometric delay.
The procedure for computing the control delay at a roundabout at the
downstream end of a segment is provided in Chapter 22, which determines the
control delay for a roundabout on a lane-by-lane basis. For an approach with one
lane, the through control delay is equal to the control delay of the lane. For an
approach with two lanes, the through control delay is computed by allocating
the control delay in each lane in proportion to the through traffic in each lane by
𝑑𝐿𝐿 𝑣𝐿𝐿 𝑃𝐿𝐿,𝑇 + 𝑑𝑅𝐿 𝑣𝑅𝐿 𝑃𝑅𝐿,𝑇
Equation 30-88 𝑑control,𝑡 =
𝑣𝑡ℎ
where
dcontrol,t = through control delay (s/veh),
vth = through-demand flow rate (veh/h),
dLL = control delay in left lane (s/veh),
vLL = demand flow rate in left lane (veh/h),
dRL = control delay in right lane (s/veh),
vRL = demand flow rate in right lane (veh/h),
PLL,T = proportion of through-movement vehicles in the left lane (decimal), and
PRL,T = proportion of through-movement vehicles in the right lane (decimal).
Geometric delay is calculated separately for the presence of a roundabout on
the two subsegments. If a roundabout is present on the upstream end of
Subsegment 1 (regardless of the control present at the downstream end of
Subsegment 2), the geometric delay for the upstream portion of the segment dgeom,1
is calculated with Equation 30-89. If the upstream end of the segment is controlled
by a signalized or stop-controlled intersection or is uncontrolled, dgeom,1 = 0.
Equation 30-89
1 1
𝑑𝑔𝑒𝑜𝑚,1 = max [−2.63 + 0.09𝑆𝑓 + 0.625𝐼𝐶𝐷1 ( − ) , 0]
𝑆𝑐,1 𝑆𝑓
where dgeom,1 is the geometric delay for Subsegment 1 (s/veh).

If a roundabout is present on the downstream end of the segment (regardless

of the control present at the upstream end), the geometric delay for the
downstream portion of the segment dgeom,2 is calculated with Equation 30-90. If the
upstream end of the segment is controlled by a signalized or stop-controlled
intersection or is uncontrolled, dgeom,2 = 0.
𝑑𝑔𝑒𝑜𝑚,2 = max(1.57 + 0.11𝑆𝑓 − 0.21𝑆𝑐,2 , 0) Equation 30-90
where dgeom,2 is the geometric delay for Subsegment 2 (s/veh).

The analyst is cautioned that if these calculations result in one or more
geometric delay estimates outside the validity range presented in Exhibit 30-40,
the calculation is an extrapolation of the model.
The through delay dt is computed as the sum of control and geometric
delays, as given in Equation 30-91.
𝑑𝑡 = 𝑑control,𝑡 + 𝑑𝑔𝑒𝑜𝑚,1 + 𝑑𝑔𝑒𝑜𝑚,2 Equation 30-91
Step 6: Determine Through Stop Rate

As noted in Chapter 18, the stop rate at a YIELD-controlled approach will vary
with conflicting demand. It can be estimated (in stops per vehicle) as equal to the
volume-to-capacity ratio of the through movement at the boundary intersection.
This approach recognizes that YIELD control does not require drivers to come to a
complete stop when there is no conflicting traffic. The through stop rate h is
computed as given in Equation 30-92. The methodology does not apply for
volume-to-capacity ratios exceeding 1.0.
𝑣𝑡ℎ
ℎ = min ( , 1.00) Equation 30-92
𝑐𝑡ℎ
Step 7: Determine Travel Speed

This step is the same as for Chapter 18.
Step 8: Determine Spatial Stop Rate

This step is the same as for Chapter 18.

This step is the same as for Chapter 18. The base free-flow speed for the
estimation of LOS is the same base free-flow speed as determined in Chapter 18.
Step 10: Determine Motor Vehicle Traveler Perception Score

Research has not been conducted on the traveler’s perception of service
quality for roundabouts in a manner that can be integrated into this
methodology. As a result, the motor vehicle traveler perception score for a
segment bounded by a roundabout is undefined and this step is not applicable
for the evaluation of roundabout segments.

10. REFERENCES
Some of these references can 1. Van Zuylen, H. The Estimation of Turning Flows on a Junction. Traffic
Reference Library in Volume 4. Engineering and Control, Vol. 20, No. 11, 1979, pp. 539–541.
2. Wallace, C., K. Courage, M. Hadi, and A. Gan. TRANSYT-7F User’s Guide,
Vol. 4 in a Series: Methodology for Optimizing Signal Timing. University of
Florida, Gainesville, March 1998.
3. Robertson, D. TRANSYT: A Traffic Network Study Tool. RRL Report LR 253.
Road Research Laboratory, Crowthorne, Berkshire, United Kingdom, 1969.
4. Bonneson, J., M. Pratt, and M. Vandehey. Predicting the Performance of
Automobile Traffic on Urban Streets: Final Report. NCHRP Project 3-79. Texas
Transportation Institute, Texas A&M University, College Station, Jan. 2008.
6. Bonneson, J., and J. Fitts. Delay to Major Street Through Vehicles at Two-
Way Stop-Controlled Intersections. Transportation Research Part A: Policy and
Practice, Vol. 33, Nos. 3–4, 1999, pp. 237–254.
7. Bonneson, J. Delay to Major Street Through Vehicles due to Right-Turn
Activity. Transportation Research Part A: Policy and Practice, Vol. 32, No. 2,
1998, pp. 139–148.
8. Robertson, H., J. Hummer, and D. Nelson. Manual of Transportation Engineering
Studies. Institute of Transportation Engineers, Washington, D.C., 2000.
9. Kittelson & Associates, Inc.; Parsons Brinckerhoff; KFH Group, Inc.; Texas
A&M Transportation Institute; and Arup. TCRP Report 165: Transit Capacity
and Quality of Service Manual, 3rd Edition. Transportation Research Board of
the National Academies, Washington, D.C., 2013.
10. Rodegerdts, L. A., P. M. Jenior, Z. H. Bugg, B. L. Ray, B. J. Schroeder, and M.
A. Brewer. NCHRP Report 772: Evaluating the Performance of Corridors with
Roundabouts. Transportation Research Board of the National Academies,
Washington, D. C., 2014.
12. Rodegerdts, L., M. Blogg, E. Wemple, E. Myers, M. Kyte, M. P. Dixon, G. F.
List, A. Flannery, R. Troutbeck, W. Brilon, N. Wu, B. N. Persaud, C. Lyon, D.
L. Harkey, and D. Carter. NCHRP Report 572: Roundabouts in the United States.
D.C., 2007.
References Chapter 30/Urban Street Segments: Supplemental

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 31
SIGNALIZED INTERSECTIONS: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 31-1
2. CAPACITY AND PHASE DURATION ............................................................. 31-2

Actuated Phase Duration .................................................................................. 31-2
Lane Group Flow Rate on Multiple-Lane Approaches ................................31-23
Pretimed Phase Duration .................................................................................31-30
Pedestrian and Bicycle Adjustment Factors ...................................................31-35
Work Zone Presence Adjustment Factor ........................................................31-40
3. QUEUE ACCUMULATION POLYGON ......................................................... 31-42

Concepts .............................................................................................................31-42
General QAP Construction Procedure ...........................................................31-43
QAP Construction Procedure for Selected Lane Groups .............................31-45
4. QUEUE STORAGE RATIO................................................................................ 31-63

Concepts .............................................................................................................31-63
Procedure for Estimating Back of Queue for Selected Lane Groups ..........31-70
5. PLANNING-LEVEL ANALYSIS APPLICATION ......................................... 31-78

Overview of the Application............................................................................31-78
Methodology ......................................................................................................31-80
Worksheets .........................................................................................................31-94
6. FIELD MEASUREMENT TECHNIQUES ........................................................ 31-99

Field Measurement of Intersection Control Delay ........................................31-99
Field Measurement of Saturation Flow Rate ...............................................31-105
7. COMPUTATIONAL ENGINE DOCUMENTATION ................................. 31-111

Flowcharts ........................................................................................................31-111
Linkage Lists ....................................................................................................31-113
8. USE OF ALTERNATIVE TOOLS ................................................................... 31-119

Effect of Storage Bay Overflow ......................................................................31-119
Effect of Right-Turn-on-Red Operation ........................................................31-121
Chapter 31/Signalized Intersections: Supplemental Contents

Effect of Short Through Lanes ...................................................................... 31-124

Effect of Closely Spaced Intersections ......................................................... 31-125
9. CONNECTED AND AUTOMATED VEHICLES ........................................31-127

Introduction ..................................................................................................... 31-127
Concepts ........................................................................................................... 31-127
Modifications to Signalized Intersection Core Methodology Inputs....... 31-130
Service Volume Table ..................................................................................... 31-132
10. EXAMPLE PROBLEMS .................................................................................31-133

Example Problem 1: Motorized Vehicle LOS .............................................. 31-133
Example Problem 2: Pedestrian LOS ........................................................... 31-142
Example Problem 3: Bicycle LOS .................................................................. 31-148
Example Problem 4: Pedestrian Delay with Two-Stage Crossing of One
Intersection Leg ........................................................................................ 31-150
Example Problem 5: Pedestrian Delay with Two-Stage Crossing of Two
Intersection Legs ...................................................................................... 31-152
11. REFERENCES..................................................................................................31-156
Contents Chapter 31/Signalized Intersections: Supplemental

LIST OF EXHIBITS
Exhibit 31-1 Time Elements Influencing Actuated Phase Duration .................... 31-3
Exhibit 31-2 Detection Design and Maximum Allowable Headway ................... 31-8
Exhibit 31-3 Force-Off Points, Yield Point, and Phase Splits .............................. 31-14
Exhibit 31-4 Example Equivalent Maximum Green for Fixed Force Mode ...... 31-16
Exhibit 31-5 Probability of a Lane Change ............................................................ 31-24
Exhibit 31-6 Input Variables for Lane Group Flow Rate Procedure .................. 31-26
Exhibit 31-7 Example Intersection .......................................................................... 31-32
Exhibit 31-8 Conflict Zone Locations ..................................................................... 31-35
Exhibit 31-9 Work Zone on an Intersection Approach ........................................ 31-40
Exhibit 31-10 Geometric Design Input Data Requirements for
Work Zones ........................................................................................................ 31-40
Exhibit 31-11 Queue Accumulation Polygon for Protected Movements .......... 31-43
Exhibit 31-12 Unblocked Permitted Green Time .................................................. 31-46
Exhibit 31-13 QAP for Permitted Left-Turn Operation in an
Exclusive Lane ................................................................................................... 31-56
Exhibit 31-14 QAP for Permitted Left-Turn Operation in a Shared Lane ......... 31-56
Exhibit 31-15 QAP for Leading, Protected-Permitted Left-Turn Operation
in an Exclusive Lane .......................................................................................... 31-56
Exhibit 31-16 QAP for Lagging, Protected-Permitted Left-Turn Operation
Exhibit 31-17 QAP for Leading, Protected-Permitted Left-Turn Operation
in a Shared Lane ................................................................................................ 31-57
Exhibit 31-18 QAP for Lagging, Protected-Permitted Left-Turn Operation
in a Shared Lane ................................................................................................ 31-57
Exhibit 31-19 Polygon for Uniform Delay Calculation ........................................ 31-59
Exhibit 31-20 Time–Space Diagram of Vehicle Trajectory on an
Intersection Approach ...................................................................................... 31-64
Exhibit 31-21 Cumulative Arrivals and Departures During an
Oversaturated Analysis Period ........................................................................ 31-65
Exhibit 31-22 Third-Term Back-of-Queue Size with Increasing Queue ............ 31-66
Exhibit 31-23 Third-Term Back-of-Queue Size with Decreasing Queue ........... 31-66
Exhibit 31-24 Third-Term Back-of-Queue Size with Queue Clearing ............... 31-66
Exhibit 31-25 Arrival–Departure Polygon ............................................................. 31-69
Exhibit 31-26 ADP for Permitted Left-Turn Operation in an
Exclusive Lane ................................................................................................... 31-71
Exhibit 31-27 ADP for Permitted Left-Turn Operation in a Shared Lane ......... 31-72
Exhibit 31-28 ADP for Leading, Protected-Permitted Left-Turn Operation

Exhibit 31-29 ADP for Lagging, Protected-Permitted Left-Turn Operation

in an Exclusive Lane ..........................................................................................31-72
Exhibit 31-30 ADP for Leading, Protected-Permitted Left-Turn Operation
in a Shared Lane .................................................................................................31-73
Exhibit 31-31 ADP for Lagging, Protected-Permitted Left-Turn Operation
in a Shared Lane .................................................................................................31-73
Exhibit 31-32 Required Input Data for the Planning-Level Analysis
Application .........................................................................................................31-80
Exhibit 31-33 Planning-Level Analysis: Equivalency Factor for Left Turns .....31-83
Exhibit 31-34 Planning-Level Analysis: Equivalency Factor for
Right Turns .........................................................................................................31-83
Parking Activity .................................................................................................31-83
Lane Utilization ..................................................................................................31-84
Exhibit 31-37 Planning-Level Analysis: Intersection Volume-to-Capacity
Ratio Assessment Levels ...................................................................................31-90
Exhibit 31-38 Planning-Level Analysis: Progression Adjustment Factor ..........31-92
Exhibit 31-39 Planning-Level Analysis: Input Worksheet ...................................31-95
Exhibit 31-40 Planning-Level Analysis: Left-Turn Treatment Worksheet ........31-96
Exhibit 31-41 Planning Level Analysis: Intersection Sufficiency
Worksheet ...........................................................................................................31-97
Exhibit 31-42 Planning-Level Analysis: Delay and LOS Worksheet ..................31-98
Exhibit 31-43 Control Delay Field Study Worksheet .........................................31-101
Exhibit 31-44 Acceleration–Deceleration Correction Factor .............................31-103
Exhibit 31-45 Example Control Delay Field Study Worksheet .........................31-104
Exhibit 31-46 Example Worksheet with Residual Queue at End .....................31-105
Exhibit 31-47 Saturation Flow Rate Field Study Worksheet .............................31-107
Exhibit 31-48 Methodology Flowchart .................................................................31-111
Exhibit 31-49 Setup Module ..................................................................................31-112
Exhibit 31-50 Signalized Intersection Module ....................................................31-112
Exhibit 31-51 Initial Queue Delay Module ..........................................................31-113
Exhibit 31-52 Performance Measures Module ....................................................31-113
Exhibit 31-53 Setup Module Routines ..................................................................31-114
Exhibit 31-54 Signalized Intersection Module: Main Routines .........................31-115
Exhibit 31-55 Signalized Intersection Module: ComputeQAPolygon
Routines ............................................................................................................31-117
Exhibit 31-56 Performance Measures Module Routines ....................................31-118
Exhibit 31-57 Effect of Storage Bay Length on Throughput and Delay ...........31-120

Exhibit 31-58 Effect of Storage Bay Length on Capacity ................................... 31-121

Exhibit 31-59 Effect of Right-Turn-on-Red and Lane Allocation on Delay ..... 31-122
Exhibit 31-60 Effect of Right-Turn-on-Red and Right-Turn Volume
on Delay ............................................................................................................ 31-123
Exhibit 31-61 Effect of Right-Turn-on-Red and Right-Turn Protection
on Delay ............................................................................................................ 31-124
Exhibit 31-62 Closely Spaced Intersections ......................................................... 31-125
Exhibit 31-63 Effect of Closely Spaced Intersections on Capacity
and Delay .......................................................................................................... 31-126
Exhibit 31-64 Base Saturation Flow Rates for CAVs for Through
Movements at Signalized Intersections ........................................................ 31-131
Exhibit 31-65 Saturation Flow Rate CAV Adjustment for Protected Left
Turns at Signalized Intersections .................................................................. 31-131
Exhibit 31-66 Saturation Flow Rate CAV Adjustments for Permissive Left
Turns at Signalized Intersections .................................................................. 31-132
Exhibit 31-67 Illustrative Generalized Service Volume LOS E Thresholds
for Signalized Intersections with CAV Presence (veh/h) ........................... 31-132
Exhibit 31-68 Example Problems .......................................................................... 31-133
Exhibit 31-69 Example Problem 1: Intersection Plan View ............................... 31-133
Exhibit 31-70 Example Problem 1: Traffic Characteristics Data ....................... 31-134
Exhibit 31-71 Example Problem 1: Geometric Design Data .............................. 31-134
Exhibit 31-72 Example Problem 1: Signal Control Data .................................... 31-134
Exhibit 31-73 Example Problem 1: Other Data ................................................... 31-135
Exhibit 31-74 Example Problem 1: Movement Groups and Lane Groups ...... 31-136
Exhibit 31-75 Example Problem 1: Movement Group Flow Rates ................... 31-136
Exhibit 31-76 Example Problem 1: Lane Group Flow Rates ............................. 31-136
Exhibit 31-77 Example Problem 1: Adjusted Saturation Flow Rate ................. 31-137
Exhibit 31-78 Example Problem 1: Proportion Arriving During Green .......... 31-139
Exhibit 31-79 Example Problem 1: Signal Phase Duration ................................ 31-139
Exhibit 31-80 Example Problem 1: Capacity and Volume-to-Capacity
Ratio................................................................................................................... 31-140
Exhibit 31-81 Example Problem 1: Control Delay .............................................. 31-140
Exhibit 31-82 Example Problem 1: Back of Queue and Queue Storage
Ratio................................................................................................................... 31-141
Exhibit 31-83 Example Problem 1: Queue Accumulation Polygon .................. 31-141
Exhibit 31-84 Example Problem 2: Pedestrian Flow Rates ................................ 31-142
Exhibit 31-85 Example Problem 2: Vehicular Demand Flow Rates ................. 31-142
Exhibit 31-86 Example Problem 3: Vehicular Demand Flow Rates and
Cross-Section Element Widths ....................................................................... 31-148

Exhibit 31-87 Example Problem 4: Intersection Geometry and Signal

Phase Sequence ................................................................................................31-150
Exhibit 31-88 Example Problem 5: Intersection Geometry and Signal
Phase Sequence ................................................................................................31-153

1. INTRODUCTION
Chapter 31 is the supplemental chapter for Chapter 19, Signalized VOLUME 4: APPLICATIONS
GUIDE
Intersections, which is found in Volume 3 of the Highway Capacity Manual 25. Freeway Facilities:
(HCM). This chapter presents detailed information about the following aspects of Supplemental
the Chapter 19 motorized vehicle methodology: Segments: Supplemental
• Procedures are described for computing actuated phase duration and Supplemental
pretimed phase duration. 28. Freeway Merges and
• Procedures are described for computing saturation flow rate adjustment 29. Urban Street Facilities:
Supplemental
factors to account for the presence of pedestrians, bicycles, and work 30. Urban Street Segments:
zones. Supplemental
31. Signalized
• A procedure is described for computing uniform delay by using the queue Intersections:
Supplemental
accumulation polygon (QAP) concept. The procedure is extended to 32. STOP-Controlled
shared-lane lane groups and lane groups with permitted turn movements. Intersections:
Supplemental
• A procedure is described for computing queue length and queue storage 33. Roundabouts:
Supplemental
ratio. 34. Interchange Ramp
This chapter provides a simplified version of the Chapter 19 motorized Terminals: Supplemental
vehicle methodology that is suitable for planning applications. The chapter also Supplemental
describes techniques for measuring control delay and saturation flow rate in the 36. Concepts: Supplemental
field, provides details about the computational engine that implements the 38. Network Analysis
Chapter 19 motorized vehicle methodology, illustrates the use of alternative tools
to evaluate signalized intersection operation, and provides guidance on
forecasting the effects of connected and automated vehicles on signalized
intersection operation. Finally, this chapter provides five example problems that
demonstrate the application of the motorized vehicle, pedestrian, and bicycle
methodologies to a signalized intersection.
Chapter 31/Signalized Intersections: Supplemental Introduction

2. CAPACITY AND PHASE DURATION
This section describes five procedures related to the calculation of capacity

and phase duration. The first procedure is used to calculate the average duration
of an actuated phase, and the second is used to calculate the lane volume
distribution on multilane intersection approaches. The third procedure focuses
on the calculation of phase duration for pretimed intersection operation. The
fourth procedure is used to compute the pedestrian and bicycle saturation flow
rate adjustment factors, and the fifth computes the work zone saturation flow
rate adjustment factor. Each procedure is described in a separate subsection.
ACTUATED PHASE DURATION

This subsection describes a procedure for estimating the average phase
duration for an intersection that is operating with actuated control. When
appropriate, the description is extended to include techniques for estimating the
duration of noncoordinated and coordinated phases. Unless stated otherwise, a
noncoordinated phase is modeled as an actuated phase in this methodology.
This subsection consists of the following seven parts:
• Concepts,
• Volume computations,
• Queue accumulation polygon,
• Maximum allowable headway,
• Equivalent maximum green,
• Average phase duration, and
• Probability of max-out.
The last six parts in the list above describe a series of calculations that are
completed in the sequence shown to obtain estimates of average phase duration
and the probability of phase termination by extension to its maximum green
limit (i.e., max-out).
Concepts
The duration of an actuated phase is composed of five time periods, as
shown in Equation 31-1. The first period represents the time lost while the queue
reacts to the signal indication changing to green. The second interval represents
the effective green time associated with queue clearance. The third period
represents the time the green indication is extended by randomly arriving
vehicles. It ends when there is a gap in traffic (i.e., gap-out) or a max-out. The
fourth period represents the yellow change interval, and the last period
represents the red clearance interval.
Equation 31-1 𝐷𝑝 = 𝑙1 + 𝑔𝑠 + 𝑔𝑒 + 𝑌 + 𝑅𝑐
where
Dp = phase duration (s),
Capacity and Phase Duration Chapter 31/Signalized Intersections: Supplemental

l1 = start-up lost time = 2.0 (s),

Y = yellow change interval (s),
Rc = red clearance interval (s),
gs = queue service time (s),
ge = green extension time (s).
The relationship between the variables in Equation 31-1 is shown in
Exhibit 31-1 with a QAP. Key variables shown in the exhibit are defined for
Equation 31-1 and in the following list:
qr = arrival flow rate during the effective red time = (1 – P) q C/r (veh/s),
P = proportion of vehicles arriving during the green indication (decimal),
r = effective red time = C – g (s),
g = effective green time (s),
s = adjusted saturation flow rate (veh/h/ln),
qg = arrival flow rate during the effective green time = P q C/g (veh/s),
q = arrival flow rate (veh/s),
Qr = queue size at the end of the effective red time = qr r (veh),
l2 = clearance lost time = Y + Rc – e (s), and
e = extension of effective green = 2.0 (s).
Exhibit 31-1
Time Elements Influencing
Actuated Phase Duration
Exhibit 31-1 shows the relationship between phase duration and queue size
for the average signal cycle. During the red interval, vehicles arrive at a rate of qr
and form a queue. The queue reaches its maximum size l1 seconds after the green
interval starts. At this time, the queue begins to discharge at a rate equal to the
saturation flow rate s less the arrival rate during green qg. The queue clears gs
seconds after it first begins to discharge. Thereafter, random vehicle arrivals are
detected and cause the green interval to be extended. Eventually, a gap occurs in
traffic (or the maximum green limit is reached), and the green interval ends. The
end of the green interval coincides with the end of the extension time ge.
Chapter 31/Signalized Intersections: Supplemental Capacity and Phase Duration

The effective green time for the phase is computed with Equation 31-2.
Equation 31-2 𝑔 = 𝐷𝑝 − 𝑙1 − 𝑙2
= 𝑔𝑠 + 𝑔𝑒 + 𝑒
Coordinated Phase Duration

The duration of a coordinated phase is dictated by the cycle length and the
force-off settings for the noncoordinated phases. These settings define the points
in the signal cycle at which each noncoordinated phase must end. The force-off
settings are used to ensure the coordinated phases receive a green indication at a
specific time in the cycle. Presumably, this time is synchronized with the
coordinated phase time at the adjacent intersections so that traffic progresses
along the street segment. In general, the duration of a coordinated phase is equal
to the cycle length less the time allocated to the conflicting phase in the same ring
and less the time allocated to the minor-street phases. Detectors are not typically
assigned to the coordinated phase, and this phase is not typically extended by
the vehicles it serves.
Noncoordinated Phase Duration

The duration of a noncoordinated phase is dictated by traffic demand in
much the same manner as for an actuated phase. However, the noncoordinated
phase duration is typically constrained by its force-off setting (rather than a
maximum green setting). A noncoordinated phase is referred to here and
modeled as an actuated phase.
Right-Turn Overlap Duration

If a right-turn lane group is operated in a protected or protected-permitted
mode, then the protected indication is assumed to be provided as a right-turn
overlap with the complementary left-turn phase on the intersecting roadway. In
this manner, the right-turn protected interval duration is dictated by the duration
of the complementary left-turn phase (which is determined by the left-turn phase
settings, left-turn detection, and left-turn volume). The procedures described in
this subsection are used to determine the average duration of the complementary
left-turn lane phase (and thus the protected right-turn interval duration).
The right-turn permitted interval duration is dictated by the phase settings,
detection, and volume associated with the right-turn movement and its adjacent
through movement. The procedures described in this subsection are used to
determine the average duration of the phase serving the right-turn movement in
a permitted manner.
Volume Computations
This subsection describes the calculations needed to quantify the time rate of
calls submitted to the controller by the detectors. Two call rates are computed for
each signal phase. The first rate represents the flow rate of calls for green
extension that arrive during the green interval. The second call rate represents
the flow rate of calls for phase activation that arrive during the red indication.

A. Call Rate to Extend Green

The call rate to extend the green indication for a given phase is based on the
flow rate of the lane groups served by the phase. The call rate is represented in
the analysis by the flow rate parameter. This parameter represents an adjusted
flow rate that accounts for the tendency of drivers to form “bunches” (i.e.,
randomly formed platoons). The flow rate parameter for the phase is computed
as shown by Equation 31-3 with Equation 31-4 and Equation 31-5.
𝑚
∗
𝜆 = ∑ 𝜆𝑖 Equation 31-3
𝑖=1
with
𝜑𝑖 𝑞𝑖
𝜆𝑖 = Equation 31-4
1 − 𝛥𝑖 𝑞𝑖
𝜑𝑖 = 𝑒 −𝑏𝑖 𝛥𝑖 𝑞𝑖 Equation 31-5
where
λ* = flow rate parameter for the phase (veh/s);
λi = flow rate parameter for lane group i (i = 1, 2, . . . , m) (veh/s);
ϕi = proportion of free (unbunched) vehicles in lane group i (decimal);
qi = arrival flow rate for lane group i = vi/3,600 (veh/s);
vi = demand flow rate for lane group i (veh/h);
Δi = headway of bunched vehicle stream in lane group i; = 1.5 s for single-
lane lane group, 0.5 s otherwise (s/veh);
m = number of lane groups served during the phase; and
bi = bunching factor for lane group i (0.6, 0.5, and 0.8 for lane groups with
1, 2, and 3 or more lanes, respectively).
Using Equation 31-6, Equation 31-7, and Equation 31-8, it is also useful to
compute the following three variables for each phase. These variables are used in
a later step to compute green extension time.
𝑚
𝜑 ∗ = 𝑒 − ∑𝑖=1 𝑏𝑖 𝛥𝑖 𝑞𝑖 Equation 31-6
∑𝑚
𝑖=1 𝜆𝑖 𝛥𝑖
𝛥∗ = Equation 31-7
𝜆∗
𝑚
∗
𝑞 = ∑ 𝑞𝑖
Equation 31-8
𝑖=1
where
ϕ* = combined proportion of free (unbunched) vehicles for the phase
(decimal),
Δ* = equivalent headway of bunched vehicle stream served by the phase
(s/veh), and
q* = arrival flow rate for the phase (veh/s), and

all other variables are as previously defined.

The call rate for green extension for a phase that does not end at a barrier is
equal to the flow rate parameter λ*. If two phases terminate at a common barrier
(i.e., one phase in each ring) and simultaneous gap-out is enabled, then the call
rate for either phase is based on the combined set of lane groups being served by
the two phases. To model this behavior, the lane group parameters for each
phase are combined to estimate the call rate for green extension. Specifically, the
variable m in the preceding six equations is modified to represent the combined
number of lane groups served by both phases.
The following rules are evaluated to determine the number of lane groups
served m if simultaneous gap-out is enabled. They are described for the case in
which Phases 2, 6, 4, and 8 end at the barrier (as shown in Exhibit 19-2). The rules
should be modified if other phase pairs end at the barrier.
1. If Phases 2 and 6 have simultaneous gap-out enabled, then the lane groups
associated with Phase 2 are combined with the lane groups associated
with Phase 6 in applying Equation 31-3 through Equation 31-8 for Phase 6.
Similarly, the lane groups associated with Phase 6 are combined with the
lane groups associated with Phase 2 in applying these equations for
Phase 2.
2. If Phases 4 and 8 have simultaneous gap-out enabled, then the lane groups
associated with Phase 4 are combined with the lane groups associated
with Phase 8 in evaluating Phase 8. Similarly, the lane groups associated
with Phase 8 are combined with the lane groups associated with Phase 4
in evaluating Phase 4.
B. Call Rate to Activate a Phase

The call rate to activate a phase is used to determine the probability that the
phase is activated in the forthcoming cycle sequence. This rate is based on the
arrival flow rate of the traffic movements served by the phase and whether the
phase is associated with dual entry. Vehicles or pedestrians can call a phase, so a
separate call rate is computed for each traffic movement.
i. Determine Phase Vehicular Flow Rate. The vehicular flow rate associated with
a phase depends on the type of movements it serves as well as the approach lane
allocation. The following rules apply in determining the phase vehicular flow rate:
1. If the phase exclusively serves a left-turn movement, then the phase
vehicular flow rate is equal to the left-turn movement flow rate.
2. If the phase serves a through or right-turn movement and there is no
exclusive left-turn phase for the adjacent left-turn movement, then the
phase vehicular flow rate equals the approach flow rate.
3. If the phase serves a through or right-turn movement and there is an
exclusive left-turn phase for the adjacent left-turn movement, then
a. If there is a left-turn bay, then the phase vehicular flow rate equals the
sum of the through and right-turn movement flow rates.
b. If there is no left-turn bay, then the phase vehicular flow rate equals
the approach flow rate.

c. If split phasing is used, then the phase vehicular flow rate equals the
approach flow rate.
ii. Determine Activating Vehicular Call Rate. The activating vehicular call rate qv*
is equal to the phase vehicular flow rate divided by 3,600 to convert it to units of
vehicles per second. If dual entry is activated for a phase, then the activation call
rate must be modified by adding its original rate to that of both concurrent
phases. For example, if Phase 2 is set for dual entry, then the modified Phase 2
activation call rate equals the original Phase 2 activation call rate plus the
activation rate of Phase 5 and the activation rate of Phase 6. In this manner,
Phase 2 is activated when demand is present for Phase 2, 5, or 6.
iii. Determine Activating Pedestrian Call Rate. The activating pedestrian call
rate qp* is equal to the pedestrian flow rate associated with the subject approach
divided by 3,600 to convert it to units of pedestrians per second. If dual entry is
activated for a phase, then the activation call rate must be modified by adding its
original rate to that of the opposing through phase. For example, if Phase 2 is set
for dual entry, then the modified Phase 2 activation call rate equals the original
Phase 2 activation call rate plus the activation rate of Phase 6. In this manner,
Phase 2 is activated when pedestrian demand is present for Phase 2 or 6.
Queue Accumulation Polygon

This subsection summarizes the procedure used to construct the QAP
associated with a lane group. This polygon defines the queue size for a traffic
movement as a function of time during the cycle. The procedure is described
more fully in Section 3; it is discussed here to illustrate its use in calculating
queue service time.
For polygon construction, all flow rate variables are converted to common
units of vehicles per second per lane. The presentation in this subsection is based
on these units for q and s. If the flow rate q exceeds the lane capacity, then it is set
to equal this capacity.
A polygon is shown in Exhibit 31-1 for a through movement in an exclusive
lane. At the start of the effective red, vehicles arrive at a rate of qr and accumulate
to a length of Qr vehicles at the time the effective green begins. Thereafter, the
queue begins to discharge at a rate of s – qg until it clears after gs seconds. The
queue service time gs represents the time required to serve the queue present at
the end of effective red Qr plus any additional arrivals that join the queue before
it fully clears. Queue service time is computed as Qr/(s – qg). Substituting the
variable relationships in the previous variable list into this equation yields
Equation 31-9 for estimating queue service time.
𝑞 𝐶 (1 − 𝑃)
𝑔𝑠 = 𝑠 Equation 31-9
3,600 − 𝑞 𝐶 (𝑃/𝑔)
where P is the proportion of vehicles arriving during the green indication
(decimal), s is the adjusted saturation flow rate (veh/h/ln), and all other variables
are as previously defined.

The polygon in Exhibit 31-1 applies to some types of lane groups. Other
polygon shapes are possible. A detailed procedure for constructing polygons is
described in Section 3.
Maximum Allowable Headway

This subsection describes a procedure for calculating the maximum
allowable headway (MAH) for the detection associated with a phase. It consists
of two steps. Step A computes MAH for each lane group served by the subject
phase. Step B combines MAH into an equivalent MAH for the phase. The latter
step is used when a phase serves two or more lane groups or when simultaneous
gap-out is enabled.
The procedure addresses the situation in which there is one zone of detection
per lane. This type of detection is referred to here as stop-line detection because the
detection zone is typically located at the stop line. However, some agencies
prefer to locate the detection zone at a specified distance upstream from the stop
line. This procedure can be used to evaluate any single-detector-per-lane design,
provided the detector is located so that only the subject traffic movement travels
over this detector during normal operation.
The detector length and detection mode input data are specified by
movement group. When these data describe a through movement group, it is
reasonable to assume they also describe the detection in any shared-lane lane
groups that serve the through movement. This assumption allows the movement
group inputs to describe the associated lane group values, and the analysis can
proceed on a lane-group basis. However, if this assumption is not valid or if
information about the detection design for each lane is known, then the
procedure can be extended to the calculation of MAH for each lane. The lane-
specific MAHs would then be combined for the phase that serves these lanes.
Concepts
MAH represents the maximum time that can elapse between successive calls
for service without terminating the phase by gap-out. It is useful for describing
the detection design and signal settings associated with a phase. MAH depends
on the number of detectors serving the lane group, the length of these detectors,
and the average vehicle speed in the lane group.
The relationship between passage time PT, detection zone length Lds, vehicle
length Lv, average speed Sa, and MAH is shown in Exhibit 31-2. The two vehicles
shown are traveling from left to right and have a headway equal to MAH so that
the second vehicle arrives at the detector the instant the passage time is set to
time out.
Exhibit 31-2
Detection Design and
Maximum Allowable Headway

According to Exhibit 31-2, Equation 31-10 with Equation 31-11 can be derived
for estimating MAH for stop-line detection operating in the presence mode.
𝐿𝑑𝑠 + 𝐿𝑣 Equation 31-10
𝑀𝐴𝐻 = 𝑃𝑇 +
1.47 𝑆𝑎
with
𝐿𝑣 = 𝐿𝑝𝑐 (1 − 0.01 𝑃𝐻𝑉 ) + 0.01 𝐿𝐻𝑉 𝑃𝐻𝑉 − 𝐷𝑠𝑣 Equation 31-11
where
MAH = maximum allowable headway (s/veh),
PT = passage time setting (s),
Lds = length of the stop-line detection zone (ft),
Lv = detected length of the vehicle (ft),
Sa = average speed on the intersection approach (mi/h),
Lpc = stored passenger car lane length = 25 (ft),
PHV = percentage heavy vehicles in the corresponding movement group (%),
LHV = stored heavy-vehicle lane length = 45 (ft), and
Dsv = distance between stored vehicles = 8 (ft).
The average speed on the intersection approach can be estimated with
Equation 31-12.
𝑆𝑎 = 0.90 (25.6 + 0.47 𝑆𝑝𝑙 ) Equation 31-12
where Spl is the posted speed limit (mi/h).

Equation 31-10 is derived for the typical case in which the detection unit is
operating in the presence mode. If it is operating in the pulse mode, then MAH
equals the passage time setting PT.
A. Determine Maximum Allowable Headway

Equation 31-10 has been modified to adapt it to various combinations of lane
use and left-turn operation. A family of equations is presented in this step. The
appropriate equation is selected for the subject lane group and then used to
compute the corresponding MAH.
The equations presented in this step are derived for the typical case in which
the detection unit is operating in the presence mode. If a detector is operating in
the pulse mode, then MAH equals the passage time setting PT.
MAH for lane groups serving through vehicles is calculated with Equation
31-13.
𝐿𝑑𝑠,𝑡ℎ + 𝐿𝑣
𝑀𝐴𝐻𝑡ℎ = 𝑃𝑇𝑡ℎ + Equation 31-13
1.47 𝑆𝑎
where
MAHth = maximum allowable headway for through vehicles (s/veh),
PTth = passage time setting for phase serving through vehicles (s),

Lds,th = length of the stop-line detection zone in the through lanes (ft), and
Sa = average speed on the intersection approach (mi/h).
MAH for a left-turn movement served in exclusive lanes with the protected
mode (or protected-permitted mode) is based on Equation 31-13, but the
equation is adjusted as shown in Equation 31-14 to account for the slower speed
of the left-turn movement.
𝐿𝑑𝑠,𝑙𝑡 + 𝐿𝑣 𝐸𝐿 − 1
Equation 31-14 𝑀𝐴𝐻𝑙𝑡,𝑒,𝑝 = 𝑃𝑇𝑙𝑡 + +
1.47 𝑆𝑎 𝑠𝑜 /3,600
where
MAHlt,e,p = maximum allowable headway for protected left-turning vehicles in
exclusive lane (s/veh),
PTlt = passage time setting for phase serving the left-turning vehicles (s),
Lds,lt = length of the stop-line detection zone in the left-turn lanes (ft),
EL = equivalent number of through cars for a protected left-turning
vehicle = 1.05, and
so = base saturation flow rate (pc/h/ln).
MAH for left-turning vehicles served in a shared lane with the protected-
permitted mode is calculated as shown in Equation 31-15.
𝐸𝐿 − 1
Equation 31-15 𝑀𝐴𝐻𝑙𝑡,𝑠,𝑝 = 𝑀𝐴𝐻𝑡ℎ +
𝑠𝑜 /3,600
where MAHlt,s,p is the maximum allowable headway for protected left-turning
vehicles in a shared lane (s/veh).
MAH for left-turning vehicles served in an exclusive lane with the permitted
mode is adjusted to account for the longer headway of the turning vehicle. In this
case, the longer headway includes the time spent waiting for an acceptable gap
in the opposing traffic stream. Equation 31-16 addresses these adjustments.
𝐿𝑑𝑠,𝑙𝑡 + 𝐿𝑣 3,600
Equation 31-16 𝑀𝐴𝐻𝑙𝑡,𝑒 = 𝑃𝑇𝑡ℎ + + − 𝑡𝑓ℎ
1.47 𝑆𝑎 𝑠𝑙
where
MAHlt,e = maximum allowable headway for permitted left-turning vehicles in
sl = saturation flow rate in exclusive left-turn lane group with permitted
operation (veh/h/ln), and
tfh = follow-up headway = 2.5 (s).
MAH for right-turning vehicles served in an exclusive lane with the
protected mode is computed with Equation 31-17.
𝐿𝑑𝑠,𝑟𝑡 + 𝐿𝑣 𝐸𝑅 − 1
Equation 31-17 𝑀𝐴𝐻𝑟𝑡,𝑒,𝑝 = 𝑃𝑇𝑟𝑡 + +
1.47 𝑆𝑎 𝑠𝑜 /3,600

where
MAHrt,e,p = maximum allowable headway for protected right-turning vehicles in
PTrt = passage time setting for phase serving right-turning vehicles (s),
ER = equivalent number of through cars for a protected right-turning
vehicle = 1.18, and
Lds,rt = length of the stop-line detection zone in the right-turn lanes (ft).
If the variable ER in Equation 31-17 is divided by the pedestrian–bicycle
saturation flow rate adjustment factor fRpb and PTth is substituted for PTrt, then the
equation can be used to estimate MAHrt,e for permitted right-turning vehicles in
an exclusive lane.
Equation 31-18 and Equation 31-19, respectively, are used to estimate MAH
for left- and right-turning vehicles that are served in a shared lane with the
permitted mode.
3,600
𝑀𝐴𝐻𝑙𝑡,𝑠 = 𝑀𝐴𝐻𝑡ℎ + − 𝑡𝑓ℎ Equation 31-18
𝑠𝑙
(𝐸𝑅 /𝑓𝑅𝑝𝑏 ) − 1
𝑀𝐴𝐻𝑟𝑡,𝑠 = 𝑀𝐴𝐻𝑡ℎ + Equation 31-19
𝑠𝑜 /3,600
where MAHlt,s is the maximum allowable headway for permitted left-turning
vehicles in a shared lane (s/veh), and MAHrt,s is the maximum allowable
headway for permitted right-turning vehicles in a shared lane (s/veh).
B. Determine Equivalent Maximum Allowable Headway

The equivalent MAH (i.e., MAH*) is calculated for cases in which more than
one lane group is served by a phase. It is also calculated for phases that end at a
barrier and that are specified in the controller as needing to gap out at the same
time as a phase in the other ring. The following rules are used to compute the
equivalent MAH:
1. If simultaneous gap-out is not enabled, or the phase does not end at the
barrier, then
a. If the phase serves only one movement, then MAH* for the phase
equals the MAH computed for the corresponding lane group.
b. This rule subset applies when the phase serves all movements and
there is no exclusive left-turn phase for the approach (i.e., it operates
with the permitted mode). The equations shown apply to the most
general case in which a left-turn, through, and right-turn movement
exist and a through lane group exists. If any of these movements or
lane groups do not exist, then their corresponding flow rate
parameter equals 0.0 veh/s.
i. If there is no left-turn lane group or right-turn lane group (i.e.,
shared lanes), then MAH* for the phase is computed from
Equation 31-20.

𝑃𝐿 𝜆𝑠𝑙 𝑀𝐴𝐻𝑙𝑡,𝑠 + [(1 − 𝑃𝐿 )𝜆𝑠𝑙 + 𝜆𝑡 + (1 − 𝑃𝑅 )𝜆𝑠𝑟 ]𝑀𝐴𝐻𝑡ℎ + 𝑃𝑅 𝜆𝑠𝑟 𝑀𝐴𝐻𝑟𝑡,𝑠

Equation 31-20 𝑀𝐴𝐻∗ =
𝜆𝑠𝑙 + 𝜆𝑡 + 𝜆𝑠𝑟
where
λsl = flow rate parameter for shared left-turn and through lane group
(veh/s),
λt = flow rate parameter for exclusive through lane group (veh/s),
λsr = flow rate parameter for shared right-turn and through lane group
(veh/s),
PL = proportion of left-turning vehicles in the shared lane (decimal), and
PR = proportion of right-turning vehicles in the shared lane (decimal).
ii. If there is a right-turn lane group but no left-turn lane group, then
Equation 31-21 is applicable.
𝑃𝐿 𝜆𝑠𝑙 𝑀𝐴𝐻𝑙𝑡,𝑠 + [(1 − 𝑃𝐿 ) 𝜆𝑠𝑙 + 𝜆𝑡 ] 𝑀𝐴𝐻𝑡ℎ + 𝜆𝑟 𝑀𝐴𝐻𝑟𝑡,𝑒
𝜆𝑠𝑙 + 𝜆𝑡 + 𝜆𝑟
where λr is the flow rate parameter for the exclusive right-turn lane group (veh/s).
iii. If there is a left-turn lane group but no right-turn lane group, then
MAH* for the phase is computed with Equation 31-22.
𝜆𝑙 𝑀𝐴𝐻𝑙𝑡,𝑒 + [𝜆𝑡 + (1 − 𝑃𝑅 ) 𝜆𝑠𝑟 ] 𝑀𝐴𝐻𝑡ℎ + 𝑃𝑅 𝜆𝑠𝑟 𝑀𝐴𝐻𝑟𝑡,𝑠
Equation 31-22 𝑀𝐴𝐻 ∗ =
𝜆𝑙 + 𝜆𝑡 + 𝜆𝑠𝑟
where λl is the flow rate parameter for the exclusive left-turn lane group (veh/s).
iv. If there is a left-turn lane group and a right-turn lane group, then
𝜆𝑙 𝑀𝐴𝐻𝑙𝑡,𝑒 + 𝜆𝑡 𝑀𝐴𝐻𝑡ℎ + 𝜆𝑟 𝑀𝐴𝐻𝑟𝑡,𝑒
𝜆𝑙 + 𝜆𝑡 + 𝜆𝑟
c. If the phase serves only a through lane group, right-turn lane group,
or both, then
i. If there is a right-turn lane group and a through lane group, then
𝜆𝑡 𝑀𝐴𝐻𝑡ℎ + 𝜆𝑟 𝑀𝐴𝐻𝑟𝑡,𝑒
𝜆𝑡 + 𝜆𝑟
ii. If there is a shared right-turn and through lane group, then MAH*
for the phase is computed with Equation 31-25.
[𝜆𝑡 + (1 − 𝑃𝑅 ) 𝜆𝑠𝑟 ] 𝑀𝐴𝐻𝑡ℎ + 𝑃𝑅 𝜆𝑠𝑟 𝑀𝐴𝐻𝑟𝑡,𝑠
𝜆𝑡 + 𝜆𝑠𝑟
d. If the phase serves all approach movements using split phasing, then
i. If there is one lane group (i.e., a shared lane), then MAH* for the
phase equals the MAH computed for the lane group.

ii. If there is more than one lane group, then MAH* is computed with
the equations in previous Rule 1.b, but MAHlt,e,p is substituted for
MAHlt,e, and MAHlt,s,p is substituted for MAHlt,s.
e. If the phase has protected-permitted operation with a shared left-turn
and through lane, then the equations in previous Rule 1.b (i.e., 1.b.i
and 1.b.ii) apply. The detection for this operation does not influence
the duration of the left-turn phase. The left-turn phase will be set to
minimum recall and will extend to its minimum value before
terminating.
2. If simultaneous gap-out is enabled and the phase ends at the barrier, then
MAH* for the phase is computed with Equation 31-26, where the
summations shown are for all lane groups served by the subject (or
concurrent) phase.
𝑀𝐴𝐻𝑖 ∑ 𝜆𝑖 + 𝑀𝐴𝐻𝑐 ∑ 𝜆𝑐,𝑖 Equation 31-26
𝑀𝐴𝐻 ∗ =
∑ 𝜆𝑖 + ∑ 𝜆𝑐,𝑖
where
MAH* = equivalent maximum allowable headway for the phase (s/veh),
MAHi = equivalent maximum allowable headway computed in Step 1 for the
subject phase (s/veh),
MAHc = equivalent maximum allowable headway computed in Step 1 for the
concurrent phase that also ends at the barrier (s/veh), and
λc,i = flow rate parameter for lane group i served in the concurrent phase
that also ends at the barrier (veh/s).
When there is split phasing, there are no concurrent phases, and Equation
31-26 does not apply.
Equivalent Maximum Green

In coordinated-actuated operation, the force-off points are used to constrain
the duration of the noncoordinated phases. Although the maximum green setting
is also available to provide additional constraint, it is not commonly used. In fact,
the default mode in most modern controllers is to inhibit the maximum green
timer when the controller is used in a coordinated signal system.
The relationship between the force-off points, yield point, and phase splits is
shown in Exhibit 31-3. The yield point is associated with the coordinated phases
(i.e., Phases 2 and 6). It coincides with the start of the yellow change interval. If a
call for service by one of the noncoordinated phases arrives after the yield point
is reached, then the coordinated phases begin the termination process by
presenting the yellow indication. Calls that arrive before the yield point are not
served until the yield point is reached.
The force-off and yield points for common phase pairs are shown in Exhibit
31-3 to occur at the same time. This approach is shown for convenience of
illustration. In practice, the two phases may have different force-off or yield points.

A permissive period typically follows the yield point. If a conflicting call

arrives during the permissive period, then the phase termination process begins
immediately, and all phases associated with conflicting calls are served in
sequence. Permissive periods are typically long enough to ensure that all calls for
service are met during the signal cycle. This methodology does not explicitly
model permissive periods. It is assumed the permissive period begins at the
yield point and is sufficiently long that all conflicting calls are served in sequence
each cycle.
One force-off point is associated with each of Phases 1, 3, 4, 5, 7, and 8. If a
phase is extended to its force-off point, the phase begins the termination process
by presenting the yellow indication (phases that terminate at a barrier must be in
agreement to terminate before the yellow indication will be presented). Modern
controllers compute the force-off points and yield point by using the entered
phase splits and change periods. These computations are based on the
relationships shown in Exhibit 31-3.
Exhibit 31-3
Force-Off Points, Yield Point,
and Phase Splits
The concept of equivalent maximum green is useful for modeling

noncoordinated phase operation. This maximum green replicates the effect of a
force-off or yield point on phase duration. The procedure described in this
subsection is used to compute the equivalent maximum green for coordinated-
actuated operation. Separate procedures are described for the fixed force mode
and the floating force mode.
A. Determine Equivalent Maximum Green for Floating Force Mode

This step is applicable if the controller is set to operate in the floating force
mode. With this mode, each noncoordinated phase has its force-off point set at
the split time after the phase first becomes active. The force-off point for a phase

is established when the phase is first activated. Thus, the force-off point “floats,”
or changes, each time the phase is activated. This operation allows unused split
time to revert to the coordinated phase via an early return to green. The
equivalent maximum green for this mode is computed as being equal to the
phase split less the change period. This relationship is shown in Exhibit 31-3 for
Phases 4 and 8.
B. Determine Equivalent Maximum Green for Fixed Force Mode

This step is applicable if the controller is set to operate in the fixed force
mode. With this mode, each noncoordinated phase has its force-off point set at a
fixed time in the cycle relative to time zero on the system master. The force-off
points are established whenever a new timing plan is selected (e.g., by time of
day) and remains “fixed” until a new plan is selected. This operation allows
unused split time to revert to the following phase.
The equivalent maximum green for this mode is computed for each phase by
first establishing the fixed force-off points (as shown in Exhibit 31-3) and then
computing the average duration of each noncoordinated phase. The calculation
process is iterative. For the first iteration, the equivalent maximum green is set
equal to the phase split less the change period. Thereafter, the equivalent
maximum green for a specific phase is computed as the difference between its
force-off point and the sum of the previous phase durations, starting with the
first noncoordinated phase. Equation 31-27 illustrates this computation for Phase
4, using the ring structure shown in Exhibit 19-2. A similar calculation is
performed for the other phases.
𝐺𝑚𝑎𝑥,4 = 𝐹𝑂4 − (𝑌𝑃2 + 𝐶𝑃2 + 𝐺3 + 𝐶𝑃3) Equation 31-27
where
Gmax,4 = equivalent maximum green for Phase 4 (s),
FO4 = force-off point for Phase 4 (s),
YP2 = yield point for Phase 2 (s),
G3 = green interval duration for Phase 3 (s), and
CP3 = change period (yellow change interval plus red clearance interval) for
Phase 3 (s).
The maximum green obtained from Equation 31-27 is shown in Exhibit 31-4
for the ring that serves Phases 1, 2, 3, and 4. Unlike Exhibit 31-3, Exhibit 31-4
illustrates the actual average phase durations for a given cycle. In this example,
Phase 3 timed to its minimum green and terminated. It never reached its force-off
point. The unused time from Phase 3 was made available to Phase 4, which
resulted in a larger maximum green than was obtained with the floating mode
(see Exhibit 31-3). If every noncoordinated phase extends to its force-off point,
then the maximum green from the fixed force mode equals that obtained from
the floating force mode.

Average Phase Duration

This subsection describes the sequence of calculations needed to estimate the
average duration of a phase. In fact, the process requires the combined
calculation of the duration of all phases together because of the constraints
imposed by the controller ring structure and associated barriers.
The calculation process is iterative because several intermediate equations
require knowledge of the green interval duration. Specifically, the green interval
duration is required in calculating lane group flow rate, queue service time,
permitted green time, left-turn volume served during the permitted portion of a
protected-permitted mode, and equivalent maximum green. To overcome this
circular dependency, the green interval for each phase is initially estimated, and
then the procedure is implemented by using this estimate. When completed, the
procedure provides a new initial estimate of the green interval duration. The
calculations are repeated until the initial estimate and computed green interval
duration are effectively equal.
Exhibit 31-4
Example Equivalent Maximum
Green for Fixed Force Mode
The calculation steps that constitute the procedure are described in the
following paragraphs.
A. Compute Effective Change Period

The change period is computed for each phase. It is equal to the sum of the
yellow change interval and the red clearance interval (i.e., Y + Rc). For phases that
end at a barrier, the longer change period of the two phases that terminate at a
barrier is used to define the effective change period for both phases.

B. Estimate Green Interval

An initial estimate of the green interval duration is provided for each phase.
For the first iteration with fully actuated control, the initial estimate is equal to
the maximum green setting. For the first iteration with coordinated-actuated
control, the initial estimate is equal to the input phase split less the change period.
C. Compute Equivalent Maximum Green (Coordinated-Actuated)

If the controller is operating as coordinated-actuated, then the equivalent
maximum green is computed for each phase. It is based on the estimated green
interval duration, phase splits, and change periods. The previous subsection
titled Equivalent Maximum Green describes how to compute this value.
D. Construct Queue Accumulation Polygon

The QAP is constructed for each lane group and corresponding phase by
using the known flow rates and signal timing. The procedure for constructing
this polygon is summarized in the previous subsection titled Queue
Accumulation Polygon. It is described in more detail in Section 3.
E. Compute Queue Service Time

The queue service time gs is computed for each QAP constructed in the
previous step. For through movements or left-turn movements served during a
left-turn phase, the polygon in Exhibit 31-1 applies and Equation 31-9 can be
used. The procedure described in Section 3 is applicable to more complicated
polygon shapes.
F. Compute Call Rate to Extend Green

The extending call rate is represented as the flow rate parameter λ. This
parameter is computed for each lane group served by an actuated phase and is
then aggregated to a phase-specific value. The procedure for computing this
parameter is described in the previous subsection titled Volume Computations.
G. Compute Equivalent Maximum Allowable Headway

The equivalent maximum allowable headway MAH* is computed for each
actuated phase. The procedure for computing MAH* is described in the previous
subsection titled Maximum Allowable Headway.
H. Compute Number of Extensions Before Max-Out

The average number of extensions before the phase terminates by max-out is
computed for each actuated phase with Equation 31-28.
𝑛 = 𝑞∗ [𝐺𝑚𝑎𝑥 − (𝑔𝑠 + 𝑙1 )] ≥ 0.0 Equation 31-28
where n is the number of extensions before the green interval reaches its
maximum limit, Gmax is the maximum green setting (s), (gs + l1) is the maximum
sum of queue service time and start-up lost time from all lane groups of the
subject phase, and all other variables are as previously defined.

I. Compute Probability of Green Extension

The probability of the green interval being extended by randomly arriving
vehicles is computed for each actuated phase with Equation 31-29.
∗ ∗ −𝛥∗ )
Equation 31-29 𝑝 = 1 − 𝜑 ∗ 𝑒 −𝜆 (𝑀𝐴𝐻
where p is the probability of a call headway being less than the maximum
allowable headway.
J. Compute Green Extension Time

The average green extension time is computed for each actuated phase with
Equation 31-30.
𝑝2 (1 − 𝑝𝑛 )
Equation 31-30 𝑔𝑒 =
𝑞∗ (1 − 𝑝)
K. Compute Activating Call Rate

The call rate to activate a phase is computed for each actuated phase. A
separate rate is computed for vehicular traffic and for pedestrian traffic. The rate
for each travel mode is based on its flow rate and the use of dual entry. The
procedure for computing this rate is described in the previous subsection titled
Volume Computations.
L. Compute Probability of Phase Call

The probability that an actuated phase is called depends on whether it is set
on recall in the controller. If it is on recall, then the probability that the phase is
called equals 1.0. If the phase is not on recall, then the probability that it is called
can be estimated by using Equation 31-31 with Equation 31-32 and Equation 31-33.
Equation 31-31 𝑝𝑐 = 𝑝𝑣 (1 − 𝑝𝑝 ) + 𝑝𝑝 (1 − 𝑝𝑣 ) + 𝑝𝑣 𝑝𝑝
with
∗
Equation 31-32 𝑝𝑣 = 1 − 𝑒 −𝑞𝑣 𝐶
∗
Equation 31-33 𝑝𝑝 = 1 − 𝑒 −𝑞𝑝 𝑃𝑝 𝐶
where
pc = probability that the subject phase is called,
pv = probability that the subject phase is called by a vehicle detection,
pp = probability that the subject phase is called by a pedestrian detection,
qv* = activating vehicular call rate for the phase (veh/s),
qp* = activating pedestrian call rate for the phase (p/s), and
Pp = probability of a pedestrian pressing the detector button = 0.51.
The probability of a pedestrian pressing the detector button reflects the
tendency of some pedestrians to decline from using the detector button before
crossing a street. Research indicates about 51% of all crossing pedestrians will
push the button to place a call for pedestrian service (1).

M. Compute Unbalanced Green Duration

The unbalanced average green interval duration is computed for each
actuated phase by using Equation 31-34 with Equation 31-35 and Equation 31-36.
𝐺𝑢 = 𝐺|𝑣𝑒ℎ,𝑐𝑎𝑙𝑙 𝑝𝑣 (1 − 𝑝𝑝 ) + 𝐺|𝑝𝑒𝑑,𝑐𝑎𝑙𝑙 𝑝𝑝 (1 − 𝑝𝑣 ) + Equation 31-34
max(𝐺|𝑣𝑒ℎ,𝑐𝑎𝑙𝑙 , 𝐺|𝑝𝑒𝑑,𝑐𝑎𝑙𝑙 ) 𝑝𝑣 𝑝𝑝 ≤ 𝐺𝑚𝑎𝑥

with
𝐺|𝑣𝑒ℎ,𝑐𝑎𝑙𝑙 = max(𝑙1 + 𝑔𝑠 + 𝑔𝑒 , 𝐺𝑚𝑖𝑛 ) Equation 31-35
𝐺|𝑝𝑒𝑑,𝑐𝑎𝑙𝑙 = 𝑊𝑎𝑙𝑘 + 𝑃𝐶 Equation 31-36
where
Gu = unbalanced green interval duration for a phase (s),
G|veh,call = average green interval given that the phase is called by a vehicle
detection (s),
Gmin = minimum green setting (s),
G|ped,call = average green interval given that the phase is called by a pedestrian
detection (s),
Walk = pedestrian walk setting (s), and
PC = pedestrian clear setting (s).
If maximum recall is set for the phase, then Gu is equal to Gmax. If the phase
serves a left-turn movement that operates in the protected mode, then the
probability that it is called by pedestrian detection pp is equal to 0.0.
If the phase serves a left-turn movement that operates in the protected-
permitted mode and the left-turn movement shares a lane with through vehicles,
then the green interval duration is equal to the phase’s minimum green setting.
The green interval duration obtained from this step is “unbalanced” because
it does not reflect the constraints imposed by the controller ring structure and
associated barriers. These constraints are imposed in Step O or Step P, depending
on the type of control used at the intersection.
It is assumed the rest-in-walk mode is not enabled.
N. Compute Unbalanced Phase Duration

The unbalanced average phase duration is computed for each actuated phase
by adding the unbalanced green interval duration and the corresponding change
period components. This calculation is completed with Equation 31-37.
𝐷𝑢𝑝 = 𝐺𝑢 + 𝑌 + 𝑅𝑐 Equation 31-37
where Dup is the unbalanced phase duration (s).

If simultaneous gap-out is enabled, the phase ends at a barrier, and the
subject phase experiences green extension when the concurrent phase has
reached its maximum green limit, then both phases are extended, but only due to
the call flow rate of the subject phase. Hence, the green extension time computed
in Step J is too long. The effect is accounted for in the current step by multiplying

the green extension time from Step J by a “flow rate ratio.” This ratio represents
the sum of the flow rate parameter for each lane group served by the subject
phase divided by the sum of the flow rate parameter for each group served by
the subject phase and served by the concurrent phase (the latter sum equals the
call rate from Step F).
O. Compute Average Phase Duration—Fully Actuated Control

For this discussion, it is assumed Phases 2 and 6 are serving Movements 2
and 6, respectively, on the major street (see Exhibit 19-2). If the left-turn
movements on the major street operate in the protected mode or the protected-
permitted mode, then Movements 1 and 5 are served during Phases 1 and 5,
respectively. Similarly, Phases 4 and 8 are serving Movements 4 and 8,
respectively, on the minor street. If the left-turn movements on the minor street
are protected or protected-permitted, then Phases 3 and 7 are serving
Movements 3 and 7, respectively. If a through movement phase occurs first in a
phase pair, then the other phase (i.e., the one serving the opposing left-turn
movement) is a lagging left-turn phase.
The following rules are used to estimate the average duration of each phase:
1. Given two phases that occur in sequence between barriers (i.e., phase a
followed by phase b), the duration of Dp,a is equal to the unbalanced phase
duration of the first phase to occur (i.e., Dp,a = Dup,a). The duration of Dp,b is
based on Equation 31-38 for the major-street phases.
Equation 31-38 𝐷𝑝,𝑏 = max(𝐷𝑢𝑝,1 + 𝐷𝑢𝑝,2 , 𝐷𝑢𝑝,5 + 𝐷𝑢𝑝,6 ) − 𝐷𝑝,𝑎
where
Dp,b = phase duration for phase b, which occurs just after phase a (s);
Dp,a = phase duration for phase a, which occurs just before phase b (s); and
Dup,i = unbalanced phase duration for phase i; i = 1, 2, 5, and 6 for major street,
and i = 3, 4, 7, and 8 for minor street (s).
Equation 31-39 applies for the minor-street phases.
Equation 31-39 𝐷𝑝,𝑏 = max(𝐷𝑢𝑝,3 + 𝐷𝑢𝑝,4 , 𝐷𝑢𝑝,7 + 𝐷𝑢𝑝,8 ) − 𝐷𝑝,𝑎
For example, if the phase pair consists of Phase 3 followed by Phase 4 (i.e.,
a leading left-turn arrangement), then Dp,3 is set to equal Dup,3 and Dp,4 is
computed from Equation 31-39. In contrast, if the pair consists of Phase 8
followed by Phase 7 (i.e., a lagging left-turn arrangement), then Dp,8 is set
to equal Dup,8 and Dp,7 is computed from Equation 31-39.
2. If an approach is served with one phase operating in the permitted mode
(but not split phasing), then Dp,a equals 0.0, and the equations above are
used to estimate the duration of the phase (i.e., Dp,b).
3. If split phasing is used, then Dp,a equals the unbalanced phase duration for
one approach and Dp,b equals the unbalanced phase duration for the other
approach.

P. Compute Average Phase Duration—Coordinated-Actuated Control

For this discussion, it is assumed Phases 2 and 6 are the coordinated phases
serving Movements 2 and 6, respectively (see Exhibit 19-2). If the left-turn
movements operate in the protected mode or the protected-permitted mode, then
the opposing left-turn movements are served during Phases 1 and 5. If a
coordinated phase occurs first in the phase pair, then the other phase (i.e., the
one serving the opposing left-turn movement) is a lagging left-turn phase.
The following rules are used to estimate the average duration of each phase:
1. If the phase is associated with the street serving the coordinated
movements, then
a. If a left-turn phase exists for the subject approach, then its duration
Dp,l equals Dup,l, and the opposing through phase has a duration Dp,t,
which is calculated by using Equation 31-40.
𝐷𝑝,𝑡 = 𝐶 − max(𝐷𝑢𝑝,3 + 𝐷𝑢𝑝,4 , 𝐷𝑢𝑝,7 + 𝐷𝑢𝑝,8 ) − 𝐷𝑝,𝑙 Equation 31-40
where Dp,t is the phase duration for coordinated phase t (t = 2 or 6) (s),

Dp,l is the phase duration for left-turn phase l (l = 1 or 5) (s), and all
other variables are as previously defined.
If Equation 31-40 is applied to Phase 2, then t equals 2 and l equals 1.
If it is applied to Phase 6, then t equals 6 and l equals 5.
b. If a left-turn phase does not exist for the subject approach, then Dp,l
equals 0.0, and Equation 31-40 is used to estimate the duration of the
coordinated phase.
This procedure for determining average phase duration accommodates
split phasing only on the street that does not serve the coordinated
movements.
If Dp,t obtained from Equation 31-40 is less than the minimum phase
duration (= Gmin + Y + Rc), then the phase splits are too generous and do
not leave adequate time for the coordinated phases.
2. If the phase is associated with the street serving the noncoordinated
movements, then the rules described in Step O are used to determine the
phase’s average duration.
Q. Compute Green Interval Duration

The average green interval duration is computed for each phase by subtracting
the yellow change and red clearance intervals from the average phase duration.
𝐺 = 𝐷𝑝 − 𝑌 − 𝑅𝑐 Equation 31-41
where G is the green interval duration (s).
R. Compare Computed and Estimated Green Interval Durations

If the intersection is semiactuated or fully actuated, then the equilibrium
cycle length is computed with Equation 31-42.

Equation 31-42 𝐶𝑒 = ∑ 𝐷𝑝,𝑖

𝑖=1
where Ce is the equilibrium cycle length (s) and i is the phase number. The sum in
this equation includes all phases in Ring 1. The equilibrium cycle length is used
in all subsequent calculations in which cycle length C is an input variable.
The green interval duration from the previous step is compared with the
value estimated in Step B. If the two values differ by 0.1 s or more, then the
computed green interval becomes the new initial estimate, and the sequence of
calculations is repeated starting with Step C. This process is repeated until the
two green intervals differ by less than 0.1 s. The equilibrium cycle length,
computed with Equation 31-42, is used for subsequent iterations of determining
green interval durations.
Probability of Max-Out
When the green indication is extended to its maximum green limit, the
associated phase is considered to have terminated by max-out. The probability of
max-out provides useful information about phase performance. When max-out
occurs, the phase ends without consideration of whether the queue is served or
vehicles are in the dilemma zone. Hence, a phase that frequently terminates by
max-out may have inadequate capacity and may be associated with more
frequent rear-end crashes.
The probability of max-out can be equated to the joint probability of there
being a sequence of calls to the phase in service, each call having a headway that
is shorter than the equivalent maximum allowable headway for the phase. This
probability can be stated mathematically by using Equation 31-43 with Equation
31-44 and Equation 31-45.
Equation 31-43 𝑝𝑥 = 𝑝𝑛𝑥
with
𝐺𝑚𝑎𝑥 − 𝑀𝐴𝐻 ∗ − (𝑔𝑠 + 𝑙1 )
Equation 31-44 𝑛𝑥 = ≥ 0.0
ℎ
∗ ∗ ∗
𝛥∗ + (𝜑∗ /𝜆∗ ) − (𝑀𝐴𝐻 ∗ + [1/𝜆∗ ])𝜑∗ 𝑒 −𝜆 (𝑀𝐴𝐻 −𝛥 )
Equation 31-45 ℎ= ∗ ∗ ∗
1 − 𝜑 ∗ 𝑒 −𝜆 (𝑀𝐴𝐻 −𝛥 )
where
px = probability of phase termination by extension to the maximum green
limit,
h = average call headway for all calls with headways less than MAH* (s),
and
nx = number of calls necessary to extend the green to max-out.

LANE GROUP FLOW RATE ON MULTIPLE-LANE APPROACHES

Introduction
When drivers approach an intersection, their primary criterion for lane choice
is movement accommodation (i.e., left, through, or right). If multiple exclusive
lanes are available to accommodate their movement, they tend to choose the lane
that minimizes their service time (i.e., the time required to reach the stop line, as
influenced by the number and type of vehicles between them and the stop line).
This criterion tends to result in relatively equal lane use under most circumstances.
If one of the lanes being considered is a shared lane, then service time is
influenced by the distribution of turning vehicles in the shared lane. Turning
vehicles tend to have a longer service time because of the turn maneuver.
Moreover, when turning vehicles operate in the permitted mode, their service
time can be lengthy because of the gap search process.
Observation of driver lane-choice behavior indicates there is an equilibrium
lane flow rate that characterizes the collective choices of the population of
drivers. Research indicates the equilibrium flow rate can be estimated from the
lane volume distribution that yields the minimum service time for the
population of drivers having a choice of lanes (2).
A model for predicting the equilibrium lane flow rate on an intersection
approach is described in this subsection. The model is based on the principle that
through drivers will choose the lane that minimizes their perceived service time.
As a result of this lane selection process, each lane will have the same minimum
service time. The principle is represented mathematically by (a) defining service
time for each lane as the product of lane flow rate and saturation headway, (b)
representing this product as the lane demand–to–saturation flow rate ratio (i.e.,
v/s ratio), and (c) making the v/s ratios equal among alternative approach lanes.
Equation 31-46 is derived from this representation.
𝑁𝑡ℎ
𝑣𝑖 ∑𝑖=1 𝑣𝑖
= 𝑁𝑡ℎ Equation 31-46
𝑠𝑖 ∑ 𝑠𝑖
𝑖=1
where
vi = demand flow rate in lane i (veh/h/ln),
si = saturation flow rate in lane i (veh/h/ln), and
The “equalization of flow ratios” principle has been embodied in the HCM
since the 1985 edition. Specifically, it has been used to derive the equation for
estimating the proportion of left-turning vehicles in a shared lane PL.
During field observations of various intersection approaches, it was noted
that the principle overestimated the effect of turning vehicles in shared lanes for
very low and for very high approach flow-rate conditions (3). Under low flow-
rate conditions, it was rationalized that through drivers are not motivated to
change lanes because the frequency of turns is very low and the threat of delay is
negligible. Under high flow-rate conditions, it was rationalized that through
drivers do not have an opportunity to change lanes because of the lack of

adequate gaps in the outside lane. The field observations also indicated that most
lane choice decisions (and related lane changes) for through drivers tended to
occur upstream of the intersection, before deceleration occurs.
As a result of these field observations (3), the model was extended to include
the probability of a lane change. The probability of a lane change represents the
joint probability of there being motivation (i.e., moderate to high flow rates) and
opportunity (i.e., adequate lane-change gaps). A variable that is common to each
probability distribution is the ratio of the approach flow rate to the maximum
flow rate that would allow any lane changes. This maximum flow rate is the rate
corresponding to the minimum headway considered acceptable for a lane change
(i.e., about 3.7 s) (4). Exhibit 31-5 illustrates the modeled relationship between
lane change probability and the flow ratio in the traffic lanes upstream of the
intersection, before deceleration occurs (3).
Exhibit 31-5 1.2

Probability of a Lane Change
Probability of a Lane Change Decreasing Decreasing

Motivation Opportunity
1.0
0.8
0.6
0.4
0.2
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Lane Change Flow Ratio
Procedure
The procedure described in this subsection is generalized so it can be applied
to any signalized intersection approach with any combination of exclusive turn
lanes, shared lanes, and exclusive through lanes. At least one shared lane must
be present, and the approach must have two or more lanes (or bays) serving two
or more traffic movements. This type of generalized formulation is attractive
because of its flexibility; however, the trade-off is that the calculation process is
iterative. If a closed-form solution is desired, then one would likely have to be
uniquely derived for each lane assignment combination.
The procedure is described in the following steps. Input variables used in the
procedure are identified in the following list and are shown in Exhibit 31-6:
Nl = number of lanes in exclusive left-turn lane group (ln),
Nsl = number of lanes in shared left-turn and through lane group (ln),
Nt = number of lanes in exclusive through lane group (ln),
Nsr = number of lanes in shared right-turn and through lane group (ln),
Nr = number of lanes in exclusive right-turn lane group (ln),

Nlr = number of lanes in shared left- and right-turn lane group (ln),
vlt = left-turn demand flow rate (veh/h),
vth = through demand flow rate (veh/h),
vrt = right-turn demand flow rate (veh/h),
vl = demand flow rate in exclusive left-turn lane group (veh/h/ln),
vsl = demand flow rate in shared left-turn and through lane group (veh/h),
vt = demand flow rate in exclusive through lane group (veh/h/ln),
vsr = demand flow rate in shared right-turn and through lane group (veh/h),
vr = demand flow rate in exclusive right-turn lane group (veh/h/ln),
vlr = demand flow rate in shared left- and right-turn lane group (veh/h),
vsl,lt = left-turn flow rate in shared lane group (veh/h/ln),
vsr,rt = right-turn flow rate in shared lane group (veh/h/ln),
sl = saturation flow rate in exclusive left-turn lane group with permitted
operation (veh/h/ln),
ssl = saturation flow rate in shared left-turn and through lane group with
permitted operation (veh/h/ln),
st = saturation flow rate in exclusive through lane group (veh/h/ln),
ssr = saturation flow rate in shared right-turn and through lane group with
sr = saturation flow rate in exclusive right-turn lane group with permitted
operation (veh/h/ln),
slr = saturation flow rate in shared left- and right-turn lane group (veh/h/ln),
sth = saturation flow rate of an exclusive through lane (= base saturation
flow rate adjusted for lane width, heavy vehicles, grade, parking,
buses, area type, work zone presence, downstream lane blockage, and
spillback) (veh/h/ln),
gp = effective green time for permitted left-turn operation (s),
gf = time before the first left-turning vehicle arrives and blocks the shared
lane (s), and
gu = duration of permitted left-turn green time that is not blocked by an
opposing queue (s).
Each shared-lane lane group has one lane (i.e., Nsl = 1, Nsr = 1, and Nlr = 1).
Procedures for calculating gp, gf, and gu are provided in Section 3.

Exhibit 31-6 Approach 1 Approach 2

Input Variables for Lane
Group Flow Rate Procedure Movement Variables Lane Group Variables Lane Group Variables
vl sl Nl vl sl Nl
vlt vsl,lt vsl,lt
vsl ssl Nsl v s Nlr
vsr,rt lr lr
vth s th vt st Nt vr sr Nr
vrt vsr ssr Nsr

vsr,rt
Variables
vr sr Nr v = demand flow rate
s = saturation flow rate
N = number of lanes
A. Compute Modified Through-Car Equivalents

Three modified through-car equivalent factors are computed for the left-turn
movement. These factors are computed with Equation 31-47 through Equation
31-51.
Equation 31-47 𝐸𝐿,𝑚 = (𝐸𝐿 − 1)𝑃𝑙𝑐 + 1
𝐸𝐿1
Equation 31-48 𝐸𝐿1,𝑚 = ( − 1) 𝑃𝑙𝑐 + 1
𝑓𝐿𝑝𝑏
𝐸𝐿2
Equation 31-49 𝐸𝐿2,𝑚 = ( − 1) 𝑃𝑙𝑐 + 1
𝑓𝐿𝑝𝑏
with
𝑣𝑎𝑝𝑝 2
Equation 31-50
𝑃𝑙𝑐 = 1 − ([2 ] − 1) ≥ 0.0
𝑠𝑙𝑐
Equation 31-51 𝑣𝑎𝑝𝑝 =
𝑁𝑠𝑙 + 𝑁𝑡 + 𝑁𝑠𝑟
where
EL,m = modified through-car equivalent for a protected left-turning vehicle,
EL1,m = modified through-car equivalent for a permitted left-turning vehicle,
EL = equivalent number of through cars for a protected left-turning vehicle
(= 1.05),
EL2,m = modified through-car equivalent for a permitted left-turning vehicle
when opposed by a queue on a single-lane approach,
EL2 = equivalent number of through cars for a permitted left-turning vehicle
fLpb = pedestrian adjustment factor for left-turn groups,
Plc = probability of a lane change among the approach through lanes,
vapp = average demand flow rate per through lane (upstream of any turn
bays on the approach) (veh/h/ln),

slc = maximum flow rate at which a lane change can occur = 3,600/tlc
(veh/h/ln), and
tlc = critical merge headway = 3.7 (s).
The factor obtained from Equation 31-49 is applicable when permitted left-
turning vehicles are opposed by a queue on a single-lane approach. Equations for
calculating EL1 and EL2 are provided in Section 3. A procedure for calculating fLpb
is provided later in this section.
If the approach has a shared left- and right-turn lane (as shown in Approach
2 in Exhibit 31-6), then Equation 31-52 is used to compute the average demand
flow rate per lane (with Nlr = 1.0).
𝑣𝑎𝑝𝑝 = (𝑣𝑙𝑡 + 𝑣𝑟𝑡 )/𝑁𝑙𝑟 Equation 31-52
The modified through-car equivalent for permitted right-turning vehicles is

𝐸𝑅
𝐸𝑅,𝑚 = ( − 1) 𝑃𝑙𝑐 + 1 Equation 31-53
𝑓𝑅𝑝𝑏
where ER,m is the modified through-car equivalent for a protected right-turning
vehicle, fRpb is the pedestrian–bicycle adjustment factor for right-turn groups, ER
is the equivalent number of through cars for a protected right-turning vehicle
(= 1.18), and all other variables are as previously defined.
A procedure for calculating fRpb is provided later in this section.
If the opposing approach has two lanes serving through vehicles and the
inside lane serves through and left-turn vehicles, then Equation 31-54 is used to
compute the adjusted duration of permitted left-turn green time that is not
blocked by an opposing queue gu*. This variable is then used in Equation 31-59 in
replacement of the variable gu. This adjustment is intended to reflect the
occasional hesitancy of drivers to shift from the inside lane to the outside lane
during higher-volume conditions for this approach-lane geometry. In all other
cases of opposing approach-lane geometry, the variable g*u is not computed and
Equation 31-59 is used as described in the text.
𝑔𝑢∗ = 𝑔𝑢 + (𝑔diff × 𝑃𝑙𝑐 ) Equation 31-54
where
g*u = adjusted duration of permitted left-turn green time that is not blocked
by an opposing queue (s), and
gdiff = supplemental service time (s).
Equation 31-107 in Section 3 can be used to calculate gdiff.
B. Estimate Shared-Lane Lane Group Flow Rate

The procedure to estimate the shared-lane lane group flow rate requires an
initial estimate of the demand flow rate for each traffic movement in each
shared-lane lane group on the subject approach. For the shared lane serving left-
turn and through vehicles, the left-turn flow rate in the shared lane vsl,lt is initially

estimated as 0.0 veh/h, and the total lane group flow rate vsl is estimated as equal
to the average flow rate per through lane vapp. For the shared lane serving right-
turn vehicles, the right-turn flow rate in the shared lane vsr,rt is estimated as 0.0
veh/h, and the total lane group flow rate vsr is estimated as equal to the average
flow rate per through lane vapp. These estimates are updated in a subsequent step.
C. Compute Exclusive Lane-Group Flow Rate

The demand flow rate in the exclusive left-turn lane group vl is computed
with Equation 31-55, where all variables are as previously defined.
𝑣𝑙𝑡 − 𝑣𝑠𝑙,𝑙𝑡
Equation 31-55
𝑣𝑙 = ≥ 0.0
𝑁𝑙
A similar calculation is completed to estimate the demand flow rate in the
exclusive right-turn lane group vr. The flow rate in the exclusive through lane
group is then computed with Equation 31-56.
𝑣𝑡ℎ − (𝑣𝑠𝑙 − 𝑣𝑠𝑙,𝑙𝑡 ) − (𝑣𝑠𝑟 − 𝑣𝑠𝑟,𝑟𝑡 )
Equation 31-56 𝑣𝑡 = ≥ 0.0
𝑁𝑡
D. Compute Proportion of Turns in Shared-Lane Lane Groups

The proportion of left-turning vehicles in the shared left-turn and through
lane is computed with Equation 31-57.
𝑣𝑠𝑙,𝑙𝑡
Equation 31-57 𝑃𝐿 = ≤ 1.0
𝑣𝑠𝑙
where PL is the proportion of left-turning vehicles in the shared lane. Substitution
of vsr,rt for vsl,lt and vsr for vsl in Equation 31-57 yields an estimate of the proportion
of right-turning vehicles in the shared lane PR.
The proportion of left-turning vehicles in the shared left- and right-turn lane
is computed with Equation 31-58.
𝑣𝑠𝑙,𝑙𝑡
Equation 31-58
𝑃𝐿 = ≤ 1.0
𝑣𝑙𝑟
Substituting vsr,rt for vsl,lt in Equation 31-58 yields an estimate of the
proportion of right-turning vehicles in the shared lane PR.
E. Compute Lane Group Saturation Flow Rate

The saturation flow rate for the lane group shared by the left-turn and
through movements is computed by using Equation 31-59 with Equation 31-60.
𝑠𝑡ℎ 𝑔diff min [𝑔𝑝 − 𝑔𝑓 , 𝑔𝑢 ] 3,600 𝑛𝑠∗ 𝑓𝑚𝑠 𝑓𝑠𝑝
Equation 31-59 𝑠𝑠𝑙 = (𝑔𝑓 + + + )
𝑔𝑝 1 + 𝑃𝐿 [𝐸𝐿2,𝑚 − 1] 1 + 𝑃𝐿 [𝐸𝐿1,𝑚 − 1] 𝑠𝑡ℎ
with
𝑃𝐿 𝑛
(1 − 𝑃𝐿 𝑠 ) if 𝑃𝐿 < 0.999
Equation 31-60 𝑛𝑠∗ = {1 − 𝑃𝐿
𝑛𝑠 𝑃𝐿 if 𝑃𝐿 ≥ 0.999

where gdiff is the supplemental service time (s), n*s is the expected number of
sneakers per cycle in a shared left-turn lane, fms is the adjustment factor for
downstream lane blockage, fsp is the adjustment factor for sustained spillback,
and all other variables are as previously defined.
Equation 31-107 in Section 3 can be used to calculate gdiff.
Equation 31-61 is used to compute the saturation flow rate in a shared right-
turn and through lane group ssr .
𝑠𝑡ℎ
𝑠𝑠𝑟 = Equation 31-61
1 + 𝑃𝑅 (𝐸𝑅,𝑚 − 1)
where PR is the proportion of right-turning vehicles in the shared lane (decimal).
The saturation flow rate for the lane group serving left-turning vehicles in an
exclusive lane sl is computed with Equation 31-59, with PL = 1.0, gdiff = 0.0, gf = 0.0,
and sth replaced by slt (see Equation 31-112). Similarly, the saturation flow rate in an
exclusive right-turn lane group sr is computed with Equation 31-61, with PR = 1.0.
The saturation flow rate for the lane group serving through vehicles in an
exclusive lane is computed with Equation 31-62.
𝑠𝑡 = 𝑠𝑡ℎ 𝑓𝑠 Equation 31-62
where fs is the adjustment factor for all lanes serving through vehicles on an
approach with a shared left-turn and through lane group (= 0.91 if Nsl = 1 and the
left-turn has permitted operation; 1.0 otherwise).
The saturation flow rate for the shared left- and right-turn lane is computed
𝑠𝑡ℎ
𝑠𝑙𝑟 = Equation 31-63
1 + 𝑃𝐿 (𝐸𝐿,𝑚 − 1) + 𝑃𝑅 (𝐸𝑅,𝑚 − 1)
F. Compute Flow Ratio

The flow ratio for the subject intersection approach is computed with
Equation 31-64.
𝑣𝑙 𝑁𝑙 + 𝑣𝑠𝑙 𝑁𝑠𝑙 + 𝑣𝑡 𝑁𝑡 + 𝑣𝑠𝑟 𝑁𝑠𝑟 + 𝑣𝑟 𝑁𝑟 + 𝑣𝑙𝑟 𝑁𝑙𝑟
𝑦∗ = Equation 31-64
𝑠𝑙 𝑁𝑙 + 𝑠𝑠𝑙 𝑁𝑠𝑙 + 𝑠𝑡 𝑁𝑡 + 𝑠𝑠𝑟 𝑁𝑠𝑟 + 𝑠𝑟 𝑁𝑟 + 𝑠𝑙𝑟 𝑁𝑙𝑟
where y* is the flow ratio for the approach. If a shared left- and right-turn lane
exists on the subject approach, then Nsl = 0, Nt = 0, Nsr = 0, and Nlr = 1; otherwise,
Nsl = 1, Nt ≥ 0, Nsr = 1, and Nlr = 0.
G. Compute Revised Lane Group Flow Rate

The flow ratio from Step F is used to compute the demand flow rate in the
exclusive left-turn lane group with Equation 31-65.
𝑣𝑙 = 𝑠𝑙 𝑦 ∗ Equation 31-65
In a similar manner, the demand flow rate for the other lane groups is
estimated by multiplying the flow ratio y* by the corresponding lane group
saturation flow rate.

H. Compute Turn Movement Flow Rate in Shared-Lane Lane Groups

The left-turn demand flow rate in the shared lane group is computed with
Equation 31-66.
Equation 31-66 𝑣𝑠𝑙,𝑙𝑡 = 𝑣𝑙𝑡 − 𝑣𝑙 ≥ 0.0
Equation 31-66 can be used to compute the right-turn demand flow rate in
the shared lane group by substituting vsr,rt for vsl,lt, vrt for vlt,, and vr for vl.
The demand flow rate in each shared-lane lane group is now compared with
the rate estimated in Step B. If they differ by less than 0.1 veh/h, then the
procedure is complete and the flow rates estimated in Steps G and H represent
the best estimate of the flow rate for each lane group.
If there is disagreement between the lane group demand flow rates, then the
calculations are repeated, starting with Step C. However, for this iteration, the
flow rates computed in Steps G and H are used in the new calculation sequence.
The calculations are complete when the flow rates used at the start of Step C
differ from those obtained in Step H by less than 0.1 veh/h.
PRETIMED PHASE DURATION

The design of a pretimed timing plan can be a complex and iterative process
that is generally carried out with the assistance of software. Several software
products are available for this purpose. This subsection describes various
strategies for pretimed signal-timing design and provides a procedure for
implementing one of these strategies.
Design Strategies
Several aspects of signal-timing design, such as the choice of the timing
strategy, are beyond the scope of this manual. Three basic strategies are
commonly used for pretimed signals.
One strategy is to equalize the volume-to-capacity ratios for critical lane
groups. It is the simplest strategy and the only one that can be calculated without
excessive iteration. Under this strategy, the green time is allocated among the
various signal phases in proportion to the flow ratio of the critical lane group for
each phase. This strategy is described briefly in the next subsection. It is also
used in the planning-level analysis application described in Section 5.
A second strategy is to minimize the total delay to all vehicles. This strategy
is generally proposed as the optimal solution to the signal-timing problem.
Variations of this strategy often combine other performance measures (e.g., stop
rate, fuel consumption) in the optimization function. Many signal-timing
software products offer this optimization feature. Some products use a delay
estimation procedure identical to that in the motorized vehicle methodology in
Chapter 19, but other products use minor departures from it.
A third strategy is to equalize the level of service (LOS) for all critical lane
groups. This strategy promotes a LOS on all approaches that is consistent with
the overall intersection LOS. It improves on the first and second strategies
because they tend to produce a higher delay per vehicle for the minor
movements at the intersection (and therefore a less favorable LOS).

Determining Phase Duration on the Basis of Vehicle Demand

Signal timing based on equalization of the volume-to-capacity ratio is
described in this subsection. Equation 31-67, Equation 31-68, and Equation 31-69
are used to estimate the cycle length and effective green time for each critical
phase. Conversion to green interval duration follows by applying the
appropriate lost-time increments.
𝐶
𝑋𝑐 = ( ) ∑ 𝑦𝑐,𝑖 Equation 31-67
𝐶−𝐿
𝑖∈𝑐𝑖
𝐿 𝑋𝑐
𝐶= Equation 31-68
𝑋𝑐 − ∑𝑖∈𝑐𝑖 𝑦𝑐,𝑖
𝑣𝑖 𝐶 𝑣 𝐶
𝑔𝑖 = =( ) ( ) Equation 31-69
𝑁𝑖 𝑠𝑖 𝑋𝑖 𝑁 𝑠 𝑖 𝑋𝑖
where
L = cycle lost time (s),
Xc = critical intersection volume-to-capacity ratio,
yc,i = critical flow ratio for phase i = vi/(N si),
ci = set of critical phases on the critical path,
Xi = volume-to-capacity ratio for lane group i,
vi = demand flow rate for lane group i (veh/h),
Ni = number of lanes in lane group i (ln),
si = saturation flow rate for lane group i (veh/h/ln), and
gi = effective green time for lane group i (s).
The summation term in each of these equations represents the summation of
a specific variable for the set of critical phases. A critical phase is one phase of a
set of phases that occurs in sequence whose combined flow ratio is the largest for
the signal cycle.
Procedure
The following steps summarize the procedure for estimating the cycle length
and effective green time for the critical phases:
1. Compute the flow ratio [= vi/(N si)] for each lane group and identify the
critical flow ratio for each phase. When there are several lane groups on
the approach and they are served during a common phase, then the lane
group with the largest flow ratio represents the critical flow ratio for the
phase. A procedure for identifying the critical phases and associated flow
ratios is described in Section 4 of Chapter 19, Signalized Intersections.
2. If signal-system constraints do not dictate the cycle length, then estimate
the minimum cycle length with Equation 31-68 by setting Xc equal to 1.0.

3. If signal-system constraints do not dictate the cycle length, then estimate

the desired cycle length with Equation 31-68 by substituting a target
volume-to-capacity ratio Xt for the critical ratio Xc. A value of Xt in the
range of 0.80 to 0.90 is recommended for this purpose.
4. If signal-system constraints do not dictate the cycle length, then use the
results of Steps 2 and 3 to select an appropriate cycle length for the signal.
Otherwise, the cycle length is that dictated by the signal system.
5. Estimate the effective green time for each phase with Equation 31-69 and
the target volume-to-capacity ratio.
6. Check the timing to ensure the effective green time and the lost time for
each phase in a common ring sum to the cycle length.
Example Application
The procedure is illustrated by a sample calculation. Consider the intersection
Exhibit 31-7 N
Example Intersection (0.20)
(0.40)
(0.45)
(x.xx) = flow ratio
(0.35)
Phases 2 and 6 serve the eastbound and westbound approaches, respectively.

Phases 4 and 8 serve the southbound and northbound approaches, respectively.
One phase from each pair will represent the critical phase and dictate the
duration of both phases. It is assumed the lost time for each phase equals the
change period (i.e., the yellow change interval plus the red clearance interval).
Thus, the lost time for each critical phase is 4 s, or 8 s for the cycle.
In this simple example, only one lane group is served on each approach, so
the critical flow ratios can be identified by inspection of Exhibit 31-7. Specifically,
the critical flow ratio for the east–west phases is that associated with the
eastbound approach (i.e., Phase 2) at a value of 0.45. Similarly, the critical flow
ratio for the north–south phases is that associated with the northbound approach
(i.e., Phase 8).
The minimum cycle length that will avoid oversaturation is computed by
Equation 31-68 with Xc = 1.00.
8(1.0) 8
𝐶(minimum) = = = 40 s
1.0 − (0.45 + 0.35) 0.2
A target volume-to-capacity ratio of 0.80 is used to estimate the target cycle
length.

8(0.8) 6.4
𝐶= = = infinity
0.8 − (0.45 + 0.35) 0
This computation indicates a critical volume-to-capacity ratio of 0.8 cannot
be provided with the present demand levels at the intersection.
As a second trial estimate, a target volume-to-capacity ratio of 0.92 is selected
and used to estimate the target cycle length.
8(0.92)
𝐶= = 61 s
0.92 − (0.45 + 0.35)
The estimate is rounded to 60 s for practical application. Equation 31-67 is
then used to estimate the critical volume-to-capacity ratio of 0.923 for the selected
cycle length of 60 s.
With Equation 31-69, the effective green time is allocated so the volume-to-
capacity ratio for each critical lane group is equal to the target volume-to-
capacity ratio. Thus, for the example problem, the target volume-to-capacity ratio
for each phase is 0.923. The effective green times are computed with Equation 31-
69. The results of the calculations are listed below:
𝑔2 = 0.45(60/0.923) = 29.3 s
𝑔8 = 0.35(60/0.923) = 22.7 s
𝑔2 + 𝑔8 + 𝐿 = 29.3 + 22.7 + 8.0 = 60.0 s
The duration of the effective green interval for Phase 6 is the same as for
Phase 2, given that they have the same phase lost time. Similarly, the effective
green interval for Phase 4 is the same as for Phase 8.
Determining Phase Duration on the Basis of Pedestrian Considerations

Two pedestrian considerations are addressed in this subsection as they relate
to pretimed phase duration. One consideration addresses the time a pedestrian
needs to perceive the signal indication and traverse the crosswalk. A second
consideration addresses the time needed to serve cyclic pedestrian demand.
When available, local guidelines or practice should be used to establish phase
duration on the basis of pedestrian considerations.
A minimum green interval duration that allows a pedestrian to perceive the
indication and traverse the crosswalk can be computed with Equation 31-70.
𝐿𝑐𝑐
𝐺𝑝,𝑚𝑖𝑛 = 𝑡𝑝𝑟 + − 𝑌 − 𝑅𝑐 Equation 31-70
𝑆𝑝
where
Gp,min = minimum green interval duration based on pedestrian crossing time (s),
tpr = pedestrian perception of signal indication and curb departure time
= 7.0 (s),
Lcc = curb-to-curb crossing distance (ft),
Sp = pedestrian walking speed = 3.5 (ft/s),
Y = yellow change interval (s), and
Rc = red clearance interval (s).

The variable tpr in this equation represents the time pedestrians need to
perceive the start of the phase and depart from the curb. A value of 7.0 s
represents a conservatively long value that is adequate for most pedestrian
crossing conditions. The variable Sp represents the pedestrian walking speed in a
crosswalk. A value of 3.5 ft/s represents a conservatively slow value that most
pedestrians will exceed.
If a permitted or protected-permitted left-turn operation is used for the left-
turn movement that crosses the subject crosswalk, then the subtraction of the
yellow change interval and the red clearance interval in Equation 31-70 may
cause some conflict between pedestrians and left-turning vehicles. If this conflict
can occur, then the minimum green interval duration should be computed as
Gp,min = tpr + (Lcc/Sp).
The second pedestrian consideration in timing design is the time required to
serve pedestrian demand. The green interval duration should equal or exceed
this time to ensure pedestrian demand is served each cycle. The time needed to
serve this demand is computed with either Equation 31-71 or Equation 31-72,
along with Equation 31-73.
If the crosswalk width W is greater than 10 ft, then
𝐿𝑐𝑐 𝑁𝑝𝑒𝑑
Equation 31-71 𝑡𝑝𝑠 = 3.2 + + 2.7
𝑆𝑝 𝑊
If the crosswalk width W is less than or equal to 10 ft, then
𝐿𝑐𝑐
Equation 31-72 𝑡𝑝𝑠 = 3.2 + + 0.27 𝑁𝑝𝑒𝑑
𝑆𝑝
with
𝑣𝑝𝑒𝑑,𝑖
Equation 31-73 𝑁𝑝𝑒𝑑 = 𝐶
3,600
where
tps = pedestrian service time (s),
W = effective width of crosswalk (ft),
vped,i = pedestrian flow rate in the subject crossing for travel direction i (p/h),
and
Nped = number of pedestrians crossing during an interval (p).
Equation 31-73 assumes pedestrians always cross at the start of the phase.
Thus, it yields a conservatively large estimate of Nped because some pedestrians
arrive and cross during the green indication.
Equation 31-73 is specific to the pedestrian flow rate in one direction of travel
along the subject crosswalk. If the pedestrian flow rate varies significantly during
the analysis period for the crosswalk’s two travel directions, then tps should be
calculated for both travel directions, and the larger value should be used to
estimate the green interval duration needed to serve pedestrian demand.

PEDESTRIAN AND BICYCLE ADJUSTMENT FACTORS

Exhibit 31-8 shows sample conflict zones where intersection users compete
for space. This competition reduces the saturation flow rate of the turning
vehicles. Its effect is quantified in the pedestrian and bicycle adjustment factors.
This subsection describes a procedure for calculating these factors, which are
used in the procedure for calculating the adjusted saturation flow rate that is
described in Section 3 of Chapter 19.
Opposing Lanes Exhibit 31-8

Conflict Zone Locations
Receiving
Receiving Lanes
Lanes
Pedestrian-Vehicle
Conflict Zone
Pedestrians Pedestrians
Bicycle-Vehicle
Conflict Zone
Bicycles
Subject Approach
This subsection consists of two subsections. The first subsection describes the
procedure for computing (a) the pedestrian–bicycle adjustment factor for right-
turn lane groups and (b) the pedestrian adjustment factor for left-turn lane
groups from a one-way street. The second subsection describes the procedure for
computing the pedestrian adjustment factor for left-turn groups served by
permitted or protected-permitted operation.
The following guidance is used to determine the pedestrian adjustment
factor for lane groups serving left-turn movements fLpb:
• If there are no conflicting pedestrians, then fLpb is equal to 1.0.
• If the lane group is on a two-way street and the protected mode or split
phasing is used, then fLpb is equal to 1.0.
• If the lane group is on a one-way street, then the procedure described in
the first subsection below is used to compute fLpb.
• If the lane group is on a two-way street and either the permitted mode or
the protected-permitted mode is used, then the procedure described in
the second subsection below is used to calculate fLpb.
The following guidance is used to determine the pedestrian–bicycle
adjustment factor for lane groups serving right-turn movements fRpb:

• If there are no conflicting pedestrians or bicycles, then fRpb is equal to 1.0.

• If the protected mode is used, then fRpb is equal to 1.0.
• If the permitted mode or the protected-permitted mode is used, then the
procedure described in the first subsection below is used to compute fRpb.
Right-Turn Movements and Left-Turn Movements from One-Way Street

A. Determine Pedestrian Flow Rate During Service
This procedure requires knowledge of the phase duration and cycle length. If
these variables are not known and the intersection is pretimed, then they can be
estimated by using the procedure described in the previous subsection titled
Pretimed Phase Duration. If the intersection is actuated, then the average phase
duration and cycle length can be computed by using the procedure described in
the previous subsection titled Actuated Phase Duration.
The pedestrian flow rate during the pedestrian service time is computed
𝐶
Equation 31-74 𝑣𝑝𝑒𝑑𝑔 = 𝑣𝑝𝑒𝑑 ≤ 5,000
𝑔𝑝𝑒𝑑
where
vpedg = pedestrian flow rate during the pedestrian service time (p/h),
vped = pedestrian flow rate in the subject crossing (walking in both directions)
(p/h),
C = cycle length (s), and
gped = pedestrian service time (s).
If the phase providing service to pedestrians is actuated, has a pedestrian
signal head, and rest-in-walk is not enabled, then the pedestrian service time is
equal to the smaller of (a) the effective green time for the phase or (b) the sum of
the walk and pedestrian clear settings [i.e., gped = min(g, Walk + PC)]. Otherwise,
the pedestrian service time can be assumed to equal the effective green time for
the phase (i.e., gped = g).
B. Determine Average Pedestrian Occupancy

If the pedestrian flow rate during the pedestrian service time is 1,000 p/h or
less, then the pedestrian occupancy is computed with Equation 31-75.
𝑣𝑝𝑒𝑑𝑔
Equation 31-75 𝑂𝐶𝐶𝑝𝑒𝑑𝑔 =
2,000
where OCCpedg is the pedestrian occupancy.
If the pedestrian flow rate during the pedestrian service time exceeds 1,000
p/h, then Equation 31-76 is used.
𝑣𝑝𝑒𝑑𝑔
Equation 31-76 𝑂𝐶𝐶𝑝𝑒𝑑𝑔 = 0.4 + ≤ 0.90
10,000
A practical upper limit on vpedg of 5,000 p/h should be maintained when

C. Determine Bicycle Flow Rate During Green

The bicycle flow rate during the green indication is computed with Equation
31-77.
𝐶
𝑣𝑏𝑖𝑐𝑔 = 𝑣𝑏𝑖𝑐 ≤ 1,900 Equation 31-77
𝑔
where
vbicg = bicycle flow rate during the green indication (bicycles/h),
vbic = bicycle flow rate (bicycles/h),
C = cycle length (s), and
g = effective green time (s).
D. Determine Average Bicycle Occupancy

The average bicycle occupancy is computed with Equation 31-78.
𝑣𝑏𝑖𝑐𝑔
𝑂𝐶𝐶𝑏𝑖𝑐𝑔 = 0.02 + Equation 31-78
2,700
where OCCbicg is the bicycle occupancy, and vbicg is the bicycle flow rate during the
green indication (bicycles/h).
A practical upper limit on vbicg of 1,900 bicycles/h should be maintained when
E. Determine Relevant Conflict Zone Occupancy

Equation 31-79 is used for right-turn movements with no bicycle interference
or for left-turn movements from a one-way street. This equation is based on the
assumptions that (a) pedestrian crossing activity takes place during the time
period associated with gped, and (b) no crossing occurs during the green time
period g – gped, when this time period exists.
𝑔𝑝𝑒𝑑 Equation 31-79
𝑂𝐶𝐶𝑟 = 𝑂𝐶𝐶𝑝𝑒𝑑𝑔
𝑔
where OCCr is the relevant conflict zone occupancy.
Alternatively, Equation 31-80 is used for right-turn movements with
pedestrian and bicycle interference, with all variables as previously defined.
𝑔𝑝𝑒𝑑 𝑔𝑝𝑒𝑑
𝑂𝐶𝐶𝑟 = ( 𝑂𝐶𝐶𝑝𝑒𝑑𝑔 ) + 𝑂𝐶𝐶𝑏𝑖𝑐𝑔 − ( 𝑂𝐶𝐶𝑝𝑒𝑑𝑔 𝑂𝐶𝐶𝑏𝑖𝑐𝑔 ) Equation 31-80
𝑔 𝑔
F. Determine Unoccupied Time

If the number of cross-street receiving lanes is equal to the number of turn
lanes, then turning vehicles will not be able to maneuver around pedestrians or
bicycles. In this situation, the time the conflict zone is unoccupied is computed
𝐴𝑝𝑏𝑇 = 1 − 𝑂𝐶𝐶𝑟 Equation 31-81
where ApbT is the unoccupied time, and OCCr is the relevant conflict zone
occupancy.

Alternatively, if the number of cross-street receiving lanes exceeds the

number of turn lanes, turning vehicles will more likely maneuver around
pedestrians or bicycles. In this situation, the effect of pedestrians and bicycles on
saturation flow is lower, and the time the conflict zone is unoccupied is
Equation 31-82 𝐴𝑝𝑏𝑇 = 1 − 0.6 𝑂𝐶𝐶𝑟
Either Equation 31-81 or Equation 31-82 is used to compute ApbT. The choice
of which equation to use should be based on careful consideration of the number
of turn lanes and the number of receiving lanes. At some intersections, drivers
may consistently and deliberately make illegal turns from an exclusive through
lane. At other intersections, proper turning cannot be executed because the
receiving lane is blocked by double-parked vehicles. For these reasons, the number
of turn lanes and receiving lanes should be determined from field observation.
G. Determine Saturation Flow Rate Adjustment Factor

For permitted right-turn operation in an exclusive lane, Equation 31-83 is
used to compute the pedestrian–bicycle adjustment factor.
Equation 31-83 𝑓𝑅𝑝𝑏 = 𝐴𝑝𝑏𝑇
where fRpb is the pedestrian–bicycle adjustment factor for right-turn groups, and
ApbT is the unoccupied time.
For protected-permitted operation in an exclusive lane, the factor from
Equation 31-83 is used to compute the adjusted saturation flow rate during the
permitted period. The factor has a value of 1.0 when used to compute the
adjusted saturation flow rate for the protected period.
For left-turn movements from a one-way street, Equation 31-84 is used to
compute the pedestrian adjustment factor.
Equation 31-84 𝑓𝐿𝑝𝑏 = 𝐴𝑝𝑏𝑇
where fLpb is the pedestrian adjustment factor for left-turn groups, and ApbT is the
unoccupied time.
Permitted and Protected-Permitted Left-Turn Movements

This subsection describes a procedure for computing the adjustment factor
for left-turn movements on a two-way street that are operating in either the
permitted mode or the protected-permitted mode. The calculations in this
subsection supplement the procedure described in the previous subsection. The
calculations described in Steps A and B in the previous subsection must be
completed first (substitute the effective permitted green time gp for g in Step A),
after which the calculations described in this subsection are completed.
This procedure does not account for vehicle–bicycle conflict during the left-
turn maneuver.
A. Compute Pedestrian Occupancy After Queue Clears

The pedestrian occupancy after the opposing queue clears is computed with
Equation 31-85 or Equation 31-86. The opposing-queue service time gq is

computed as the effective permitted green time gp less the duration of permitted
left-turn green time that is not blocked by an opposing queue gu (i.e., gq = gp – gu).
If gq < gped, then
0.5 𝑔𝑞
𝑂𝐶𝐶𝑝𝑒𝑑𝑢 = 𝑂𝐶𝐶𝑝𝑒𝑑𝑔 (1 − ) Equation 31-85
𝑔𝑝𝑒𝑑
otherwise
𝑂𝐶𝐶𝑝𝑒𝑑𝑢 = 0.0 Equation 31-86
where OCCpedu is the pedestrian occupancy after the opposing queue clears, gq is
the opposing-queue service time (= gs for the opposing movement) (s), and all
If the opposing-queue service time gq equals or exceeds the pedestrian service
time gped, then the opposing queue consumes the entire pedestrian service time.
B. Determine Relevant Conflict Zone Occupancy

After the opposing queue clears, left-turning vehicles complete their
maneuvers on the basis of accepted gap availability in the opposing traffic
stream. Relevant conflict zone occupancy is a function of the probability of
accepted gap availability and pedestrian occupancy. It is computed with
Equation 31-87.
𝑔𝑝𝑒𝑑 − 𝑔𝑞
𝑂𝐶𝐶𝑟 = (𝑂𝐶𝐶𝑝𝑒𝑑𝑢 ) 𝑒 −5.00 𝑣𝑜/3,600 Equation 31-87
𝑔𝑝 − 𝑔𝑞
where vo is the opposing demand flow rate (veh/h), gp is the effective green time
for permitted left-turn operation (s), and all other variables are as previously
defined.
The opposing demand flow rate vo is determined to be one of two cases. In
Case 1, vo equals the sum of the opposing through and right-turn volumes. In
Case 2, vo equals the opposing through volume. Case 2 applies when there is a
through movement on the opposing approach and one of the following
conditions applies: (a) there is an exclusive right-turn lane on the opposing
approach and the analyst optionally indicates that this lane does not influence
the left-turn drivers’ gap acceptance, or (b) there is no right-turn movement on
the opposing approach. Case 1 applies whenever Case 2 does not apply.
When an exclusive right-turn lane exists on the opposing approach, the
default condition is to assume this lane influences the subject left-turn drivers’
gap acceptance. The determination that the exclusive right-turn lane does not
influence gap acceptance should be based on knowledge of local driver behavior,
traffic conditions, and intersection geometry.
C. Determine Unoccupied Time

Either Equation 31-81 or Equation 31-82 from the previous subsection (i.e.,
Step F above) is used to compute ApbT. The choice of which equation to use
should be based on a consideration of the number of left-turn lanes and the
number of receiving lanes.
D. Determine Saturation Flow Rate Adjustment Factor

Equation 31-88 is used to compute the pedestrian adjustment factor fLpb from
ApbT, the unoccupied time.
Equation 31-88 𝑓𝐿𝑝𝑏 = 𝐴𝑝𝑏𝑇
WORK ZONE PRESENCE ADJUSTMENT FACTOR

The procedure described in this subsection can be used to evaluate
signalized intersection operation when a work zone is present on the intersection
approach. The work zone is considered to be on the intersection approach if
some (or all) of the work zone is located between the stop line and a point 250 ft
upstream of the stop line. The work zone may be located on the shoulder, or it
may include the closure of one or more lanes. An intersection with a work zone
located on the eastbound approach is shown in Exhibit 31-9.
Exhibit 31-9
Work Zone on an Intersection
Approach
Required Input Data

The input data that are needed to estimate the effect of work zone presence
on saturation flow rate are listed in Exhibit 31-10. The two data elements listed
are described in this subsection. The contents of Exhibit 31-10 are in addition to
those listed in Exhibit 19-11.
Exhibit 31-10 Input Data Element and Units Basis

Geometric Design Input Data Number of lanes open on the approach in the work zone (ln) Approach
Requirements for Work Zones Approach lane width during work zone (ft) Approach
Note: Approach = one value or condition for the intersection approach.
Number of Lanes Open on the Approach in the Work Zone

The number of lanes open on the approach in the work zone represents the
count of left-turn and through lanes that are open during work zone presence.
The count does not include any exclusive right-turn lanes that may exist. The
count is taken in the work zone (not upstream or downstream of the work zone).
If the number of lanes in the work zone varies, then the smallest number of lanes
provided to motorists is used for this input variable.

Approach Lane Width During Work Zone

The approach lane width represents the total width of all open left-turn,
through, and right-turn lanes on the intersection approach when the work zone
is present.
Computational Steps
The saturation flow rate adjustment factor for the case in which a work zone
is located at the intersection can be computed by using Equation 31-89 with
Equation 31-90 and Equation 31-91.
𝑓𝑤𝑧 = 0.858 × 𝑓𝑤𝑖𝑑 × 𝑓reduce ≤ 1.0 Equation 31-89
with
1
𝑓𝑤𝑖𝑑 = Equation 31-90
1 − 0.0057 (𝑎𝑤 − 12)
1
𝑓reduce = Equation 31-91
1 + 0.0402 (𝑛𝑜 − 𝑛𝑤𝑧 )
where
fwz = adjustment factor for work zone presence at the intersection,
fwid = adjustment factor for approach width,
freduce = adjustment factor for reducing lanes during work zone presence,
aw = approach lane width during work zone (= total width of all open left-
turn, through, and right-turn lanes) (ft),
no = number of left-turn and through lanes open during normal operation
(ln), and
nwz = number of left-turn and through lanes open during work zone
presence (ln).
This factor is computed during Step 4, Determine Adjusted Saturation Flow
Rate, of the motorized vehicle methodology in Chapter 19, Signalized
Intersections. One value is computed for (and is applicable to) all lane groups on
the subject intersection approach.

3. QUEUE ACCUMULATION POLYGON
This section describes a procedure for using the queue accumulation polygon
(QAP) to estimate delay. The section consists of three subsections. The first
subsection provides a review of concepts related to the QAP. The second
subsection describes a general procedure for developing the QAP, and the third
subsection extends the general procedure to the evaluation of left-turn lane groups.
The discussion in this section describes basic principles for developing
polygons for selected types of lane assignment, lane grouping, left-turn
operation, and phase sequence. The analyst is referred to the computational
engine for specific calculation details, especially as they relate to assignments,
groupings, left-turn operations, and phase sequences not addressed in this
section. This engine is described in Section 7.
CONCEPTS
The QAP is a graphic tool for describing the deterministic relationship
between vehicle arrivals, departures, queue service time, and delay. The QAP
defines the queue size for a traffic movement as a function of time during the
cycle. The shape of the polygon is defined by the following factors: arrival flow
rate during the effective red and green intervals, saturation flow rate associated
with each movement in the lane group, signal indication status, left-turn
operation mode, and phase sequence. Once constructed, the polygon can be used
to compute the queue service time, capacity, and uniform delay for the
corresponding lane group.
A QAP is shown in Exhibit 31-11. The variables shown in the exhibit are
defined in the following list:
gs = queue service time = Qr/(s – qg) (s),
ge = green extension time (s),
q = arrival flow rate = v/3,600 (veh/s),
v = demand flow rate (veh/h),
qg = arrival flow rate during the effective green time = P q C/g (veh/s),
Qr = queue size at the end of the effective red time = qr r (veh),
and
s = adjusted saturation flow rate (veh/h/ln).
Queue Accumulation Polygon Chapter 31/Signalized Intersections: Supplemental

Exhibit 31-11
r g
Number of Vehicles in Queue
gs ge for Protected Movements
Qr
1 1
qr s - qg
0
0 Time (s)
In application, all flow rate variables are converted to common units of

vehicles per second per lane. The presentation in this section is based on these
units for q and s.
The polygon in Exhibit 31-11 applies to either a through lane group or a left-
or right-turn lane group with exclusive lanes operating with the protected mode.
Other polygon shapes are possible, depending on whether the lane group includes
a shared lane and whether the lane group serves a permitted (or protected-
permitted) left-turn movement. In general, a unique polygon shape will be dictated
by each combination of left-turn operational mode (i.e., permitted, protected, or
protected-permitted) and phase sequence (i.e., lead, lag, or split). A general
procedure for constructing these polygons is described in the next subsection.
GENERAL QAP CONSTRUCTION PROCEDURE

This subsection describes a general procedure for constructing a QAP for a
lane group at a signalized intersection. It is directly applicable to left-turn lane
groups that have exclusive lanes and protected operation, through lane groups
with exclusive lanes, and right-turn lane groups with exclusive lanes. Variations
that extend this procedure to turn lane groups with shared lanes, permitted
operation, or protected-permitted operation are described in the next subsection.
The construction of a QAP is based on identification of flow rates and service
times during the average signal cycle. These rates and times define periods of
queue growth, queue service, and service upon arrival. As shown in Exhibit 31-
11, the rates and times define queue size as it varies during the cycle. The resulting
polygon formed by the queue size profile can be decomposed into a series of
trapezoid or triangle shapes, with each shape having a known time interval.
Collectively, the areas of the individual shapes can be added to equal the area of
the polygon, and the time intervals can be added to equal the cycle length.
The QAP calculation sequence follows the order of interval occurrence over
time, and the results can be recorded graphically (as in Exhibit 31-11) or in a
tabular manner (i.e., row by row, where each row represents one time interval).
A time interval is defined to begin and end at points when either the departure
rate or the arrival rate changes. For the duration of the interval, these rates are
assumed to be constant.
Chapter 31/Signalized Intersections: Supplemental Queue Accumulation Polygon

The following text outlines the calculation sequence used to construct a QAP
for a specified lane group. The sequence is repeated for each lane group at the
intersection, with the through lane groups evaluated first so the saturation flow
rate of permitted left-turn lane groups can be based on the known queue service
time for the opposing traffic movements.
1. The QAP calculations for a given lane group start with the end of the
effective green period for the phase serving the subject lane group in a
protected manner. The initial queue Qi is assumed to equal 0.0 vehicles.
2. Determine the points in the cycle when the arrival flow rate or the
discharge rate changes. The arrival rate may change because of platoons
formed in response to an upstream signal, so it is expressed in terms of
the arrival rate during green qg and during red qr. The discharge rate may
change because of the start or end of effective green, a change in the
saturation flow rate, the depletion of the subject queue, the depletion of
the opposing queue, or the departure of left-turn vehicles as sneakers.
3. For the time interval between the points identified in Step 2, number each
interval and compute its duration. Next, identify the arrival rate and
discharge rate associated with the interval. Finally, confirm that the sum
of all interval durations equals the cycle length.
4. Calculate the capacity of each interval for which there is some discharge,
including sneakers when applicable. The sum of these capacities equals
the total lane group capacity. Calculate the demand volume for each
interval for which there are some arrivals. The sum of these volumes
equals the total lane group volume.
5. Calculate the volume-to-capacity ratio X for the lane group by dividing
the lane group’s total volume by its total capacity. If the volume-to-
capacity ratio exceeds 1.0, then calculate the adjusted arrival flow rate q’
for each interval by dividing the original flow rate q by X (i.e., q’ = q/X).
6. Calculate the queue at the end of interval i with Equation 31-92.
𝑠 𝑞
Equation 31-92 𝑄𝑖 = 𝑄𝑖−1 − ( − ) 𝑡𝑑,𝑖 ≥ 0.0
3,600 𝑁
where Qi is the queue size at the end of interval i (veh), td,i is the duration
of time interval i during which the arrival flow rate and saturation flow
rate are constant (s), and all other variables are as previously defined.
7. If the queue at the end of interval i equals 0.0 vehicles, then compute the
duration of the trapezoid or triangle with Equation 31-93. The subject
interval should be divided into two intervals, with the first interval having
a duration of tt,i and the second interval having a duration of td,i – tt,i. The
second interval has starting and ending queues equal to 0.0 vehicles.
Equation 31-93
𝑡𝑡,𝑖 = min (𝑡𝑑,𝑖 , 𝑄𝑖−1 /𝑤𝑞 )
where tt,i is the duration of trapezoid or triangle in interval i (s), wq is the
queue change rate (= discharge rate minus arrival rate) (veh/s), and all

8. Steps 6 and 7 are repeated for each interval in the cycle.

9. When all intervals are completed, the assumption of a zero starting queue
(made in Step 1) is checked. The queue size computed for the last interval
should always equal the initially assumed value. If this is not the case,
then Steps 6 through 8 are repeated by using the ending queue size of the
last interval as the starting queue size for the first interval.
10. When all intervals have been evaluated and the starting and ending
queue sizes are equal, then the uniform delay can be calculated. This
calculation starts with computing the area of each trapezoid or triangle.
These areas are then added to determine the total delay. Finally, the total
delay is divided by the number of arrivals per cycle to produce uniform
delay. Equations for calculating uniform delay by using the QAP are
described in Step 7 of the next subsection.
QAP CONSTRUCTION PROCEDURE FOR SELECTED LANE GROUPS

This subsection describes a seven-step procedure for constructing a QAP for
selected lane groups. The focus is on left-turn movements in lane groups with
shared lanes, permitted operation, or protected-permitted operation. However,
there is some discussion of other lane groups, lane assignments, and operation.
The procedure described in this subsection represents an extension of the general
procedure described in the previous subsection.
Step 1. Determine Permitted Green Time

This step applies when the subject left-turn movement is served by using the
permitted mode or the protected-permitted mode. Two effective green times are
computed. One is the effective green time for permitted left-turn operation gp.
This green time occurs during the period when the adjacent and opposing
through movements both have a circular green indication (after adjustment for
lost time).
The other effective green time represents the duration of permitted left-turn
green time that is not blocked by an opposing queue gu. This green time
represents the time during the effective green time for permitted left-turn
operation gp that is not used to serve the opposing queue. This time is available
to the subject left-turn movement to filter through the conflicting traffic stream.
Exhibit 31-12 provides equations for computing the unblocked permitted
green time for left-turn Movement 1 (see Exhibit 19-1) when Dallas left-turn
phasing is not used. Similar equations can be derived for the other left-turn
movements or when Dallas phasing is used. The variables defined in this exhibit
are provided in the following list:
opposing queue (s),
GU = displayed green interval corresponding to gu (s),
e = extension of effective green = 2.0 (s),

Gq = displayed green interval corresponding to gq (s),

Rc = red clearance interval (s),
Y = yellow change interval (s), and
gq = opposing-queue service time (= gs for the opposing movement) (s).
Exhibit 31-12 Phase Sequence Permitted Permitted

Unblocked Permitted Green (phase numbers Displayed Unblocked Permitted Start-Up Lost Extension
Time shown in boxes) Green Time GU (s)a Time l1,p (s)b Time ep (s)c
Lead– 1 2 GU 1 = min[Dp 1 + Dp 2 – Dp 5 – Y6 – Rc 6, GU 1* ]
Lead l 1,1* e1
5 6 with GU *
1 = Dp 2 – Y6 – Rc 6 – Gq 2
1 2
GU 1 = Dp 2 – Y6 – Rc 6 – Gq 2 l 1,1* e1
5 6
Lead– 1 2
Lag GU 1 = Dp 6 – Y6 – Rc 6 – Dp 1 – Gq 2 0.0 e1
6 5
or
Lead– 1 2 Not Not
No permitted period
Perm 6 5 applicable applicable
1 2
GU 1 = Dp 6 – Y6 – Rc 6 – Dp 1 – Gq 2 0.0 e1
6
Lag– 2 1 Not Not
Lead No permitted period
5 6 applicable applicable
or
Lag– 2 1
GU 1 = Dp 2 – Y2 – Rc 2 – max[Dp 5, Gq 2] l 1,1 0.0
Perm 5 6
2 1 GU 1 = min[Dp 2 – Y2 – Rc 2, Dp 6 – Y6 – Rc 6]
l 1,1 0.0
6 – Gq 2
Perm– 2
Lead GU 1 = Dp 2 – Y2 – Rc 2 – max[Dp 5, Gq 2] l 1,1 e1
5 6
Perm– 2 GU 1 = min[Dp 2 – Y2 – Rc 2, Dp 6 – Y6 – Rc 6]
Lag l 1,1 e1
6 5 – Gq 2
Perm– 2
Perm GU 1 = Dp 2 – Y6 – Rc 6 – Gq 2 l 1,1 e1
6
Lag– 2 1 GU 1 = min[Dp 2 – Y2 – Rc 2, Dp 6 – Y6 – Rc 6]
Lag l 1,1 e 1*
6 5 – Gq 2
2 1 GU 1 = min[Dp 2 – Y2 – Rc 2, Dp 6 – Y6 – Rc 6]
l 1,1 e 1*
6 5 – Gq 2
a
Notes: Gq 2 is computed for each opposing lane (excluding any opposing shared left-turn lane), and the value
used corresponds to the lane requiring the longest time to clear. In general, if the opposing lanes serve
through movements exclusively, then Gq 2 = gq + l 1. If an opposing lane is shared, then Gq 2 = gp – ge + l 1,
where gp is the effective green time for permitted operation (s), ge is the green extension time (s), and l 1 is
the start-up lost time (s).
b
If Dp 5 > (Dp 1 – Y1 – Rc 1), then l 1* = Dp 5 – (Dp 1 – Y1 – Rc 1) + l 1 – e 1; otherwise, l 1* = 0.0. Regardless, the
result should not be less than 0.0 or more than l 1.
c
e 1* = Dp 2 – (Dp 6 – Y6 – Rc 6), provided the result is not less than 0.0 or more than e1.
Perm = permitted.
For the first four variables in the preceding list, the subscript “1” is added to
the variable when it is used in an Exhibit 31-12 equation. This subscript denotes
Movement 1. For the next four variables in the list, a numeric subscript is added
to the variable when it is used in an equation from the exhibit. This subscript
denotes the phase number associated with the variable. Exhibit 31-12 applies
only to left-turn Movement 1. The subscripts need to be changed to apply the
equations to other left-turn movements.

The equations shown in Exhibit 31-12 indicate that the effective green time
for the permitted operation of Phase 1 depends on the duration of Phase 2 and
sometimes the duration of Phase 5. In all instances, Movement 1 has permitted
operation during all, or a portion of, Phase 6.
For a given left-turn lane group, one of the equations in the second column
(Displayed Unblocked Permitted Green Time) of Exhibit 31-12 will apply. It is
used to compute the displayed green interval corresponding to gu (i.e., GU). The
computed GU is required to have a nonnegative value. If the calculation yields a
negative value, then GU is set to 0.0.
The same equation can be used to compute the displayed green interval
corresponding to gp (i.e., Gp) by substituting Gp for GU and 0.0 for Gq. Again, the
computed Gp is required to have a nonnegative value. If the calculation yields a
negative value, then Gp is set to 0.0.
Equation 31-94 is used to compute the effective green time for permitted left-
turn operation.
𝑔𝑝 = 𝐺𝑝 − 𝑙1,𝑝 + 𝑒𝑝 ≥ 0.0 Equation 31-94
where
Gp = displayed green interval corresponding to gp (s),
l1,p = permitted start-up lost time (s), and
ep = permitted extension of effective green (s).
The values of l1,p and ep used in Equation 31-94 are obtained from the two
right-hand columns (Permitted Start-Up Lost Time and Permitted Extension
Time, respectively) of Exhibit 31-12.
The start-up lost time for gu is considered to occur coincident with the start-
up lost time associated with gp. Hence, if the opposing-queue service time
consumes an initial portion of gp, then there is no start-up lost time associated
with gu. The rationale for this approach is that left-turn drivers waiting for the
opposing queue to clear will be anticipating queue clearance and may be moving
forward slowly (perhaps already beyond the stop line) so that there is negligible
start-up lost time at this point. This approach also accommodates the
consideration of multiple effective green-time terms when there is a shared lane
(e.g., gf), and it avoids inclusion of multiple start-up lost times during gp. In
accordance with this rationale, Equation 31-95 is used to compute the permitted
left-turn green time that is not blocked by an opposing queue gu, where all other
variables are as previously defined.
𝑔𝑢 = 𝐺𝑢 + 𝑒𝑝 ≤ 𝑔𝑝 Equation 31-95
If protected-permitted operation exists and Dallas phasing is used, then the

displayed green interval corresponding to gu (i.e., GU) is equal to the opposing
through phase duration minus the queue service time and change period of the
opposing through phase (i.e., GU1 = Dp2 – Y2 – Rc2 – Gq2). The permitted start-up

lost time l1,p and permitted extension of effective green ep are equal to l1 and e,
respectively. Otherwise, all the calculations described previously apply.
Step 2. Determine Time Before First Left-Turn Vehicle Arrives

This step applies when the left-turn movement is served by using the
permitted mode on a shared-lane approach. The variable of interest represents
the time that elapses from the start of the permitted green to the arrival of the
first left-turning vehicle at the stop line. During this time, through vehicles in the
shared lane are served at the saturation flow rate of an exclusive through lane.
Considerations of vehicle distribution impose an upper limit on the time
before the first left-turn vehicle arrives when it is used to define a period of
saturation flow. This limit is computed with Equation 31-96.
(1 − 𝑃𝐿 )
Equation 31-96 𝑔𝑓,𝑚𝑎𝑥 = (1 − [1 − 𝑃𝐿 ]0.5 𝑔𝑝 ) − 𝑙1,𝑝 ≥ 0.0
0.5 𝑃𝐿
where gf,max is the maximum time before the first left-turning vehicle arrives and
within which there are sufficient through vehicles to depart at saturation (s), PL is
the proportion of left-turning vehicles in the shared lane (decimal), and all other
The value of 0.5 in two locations in Equation 31-96 represents the
approximate saturation flow rate (in vehicles per second) of through vehicles in
an exclusive lane. This approximation simplifies the calculation and provides
sufficient accuracy in the estimate of gf,max.
The time before the first left-turning vehicle arrives and blocks the shared lane
is computed with Equation 31-97 or Equation 31-98, along with Equation 31-99.
If the approach has one lane, then
0.629
Equation 31-97 𝑔𝑓 = max (𝐺𝑝 𝑒 −0.860 𝐿𝑇𝐶 − 𝑙1,𝑝 , 0.0) ≤ 𝑔𝑓,𝑚𝑎𝑥
otherwise
0.717
Equation 31-98 𝑔𝑓 = max (𝐺𝑝 𝑒 −0.882 𝐿𝑇𝐶 − 𝑙1,𝑝 , 0.0) ≤ 𝑔𝑓,𝑚𝑎𝑥
with
𝑣𝑙𝑡 𝐶
Equation 31-99 𝐿𝑇𝐶 =
3,600
where
lane (s),
LTC = left-turn flow rate per cycle (veh/cycle), and
vlt = left-turn demand flow rate (veh/h).
The approach is considered to have one lane for this step if (a) there is one
lane serving all vehicles on the approach and (b) the left-turn movement on this
approach shares the one lane.

Step 3. Determine Permitted Left-Turn Saturation Flow Rate

This step applies when left-turning vehicles are served by using the permitted
mode or the protected-permitted mode from an exclusive lane. The saturation
flow rate for permitted left-turn operation is calculated with Equation 31-100.
𝑣𝑜 𝑒 −𝑣𝑜 𝑡𝑐𝑔/3,600
𝑠𝑝 = Equation 31-100
1 − 𝑒 −𝑣𝑜 𝑡𝑓ℎ/3,600
where
sp = saturation flow rate of a permitted left-turn movement (veh/h/ln),
vo = opposing demand flow rate (veh/h),
tcg = critical headway = 4.5 (s), and
The opposing demand flow rate vo is determined to be one of two cases. In
Case 1, vo equals the sum of the opposing through and right-turn volumes. In
Case 2, vo equals the opposing through volume. Case 2 applies when there is a
through movement on the opposing approach and one of the following
conditions applies: (a) there is an exclusive right-turn lane on the opposing
approach and the analyst optionally indicates that this lane does not influence
the left-turn drivers’ gap acceptance, or (b) there is no right-turn movement on
the opposing approach. Case 1 applies whenever Case 2 does not apply.
In those instances in which the opposing volume equals 0.0 veh/h during the
analysis period, the opposing volume is set to a value of 0.1 veh/h.
The opposing demand flow rate is not adjusted for unequal lane use in this
equation. Increasing this flow rate to account for unequal lane use would
misrepresent the frequency and size of headways in the opposing traffic stream.
Thus, this adjustment would result in the left-turn saturation flow rate being
underestimated.
Step 4. Determine Through-Car Equivalent

This step applies when left-turning vehicles are served by using the
permitted mode or the protected-permitted mode. Two variables are computed
to quantify the relationship between left-turn saturation flow rate and the base
saturation flow rate. The first variable represents the more common case in
which left-turning vehicles filter through an oncoming traffic stream. It is
computed from Equation 31-101.
𝑠𝑜
𝐸𝐿1 =
𝑠𝑝 Equation 31-101

where
so = base saturation flow rate (pc/h/ln), and
sp = saturation flow rate of a permitted left-turn movement (veh/h/ln).
The second variable to be computed represents the case in which the
opposing approach has one lane. It describes the saturation flow rate during the
time interval coincident with the queue service time of the opposing queue. For
this case, the saturation flow rate during the period after the arrival of the first
blocking left-turning vehicle and before the end of the opposing-queue service
time is influenced by the proportion of left-turning vehicles in the opposing
traffic stream. These vehicles create artificial gaps in the opposing traffic stream
through which the blocking left-turning vehicles on the subject approach can
turn. This effect is considered through calculation of the following through-car
equivalency factor by using Equation 31-102 with Equation 31-103.
1 − (1 − 𝑃𝑙𝑡𝑜 )𝑛𝑞
Equation 31-102 𝐸𝐿2 = ≥ 𝐸𝐿
𝑃𝑙𝑡𝑜
with
Equation 31-103 𝑛𝑞 = 0.278(𝑔𝑝 − 𝑔𝑢 − 𝑔𝑓 ) ≥ 0.0
where
EL2 = equivalent number of through cars for a permitted left-turning vehicle
Plto = proportion of left-turning vehicles in the opposing traffic stream
(decimal),
nq = maximum number of opposing vehicles that could arrive after gf and
before gu (veh), and
The value of 0.278 in Equation 31-103 represents the approximate saturation
flow rate (in vehicles per second) of vehicles in the opposing shared lane. This
approximation simplifies the calculation and provides sufficient accuracy in the
estimation of nq.
There is one lane on the opposing approach when this approach has one lane
serving through vehicles, a left-turn movement that shares the through lane, and
one of the following conditions applies: (a) there is an exclusive right-turn lane
on the opposing approach and the analyst optionally indicates that this lane does
not influence the left-turn drivers’ gap acceptance, (b) there is a right-turn
movement on the opposing approach and it shares the through lane, or (c) there
is no right-turn movement on the opposing approach.

Step 5. Determine Proportion of Turns in a Shared Lane

This step applies when turning vehicles share a lane with through vehicles
and the approach has two or more lanes. The proportion of turning vehicles in
the shared lane is used in the next step to determine the saturation flow rate for
the shared lane.
The proportion of left-turning vehicles in the shared lane PL is computed if
the shared lane includes left-turning vehicles. The proportion of right-turning
vehicles in the shared lane PR is computed if the shared lane includes right-
turning vehicles. Guidance for computing these two variables is provided in
Section 2.
If the approach has one traffic lane, then PL equals the proportion of left-
turning vehicles on the subject approach Plt, and PR equals the proportion of
right-turning vehicles on the subject approach Prt.
Step 6. Determine Lane Group Saturation Flow Rate

The saturation flow rate for the lane group is computed during this step.
When the lane group consists of an exclusive lane operating in the protected
mode, then it has one saturation flow rate. This rate equals the adjusted
saturation flow rate computed by the procedure described in the motorized
vehicle methodology in Section 3 of Chapter 19.
The focus of discussion in this step is the calculation of saturation flow rate
for lane groups that are not in an exclusive lane or operating in the protected
mode. Thus, the discussion in this step focuses on shared-lane lane groups and
lane groups for which the permitted or protected-permitted mode is used. As the
discussion indicates, these lane groups often have two or more saturation flow
rates, depending on the phase sequence and operational mode of the turn
movements.
Permitted Right-Turn Operation in Exclusive Lane

The saturation flow rate for a permitted right-turn operation in an exclusive
𝑠𝑟 = 𝑠𝑜 𝑓𝑤 𝑓𝐻𝑉𝑔 𝑓𝑝 𝑓𝑏𝑏 𝑓𝑎 𝑓𝐿𝑈 𝑓𝑅𝑇 𝑓𝑅𝑝𝑏 𝑓𝑤𝑧 𝑓𝑚𝑠 𝑓𝑠𝑝 Equation 31-104
where sr is the saturation flow rate in an exclusive right-turn lane group with
permitted operation (veh/h/ln), and the other variables are defined following
Equation 19-8 in Chapter 19.
Permitted Right-Turn Operation in Shared Lane

The saturation flow rate for permitted right-turn operation in a shared lane is
𝑠𝑡ℎ Equation 31-105
𝑠𝑠𝑟 =
𝐸
1 + 𝑃𝑅 ( 𝑅 − 1)
𝑓𝑅𝑝𝑏

where
ssr = saturation flow rate in shared right-turn and through lane group with
buses, area type, work zone presence, downstream lane blockage, and
spillback) (veh/h/ln),
PR = proportion of right-turning vehicles in the shared lane (decimal),
ER = equivalent number of through cars for a protected right-turning
vehicle = 1.18, and
fRpb = pedestrian–bicycle adjustment factor for right-turn groups.
The value of fRpb is obtained by the procedure described in Section 2.
Protected-Permitted Right-Turn Operation in Exclusive Lane

Two saturation flow rates are associated with protected-permitted operation.
The saturation flow rate during the protected period srt is computed with
Equation 31-106.
Equation 31-106 𝑠𝑟𝑡 = 𝑠𝑜 𝑓𝑤 𝑓𝐻𝑉𝑔 𝑓𝑝 𝑓𝑏𝑏 𝑓𝑎 𝑓𝐿𝑈 𝑓𝑅𝑇 𝑓𝑤𝑧 𝑓𝑚𝑠 𝑓𝑠𝑝
where srt is the saturation flow rate of an exclusive right-turn lane with protected
operation (veh/h/ln), and the other variables are defined following Equation 19-8
in Chapter 19.
The saturation flow rate during the permitted period is computed with
Equation 31-104.
Permitted Left-Turn Operation in Shared Lane

There are three possible saturation flow periods during the effective green
time associated with permitted left-turn operation in a shared lane. The first
period occurs before the arrival of the first left-turning vehicle in the shared lane.
This left-turning vehicle will block the shared lane until the opposing queue
clears and a gap is available in the opposing traffic stream. The duration of this
flow period is gf. The saturation flow during this period is equal to sth.
The second period of flow begins after gf and ends with clearance of the
opposing queue. It is computed with Equation 31-107.
Equation 31-107 𝑔𝑑𝑖𝑓𝑓 = 𝑔𝑝 − 𝑔𝑢 − 𝑔𝑓 ≥ 0.0
where gdiff is the supplemental service time (s), and all other variables are as
previously defined. This period may or may not exist, depending on the values
of gu and gf.
If there are two or more opposing traffic lanes, then the saturation flow
during the second period ssl2 equals 0.0 veh/h/ln. However, if the opposing
approach has only one traffic lane, then the flow during this period occurs at a
reduced rate that reflects the blocking effect of left-turning vehicles as they await
an opposing left-turning vehicle. Left-turning vehicles during this period are

assigned a through-car equivalent EL2. The saturation flow rate for the shared
𝑠𝑠𝑙2 =
𝐸
1 + 𝑃𝐿 ( 𝐿2 − 1)
𝑓𝐿𝑝𝑏
where ssl2 is the saturation flow rate in the shared left-turn and through lane
group during Period 2 (veh/h/ln), PL is the proportion of left-turning vehicles in
the shared lane (decimal), and all other variables are as previously defined.
There is one lane on the opposing approach when this approach has one lane
serving through vehicles, a left-turn movement that shares the through lane, and
one of the following conditions applies: (a) there is an exclusive right-turn lane
on the opposing approach and the analyst optionally indicates that this lane does
not influence the left-turn drivers’ gap acceptance, (b) there is a right-turn
movement on the opposing approach and it shares the through lane, or (c) there
is no right-turn movement on the opposing approach.
The third period of flow begins after clearance of the opposing queue or
arrival of the first blocking left-turn vehicle, whichever occurs last. Its duration
equals the smaller of gp – gf or gu. The saturation flow rate for this period is
𝑠𝑡ℎ
𝑠𝑠𝑙3 = Equation 31-109
𝐸
1 + 𝑃𝐿 ( 𝐿1 − 1)
𝑓𝐿𝑝𝑏
group during Period 3 (veh/h/ln).
For multiple-lane approaches, the impact of the shared lane is extended to
include the adjacent through traffic lanes. Specifically, queued drivers are
observed to maneuver from lane to lane on the approach to avoid delay
associated with the left-turning vehicles in the shared lane. The effect of this
impact is accounted for by multiplying the saturation flow rate of the adjacent
lanes by a factor of 0.91.
Permitted Left-Turn Operation in Exclusive Lane

There are two possible saturation flow periods during the effective green
time associated with permitted left-turn operation in an exclusive lane. The two
flow periods are discussed in reverse order, with the second period of flow
discussed first.
The second period of flow begins after clearance of the opposing queue. Its
duration is gu. The saturation flow rate for this period is computed with Equation
31-110.

Equation 31-110 𝑠𝑙 = 𝑠𝑝 𝑓𝑤 𝑓𝐻𝑉𝑔 𝑓𝑝 𝑓𝑏𝑏 𝑓𝑎 𝑓𝐿𝑈 𝑓𝐿𝑝𝑏 𝑓𝑤𝑧 𝑓𝑚𝑠 𝑓𝑠𝑝

where sl is the saturation flow rate in an exclusive left-turn lane group with
permitted operation (veh/h/ln), and all other variables are defined following
Equation 19-8 in Chapter 19.
The first period of flow begins with the start of the effective green period and
ends with the clearance of the opposing queue. It is computed by using Equation
31-107 with the variable gf equal to 0.0.
If there are two or more opposing traffic lanes, then the saturation flow during
the first period sl1 equals 0.0 veh/h/ln. However, if the opposing approach has only
one traffic lane, then the saturation flow rate is computed with Equation 31-111.
Equation 31-111 𝑠𝑙
𝑠𝑙1 =
𝐸
( 𝐿2 )
𝑓𝐿𝑝𝑏
where sl1 is the saturation flow rate in the exclusive left-turn lane group during
Period 1 (veh/h/ln). The discussion following Equation 31-108 provides guidance
for determining whether the opposing approach has only one traffic lane.
Protected-Permitted Left-Turn Operation in Exclusive Lane

Two saturation flow rates are associated with protected-permitted operation.
The saturation flow rate during the protected period slt is computed with
Equation 31-112.
Equation 31-112 𝑠𝑙𝑡 = 𝑠𝑜 𝑓𝑤 𝑓𝐻𝑉𝑔 𝑓𝑝 𝑓𝑏𝑏 𝑓𝑎 𝑓𝐿𝑈 𝑓𝐿𝑇 𝑓𝑤𝑧 𝑓𝑚𝑠 𝑓𝑠𝑝
where slt is the saturation flow rate of an exclusive left-turn lane with protected
operation (veh/h/ln), and all other variables are defined following Equation 19-8
in Chapter 19.
The saturation flow rate during the permitted period is computed with
Equation 31-110. The duration of the permitted period is equal to gu.
Protected-Permitted Left-Turn Operation in Shared Lane

The use of a protected-permitted operation in a shared lane has some special
requirements to ensure safe and efficient operation. This operational mode
requires display of the green ball when the left-turn green arrow is displayed
(i.e., the green arrow is not displayed without also displaying the circular green).
The following conditions are applied for actuated, protected-permitted operation
in a shared lane:
• The left-turn phase is set to minimum recall.
• The maximum green setting for the left-turn phase must be less than or
equal to the minimum green for the adjacent through phase.
• If both opposing approaches have protected-permitted operation in a
shared lane, then the phase sequence must be lead–lag.
• No vehicle detection is assigned to the left-turn phase.
• Vehicle detection in the shared lane is assigned to the adjacent through
movement phase.

There are four possible saturation flow periods during the effective green
time associated with protected-permitted left-turn operation in a shared lane.
The first three periods are the same as those for permitted left-turn operation in a
shared lane (as described above).
The fourth period of flow coincides with the left-turn phase (i.e., the
protected period). Its duration is equal to the effective green time for the left-turn
phase gl. The flow rate during this period is computed with Equation 31-113.
𝑠𝑡ℎ
𝑠𝑠𝑙4 = Equation 31-113
1 + 𝑃𝐿 (𝐸𝐿 − 1)
group during Period 4 (veh/h/ln).
For multiple-lane approaches, the impact of the shared lane is extended to
include the adjacent through lanes. This impact is accounted for by multiplying
the saturation flow rate of the adjacent lanes by a factor of 0.91.
Protected Left- and Right-Turn Operation in a Shared Lane

The saturation flow rate in a shared left- and right-turn lane group with
protected operation is computed with Equation 31-114.
𝑠𝑙𝑟 =
1 + 𝑃𝐿 (𝐸𝐿 − 1) + 𝑃𝑅 (𝐸𝑅 − 1)
where slr is the saturation flow rate in the shared left- and right-turn lane group
(veh/h/ln).
Step 7. Define Queue Accumulation Polygon

During this step, the green times and saturation flow rates are used to
construct the QAP associated with each lane group. The polygon is then used to
estimate uniform delay and queue service time. The lane group with the longest
queue service time dictates the queue service time for the phase.
The QAP in Exhibit 31-11 applies to either a through lane group or a left- or
right-turn lane group with exclusive lanes operating with the protected mode.
This polygon also applies to split phasing and to shared lane groups serving
through and right-turning vehicles operating with the permitted mode. For split
phasing, each approach is evaluated separately to determine its queue service
time and uniform delay. If the approach has left- or right-turn lanes, then a
separate polygon is constructed for each turn lane group.
More complicated combinations of lane assignment, phase sequence, and
left-turn operational mode dictate more complicated polygons. A polygon (or its
tabular equivalent) must be derived for each combination. The most common
combinations are illustrated in Exhibit 31-13 through Exhibit 31-16.

Exhibit 31-13
QAP for Permitted Left-Turn
Operation in an Exclusive
Lane
Exhibit 31-14
QAP for Permitted Left-Turn
Operation in a Shared Lane
Exhibit 31-15
QAP for Leading, Protected-
Permitted Left-Turn Operation
in an Exclusive Lane
The concept is extended to shared left-turn and through lane groups with
protected-permitted operation in Exhibit 31-17 and Exhibit 31-18. Other polygon
shapes exist, depending on traffic flow rates, phase sequence, lane use, and left-
turn operational mode. The concept of polygon construction must be extended to
these other combinations to accurately estimate queue service time and uniform
delay.

Exhibit 31-16
QAP for Lagging, Protected-
Exhibit 31-17
QAP for Leading, Protected-
in a Shared Lane
Exhibit 31-18
QAP for Lagging, Protected-
in a Shared Lane
Most of the variables shown in the following exhibits are defined in a

previous subsection. Other variables are defined as follows:
gl = effective green time for left-turn phase (s);
gps = queue service time during permitted left-turn operation (s);
Qq = queue size at the start of gu (veh);
Qp = queue size at the end of permitted service time (veh);
Qp’ = queue size at the end of permitted service time, adjusted for sneakers
(veh); and
Qf = queue size at the end of gf (veh).

The polygon in Exhibit 31-13 applies to the left-turn lane group with an
exclusive lane that operates in the permitted mode during the adjacent through
phase. If the phase extends to max-out, then some left-turning vehicles will be
served as sneakers. The expected number of sneakers for this mode is reduced if
downstream lane blockage or spillback is present [i.e., sneakers = ns fms fsp, where
ns is the number of sneakers per cycle = 2.0 (veh), fms is the adjustment factor for
downstream lane blockage, and fsp is the adjustment factor for sustained spillback].
The polygon in Exhibit 31-14 applies to the left-turn and through lane group
on a shared lane approach with permitted operation. If the phase extends to max-
out, then some left-turning vehicles will be served as sneakers. The expected
number of sneakers (shown as 1 + PL) is computed as (1 + PL) fms fsp, where PL is
the proportion of left-turning vehicles in the shared lane.
The polygon in Exhibit 31-15 applies to left-turn movements that have
protected-permitted operation with a leading left-turn phase and an exclusive
lane. The polygon in Exhibit 31-16 applies to left-turn movements that have
protected-permitted operation with a lagging left-turn phase and an exclusive
lane. If a queue exists at the end of the permitted period for either polygon, then
the queue is reduced by the number of sneakers (where sneakers = ns fms fsp).
protected-permitted operation with a leading left-turn phase and a shared left-
turn and through lane group. The polygon in Exhibit 31-18 applies to the same
movements and operation but with a lagging left-turn phase. If a queue exists at
the end of the permitted period for either polygon, then the queue is reduced by
the expected number of sneakers [which is computed as (1 + PL) fms fsp].
As noted above, all polygons are based on the requirement that lane volume
cannot exceed lane capacity for the purpose of estimating the queue service time.
This requirement is met in the polygons shown because the queue size equals
0.0 vehicles at some point during the cycle.
Exhibit 31-14 through Exhibit 31-18 are shown to indicate that queue size
equals 0.0 vehicles at the start of the cycle (i.e., time = 0.0 s). In fact, the queue
may not equal 0.0 vehicles at the start of the cycle for some signal timing and
traffic conditions. Rather, there may be a nonzero queue at the start of the cycle,
and a queue of 0.0 vehicles may not be reached until a different time in the cycle.
Thus, in modeling any of the polygons in Exhibit 31-14 through Exhibit 31-18, an
iterative process is required. For the first iteration, the queue is assumed to equal
0.0 vehicles at the start of the cycle. The polygon is then constructed, and the
queue status is checked at the end of the cycle. If the queue at the end of the cycle
is not 0.0 vehicles, then this value is used as a starting point in a second polygon
construction. The second polygon will result in a queue at the end of the cycle
that equals the queue used at the start of the cycle. Moreover, a queue value of
0.0 vehicles will occur at some point in the cycle.
A. Compute Uniform Delay and Queue Service Time

The procedure for calculating uniform delay and queue service time is
described in this step. Exhibit 31-19 is used for this purpose.

Exhibit 31-19
Polygon for Uniform Delay
Calculation
The area bounded by the polygon represents the total delay incurred during
the average cycle. The total delay is then divided by the number of arrivals per
cycle to estimate the average uniform delay. These calculations are summarized
in Equation 31-115 with Equation 31-116.
0.5 ∑𝑖=1(𝑄𝑖−1 + 𝑄𝑖 ) 𝑡𝑡,𝑖
𝑑1 = Equation 31-115
𝑞𝐶
with
𝑡𝑡,𝑖 = min (𝑡𝑑,𝑖 , 𝑄𝑖−1 /𝑤𝑞 ) Equation 31-116
where d1 is the uniform delay (s/veh), tt,i is the duration of trapezoid or triangle in
interval i (s), wq is the queue change rate (i.e., slope of the upper boundary of the
trapezoid or triangle) (veh/s), and all other variables are as previously defined.
The summation term in Equation 31-115 includes all intervals for which there
is a nonzero queue. In general, tt,i will equal the duration of the corresponding
interval. However, during some intervals, the queue will decrease to 0.0 vehicles
and tt,i will be only as long as the time required for the queue to dissipate
(= Qi–1/wq). This condition is shown to occur during Time Interval 4 in Exhibit 31-19.
The time required for the queue to dissipate represents the queue service
time. The queue can dissipate during one or more intervals for turn movements
that operate in the protected-permitted mode and for shared-lane lane groups.
For lane groups with exclusive lanes and protected operation, there is one
queue service time. It is followed by the green extension time.
For permitted left-turn operation in an exclusive lane, there is one queue
service time. It is followed by the green extension time.
For permitted left-turn operation in a shared lane, there can be two queue
service times. The green extension time follows the last service time to occur.
For protected-permitted left-turn operation in an exclusive lane, there can be
two queue service times. The service time that ends during the protected period
is followed by the green extension time.

For protected-permitted left-turn operation in a shared lane, there can be

three queue service times. The green extension time can follow the service time
that ends during the protected period, but it is more likely to follow the last
service time to occur during the permitted period.
For phases serving through or right-turning vehicles in two or more lane
groups, the queue service time is measured from the start of the phase to the time
when the queue in each lane group has been serviced (i.e., the longest queue
service time controls). This consideration is extended to lane groups with shared
through and left-turning vehicles.
B. Calculate Lane Group Capacity

This step describes the procedure used to calculate lane group capacity. It is
based on the QAP and considers all opportunities for service during the cycle.
The equations vary, depending on the left-turn operational mode, phase
sequence, and lane assignments for the subject lane group.
Protected Left-Turn Operation in Exclusive Lane

The capacity for a protected left-turn operation in an exclusive-lane lane
group is computed with Equation 31-117.
𝑔𝑙 𝑠𝑙𝑡
Equation 31-117 𝑐𝑙,𝑒,𝑝 = 𝑁𝑙
𝐶
where cl,e,p is the capacity of an exclusive-lane lane group with protected left-turn
operation (veh/h), gl is the effective green time for the left-turn phase (s), Nl is the
number of lanes in the exclusive left-turn lane group (ln), and all other variables
The available capacity for the lane group is computed with Equation 31-118.
𝐺𝑚𝑎𝑥 𝑠𝑙𝑡
Equation 31-118 𝑐𝑎,𝑙,𝑒,𝑝 = 𝑁𝑙
𝐶
where ca,l,e,p is the available capacity of an exclusive-lane lane group with
protected left-turn operation (veh/h), Gmax is the maximum green setting (s), and
Equation 31-117 and Equation 31-118 can also be used to calculate the
capacity of lane groups composed of through lanes and lane groups composed of
right-turn lanes with proper substitution of saturation flow rate, number of lanes,
and maximum green variables.
Permitted Left-Turn Operation in Exclusive Lane

The capacity for a permitted left-turn operation in an exclusive-lane lane
group is computed with Equation 31-119.
𝑔𝑢 𝑠𝑙 + 3,600 𝑛𝑠 𝑓𝑚𝑠 𝑓𝑠𝑝

Equation 31-119 𝑐𝑙,𝑒 = 𝑁𝑙
𝐶
where cl,e is the capacity of an exclusive-lane lane group with permitted left-turn
operation (veh/h), ns is the number of sneakers per cycle = 2.0 (veh), and all other

(𝐺𝑚𝑎𝑥 − 𝑔)𝑠𝑙
𝑐𝑎,𝑙,𝑒 = 𝑐𝑙,𝑒 + 𝑁𝑙 Equation 31-120
𝐶
where ca,l,e is the available capacity of an exclusive-lane lane group with permitted
left-turn operation (veh/h), and all other variables are as previously defined.
The saturation flow rate sl is specifically included in the term with the
maximum green setting Gmax in Equation 31-120 because this rate represents the
saturation flow rate present at the end of the green interval. That is, it is the
saturation flow rate that would occur when the green is extended to its maximum
green limit as a result of cycle-by-cycle fluctuations in the demand flow rate.
Permitted Left-Turn Operation in Shared Lane

The capacity for a permitted left-turn operation in a shared-lane lane group
is computed with Equation 31-121.
𝑔𝑝 𝑠𝑠𝑙 + 3,600(1 + 𝑃𝐿 ) 𝑓𝑚𝑠 𝑓𝑠𝑝
𝑐𝑠𝑙 = Equation 31-121
𝐶
where csl is the capacity of a shared-lane lane group with permitted left-turn
operation (veh/h), ssl is the saturation flow rate in a shared left-turn and through
lane group with permitted operation (veh/h/ln), and all other variables are as
previously defined.
The saturation flow rate in Equation 31-121 is computed with Equation 31-
122 (all variables are as previously defined).
𝑠𝑡ℎ 𝑔diff min(𝑔𝑝 − 𝑔𝑓 , 𝑔𝑢 )

𝑠𝑠𝑙 = (𝑔𝑓 + + ) Equation 31-122
𝑔𝑝 𝐸 𝐸
1 + 𝑃𝐿 [ 𝐿2 − 1] 1 + 𝑃𝐿 [ 𝐿1 − 1]
𝑓𝐿𝑝𝑏 𝑓𝐿𝑝𝑏
(𝐺𝑚𝑎𝑥 − 𝑔𝑝 ) 𝑠𝑠𝑙3
𝑐𝑎,𝑠𝑙 = 𝑐𝑠𝑙 + Equation 31-123
𝐶
where ca,sl is the available capacity of a shared-lane lane group with permitted
left-turn operation (veh/h).
The saturation flow rate ssl3 is specifically included in the term with the
maximum green setting Gmax in Equation 31-123 because this rate represents the
saturation flow rate present at the end of the green interval.
Protected-Permitted Left-Turn Operation in Exclusive Lane

The capacity for a protected-permitted left-turn operation in an exclusive-
lane lane group is computed with Equation 31-124.
𝑔𝑙 𝑠𝑙𝑡 𝑔𝑢 𝑠𝑙 + 3,600 𝑛𝑠 𝑓𝑚𝑠 𝑓𝑠𝑝
𝑐𝑙,𝑒,𝑝𝑝 = ( + ) 𝑁𝑙 Equation 31-124
𝐶 𝐶
where cl,e,pp is the capacity of an exclusive-lane lane group with protected-permitted

𝐺𝑚𝑎𝑥 𝑠𝑙𝑡 𝑔𝑢 𝑠𝑙 + 3,600 𝑛𝑠 𝑓𝑚𝑠 𝑓𝑠𝑝
Equation 31-125 𝑐𝑎,𝑙,𝑒,𝑝𝑝 = ( + ) 𝑁𝑙
𝐶 𝐶
where ca,l,e,pp is the available capacity of an exclusive-lane lane group with
protected-permitted left-turn operation (veh/h) and all other variables are as
previously defined.
Protected-Permitted Left-Turn Operation in Shared Lane

The capacity for a protected-permitted left-turn operation in a shared-lane
lane group is computed with Equation 31-126.
𝑔𝑙 𝑠𝑠𝑙4 𝑔𝑝 𝑠𝑠𝑙 + 3,600(1 + 𝑃𝐿 ) 𝑓𝑚𝑠 𝑓𝑠𝑝
Equation 31-126 𝑐𝑠𝑙,𝑝𝑝 = +
𝐶 𝐶
where csl,pp is the capacity of a shared-lane lane group with protected-permitted
If the lane group is associated with a leading left-turn phase, then the
available capacity for the lane group is computed with Equation 31-127.
(𝐺𝑚𝑎𝑥 − 𝑔𝑝 ) 𝑠𝑠𝑙3
Equation 31-127 𝑐𝑎,𝑠𝑙,𝑝𝑝 = 𝑐𝑠𝑙,𝑝𝑝 +
𝐶
where ca,sl,pp is the available capacity of a shared-lane lane group with protected-
permitted left-turn operation (veh/h).
When the lane group is associated with a lagging left-turn phase, then the
variable ssl3 in Equation 31-127 is replaced by ssl4.
Protected-Permitted Right-Turn Operation in Exclusive Lane

The capacity for a protected-permitted right-turn operation in an exclusive-
lane lane group is computed with Equation 31-128.
𝑔𝑙 𝑠𝑟𝑡 𝑔𝑟 𝑠𝑟
Equation 31-128 𝑐𝑟,𝑒,𝑝𝑝 = ( + ) 𝑁𝑟
𝐶 𝐶
where cr,e,pp is the capacity of an exclusive-lane lane group with protected-
permitted right-turn operation (veh/h), gl is the effective green time for the
complementary left-turn phase (s), gr is the effective green time for the phase
serving the subject right-turn movement during its permitted period, and all
𝐺𝑚𝑎𝑥,𝑟 𝑠𝑟𝑡 𝑔𝑟 𝑠𝑟
Equation 31-129 𝑐𝑎,𝑟,𝑒,𝑝𝑝 = ( + ) 𝑁𝑟
𝐶 𝐶
where ca,r,e,pp is the available capacity of an exclusive-lane lane group with
protected-permitted right-turn operation (veh/h), and Gmax,r is the maximum
green setting for the phase serving the subject right-turn movement during its
permitted period (s).

4. QUEUE STORAGE RATIO
This section discusses queue storage ratio as a performance measure at a

signalized intersection. This measure represents the ratio of the back-of-queue
size to the available vehicle storage length. The first subsection reviews concepts
related to back-of-queue estimation. The second subsection describes a
procedure for estimating the back-of-queue size and queue storage ratio.
The discussion in this section describes basic principles for quantifying the
back of queue for selected types of lane assignment, lane grouping, left-turn
operation, and phase sequence. The analyst is referred to the computational
engine for specific calculation details, especially as they relate to assignments,
groupings, left-turn operation, and phase sequences not addressed in this
section. This engine is described in Section 7.
CONCEPTS
The back of queue represents the maximum backward extent of queued
vehicles during a typical cycle, as measured from the stop line to the last queued
vehicle. The back-of-queue size is typically reached after the onset of the green
indication. The point when it is reached occurs just before the most distant
queued vehicle begins forward motion as a consequence of the green indication
and in response to the forward motion of the vehicle ahead.
A queued vehicle is defined as a vehicle that is fully stopped as a consequence
of the signal. A full stop is defined to occur when a vehicle slows to zero (or a
crawl speed, if in queue) as a consequence of the change in signal indication from
green to red, but not necessarily in direct response to an observed red indication.
The back-of-queue size that is estimated by the equations described here
represents an overall average for the analysis period. It is represented in units of
vehicles.
Background
Queue size is defined here to include only fully stopped vehicles. Vehicles
that slow as they approach the back of the queue are considered to incur a partial
stop but are not considered to be part of the queue. The distinction between a full
and a partial stop is shown in Exhibit 31-20. This exhibit illustrates the trajectory
of several vehicles as they traverse an intersection approach during one signal
cycle. There is no residual queue at the end of the cycle.
Each thin line in Exhibit 31-20 that slopes upward from left to right
represents the trajectory of one vehicle. The average time between trajectories
represents the headway between vehicles (i.e., the inverse of flow rate q). The
slope of the trajectory represents the vehicle’s speed. The curved portion of a
trajectory indicates deceleration or acceleration. The horizontal portion of a
trajectory indicates a stopped condition. The effective red r and effective green g
times are shown at the top of the exhibit. The other variables shown are defined
in the discussion below.
Chapter 31/Signalized Intersections: Supplemental Queue Storage Ratio

Exhibit 31-20
Time–Space Diagram of
Vehicle Trajectory on an
Intersection Approach
Exhibit 31-20 shows the trajectories of eight vehicles. The first five trajectories
(counting from left to right) have a horizontal component to their trajectory that
indicates they have reached a full stop as a result of the red indication. The sixth
trajectory has some deceleration and acceleration but the vehicle does not stop.
This trajectory indicates a partial stop was incurred for the associated vehicle.
The last two trajectories do not incur deceleration or acceleration, and the
associated vehicles do not slow or stop. Thus, the number of full stops Nf is 5 and
the number of partial stops Np is 1. The total number of stops Nt is 6. The back-of-
queue size is equal to the number of full stops.
The back-of-queue size (computed by the procedure described in the next
subsection) represents the average back-of-queue size for the analysis period. It is
based only on those vehicles that arrive during the analysis period and join the
queue. It includes the vehicles that are still in queue after the analysis period ends.
The back-of-queue size for a given lane group is computed with Equation 31-130.
Equation 31-130 𝑄 = 𝑄1 + 𝑄2 + 𝑄3
where
Q = back-of-queue size (veh/ln),
Q1 = first-term back-of-queue size (veh/ln),
Q2 = second-term back-of-queue size (veh/ln), and
Q3 = third-term back-of-queue size (veh/ln).
The first-term back-of-queue estimate quantifies the queue size described in
Exhibit 31-20. It represents the queue caused by the signal cycling through its
phase sequence.
The second-term back-of-queue estimate consists of two queue components.
One component accounts for the effect of random, cycle-by-cycle fluctuations in
demand that occasionally exceed capacity. This fluctuation results in the
occasional overflow queue at the end of the green interval (i.e., cycle failure). The
second component accounts for queuing due to a sustained oversaturation
during the analysis period. This queuing occurs when aggregate demand during
the analysis period exceeds aggregate capacity. It is sometimes referred to as the
deterministic queue component and is shown as variable Q2,d in Exhibit 31-21.
Queue Storage Ratio Chapter 31/Signalized Intersections: Supplemental

Exhibit 31-21
Cumulative Arrivals and
Departures During an
Oversaturated Analysis Period
Exhibit 31-21 illustrates the queue growth that occurs as vehicles arrive at a
demand flow rate v during the analysis period T, which has capacity c. The
deterministic delay component is represented by the triangular area bounded by
the thick line and is associated with an average delay per vehicle represented by
the variable d2,d. The average queue size associated with this delay is shown in
the exhibit as Q2,d. The queue present at the end of the analysis period [= T(v – c)]
is referred to as the residual queue.
The equation used to estimate the second-term queue is based on the
assumption that no initial queue is present at the start of the analysis period. The
third-term back-of-queue estimate is used to account for the additional queuing
that occurs during the analysis period because of an initial queue. This queue is a
result of unmet demand in the previous analysis period. It does not include any
vehicles that may be in queue due to random, cycle-by-cycle fluctuations in
demand that occasionally exceed capacity. When a multiple-period analysis is
undertaken, the initial queue for the second and subsequent analysis periods is
equal to the residual queue from the previous analysis period.
Exhibit 31-22 illustrates the queue due to an initial queue as a trapezoid
shape bounded by thick lines. The average queue is represented by the variable
Q3. The initial queue size is shown as consisting of Qb vehicles. The duration of
time during the analysis period for which the effect of the initial queue is still
present is represented by the variable t. This duration is shown to equal the
analysis period in Exhibit 31-22. However, it can be less than the analysis period
duration for some lower-volume conditions.

Exhibit 31-22
Third-Term Back-of-Queue
Size with Increasing Queue
Exhibit 31-22 illustrates the case in which the demand flow rate v exceeds the
capacity c during the analysis period. In contrast, Exhibit 31-23 and Exhibit 31-24
illustrate alternative cases in which the demand flow rate is less than the capacity.
Exhibit 31-23
Size with Decreasing Queue
Exhibit 31-24
Size with Queue Clearing

In this chapter, initial queue is always used in reference to the initial queue
due to unmet demand in the previous analysis period. It never refers to vehicles
in queue due to random, cycle-by-cycle fluctuations in demand.
Acceleration–Deceleration Delay
The acceleration–deceleration delay da term shown in Exhibit 31-20 is used to
distinguish between a fully and a partially stopped vehicle. This delay term
represents the time required to decelerate to a stop and then accelerate back to
the initial speed, less the time it would have taken to traverse the equivalent
distance at the initial speed.
Various definitions are used to describe when a vehicle is stopped for the
purpose of field measurement. These definitions typically allow the observed
vehicle to be called “stopped” even if it has a slow speed (e.g., 2 to 5 mi/h) while
moving up in the queue. Many stochastic simulation programs also have a
similar allowance. These practical considerations in the count of stopped vehicles
require the specification of a threshold speed that can be used to identify when a
vehicle is effectively stopped. The acceleration–deceleration delay for a specified
threshold speed is estimated with Equation 31-131.
[1.47 (𝑆𝑎 − 𝑆𝑠 )]2 1 1 Equation 31-131
𝑑𝑎 = ( + )
2 (1.47 𝑆𝑎 ) 𝑟𝑎 𝑟𝑑
where
da = acceleration–deceleration delay (s),
Sa = average speed on the intersection approach (mi/h),
Ss = threshold speed defining a stopped vehicle = 5.0 (mi/h),
ra = acceleration rate = 3.5 (ft/s2), and
rd = deceleration rate = 4.0 (ft/s2).
The average speed on the intersection approach Sa is representative of
vehicles that would pass unimpeded through the intersection if the signal were
green for an extended period. It can be estimated with Equation 31-132.
𝑆𝑎 = 0.90 (25.6 + 0.47 𝑆𝑝𝑙 ) Equation 31-132
where Spl is the posted speed limit (mi/h).

The threshold speed Ss represents the speed at or below which a vehicle is
said to be effectively stopped while in queue or when joining a queue. The
strictest definition of this speed is 0.0 mi/h, which coincides with a complete stop.
However, vehicles sometimes move up in the queue while drivers wait for the
green indication. A vehicle that moves up in the queue and then stops again does
not incur an additional full stop. The threshold speed that is judged to
differentiate between vehicles that truly stop and those that are just moving up
in the queue is 5 mi/h.
Acceleration–deceleration delay values from Equation 31-131 typically range
from 8 to 14 s, with larger values in this range corresponding to higher speeds.

Arrival–Departure Polygon
The arrival–departure polygon (ADP) associated with a lane is a graphic tool
for computing the number of full stops Nf. The number of full stops has been
shown to be equivalent to the first-term back-of-queue size (5).
The ADP separately portrays the cumulative number of arrivals and
departures associated with a traffic movement as a function of time during the
average cycle. It is related but not identical to the QAP. The main difference is
that the polygon sides in the ADP represent an arrival rate or a discharge rate but
not both. In contrast, the polygon sides in the QAP represent the combined
arrival and discharge rates that may occur during a common time interval.
The ADP is useful for estimating the stop rate and back-of-queue size, and
the QAP is useful for estimating delay and queue service time.
The ADP for a through movement is presented in Exhibit 31-25, which shows
the polygon for a typical cycle. The red and green intervals are ordered from left
to right in the sequence of presentation so that the last two time periods
correspond to the queue service time gs and green extension time ge of the subject
phase. The variables shown in the exhibit are defined in the following list:
tf = service time for fully stopped vehicles (s),
Nf = number of fully stopped vehicles (veh/ln),
gs = queue service time (s),
ge = green extension time (s),
q = arrival flow rate = v/3,600 (veh/s),
v = demand flow rate (veh/h),
qg = arrival flow rate during the effective green time = P q C/g (veh/s), and
Qr = queue size at the end of the effective red time = qr r (veh).
In application, all flow rate variables are converted to common units of
vehicles per second per lane. The presentation in this section is based on these
units for q and s. If the flow rate q exceeds the lane capacity, then it is set to equal
this capacity.

Exhibit 31-25
Arrival–Departure Polygon
The upper solid trend line in Exhibit 31-25 corresponds to vehicles arriving
at the intersection. The lower solid trend line corresponds to queued vehicles
departing the stop line. The lower trend line is horizontal during the effective
red, denoting no departures. The vertical distance between these two lines at any
instant in time represents the number of vehicles in the queue.
At the start of the effective red, vehicles begin to queue at a rate of qr and
accumulate to a length of Qr vehicles at the time the effective green begins.
Thereafter, the rate of arrival is qg until the end of the effective green period. The
queue service time gs represents the time required to serve the queue present at
the end of the effective red Qr plus any additional arrivals that join the queue
before it fully clears. The dashed line in this exhibit represents only those
vehicles that complete a full stop. The dashed line lags behind the solid arrival
line by one-half the value of da (i.e., da/2). In contrast, the dashed line corresponding
to initiation of the departure process leads the solid departure line by da/2.
One-half the acceleration–deceleration delay da (i.e., da/2) occurs at both the
end of the arrival process and the start of the discharge process. This assumption
is made for convenience in developing the polygon. The derivation of the stop
rate and queue length equations indicates that the two components are always
combined as da. Thus, the assumed distribution of this delay to each of the two
occurrences does not influence the accuracy of the estimated back-of-queue size.
The number of fully stopped vehicles Nf represents the number of vehicles
that arrive before the queue of stopped vehicles has departed. Equation 31-133 is
used for computing this variable (all other variables are as previously defined).
Equation 31-133
𝑁𝑓 = 𝑞𝑟 𝑟 + 𝑞𝑔 (𝑡𝑓 − 𝑑𝑎 )
Equation 31-134 can also be used for estimating Nf.
𝑠 𝑡𝑓
𝑁𝑓 = Equation 31-134
3,600
Combining Equation 31-133 and Equation 31-134 to eliminate Nf and solve
for tf yields Equation 31-135.
𝑞𝑟 𝑟 − 𝑞𝑔 𝑑𝑎
𝑡𝑓 = Equation 31-135
𝑠 − 𝑞𝑔

Equation 31-135 can be used with Equation 31-133 to obtain an estimate of Nf.
The first-term back-of-queue size is then computed with Equation 31-136.
Equation 31-136 𝑄1 = 𝑁𝑓
The polygon in Exhibit 31-25 applies to either a through lane group or a left-
or right-turn lane group with exclusive lanes operating with the protected mode.
Other shapes are possible, depending on whether the lane group includes a
shared lane and whether the lane group serves a permitted (or protected-
permitted) left-turn movement. In general, a unique shape is dictated by each
combination of left-turn operational mode (i.e., permitted, protected, or
protected-permitted) and phase sequence (i.e., lead, lag, or split). A general
procedure for constructing these polygons is described in the next subsection.
PROCEDURE FOR ESTIMATING BACK OF QUEUE FOR SELECTED

LANE GROUPS
This subsection describes a procedure for estimating the back-of-queue size
for a lane group at a signalized intersection. The procedure is described in a
narrative format and does not define every equation needed to develop a
polygon for every combination of lane allocation, left-turn operational mode, and
phase sequence. This approach is taken because of the large number of equations
required to address the full range of combinations found at intersections in most
cities. However, all these equations have been developed and are automated in
the computational engine that is described in Section 7. Some of the equations
presented in the previous section are repeated in this subsection for reader
convenience.
The procedure requires the previous construction of the QAP. The
construction of the QAP is described in Section 3.
Step 1. Determine Acceleration–Deceleration Delay

The acceleration–deceleration delay term is used to distinguish between
fully and partially stopped vehicles. It is computed with Equation 31-131.
Step 2. Define Arrival–Departure Polygon

During this step, the green times and flow rates used previously to construct
the QAP are now used to construct the ADP associated with each lane group
served during a phase.
The ADP in Exhibit 31-25 applies to either a through lane group or a left- or
right-turn lane group with exclusive lanes operating with the protected mode.
This polygon is also applicable to split phasing and to shared lane groups
serving through and right-turning vehicles operating with the permitted mode.
For split phasing, each approach is evaluated separately to determine its overall
stop rate. If the approach has a turn lane, then a separate polygon is constructed
for both the turn and the through lane groups.
More complicated combinations of phase sequence and left-turn operational
mode dictate more complicated polygons. A polygon must be derived for each
combination. The most common combinations are illustrated in Exhibit 31-26
through Exhibit 31-29.

The concept is extended to shared left-turn and through lane groups with
protected-permitted operation in Exhibit 31-30 and Exhibit 31-31. Other polygon
shapes exist, depending on traffic flow rates, phase sequence, lane use, and left-
turn operational mode. The concept of construction must be extended to these
other shapes to estimate accurately the back-of-queue size.
Most variables shown in these exhibits were defined in previous
subsections. The following variables are also defined:
opposing queue (s),
lane (s),
gl = effective green time for left-turn phase (s),
gps = queue service time during permitted left-turn operation (s),
sp = saturation flow rate of a permitted left-turn movement (veh/h/ln),
slt = saturation flow rate of an exclusive left-turn lane with protected
operation = sth/EL (veh/h/ln),
EL = equivalent number of through cars for a protected left-turning vehicle
= 1.05,
buses, and area type) (veh/h/ln), and
PL = proportion of left-turning vehicles in the shared lane (decimal).
Exhibit 31-26
ADP for Permitted Left-Turn
Operation in an Exclusive
Lane

Exhibit 31-27
ADP for Permitted Left-Turn
Operation in a Shared Lane
Exhibit 31-28
ADP for Leading, Protected-
Exhibit 31-29
ADP for Lagging, Protected-

Exhibit 31-30
ADP for Leading, Protected-
in a Shared Lane
Exhibit 31-31
ADP for Lagging, Protected-
in a Shared Lane
The polygon in Exhibit 31-26 applies to the left-turn lane group served by an
exclusive lane that operates in the permitted mode during the adjacent through
phase. If the phase extends to max-out, then some left-turning vehicles will be
served as sneakers. The expected number of sneakers for this mode is reduced if
downstream lane blockage or spillback is present [i.e., sneakers = ns fms fsp, where
ns is the number of sneakers per cycle = 2.0 (veh), fms is the adjustment factor for
downstream lane blockage, and fsp is the adjustment factor for sustained spillback].
The polygon in Exhibit 31-27 applies to the left-turn and through lane group
on a shared-lane approach with permitted operation. If the phase extends to
max-out, then some left-turning vehicles will be served as sneakers. The expected
number of sneakers (shown as 1 + PL) is computed as (1 + PL) fms fsp, where PL is
the proportion of left-turning vehicles in the shared lane, and all other variables
protected-permitted operation with a leading left-turn phase and an exclusive
left-turn lane. The polygon in Exhibit 31-29 applies to the same movements and
operation but with a lagging left-turn phase. If a queue exists at the end of the
permitted period for either polygon, then the queue is reduced by the number of
sneakers (where sneakers = ns fms fsp).


protected-permitted operation with a leading left-turn phase and a shared left-
turn and through lane group. The polygon in Exhibit 31-31 applies to the same
movements and operation but with a lagging left-turn phase. If a queue exists at
the end of the permitted period for either polygon, then the queue is reduced by
the expected number of sneakers [which is computed as (1 + PL) fms fsp].
As noted above, all polygons are based on the requirement that lane volume
cannot exceed lane capacity for the purpose of estimating the queue service time.
This requirement is met in the polygons shown because the queue size equals
0.0 vehicles at some point during the cycle.
Step 3. Define Arrival–Departure Polygon for Fully Stopped Vehicles

During this step, the polygon defined in the previous step is enhanced to
include the polygon shape for the fully stopped vehicles. The fully stopped
vehicle polygon is defined by dashed lines in Exhibit 31-25 through Exhibit 31-31.
Two rules guide the development of this polygon feature. First, the dashed
line that corresponds to arrivals at the stopped queue lags behind the solid
arrival line by da/2 s. Second, the dashed line that corresponds to initiation of the
departure process leads the solid departure line by da/2 s.
Step 4. Compute Service Time for Fully Stopped Vehicles

The service time tf is computed for each polygon constructed in the previous
step. When the polygon in Exhibit 31-25 applies, then either Equation 31-137 or
Equation 31-138 can be used to compute this time.
If da ≤ (1 – P) g X, then
𝑞 𝐶 (1 − 𝑃 − 𝑃 𝑑𝑎 /𝑔)
Equation 31-137 𝑡𝑓 =
𝑠 [1 − min(1, 𝑋) 𝑃]
otherwise
𝑞 𝐶 (1 − 𝑃)(𝑟 − 𝑑𝑎 )
Equation 31-138 𝑡𝑓 =
𝑠 [𝑟 − min(1, 𝑋) (1 − 𝑃)𝑔]
where X is the volume‐to‐capacity ratio.
The saturation flow rate s used in Equation 31-137 and Equation 31-138
represents the adjusted saturation flow rate that is computed by the procedure
described in Section 3 of Chapter 19, Signalized Intersections.
Step 5. Compute the Number of Fully Stopped Vehicles

The number of fully stopped vehicles Nf is computed for each polygon
constructed in Step 3. When the polygon in Exhibit 31-25 applies, then Equation
31-139 or Equation 31-140 can be used to compute the number of stops.
If da ≤ (1 – P) g X, then
Equation 31-139 𝑁𝑓 = 𝑞𝑟 𝑟 + 𝑞𝑔 (𝑡𝑓 − 𝑑𝑎 )
otherwise
Equation 31-140 𝑁𝑓 = 𝑞𝑟 (𝑟 − 𝑑𝑎 + 𝑡𝑓 )

Step 6. Compute the First-Term Back-of-Queue Size

The first-term back-of-queue estimate Q1 (in vehicles per lane) is computed
by using the number of fully stopped vehicles from the previous step. It is
computed with Equation 31-141, where Nf is the number of fully stopped vehicles.
𝑄1 = 𝑁𝑓 Equation 31-141
For some of the more complex ADPs that include left-turn movements
operating with the permitted mode, the queue may dissipate at two or more
points during the cycle. If this occurs, then Nf,i is computed for each of the i
periods between queue dissipation points. The first-term back-of-queue estimate
is then equal to the largest of the Nf,i values computed in this manner.
Step 7. Compute the Second-Term Back-of-Queue Size

Equation 31-142 is used to compute the second-term back-of-queue estimate
Q2 for lane groups served by an actuated phase.
𝑐𝐴 Equation 31-142
𝑄2 = 𝑑
3,600 𝑁 2
where
Q2 = second-term back-of-queue size (veh/ln),
cA = average capacity (veh/h),
d2 = incremental delay (s/veh), and
N = number of lanes in lane group (ln).
If there is no initial queue, then the average capacity cA is equal to the lane
group capacity c. The procedure for computing this capacity is described in
Section 3 of Chapter 19. If there is an initial queue, then the average capacity is
computed with the procedure described in Section 4 of Chapter 19.
Step 8. Compute the Third-Term Back-of-Queue Size

The third-term back-of-queue estimate Q3 is calculated with Equation 31-143
through Equation 31-148.
1 𝑄𝑏 + 𝑄𝑒 − 𝑄𝑒𝑜
𝑄3 = (𝑡𝐴 ) Equation 31-143
𝑁𝑇 2
with
𝑄𝑒 = 𝑄𝑏 + 𝑡𝐴 (𝑣 − 𝑐𝐴 ) Equation 31-144
If v ≥ cA, then
𝑄𝑒𝑜 = 𝑇(𝑣 − 𝑐𝐴 ) Equation 31-145
𝑡𝐴 = 𝑇 Equation 31-146
If v < cA, then

𝑄𝑒𝑜 = 0.0 veh Equation 31-147
𝑡𝐴 = 𝑄𝑏 /(𝑐𝐴 − 𝑣) ≤ 𝑇 Equation 31-148

where
Q3 = third-term back-of-queue size (veh/ln),
tA = adjusted duration of unmet demand in the analysis period (h),
T = analysis period duration (h),
Qb = initial queue at the start of the analysis period (veh),
Qe = queue at the end of the analysis period (veh), and
Qeo = queue at the end of the analysis period when v ≥ cA and Qb = 0.0 (veh).
Step 9. Compute the Back-of-Queue Size

The average back-of-queue estimate Q for a lane group (in vehicles per lane)
is computed with Equation 31-149 (all other variables are as previously defined).
Equation 31-149 𝑄 = 𝑄1 + 𝑄2 + 𝑄3
If desired, a percentile back-of-queue estimate Q% can be computed with
Equation 31-150, and Equation 31-151 through Equation 31-153 can be used to
compute the percentile back-of-queue factor fB%.
Equation 31-150 𝑄% = (𝑄1 + 𝑄2 )𝑓𝐵% + 𝑄3
with
If v ≥ cA, then
𝐼 𝑔 0.33
Equation 31-151 𝑓𝐵% = min (1.8, 1.0 + 𝑧√ + 0.60 𝑧 0.24 ( ) (1.0 − 𝑒 2−2 𝑋𝐴 ))
𝑄1 + 𝑄2 𝐶
Equation 31-152 𝑋𝐴 = 𝑣/𝑐𝐴

If v < cA, then
𝐼
Equation 31-153 𝑓𝐵% = min (1.8, 1.0 + 𝑧√ )
𝑄1 + 𝑄2
where
Q% = percentile back-of-queue size (veh/ln);
fB% = percentile back-of-queue factor;
z = percentile parameter = 1.04 for 85th percentile queue, 1.28 for 90th
percentile queue, and 1.64 for 95th percentile queue;
I = upstream filtering adjustment factor; and
XA = average volume-to-capacity ratio.

Step 10. Compute Queue Storage Ratio

If the lane group is served by a bay or lane of limited storage length, then the
queue storage ratio can be computed by using Equation 31-154 with Equation
31-155.
𝐿ℎ 𝑄
𝑅𝑄 = Equation 31-154
𝐿𝑎
with
𝐿ℎ = 𝐿𝑝𝑐 (1 − 0.01 𝑃𝐻𝑉 ) + 0.01 𝐿𝐻𝑉 𝑃𝐻𝑉 Equation 31-155
where
RQ = queue storage ratio,
La = available queue storage distance (ft/ln),
Lh = average vehicle spacing in stationary queue (ft/veh),
Lpc = stored passenger car lane length = 25 (ft),
LHV = stored heavy-vehicle lane length = 45 (ft), and
PHV = percentage heavy vehicles in the corresponding movement group (%).
Average vehicle spacing is the average length between the front bumpers of
two successive vehicles in a stationary queue. The available queue storage
distance is equal to the turn bay (or lane) length.
The queue storage ratio is useful for quantifying the potential blockage of the
available queue storage distance. If the queue storage ratio is less than 1.0, then
blockage will not occur during the analysis period. Blockage will occur if the
queue storage ratio is equal to or greater than 1.0.
If desired, a percentile queue storage ratio can be computed with Equation
31-156.
𝐿ℎ 𝑄%
𝑅𝑄% = Equation 31-156
𝐿𝑎
where RQ% is the percentile queue storage ratio.

5. PLANNING-LEVEL ANALYSIS APPLICATION
The planning-level analysis application described in this section is intended

to provide the user a means for conducting a simplified and approximate
analysis of signalized intersection operations for motorized vehicles. Chapter 19,
Signalized Intersections, provides a more detailed methodology. The objective of
the planning-level analysis application is to assess whether an intersection’s
geometric conditions are sufficient to handle the projected demand volume.
Within this framework, many of the data required for a full operational analysis
are not needed. This method has several potential uses and applications:
• Conducting sketch-level analyses to quickly assess whether an
intersection’s lane geometry is sufficient to accommodate a given set of
turn-movement demand volumes;
• Evaluating intersection geometry and lane widening alternatives;
• Estimating signal phasing and timing;
• Comparing analysis results against traffic operational performance results
produced by other methods; and
• Educating students, transportation professionals, and nontransportation
professionals about the fundamentals of traffic signal operational
performance.
OVERVIEW OF THE APPLICATION

This subsection provides an overview of the two parts of the planning-level
analysis application. Part I provides an estimate of intersection capacity
sufficiency. Part II extends the analysis from Part I to provide an estimate of
delay and level of service (LOS).
The planning-level analysis application is designed to evaluate the
performance of designated groups of lanes, an intersection approach, and the
entire intersection. A group of lanes designated for separate analysis is referred
to as a lane group. Lane groups form the basis for intersection analysis in the
planning-level analysis application and in the motorized vehicle methodology
described in Chapter 19. However, the criteria for defining a lane group are
different between the two methodologies.
For the planning-level analysis application, all traffic movements for a given
approach (i.e., left, through, and right) must be assigned to at least one lane
group. A lane group can consist of one or more lanes. There are two guidelines to
follow for assigning traffic movements to lane groups:
1. When a traffic movement uses only an exclusive lane (or lanes), it is
analyzed as an exclusive lane group.
2. When two or more traffic movements share a lane, all lanes that convey
those traffic movements are analyzed as a mixed lane group.
When a right-turn movement is shared with a through movement, it is
considered to be a part of the through movement lane group. When a right-turn
movement is shared with a left-turn movement (such as at a T-intersection), it is
Planning-Level Analysis Application Chapter 31/Signalized Intersections: Supplemental

considered to be a part of the left-turn movement lane group. The concept of lane
group is discussed in more detail in the Methodology subsection.
Part I: Intersection Sufficiency Assessment

Part I provides an estimate of the intersection’s volume-to-capacity ratio,
which can be used to assess whether the intersection is likely to operate under,
near, or over capacity during the analysis period. This assessment is predicated
on the critical movement analysis technique developed originally as part of
Transportation Research Circular 212 (6).
Part I generally requires only two inputs: turn movement volume and lane
geometry. Other input data are allowed, but they can also be set to default values
if they are not explicitly known. Part I can be applied by using manual
calculations; it does not require software to implement.
Part I consists of the following steps:
1. Determine left-turn operation.
2. Convert movement volumes to through passenger-car equivalents.
3. Assign flow rates to lane groups.
4. Determine critical lane groups.
5. Determine intersection sufficiency.
Part II: Delay and Level of Service Assessment

Part II extends the results from Part I to produce estimates of volume-to-
capacity ratio, delay, and LOS. For practical purposes, Part II requires a
spreadsheet or other software to compute estimates of delay and LOS. A Part II
analysis requires the initial completion of Steps 1 to 5 of Part I. It then continues
with the following steps:
6. Calculate capacity.
7. Determine delay and LOS.
Limitations
The planning-level analysis application has the following limitations:
• It only considers the performance of motorized vehicles;
• It is based on pretimed operation and thus does not account for the effects
of actuated control;
• It does not analyze all potential combinations of left-turn operation for
opposing approaches (e.g., protected left-turn operation opposed by
permitted left-turn operation is not addressed by the application);
• It does not explicitly consider the effects of poorly timed signals;
• It does not account for upstream or downstream impedances and effects
of short lanes; and
• It does not consider the effects of grade, lane width, bus activity, area
type, pedestrian–vehicle conflicts, or pedestrian–bicycle conflicts;
Chapter 31/Signalized Intersections: Supplemental Planning-Level Analysis Application

however, an “equivalency factor for other conditions” is provided to

allow the analyst to account for these (or other) nonideal conditions.

Exhibit 31-32 describes the input data requirements for conducting an
analysis using the planning-level analysis application.
Exhibit 31-32 Data Item Comments

Required Input Data for the Part I
Planning-Level Analysis
Number of lanes and lane use Required. Exclusive or shared lane use.
Application
Turn movement volumes Required
Intersection peak hour factor Use default value of 0.92 if not known.
Percentage heavy vehicles Use default value of 3% if not known.
On-street parking presence No (default)
Level of pedestrian activity None (default)
Low – 50 p/h
Medium – 200 p/h
High – 400 p/h
Extreme – 800 p/h
Left-turn operation and phase Protected operation—with left-turn phase
sequence Permitted operation—no left-turn phase
Protected operation—split phasing
Protected-permitted operation–with left-turn phase
(Can be estimated—use guidance provided in the application)
Base saturation flow rate (Can be estimated—use guidance provided in the application)
Cycle length (Can be estimated—use guidance provided in the application)
Effective green time Required to evaluate protected-permitted operation, if present
(Can be estimated—use guidance provided in the application)
Part II
Effective green time (Can be estimated—use guidance provided in the application)
Progression quality Good progression
Random arrivals (default)
Poor progression
The analyst is required to specify values for two data items: (a) the volume
for each movement and (b) the number of lanes (and the turn designation for
each lane) on each approach. The effective green time is also required if
protected-permitted left-turn operation is to be evaluated. Default values can be
assumed for the other input data, or the user can specify these values if they are
known.
METHODOLOGY
Part I: Intersection Sufficiency Assessment
The first part of the application consists of five steps. These steps are
completed in sequence to evaluate the capacity sufficiency of the intersection.
Step 1: Determine Left-Turn Operation

For approaches with left-turn movements, the left-turn operational mode
and phase sequence must be defined. The following mode and sequence
combinations are addressed in the planning-level analysis application:
• Protected operation—with left-turn phase. This combination enables the
subject left-turn movement to proceed concurrently with either the
adjacent through movement or the opposing left-turn movement.

• Permitted operation—no left-turn phase. This combination enables the

subject left-turn movement to proceed through the intersection during the
same phase indication as the opposing through movement. It generally
results in higher capacity for the intersection than other combinations.
However, it also produces the highest potential safety conflicts.
• Protected operation—split phasing. With split phasing, the through and left-
turn movements on the subject approach are served in a protected
manner during a common phase. This combination is generally the least
efficient type of operation and is oftentimes used when geometric
properties of the intersection preclude movements on opposing
approaches from proceeding at the same time, or when traffic volumes on
opposite approaches are unbalanced.
• Protected-permitted operation—with left-turn phase. This combination serves
left turns in a protected manner during a left-turn phase and in a
permitted manner during a through phase. If this combination is to be
evaluated, the analyst should refer to the supplemental procedure in the
Protected-Permitted Left-Turn Operations section.
If the operational mode is not known, the following general rules can be
applied to determine if protected operation is appropriate for planning-level
analysis purposes. Protected operation should be assumed if any of the following
conditions are met:
1. The left-turn volume is greater than or equal to 240 veh/h.
2. The product of the left-turn volume and the opposing through volume
exceeds a given threshold (50,000 if there is one opposing through lane,
90,000 if there are two opposing through lanes, and 110,000 if there are
three or more opposing through lanes).
3. There is more than one left-turn lane on the approach.
Several other considerations for choosing a left-turn operation are not
considered to be an explicit part of a planning method. The Traffic Engineering
Handbook (7) provides additional criteria that include the speed of vehicles on the
opposing approach, restrictive sight distances, and accident rates, among others.
Therefore, protected left-turn operation may be appropriate even when the
above conditions are not satisfied.
In some cases, an intersection may have protected left-turn operation on one
approach and permitted left-turn operation on the opposite approach. When this
situation occurs, it is necessary to assume both approaches have protected
operation to use the planning-level analysis application.
Step 2: Convert Movement Volumes to Through Passenger-Car Equivalents

The objective of this step is to convert all movement volumes into through
passenger-car equivalents. The conversion considers one or more of the
following factors:
• Effect of heavy vehicles,
• Variation in flow during the hour,

• Impact of opposing through vehicles on permitted left-turn vehicles,

• Impact of pedestrians on right-turn vehicles,
• Impact of parking maneuvers, and
• Lane utilization.
Equation 31-157 provides the volume adjustment equation. Each of the
factors in this equation is described in the subsequent paragraphs.
Equation 31-157 𝑣𝑎𝑑𝑗 = 𝑉 𝐸𝐻𝑉 𝐸𝑃𝐻𝐹 𝐸𝐿𝑇 𝐸𝑅𝑇 𝐸𝑝 𝐸𝐿𝑈 𝐸other
where
vadj = equivalent through movement flow rate expressed in through
passenger cars per hour (tpc/h),
V = movement volume (veh/h),
EHV = equivalency factor for heavy vehicles,
EPHF = equivalency factor for peaking characteristics,
ERT = equivalency factor for right turns,
ELT = equivalency factor for left turns,
Ep = equivalency factor for parking activity,
ELU = equivalency factor for lane utilization, and
Eother = equivalency factor for other conditions.
Adjustment for Heavy Vehicles

The equivalency factor to convert the mixed traffic stream into passenger car
equivalents is computed with Equation 31-158.
Equation 31-158 𝐸𝐻𝑉 = 1 + 0.01 𝑃𝐻𝑉 (𝐸𝑇 − 1)
where
PHV = percentage of heavy vehicles in the corresponding lane group (%), and
ET = equivalent number of through cars for each heavy vehicle = 2.0.
The recommended passenger car equivalent ET in this method is 2.0. If the
user has more detailed or localized information about the value of ET, then this
value may be used in Equation 31-158.
Adjustment for Variation in Flow During the Hour

The movement volume is adjusted by the peak hour factor to reflect the peak
15-min flow rate, similar to the procedure used in the operational method.
Equation 31-159 is used to compute the peak hour adjustment factor.
1
Equation 31-159 𝐸𝑃𝐻𝐹 =
𝑃𝐻𝐹
where PHF is the peak hour factor (varies between 0.25 and 1.00).

Adjustment for Impedances Experienced by Turning Vehicles

The equivalency factors used to account for impedances experienced by left-
and right-turn movements are shown in Exhibit 31-33 and Exhibit 31-34.
Total Opposing Equivalency Factor for Exhibit 31-33

Volume Vo (veh/h)a Left Turns ELT Planning-Level Analysis:
Left-Turn Operation
Equivalency Factor for Left
Protected—with left-turn phase 1.05
Any Turns
Protected—split phasing
Permitted—no left-turn phase <200 1.1
200–599 2.0
600–799 3.0
800–999 4.0
≥1,000 5.0
Protected-permitted—with Refer to guidance in the Protected-Permitted Left-Turn
left-turn phase Operations section
a
Note: Includes the sum of through and right-turn volumes on the opposing approach, regardless of whether the
right-turn volume is served in an exclusive right-turn lane.
Pedestrian Volume Equivalency Factor for Exhibit 31-34

Level of Pedestrian Activity (p/h) Right Turns ERT Planning-Level Analysis:
None or low 0–199 1.2 Equivalency Factor for Right
Moderate 200–399 1.3 Turns
High 400–799 1.5
Extreme ≥800 2.1
In Exhibit 31-33, the equivalency factor that is applicable to permitted left-

turn movements is based on the opposing volume. This volume is defined as the
sum of opposing through and right-turn movements, regardless of whether the
right-turn volume is served in an exclusive right-turn lane. The equivalency
factor for right turns is a function of the pedestrian activity in the crosswalk that
conflicts with the subject right-turn movement.
Adjustment for Parking Activity

The equivalency factor for on-street parking activity is shown in Exhibit 31-
35. This factor is applicable to through and right-turn vehicles. It is also
applicable to left-turn vehicles on a one-way street when parking is allowed on
the left side.
On-Street No. of Lanes in Equivalency Factor for Exhibit 31-35

Parking Presence Lane Group Parking Activity Ep Planning-Level Analysis:
No All 1.00 Equivalency Factor for Parking
Yes 1 1.20 Activity
2 1.10
3 1.05
Adjustment for Lane Utilization

The planning-level analysis application analyzes the performance of the
heaviest-traveled lane. For lane groups with two or more lanes, the volume is
adjusted to reflect the heaviest-traveled lane. The appropriate equivalency factor
to account for lane utilization is selected from Exhibit 31-36.

Exhibit 31-36 No. of Lanes in Equivalency Factor for

Planning-Level Analysis: Lane Group Movement Lane Group Lane Utilization ELU
Equivalency Factor for Lane Through or shared 1 1.00
Utilization 2 1.05
≥3 1.10
Exclusive left turn 1 1.00
≥2 1.03
Exclusive right turn 1 1.00
≥2 1.13
Adjustment for “Other” Conditions

An adjustment factor for “other” is provided in Equation 31-157. This factor
is a placeholder to allow the user to further adjust the movement volume for
conditions that are not captured by any other adjustment factor. The analyst may
apply any combination of the saturation flow rate adjustment factors presented
in Section 3 of Chapter 19 to reflect other nonideal conditions. In this situation,
Eother is computed as the product of the inverted factors (i.e., Eother = 1/fi × 1/fj × . . .
× 1/fn, where fi, fj, and fn represent the factors in Chapter 19 that are applicable to
the subject movement).
Step 3: Assign Flow Rates to Lane Groups

Initially, lane groups should be checked to determine if a de facto turn lane
exists. A de facto turn lane occurs on approaches with multilane lane groups
where (a) either a left- or right-turn movement is shared with a through movement
and (b) the turning flow rates are sufficiently high, or the impedance to the
turning traffic is sufficiently great, to reasonably expect that the through vehicles
use only the adjacent exclusive through lane(s) and avoid the shared lane.
The presence of a de facto turn lane can be determined by comparing the
total flow rate of turning traffic (left or right) with the lane-equivalent adjusted
flow rate in the shared lane as calculated in Step 2. If the flow rate of turning
traffic is greater than the lane-equivalent adjusted flow rate, a de facto turn lane
should be assumed. De facto turn lanes should be analyzed as exclusive turn
lanes, and thus all through movements should be assigned to the through-only
lane(s).
In cases in which there are multiple turn lanes and one lane is shared with a
through movement, these lanes should be treated as a single lane group that is
designated as the through lane group. For approaches at a T-intersection where
there are only left- and right-turn movements and multiple lanes, and one of the
lanes is shared, the user has the option of coding all lanes as either the right-turn
lane group or the left-turn lane group.
Once lane groups have been defined, the lane group flow rate is divided by
the number of lanes associated with the lane group to obtain the lane flow rate.
Equation 31-160 is used for this purpose.
𝑣𝑎𝑑𝑗,𝑖
Equation 31-160 𝑣𝑖 =
𝑁𝑖
where
vi = lane flow rate for lane group i expressed in through passenger cars per
hour per lane (tpc/h/ln);

vadj,i = equivalent through movement flow rate for lane group i (tpc/h); and
Ni = number of lanes associated with lane group i, accounting for de facto
lanes (ln).
Step 4: Determine Critical Lane Groups

The critical lane groups are identified and the sum of critical-lane flow rates
is determined in this step. Critical lane groups represent the unique combination of
conflicting lane groups that have the highest total flow rate. These critical lane
groups dictate the amount of green time required during each phase. They also
dictate the total cycle length required for the intersection. The critical lane groups
for the north–south and east–west approaches are assessed independently.
This step consists of three tasks. During the first task, the right-turn flow rate
is adjusted to account for right-turn capacity during the complementary left-turn
phase. During the second task, the critical lane groups are identified. During the
third task, the critical-lane group flow rates are added to determine the sum of
critical-lane flow rates.
Step 4a. Adjust Right-Turn Flow Rate

There may be situations in which an exclusive right-turn lane could have a
higher flow rate than the adjacent through lane(s). In this situation, the right
turns that could occur simultaneously with a protected left-turn movement from
the cross street should be deducted from the right-turn flow rate. For example, if
the exclusive northbound right-turn flow rate is 300 tpc/h/ln and the protected
westbound left-turn flow rate is 125 tpc/h/ln, 125 northbound right-turn vehicles
should be assumed to depart the intersection during the westbound left-turn
phase. Thus, 125 should be deducted from the total northbound right-turn flow
rate, resulting in an adjusted northbound right-turn flow rate of 175 tpc/h/ln.
This adjustment is only necessary when the right-turn lane group is critical. If
that is the case, the rules described in Step 4b should replace the through lane
group flow rate with the right-turn lane group flow rate.
Step 4b. Identify Critical Lane Groups

The lane groups that are determined to be critical are identified in this task.
The rules for making this determination are dependent on the left-turn
operational mode and phase sequence. Each of the combinations addressed by
the planning-level analysis application is discussed in the following paragraphs.
Protected operation—with left-turn phase. When opposing approaches use
protected left-turn operation, there are two possible lane group combinations
that could determine the critical-lane flow rate. Each combination comprises a
left-turn lane group and its opposing through (or right-turn) lane group. The
flow rate for each lane group pair is added. The maximum of these two sums
defines the critical-lane flow rate. For the east–west approaches, the critical-lane
flow rate is computed with Equation 31-161.

𝑣 +𝑣
Equation 31-161 𝑉𝑐,𝑝𝑟𝑜𝑡,1 = max [ 𝑣𝐸𝐵𝑙𝑡 + 𝑣𝑊𝐵𝑡ℎ
𝑊𝐵𝑙𝑡 𝐸𝐵𝑡ℎ
where
Vc,prot,1 = critical-lane flow rate for protected left-turn operation on the east–west
approaches (tpc/h/ln), and
vi = lane flow rate for lane group i (i = EBlt: eastbound left turn, WBlt:
westbound left turn, EBth: eastbound through, WBth: westbound
through) (tpc/h/ln).
The two lane groups that add to produce the largest critical-lane flow rate in
Equation 31-161 represent the critical lane groups for the east–west street.
Similarly, for north–south approaches with protected left-turn operation, the
critical-lane flow rate is computed with Equation 31-162.
𝑣 +𝑣
Equation 31-162 𝑉𝑐,𝑝𝑟𝑜𝑡,2 = max [ 𝑣𝑁𝐵𝑙𝑡 + 𝑣 𝑆𝐵𝑡ℎ
𝑆𝐵𝑙𝑡 𝑁𝐵𝑡ℎ
where
Vc,prot,2 = critical-lane flow rate for protected left-turn operation on the north–
south approaches (tpc/h/ln), and
vi = lane flow rate for lane group i (i = NBlt: northbound left turn, SBlt:
southbound left turn, NBth: northbound through, SBth: southbound
through) (tpc/h/ln).
Equation 31-162 represent the critical lane groups for the north–south street.
Permitted operation—no left-turn phase. When opposing approaches use
permitted operation, the critical-lane flow rate will be the highest lane flow rate
of all lane groups associated with the pair of approaches. For the east–west
approaches, the critical-lane flow rate is computed with Equation 31-163.
Equation 31-163 𝑉𝑐,𝑝𝑒𝑟𝑚,1 = max(𝑣𝐸𝐵𝑙𝑡 , 𝑣𝐸𝐵𝑡ℎ , 𝑣𝐸𝐵𝑟𝑡 , 𝑣𝑊𝐵𝑙𝑡 , 𝑣𝑊𝐵𝑡ℎ , 𝑣𝑊𝐵𝑟𝑡 )
where
Vc,perm,1 = critical-lane flow rate for permitted left-turn operation on the east–
west approaches (tpc/h/ln), and
vi = lane flow rate for lane group i (i = EBlt: eastbound left turn, WBlt:
westbound left turn, EBth: eastbound through, WBth: westbound
through, EBrt: eastbound right turn, WBrt: westbound right turn)
(tpc/h/ln).
The lane group that produces the largest critical-lane flow rate in Equation
31-163 represents the critical lane group for the east–west street.
Similarly, for north–south approaches with permitted left-turn operation, the
critical-lane flow rate is computed with Equation 31-164.
Equation 31-164 𝑉𝑐,𝑝𝑒𝑟𝑚,2 = max(𝑣𝑆𝐵𝑙𝑡 , 𝑣𝑆𝐵𝑡ℎ , 𝑣𝑆𝐵𝑟𝑡 , 𝑣𝑁𝐵𝑙𝑡 , 𝑣𝑁𝐵𝑡ℎ , 𝑣𝑁𝐵𝑟𝑡 )

where
Vc,perm,2 = critical-lane flow rate for permitted left-turn operation on the north–
south approaches (tpc/h/ln), and
vi = lane flow rate for lane group i (i = SBlt: southbound left turn, NBlt:
northbound left turn, SBth: southbound through, NBth: northbound
through, SBrt: southbound right turn, NBrt: northbound right turn)
(tpc/h/ln).
The lane group that produces the largest critical-lane flow rate in Equation
31-164 represents the critical lane group for the north–south street.
Protected operation—split phasing. When opposing approaches use split
phasing (i.e., when only one approach is served during a phase), the critical-lane
flow rate for a given approach will be the highest lane flow rate of all lane groups
for that approach. The critical-lane flow rate for the two opposing approaches
will be the sum of the highest lane flow rate for each approach. For the east–west
approaches, the critical-lane flow rate is computed with Equation 31-165.
𝑉𝑐,split,1 = max(𝑣𝐸𝐵𝑙𝑡 , 𝑣𝐸𝐵𝑡ℎ , 𝑣𝐸𝐵𝑟𝑡 ) + max (𝑣𝑊𝐵𝑙𝑡 , 𝑣𝑊𝐵𝑡ℎ , 𝑣𝑊𝐵𝑟𝑡 ) Equation 31-165
where Vc,split,1 is the critical-lane flow rate for split phasing on the east–west
approaches (tpc/h/ln).
Equation 31-165 represent the critical lane groups for the east–west street.
Similarly, for the north–south approaches with split phasing, the critical-lane
flow rate is computed with Equation 31-166.
𝑉𝑐,split,2 = max(𝑣𝑆𝐵𝑙𝑡 , 𝑣𝑆𝐵𝑡ℎ , 𝑣𝑆𝐵𝑟𝑡 ) + max(𝑣𝑁𝐵𝑙𝑡 , 𝑣𝑁𝐵𝑡ℎ , 𝑣𝑁𝐵𝑟𝑡 ) Equation 31-166
where Vc,split,2 is the critical-lane flow rate for split phasing on the north–south
approaches (tpc/h/ln).
Equation 31-166 represent the critical lane groups for the north–south street.
Protected-permitted operation—with left-turn phase. If protected-permitted
operation is to be evaluated, the analyst should refer to the supplemental
procedure in the Protected-Permitted Left-Turn Operations subsection.
Step 4c. Calculate the Sum of Critical-Lane Flow Rates

Once the critical lane groups have been identified, the sum of critical-lane
flow rates for the intersection can be computed by adding the lane flow rate
associated with each critical lane group. Alternatively, the sum of critical-lane
flow rates can be computed by adding the critical-lane group flow rate for each
intersecting street, as calculated in the previous task. The following four cases
illustrate this technique for some example combinations of left-turn operation
and phase sequence using Equation 31-167 through Equation 31-170.
Case 1: East–west and north–south approaches use protected operation—
with left-turn phase.
𝑉𝑐 = 𝑉𝑐,𝑝𝑟𝑜𝑡,1 + 𝑉𝑐,𝑝𝑟𝑜𝑡,2 Equation 31-167

where Vc is the sum of the critical-lane flow rates (tpc/h/ln).

Case 2: East–west and north–south approaches use permitted operation—no
left-turn phase.
Equation 31-168 𝑉𝑐 = 𝑉𝑐,𝑝𝑒𝑟𝑚,1 + 𝑉𝑐,𝑝𝑒𝑟𝑚,2
Case 3: East–west approaches use protected operation—with left-turn phase
and north–south approaches use permitted operation—no left-turn phase.
Equation 31-169 𝑉𝑐 = 𝑉𝑐,𝑝𝑟𝑜𝑡,1 + 𝑉𝑐,𝑝𝑒𝑟𝑚,2
Case 4: East–west approaches use protected operation—with left-turn phase
and north–south approaches use protected operation—split phasing.
Equation 31-170 𝑉𝑐 = 𝑉𝑐,𝑝𝑟𝑜𝑡,1 + 𝑉𝑐,𝑠𝑝𝑙𝑖𝑡,2
Step 4d. Identify Critical Phases

The critical phases identified in this task are used in Part II. If Part II is not
part of the analysis, then this task can be skipped.
For this task, one critical phase is associated with each critical lane group, as
identified in Step 4b. The flow rate that corresponds to a critical lane group (and
critical phase i) is called the critical-lane flow rate vc,i. By definition, the sum of
these critical-lane flow rates equals the sum of critical-lane flow rates Vc.
For example, consider an intersection for which Equation 31-167 is
determined to be applicable (i.e., the intersection has protected operation—with
left-turn phase on both approaches). If the eastbound left-turn and westbound
through phases are found to yield the critical-lane flow rate Vc,prot,1, then the
eastbound left-turn phase and the westbound through phase are identified as
critical phases. The critical-lane flow rates for the east–west approaches are vc,EBlt
(= vEBlt) and vc,WBth (= vWBth).
Step 5: Determine Intersection Sufficiency

This step consists of four tasks. The first task is to determine the cycle length,
and the second is to calculate intersection capacity. The third task is to compute
the intersection volume-to-capacity ratio. The fourth task is to determine whether
the intersection is operating under, near, or over its capacity.
If local data describing cycle length and base saturation flow rate are not
available, then a default intersection capacity cI of 1,650 tpc/h/ln can be used. This
default value reflects a base saturation flow rate of 1,900 pc/h/ln, a lost time of 4.0
s per phase, and a cycle length equal to 30 s per critical phase. If the default
intersection capacity is used, then the analyst can proceed to Step 5c.
Step 5a. Calculate Cycle Length

If cycle length is known, then the analyst can proceed to Step 5b.
For purposes of conducting a planning-level analysis, the analyst can assume
a cycle length equal to 30 s for each critical phase. For example, an intersection
with a protected left-turn phase for each of the eastbound and westbound
approaches and permitted left-turn operation for the northbound and

southbound approaches could be assumed to have a 90-s cycle length. The

selection of a cycle length in practice should be based on consideration of
multiple factors including (a) local agency policies and practices and (b) needs of
nonmotorized users.
Step 5b. Calculate Intersection Capacity

Intersection capacity is calculated with Equation 31-171.
𝐶 − (𝑛𝑐𝑝 𝑙𝑡 )
𝑐𝐼 = 𝑠𝑜 Equation 31-171
𝐶
where
cI = intersection capacity (tpc/h/ln),
so = base saturation flow rate (pc/h/ln),
ncp = number of critical phases, and
lt = phase lost time (s).
A default phase lost time of 4.0 s for each critical phase is recommended. A
default value for base saturation flow rate can be obtained from Exhibit 19-11.
Step 5c. Calculate the Intersection Volume-to-Capacity Ratio

The critical intersection volume-to-capacity ratio is calculated with Equation
31-172.
𝑉𝑐
𝑋𝑐 = Equation 31-172
𝑐𝐼
where
Xc = critical intersection volume-to-capacity ratio,
Vc = sum of critical-lane flow rates (tpc/h/ln), and
cI = intersection capacity (tpc/h/ln).
Step 5d. Assess Intersection Sufficiency

The objective of this task is to assess the sufficiency of the intersection in
terms of its ability to accommodate a given demand level. Exhibit 31-37 provides
guidance for determining whether an intersection is operating under, near, or
over its available capacity.
The analyst may stop at this point or may continue with Part II to determine
delay and LOS.

Exhibit 31-37 Critical Intersection

Planning-Level Analysis: Volume-to-Capacity Capacity
Intersection Volume-to- Ratio Description Assessment
Capacity Ratio Assessment
All demand is able to be accommodated; delays are
Levels <0.85 Under
low to moderate.
Demand for critical lane groups is near capacity and
some lane groups require more than one cycle to
0.85–0.98 clear the intersection; all demand is able to be Near
processed within the analysis period; delays are
moderate to high.
Demand for critical lane groups is just able to be
accommodated within a cycle but often requires
>0.98 Over
multiple cycles to clear the intersection; delays are
high and queues are long.
Part II: Delay and Level of Service

Part II builds on the results of Part I by allowing the user to calculate
capacity, delay, and LOS.

This step consists of two tasks. For the first task, the analyst calculates the
effective green time for each critical phase. For the second task, the analyst
calculates the volume-to-capacity ratio for each lane group.
Step 6a. Calculate Effective Green Times

If the effective green time for each critical phase is known, then the analyst
can proceed to Step 6b.
The total effective green time available for all critical phases is equal to the
cycle length minus the total lost time per cycle. This calculation is shown in
Equation 31-173.
Equation 31-173 𝑔𝑡𝑜𝑡 = 𝐶 − (𝑛𝑐𝑝 𝑙𝑡 )
where
gtot = total effective green time in the cycle (s),
ncp = number of critical phases, and
lt = phase lost time (s).
A default phase lost time of 4.0 s for each critical phase is recommended.
The total effective green time is allocated to each critical phase in proportion
to the lane flow rate for each critical phase. Equation 31-174 is used to compute
the effective green time for a given critical lane group.
𝑣𝑐,𝑖
Equation 31-174
𝑔𝑐,𝑖 = 𝑔𝑡𝑜𝑡 ( )
𝑉𝑐
where
gc,i = effective green time for critical lane group i (s),
gtot = total effective green time in the cycle (s),

vc,i = lane flow rate for critical lane group i (tpc/h/ln), and
Vc = sum of the critical-lane flow rates (tpc/h/ln).
The effective green time for a noncritical lane group is set equal to the
effective green time for its counterpart critical lane group that occurs
concurrently during the same phase.
Finally, the effective green time gi for each phase i is set equal to the effective
green time that is computed for the corresponding lane group. The effective
green time computed in this manner should be reviewed against policy
requirements and other considerations (such as the minimum green time based
on driver expectancy and the time required for pedestrians to cross the approach).
Step 6b. Calculate Capacity and Volume-to-Capacity Ratios

The lane group capacity and volume-to-capacity ratio can be computed with
Equation 31-175 and Equation 31-176, respectively.
𝑔𝑖
𝑐𝑖 = 𝑠𝑜 𝑁𝑖 Equation 31-175
𝐶
𝑁𝑖 𝑣𝑖
𝑋𝑖 = Equation 31-176
𝑐𝑖
where
ci = capacity of lane group i (tpc/h);
gi = effective green time for lane group i (s);
Ni = number of lanes associated with lane group i, accounting for de facto
lanes (ln);
Xi = volume-to-capacity ratio for lane group i;
vi = lane flow rate for lane group i (tpc/h/ln); and
The capacity for each lane group is based on the base saturation flow rate so.
A default value for base saturation flow rate can be obtained from Exhibit 19-11.
This rate is not adjusted for parking activity, heavy vehicles, and so forth because
these adjustments are applied in Step 2 to the lane group flow rate.
Equation 31-177 and Equation 31-178 can be used to compute the intersection
capacity and intersection volume-to-capacity ratio, respectively.
𝑛
𝑐𝑝
∑𝑖=1 𝑔𝑐,𝑖 Equation 31-177
𝑐sum = 𝑠𝑜
𝐶
𝑉𝑐
𝑋𝑐 = Equation 31-178
𝑐sum
where csum is the intersection capacity (tpc/h/ln).
Step 7: Determine Delay and Level of Service

The control delay for each lane group is calculated by using Equation 31-179
with Equation 31-180 and Equation 31-181.

Equation 31-179 𝑑𝑖 = 𝑑1,𝑖 + 𝑑2,𝑖

with
0.5 𝐶(1 − 𝑔𝑖 /𝐶)2
Equation 31-180 𝑑1,𝑖 = 𝑃𝐹𝑖
1 − [min(1, 𝑋𝑖 ) 𝑔𝑖 /𝐶]
16 𝑋𝑖
Equation 31-181 𝑑2,𝑖 = 225 [(𝑋𝑖 − 1) + √(𝑋𝑖 − 1)2 + ]
𝑐𝑖
where
di = control delay for lane group i (s/veh),
d1,i = uniform delay for lane group i (s/veh),
d2,i = incremental delay for lane group i (s/veh),
PFi = progression adjustment factor for lane group i, and
The progression adjustment factor describes the arrival distribution for the
subject lane group, which may be influenced by an upstream traffic signal.
Recommended progression adjustment factors are shown in Exhibit 31-38.
Exhibit 31-38 Quality of Conditions That Describe Arrivals Associated with Progression
Planning-Level Analysis: Progression the Subject Lane Group Factor PF
Progression Adjustment Good progression (a) Vehicles arrive in platoons during the green interval, OR 0.70
Factor (b) most vehicles arrive during the green interval.
Random arrivals (a) The phase serving the subject lane group is not 1.00
(default) coordinated with the upstream traffic signal, OR
(b) the intersection is sufficiently distant from other
signalized intersections as to be considered isolated.
Poor progression (a) Vehicles arrive in platoons during the red interval, OR 1.25
(b) most vehicles arrive during the red indication.
Lane group delay may be aggregated for each approach and for the
intersection as a whole. The aggregation process is the same as that in the
motorized vehicle methodology in Chapter 19 using Equation 19-28 and
Equation 19-29.
Delay values may be compared with the criteria in Exhibit 19-8 to determine
the LOS for a lane group, approach, or the intersection as a whole.
Protected-Permitted Left-Turn Operations

The procedure described in this subsection applies to the analysis of
protected-permitted left-turn operation. The effective green time is a required
input data item. If it is known or can be estimated, then the supplemental
guidance in this subsection can be used with the planning-level analysis
application.
Step 2: Convert Movement Volumes to Through Passenger-Car Equivalents

The guidance provided in this subsection supplements that provided in
Step 2 of the planning-level analysis application. The objective is to compute an

equivalency factor for protected-permitted left-turn operation that reflects the

left-turn vehicle’s overall effect on operations.
A single left-turn equivalency factor is computed for both the protected and
the permitted time periods. Exhibit 31-33 is used to identify the equivalency
factor for protected left-turn operation during the left-turn phase. It is also used
to identify the equivalency factor for permitted left-turn operation during the
through phase. A single factor is calculated that weighs these two equivalency
factors in proportion to the effective green times of each time period. Equation
31-182 is used to compute the single equivalency factor for left turns.
𝐸𝐿𝑇,𝑝𝑡 𝑔𝑙𝑡,𝑝𝑡 + 𝐸𝐿𝑇,𝑝𝑚 𝑔𝑙𝑡,𝑝𝑚
𝐸𝐿𝑇 = Equation 31-182
𝑔𝑙𝑡,𝑝𝑡 + 𝑔𝑙𝑡,𝑝𝑚
where
ELT = equivalency factor for left turns,
ELT,pt = equivalency factor for protected left-turn operation,
ELT,pm = equivalency factor for permitted left-turn operation,
glt,pt = effective green time for the protected left-turn phase (s), and
glt,pm = effective green time for permitted left-turn operation during the
through phase (s).
The equivalency factor computed with Equation 31-182 is used in Equation
31-157 to compute the equivalent through movement flow rate for the left-turn
lane group. The effective green time for the first time period of the protected-
permitted operation includes the yellow interval that occurs between the two
periods.
Step 4: Determine Critical Lane Groups

The guidance provided in this subsection supplements that provided in
Step 4 of the planning-level analysis application. The objective is to compute the
left-turn lane flow rate during the protected left-turn phase and then use this
value to identify the critical lane groups.
The equivalent through-car flow rate in the left lane during the protected
left-turn phase is estimated by distributing the lane flow rate for the left-turn
lane group proportionally among the protected and permitted periods. The flow
rate for the protected left-turn period is computed with Equation 31-183.
𝑔𝑙𝑡,𝑝𝑡
𝑣𝑙𝑡,𝑝𝑡 = 𝑣𝑙𝑡 Equation 31-183
𝑔𝑙𝑡,𝑝𝑡 + 𝑔𝑙𝑡,𝑝𝑚
where
vlt,pt = lane flow rate for the left-turn lane group during the protected left-
turn phase (tpc/h/ln), and
vlt = lane flow rate for the left-turn lane group (tpc/h/ln).

In the process of identifying the critical lane groups (and related flow rate),
only the lane flow rate during the protected left-turn phase vlt,pt is used for the
left-turn lane group. The critical-lane flow rate is then determined by using the
rules described for protected operation–with left-turn phase in Step 4b above.
The remainder of the planning-level analysis application does not change. In
Step 7, the lane flow rate for the left-turn lane group vlt is used to determine the
delay and LOS.
WORKSHEETS
This subsection includes a series of worksheets that can be used to document
an application of the planning-level analysis application. These worksheets are as
follows:
• Input Worksheet (Exhibit 31-39),
• Left-Turn Treatment Worksheet (Exhibit 31-40),
• Intersection Sufficiency Worksheet (Exhibit 31-41), and
• Delay and LOS Worksheet (Exhibit 31-42).

Exhibit 31-39
PLANNING-LEVEL ANALYSIS: INPUT WORKSHEET Planning-Level Analysis: Input
Worksheet
General Information Site Information
Analyst Intersection
Agency or Company
Date Performed Jurisdiction
Analysis Period Analysis Year
Intersection Geometry
Street
Show North
= Through
= Right
= Left
= Through + Right
= Left + Through
Street
= Left + Right
= Left + Through + Right
Volume and Signal Input
EB WB NB SB
LT TH RT LT TH RT LT TH RT LT TH RT
Required Data
Volume (veh/h)
Number of lanes
Lane use (exclusive or shared)
Optional Data 1
Heavy vehicles (%)
On-street parking presence (no, yes)
Pedestrian activity (none, low, med., high, extreme)
Left-turn operation and phase sequence 2
Effective green time (s)3,4
Progression quality (good, random, poor) 4
Peak hour factor Cycle length (s) Base saturation flow rate (pc/h/ln)
Notes
1. Optional input data (guidance is provided for estimating these data if they are not known).
2. Combinations addressed: (a) protected operation—with left-turn phase, (b) permitted operation—no left-turn
....phase, (c) protected operation—split phasing, (d) protected-permitted operation—with left-turn phase
3. Data required for Part I analysis if "protected-permitted operation—with left-turn phase" is present.
4. Data required for Part II analysis.

Exhibit 31-40
Planning-Level Analysis: Left- PLANNING-LEVEL ANALYSIS: LEFT-TURN TREATMENT WORKSHEET
Turn Treatment Worksheet
General Information
Description
Check # 1. Left-Turn Lane Check
Approach EB WB NB SB
Number of left-turn lanes
Protected left turn (Y or N)?
If the number of left-turn lanes on any approach exceeds 1, then it is recommended that the left turns on that
the approach be protected. Those approaches with protected left turns need not be evaluated in subsequent checks.
Check # 2. Minimum Volume Check
Left-turn volume
If left-turn volume on any approach exceeds 240 veh/h, then it is recommended that the left turns on that the
approach be protected. Those approaches with protected left turns need not be evaluated in subsequent checks.
Check # 3. Minimum Cross-Product Check
Left-turn volume, V L (veh/h)
Opposing mainline volume, V o (veh/h)
Cross product (V L * Vo)
Opposing through lanes
Minimum Cross-Product Values for Recommending Left-Turn Protection
Number of Through Lanes Minimum Cross Product
1 50,000
2 90,000
3 110,000
If the cross product on any approach exceeds the above values, then it is recommended that the left turns on that
approach be protected. Those approaches with protected left turns need not be evaluated in subsequent checks.
Notes
1. If any approach is recommended for left-turn protection but the analyst evaluates it as having permitted operation,
then the planning-level analysis method may give overly optimistic results. The analyst should instead use the
automobile methodology described in Chapter 19, Signalized Intersections.
2. All volumes used in this worksheet are unadjusted hourly volumes.

Exhibit 31-41
PLANNING-LEVEL ANALYSIS: INTERSECTION SUFFICIENCY WORKSHEET
Planning Level Analysis:
General Information Intersection Sufficiency
Description: Worksheet
East-West Approaches
Eastbound Westbound
Left Through Right Left Through Right
Movement volume, V (veh/h)
Equivalency factor for heavy vehicles, E hv
Equivalency factor for peaking char., E P HF
Equivalency factor for right turns, E RT
Equivalency factor for left turns, E LT 1
Equivalency factor for parking activity, E p
Equivalency factor for lane utilization, E LU
Equivalency factor for other conditions, E other
Equivalent through mvmt. flow rate (tpc/h) v adj
= V E HV E P HF E LT E RT E p E LU E other
Number of lanes, N
Lane flow rate, v (tpc/h/ln) v = v adj / N
Critical lane flow rate, V c (tpc/h/ln)
Critical lane group (indicate with "X")
Critical lane group flow rate, v c (tpc/h/ln)
Supplemental Calculations for Protected-Permitted Operation
Equivalency factor for prot. left turn, E LT ,pt
Equivalency factor for perm. left turn, E LT,pm
Effective green for prot. left turn, g lt ,pt (s)
Effective green for perm. left turn, g lt ,pm (s)
Equivalency factor for left turns, E LT E LT =
(E LT ,pt g lt ,pt + E LT ,pm g lt ,pm ) / (g lt ,pt + g lt ,pm )
North-South Approaches
Northbound Southbound
Movement volume, V (veh/h)
Equivalency factor for heavy vehicles, E hv
Equivalency factor for peaking char., E P HF
Equivalency factor for right turns, E RT
Equivalency factor for left turns, E LT 1
Equivalency factor for parking activity, E p
Equivalency factor for lane utilization, E LU
Equivalency factor for other conditions, E other
Equivalent through mvmt. flow rate (tpc/h) v adj
= V E HV E P HF E LT E RT E p E LU E other
Number of lanes, N
Lane flow rate, v (tpc/h/ln) v = v adj / N
Critical lane flow rate, V c (tpc/h/ln)
Critical lane group (indicate with "X")
Critical lane group flow rate, v c (tpc/h/ln)
Supplemental Calculations for Protected-Permitted Operation
Equivalency factor for prot. left turn, E LT ,pt
Equivalency factor for perm. left turn, E LT,pm
Effective green for prot. left turn, g lt ,pt (s)
Effective green for perm. left turn, g lt ,pm (s)
Equivalency factor for left turns, E LT E LT =
(E LT ,pt g lt ,pt + E LT ,pm g lt ,pm ) / (g lt ,pt + g lt ,pm )
Intersection Sufficiency Assessment
Intersection capacity, c I (tpc/h/ln)
Number of critical phases, n cp c I = s o [C – (n cp 4.0)]/C
Critical intersection vol.-to-capacity ratio, X c
Sum of critical lane flow rates, V c (tpc/h/ln) Xc = Vc / cI
Intersection status (relationship to capacity) Under Near Over
Note
1. If the approach has protected-permitted operation, use the supplemental calculations section to compute E LT .

Exhibit 31-42
Planning-Level Analysis: Delay PLANNING-LEVEL ANALYSIS: DELAY AND LOS WORKSHEET
and LOS Worksheet
General Information
Description
Green Time Calculation
Total effective green time, gtot (s)

gtot = C - (ncp 4.0)
East-West Approaches Eastbound Westbound
Critical lane group flow rate, vc (tpc/h/ln)1
Effective green time for critical lane group,
gc (s) gc = gtot vc / Vc
Phase No. 1 Phase No. 2 Phase No. 3
Effective green time, g (s)
North-South Approaches Northbound Southbound

Critical lane group flow rate, vc (tpc/h/ln)1
Effective green time for critical lane group,
gc (s) gc = gtot vc / Vc
Phase No. 1 Phase No. 2 Phase No. 3
Control Delay and LOS
EB WB NB SB
Lane group
Green-to-cycle-length ratio, g/C
Number of lanes, N1
Lane group capacity, c (veh/h)
c = 1900 N g/C
Lane flow rate, v (tpc/h/ln)1
Volume-to-capacity ratio, X X = (N v)/c
Progression adjustment factor, PF
Uniform delay, d1 (s/veh)
Incremental delay, d2 (s/veh)
Control delay, d = d1 + d2 (s/veh)
Approach delay, dA (s/veh)
dA = Σ(d N v)/Σ(N v)
Approach flow rate, V A (veh/h)
Intersection delay, dI (s/veh) Intersection LOS (Exhibit 19-8)
dI = Σ(dA VA )/ΣVA
Intersection capacity, csum (tpc/h/ln) Critical intersection vol.-to-capacity ratio, X c
csum = 1900 (Σgc ) / C X c = Vc / csum
Notes
1. Value obtained from the Intersection Sufficiency Worksheet.

6. FIELD MEASUREMENT TECHNIQUES
This section describes two techniques for estimating key traffic

characteristics by using field data. The first subsection describes a technique for
estimating control delay. The second subsection describes a technique for
estimating saturation flow rate.
FIELD MEASUREMENT OF INTERSECTION CONTROL DELAY

Delay can be measured at existing intersections as an alternative to
estimating delay by using the motorized vehicle methodology in Chapter 19,
Signalized Intersections. Various techniques can be used for measuring delay,
including a test-car survey, vehicle path tracing, input–output analysis, and
queue counting. The first three techniques tend to require more time to
implement than the last technique, but they provide more accurate delay
estimates. They are often limited to sampling when implemented manually. They
may be more appropriate when oversaturated conditions are present. The first
two techniques can be used to estimate delay on either a movement basis or a
lane group basis. The last two techniques are more amenable to delay
measurement on a lane group basis.
The queue-count technique is recommended for control delay measurement.
It is based on direct observation of vehicle-in-queue counts for a subject lane
group. It normally requires two field personnel for each lane group surveyed.
Also needed are (a) a multifunction digital watch that includes a countdown-
repeat timer, with the countdown interval in seconds; and (b) a volume-count
board with at least two tally counters. Alternatively, a laptop computer can be
programmed to emit audio count markers at user-selected intervals, take volume
counts, and execute real-time delay computations.
The queue-count technique is applicable to all undersaturated lane groups.
Significant queue buildup can make the technique impractical for oversaturated
lane groups or lane groups with limited storage length. If queues are lengthy,
then the technique should be modified by subdividing the lane group into
manageable segments (or zones) and assigning an observer to each zone. Each
observer then counts queued vehicles in his or her assigned zone.
If queues are lengthy or the volume-to-capacity ratio is near 1.0, then care
must be taken to continue the vehicle-in-queue count past the end of the arrival
count period, as detailed in subsequent paragraphs. This extended counting
period is required for consistency with the analytic delay equation used in the
chapter text.
The queue-count technique does not directly measure delay during
deceleration and during a portion of acceleration. These delay elements are very
difficult to measure without sophisticated tracking equipment. Nevertheless, this
technique has been shown to yield a reasonable estimate of control delay by
application of appropriate adjustment factors (8, 9). One adjustment factor
accounts for sampling errors that may occur. Another factor accounts for
Chapter 31/Signalized Intersections: Supplemental Field Measurement Techniques

unmeasured acceleration–deceleration delay. This adjustment factor is a function

of the number of vehicles in queue each cycle and the approach speed.
Approach Speed
Exhibit 31-43 shows a worksheet that can be used for recording observations
and computing control delay for the subject lane group. Before starting the
survey, observers need to estimate the average approach speed during the study
period. Approach speed is the speed at which vehicles would pass unimpeded
through the intersection if the signal were green for an extended period and
volume was light. This speed may be obtained by driving through the
intersection a few times when the signal is green and there is no queue. The
approach speed is recorded at an upstream location that is least affected by the
operation of the subject signalized intersection as well as the operation of any
other signalized intersection.
Survey Period
The duration of the survey period must be clearly defined in advance so the
last arriving vehicle or vehicles that stop in the period can be identified and
counted until they exit the intersection. It is logical to define the survey period on
the basis of the same considerations used to define an evaluation analysis period
(as described in Section 3 of Chapter 19). A typical survey period is 15 min.
Count Interval
The survey technique is based on recording a vehicle-in-queue count at
specific points in time. A count interval in the range of 10 to 20 s has been found
to provide a good balance between delay estimate precision and observer
capability. The actual count interval selected from this range is based on
consideration of survey period duration and the type of control used at the
intersection.
The count interval should be an integral divisor of the survey period
duration. This characteristic ensures that a complete count of events is taken for
the full survey period. It also allows easier coordination of observer tasks during
the field study. For example, if the study period is 15 min, the count interval can
be 10, 12, 15, 18, or 20 s.
If the intersection has pretimed or coordinated-actuated control, the count
interval should not be an integral divisor of the cycle length. This characteristic
eliminates potential survey bias due to queue buildup in a cyclical pattern. For
example, if the cycle length is 120 s, the count interval can be 11, 13, 14, 16, 17, 18,
or 19 s.
If the intersection has actuated control, the count interval may be chosen as
the most convenient value for conducting the field survey with consideration of
survey period duration.
Field Measurement Techniques Chapter 31/Signalized Intersections: Supplemental

INTERSECTION CONTROL DELAY WORKSHEET

Exhibit 31-43
Control Delay Field Study
General Information Site Information Worksheet
Agency or Company Area Type CBD Other
Input Initial Parameters
Number of lanes, N Total vehicles arriving, V tot

Approach speed, S a (mi/h) Stopped-vehicle count, V stop
Survey count interval, Is (s) Cycle length, C (s)
Input Field Data

Clock Cycle Count Interval
Time Number 1 2 3 4 5 6 7 8 9 10
Total
Computations
Total vehicles in queue, ΣV iq = veh Number of cycles surveyed, Nc =

 Viq  Vstop
Time-in-queue per vehicle, d vq =  I s  0 .9
 s/veh Fraction of vehicles stopping, FVS =
 Vtot  Vtot
Vstop
No. of vehicles stopping/lane/cycle = veh/ln Accel-decel correction delay, dad = FVS * CF s/veh
(N c  N )
Accel-decel correction factor, CF s/veh Control delay, d = dv q + dad s/veh
Measurement Technique
The survey should begin at the start of the red indication associated with the
subject lane group. Two observers are required for data collection. The duties of
each observer are described in the following paragraphs.
Observer 1 Tasks
1. Observer 1 keeps track of the end of the standing queue in each lane of
the subject lane group. For purposes of the survey, a vehicle is considered
as having joined the queue when it approaches within one car length of a
stopped vehicle and is itself about to stop. This definition is used because
of the difficulty of keeping track of the moment when a vehicle comes to a
stop.
2. At the start of each count interval, Observer 1 records the number of
vehicles in queue in all lanes of the subject lane group. The countdown-
repeat timer on a digital watch can be used to signal the count time. This
count includes vehicles that arrive when the signal is actually green but
stop because queued vehicles ahead have not yet started moving. All
vehicles that join a queue are included in the vehicle-in-queue count until
they “exit” the intersection. A through vehicle exits the intersection when
its rear axle crosses the stop line. A turning vehicle exits the intersection
the instant it clears the opposing through traffic (or pedestrians to which
it must yield) and begins accelerating back to the approach speed. The
vehicle-in-queue count often includes some vehicles that have regained
speed but have not yet exited the intersection.
3. Observer 1 records the vehicle-in-queue count in the appropriate count-
interval box on the worksheet. Ten boxes are provided for each “count
cycle” (note that a count cycle is not the same as a signal cycle). Any
number of boxes can be used to define the count cycle; however, as many
as possible should be used to ensure best use of worksheet space. The clock
time at the start of the count cycle is recorded in the first (far-left) column.
The count cycle number is recorded in the second column of the sheet.
4. At the end of the survey period, Observer 1 continues taking vehicle-in-
queue counts for all vehicles that arrived during the survey period until
all of them have exited the intersection. This step requires the observer to
make a mental note of the last stopping vehicle that arrived during the
survey period in each lane of the lane group and continue the vehicle-in-
queue counts until the last stopping vehicle or vehicles, plus all vehicles
in front of the last stopping vehicle(s), exit the intersection. Stopping
vehicles that arrive after the end of the survey period are not included in
the final vehicle-in-queue counts.
Observer 2 Tasks
5. Observer 2 maintains three counts during the survey period. The first is a
count of the vehicles that arrive during the survey period. The second is a
count of the vehicles that arrive during the survey period and that stop
one or more times. A vehicle stopping multiple times is counted only
once as a stopping vehicle. The third count is the count of signal cycles, as
measured by the number of times the red indication is presented for the
subject lane group. For lane groups with a turn movement and protected
or protected-permitted operation, the protected red indication is used for
this purpose. If the survey period does not start or end at the same time as
the presentation of a red indication, then the number of count intervals
that occur in the interim can be used to estimate the fraction of the cycle
that occurred at the start or end of the survey period.
6. Observer 2 enters all counts in the appropriate boxes on the worksheet.

Data Reduction Tasks

7. Sum each column of vehicle-in-queue counts, then sum the column totals
for the entire survey period.
8. A vehicle recorded as part of a vehicle-in-queue count is assumed to be in
queue, on average, for the time interval between counts. On this basis, the
average time in queue per vehicle arriving during the survey period is
estimated with Equation 31-184.
∑ 𝑉𝑖𝑞
𝑑𝑣𝑞 = 0.9 (𝐼𝑠 ) Equation 31-184
𝑉𝑡𝑜𝑡
where
dvq = time in queue per vehicle (s/veh),
Is = interval between vehicle-in-queue counts (s),
∑Viq = sum of vehicle-in-queue counts (veh), and
Vtot = total number of vehicles arriving during the survey period (veh).
The 0.9 adjustment factor in Equation 31-184 accounts for the errors that
may occur when the queue-count technique is used to estimate delay.
Research has shown the adjustment factor value is fairly constant for a
variety of conditions (8).
9. Compute the fraction of vehicles stopping and the average number of
vehicles stopping per lane in each signal cycle, as indicated on the
worksheet.
10. Use Exhibit 31-44 to look up the correction factor appropriate to the lane
group approach speed and the average number of vehicles stopping per
lane in each cycle. This factor adjusts for deceleration and acceleration
delay, which cannot be measured directly with manual techniques (9).
Acceleration–Deceleration Correction Factor CF (s/veh) As a Exhibit 31-44

Approach Function of the Average Number of Vehicles Stopping Acceleration–Deceleration
Speed (mi/h) ≤7 veh/ln/cycle 8–19 veh/ln/cycle 20–30 veh/ln/cyclea Correction Factor
≤37 +5 +2 –1
>37–45 +7 +4 +2
>45 +9 +7 +5
a
Note: Vehicle-in-queue counts in excess of about 30 veh/ln/cycle are typically unreliable.
11. Multiply the correction factor by the fraction of vehicles stopping. Add
this product to the time-in-queue value from Task 2 to obtain the estimate
of control delay for the subject lane group.
Example Application
Exhibit 31-45 presents sample data for a lane group during a 15-min survey
period. The intersection has a 115-s cycle. A 15-s count interval is selected
because 15 is not an integral divisor of the cycle length, but it is an integral
divisor of the survey period.

Exhibit 31-45 INTERSECTION CONTROL DELAY WORKSHEET

Example Control Delay Field
Study Worksheet General Information Site Information
Analyst Intersection Cicero & Belmont

Agency or Company Area Type CBD X Other
Analysis Period PM Analysis Year 2015
Number of lanes, N 2 Total vehicles arriving, V tot 530

Approach speed, S a (mi/h) 40 Stopped-vehicle count, V stop 223
Survey count interval, I s (s) 15 Cycle length, C (s)
Input Field Data

Time Number 1 2 3 4 5 6 7 8 9 10
4:34 1 3 8 11 15 12 2 0 2
2 6 12 15 16 6 0 0 2
3 7 11 14 14 2 0 0
4 5 7 10 13 13 2 0 1
4:42 5 4 6 10 12 3 0 0 1
6 5 7 9 13 4 0 0
7 3 6 8 12 12 0 0 0
4:47 8 4 7 11 16 9 0
Total 37 64 88 111 61 4 0 6
Computations
Total vehicles in queue, Σv iq 371 veh Number of cycles surveyed, N c 7.8

 V iq  V stop
Time-in-queue per vehicle, d vq =  I s  0.9
V tot 
9.5 s/veh Fraction of vehicles stopping, FVS =
V tot
0.42

V stop
No. of vehicles stopping/lane/cycle =
(N c  N )
14 veh/ln Accel-decel correction delay, d ad = FVS × CF 1.7 s/veh
Acceleration-deceleration correction factor, CF 4 s/veh Control delay, d = d v q + d ad 11.2 s/veh
Exhibit 31-45 shows data are recorded for six, seven, or eight intervals during
each count cycle. This choice is arbitrary and based solely on best use of
worksheet space.
The data reduction results are shown at the bottom of the exhibit. A control
delay of 11.2 s/veh is estimated for the subject lane group.
Exhibit 31-46 shows how the worksheet shown in Exhibit 31-45 would have
been completed if a queue had remained at the end of the 15-min survey period.
Only the vehicles that arrived during the 15-min period would be counted.

INTERSECTION CONTROL DELAY WORKSHEET

Exhibit 31-46
Example Worksheet with
General Information Site Information Residual Queue at End
Analyst Intersection Cicero & Belmont

Agency or Company Area Type CBD X Other
Analysis Period PM Analysis Year 1999
Number of lanes, N 2 Total vehicles arriving, V tot

Approach speed, S a (mi/h) 40 Stopped-vehicle count, V stop
Survey count interval, Is (s) 15 Cycle length, C (s)
Input Field Data

Time Number 1 2 3 4 5 6 7 8 9 10
4:47 8 4 7 11 16 9 0 Queue count in previous example
4:47 8 4 4* 4* 4* 0 First four in queue have cleared by now
say 15-min. survey period ends here
4* - last stopping vehicles in survey period;

count only until they clear.
Total 37 61 81 99 52 4 0 6
Computations
Total vehicles in queue, ΣV iq = veh Number of cycles surveyed, Nc =

 Viq  Vstop
Time-in-queue per vehicle, d vq =  I s  0 .9
 s/veh Fraction of vehicles stopping, FVS =
 Vtot  Vtot
Vstop
No. of vehicles stopping/lane/cycle = veh/ln Accel-decel correction delay, dad = FVS * CF s/veh
(N c  N )
Accel-decel correction factor, CF s/veh Control delay, d = dv q + dad s/veh
FIELD MEASUREMENT OF SATURATION FLOW RATE

This subsection describes a technique for quantifying the base saturation
flow rate for local conditions. It provides a means of calibrating the saturation
flow rate calculation procedure (described in Section 3 of Chapter 19) to reflect
driver behavior at a local level. The technique is based on a comparison of field-
measured saturation flow rate with the calculated saturation flow rate for a
common set of lane groups at intersections in a given area.

Concepts
The saturation flow rate represents the maximum rate of flow in a traffic
lane, as measured at the stop line during the green indication. It is usually
achieved after 10 to 14 s of green, which corresponds to the front axle of the
fourth to sixth queued passenger car crossing the stop line.
The base saturation flow rate represents the saturation flow rate for a traffic
lane that is 12 ft wide and has no heavy vehicles, a flat grade, no parking, no
buses that stop at the intersection, even lane utilization, and no turning vehicles.
It is usually stable over a period of time in a given area and normally exhibits a
relatively narrow distribution among intersections in that area.
The prevailing saturation flow rate is the rate measured in the field for a
specific lane group at a specific intersection. It may vary significantly among
intersections with similar lane groups because of differences in lane width, traffic
composition (i.e., percentage of heavy vehicles), grade, parking, bus stops, lane
use, and turning vehicle operation. If the intersections are located in different
areas, then the prevailing saturation flow rate may also vary because of areawide
differences in the base saturation flow rate.
The adjusted saturation flow rate is the rate computed by the procedure
described in Chapter 19. It represents an estimate of the prevailing saturation
flow rate. It can vary among intersections for the same reasons as stated above
for the prevailing saturation flow rate. Any potential bias in the estimate is
minimized by local calibration of the base saturation flow rate.
The prevailing saturation flow rate and the adjusted saturation flow rate are
both expressed in units of vehicles. As a result, their value reflects the traffic
composition in the subject traffic lane. In contrast, the base saturation flow rate is
expressed in units of passenger cars and does not reflect traffic composition.
Measurement Technique
This subsection describes the technique for measuring the prevailing
saturation flow rate for a given traffic lane. In general, vehicles are recorded
when their front axles cross the stop line. The measurement period starts at the
beginning of the green interval or when the front axle of the first vehicle in the
queue passes the stop line. Saturation flow rate is calculated only from the data
recorded after the fourth vehicle in the queue passes the stop line.
The vehicle’s front axle, the stop line, and the time the fourth queued vehicle
crosses the stop line represent three key reference points for saturation flow
measurement. These three reference points must be maintained to ensure
consistency with the procedure described in Chapter 19 and to facilitate
comparability of results with other studies. The use of other reference points on
the vehicle, on the road, or in time may yield different saturation flow rates.
If the stop line is not visible or if vehicles consistently stop beyond the stop
line, then an alternative reference line must be established. This reference line
should be established just beyond the typical stopping position of the first
queued vehicle. Vehicles should consistently stop behind this line. Observation
of several cycles before the start of the study should be sufficient to identify this
substitute reference line.

The following paragraphs describe the tasks associated with a single-lane

saturation flow survey. A two-person field crew is recommended. However, one
person with a tape recorder, push-button event recorder, or a notebook computer
with appropriate software will suffice. The field notes and tasks identified in the
following paragraphs must be adjusted according to the type of equipment used.
A sample field worksheet for recording observations is included as Exhibit 31-47.
Exhibit 31-47
FIELD SATURATION FLOW RATE STUDY WORKSHEET
Saturation Flow Rate Field
General Information Site Information Study Worksheet
Agency or Company Area Type CBD Other
Lane Movement Input
grade =
street
grade = Movements Allowed

Through
Right turn
Left turn
street
grade =
Identify all lane movements and the lane studied

grade =
Input Field Measurement
Cycle 1 Cycle 2 Cycle 3 Cycle 4 Cycle 5 Cycle 6

Veh. in queue Time HV T Time HV T Time HV T Time HV T Time HV T Time HV T
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
End of saturation
End of green
No. veh. > 20
No. veh. on yellow
Glossary and Notes
HV = Heavy vehicles (vehicles with more than 4 tires on pavement)

T = Turning vehicles (L = Left, R = Right)
Pedestrians and buses that block vehicles should be noted with the time that they block traffic, for example,
P12 = Pedestrians blocked traffic for 12 s
B15 = Bus blocked for 15 s

General Tasks
Measure and record the area type as well as the width and grade of the lane
being studied. Enter these data in the lane movement input section of the field
worksheet.
Select an observation point where the roadway reference line (e.g., stop line)
for the surveyed lane and the corresponding signal heads are clearly visible.
When a vehicle crosses this line unimpeded, it has entered the intersection
conflict space for the purpose of saturation flow measurement. Left- or right-
turning vehicles yielding to opposing through traffic or yielding to pedestrians
are not recorded until they proceed through the opposing traffic or pedestrians.
Recorder Tasks
During the measurement period, note the last vehicle in the stopped queue
when the signal turns green. Describe the last vehicle to the timer. Note on the
worksheet which vehicles are heavy vehicles and which vehicles turn left or
right. Record the time called out by the timer.
Timer Tasks
Start the stopwatch at the beginning of the green indication and notify the
recorder. Count aloud each vehicle in the queue as its front axle crosses the stop
line and note the time of crossing. Call out the time of the fourth, 10th, and last
vehicle in the stopped queue as its front axle crosses the stop line.
If queued vehicles are still entering the intersection at the end of the green
interval, call out “saturation through the end of green—last vehicle was number
XX.” Note any unusual events that may have influenced the saturation flow rate,
such as buses, stalled vehicles, and unloading trucks.
The period of saturation flow begins when the front axle of the fourth vehicle
in the queue crosses the roadway reference line (e.g., stop line) and ends when
the front axle of the last queued vehicle crosses this line. The last queued vehicle
may be a vehicle that joined the queue during the green indication.
Data Reduction
Measurements are taken cycle by cycle. To reduce the data for each cycle, the
time recorded for the fourth vehicle is subtracted from the time recorded for the
last vehicle in the queue. This value represents the sum of the headways for the
fifth through nth vehicle, where n is the number of the last vehicle surveyed
(which may not be the last vehicle in the queue). This sum is divided by the
number of headways after the fourth vehicle [i.e., divided by (n – 4)] to obtain the
average headway per vehicle under saturation flow. The saturation flow rate is
3,600 divided by this average headway.
For example, if the time for the fourth vehicle was observed as 10.2 s and the
time for the 14th and last vehicle surveyed was 36.5 s, the average saturation
headway per vehicle is as follows:
(36.5 − 10.2) 26.3
= = 2.63 s/veh
(14 − 4) 10

The prevailing saturation flow rate in that cycle is as follows:

3,600
= 1,369 veh/h/ln
2.63
To obtain a statistically significant value, a minimum of 15 signal cycles (each
with more than eight vehicles in the initial queue) is typically required. The
average of the saturation headway per vehicle values from the individual cycles
is divided into 3,600 to obtain the prevailing saturation flow rate for the surveyed
lane. The percentage of heavy vehicles and turning vehicles in the sample should
be determined and noted for reference.
Calibration Technique
This subsection describes a technique for quantifying the base saturation
flow rate at a local level. It consists of three tasks. The first task entails measuring
the prevailing saturation flow rate at representative locations in the local area.
The second task requires the calculation of an adjusted saturation flow rate for
the same locations where a prevailing saturation flow rate was measured. The
third task combines the information to compute the local base saturation flow
rate.
This technique will require some resource investment by the agency.
However, it should need to be completed only once every few years. In fact, it
should be repeated only when there is evidence of a change in local driver
behavior. The benefit of this calibration activity will be realized by the agency in
terms of more accurate estimates of motorized vehicle performance, which
should translate into more effective decisions related to infrastructure investment
and system management.
Task 1. Measure Prevailing Saturation Flow Rate

This task requires measuring the prevailing saturation flow rate of one or
more lane groups at each of several representative intersections in the local area.
The minimum number of lane groups needed in the data set is difficult to judge
for all situations; however, it should reflect a statistically valid sample. The data
set should also provide a reasonable geographic and physical representation of
the population of signalized intersections in the local area.
The lane groups for which the prevailing saturation flow rate is measured
should include a representative mix of left-turn, through, and right-turn lane
groups. It should not include left-turn lane groups that operate in the permitted
or the protected-permitted mode or right-turn lane groups that have protected-
permitted operation. These lane groups are excluded because of the complex
nature of permitted and protected-permitted operation. The saturation flow rate
for these lane groups tends to have a large amount of random variation that
makes it more difficult to quantify the local base saturation flow rate with an
acceptable level of precision.
Once the set of lane groups is identified, the technique described in the
previous subsection is used to measure the prevailing saturation flow rate at
each location.

Task 2. Compute Adjusted Saturation Flow Rate

For this task, the saturation flow rate calculation procedure in Chapter 19 is
used to compute the adjusted saturation flow rate for each lane group in the data
set. If a lane group is at an intersection with actuated control for one or more
phases, the motorized vehicle methodology (as opposed to just the saturation
flow rate procedure) will be needed to compute the adjusted saturation flow rate
accurately. Regardless, the base saturation flow rate used with the procedure (or
methodology) for this task must be 1,900 pc/h/ln.
Task 3. Compute Local Base Saturation Flow Rate

The local base saturation flow rate is computed with Equation 31-185.
Equation 31-185
∑𝑚
𝑖=1 𝑠prevailing,𝑖
𝑠𝑜,local = 1,900
∑𝑚 𝑖=1 𝑠𝑖
where
so,local = local base saturation flow rate (pc/h/ln),
sprevailing,i = prevailing saturation flow rate for lane group i (veh/h/ln),
si = (adjusted) saturation flow rate for lane group i (veh/h/ln), and
m = number of lane groups.
Once the local base saturation flow rate so,local is quantified by this technique,
it is substituted thereafter for so in any equation in an HCM chapter that refers to
this variable.

7. COMPUTATIONAL ENGINE DOCUMENTATION
This section uses a series of flowcharts and linkage lists to document the
logic flow for the computational engine.
FLOWCHARTS
The methodology flowchart is shown in Exhibit 31-48. The methodology is
shown to consist of four main modules:
• Setup module,
• Signalized intersection module,
• Initial queue delay module, and
• Performance measures module.
This subsection provides a separate flowchart for each of these modules.
Exhibit 31-48
Methodology Flowchart
The setup module is shown in Exhibit 31-49. It consists of four main routines,
as shown in the large rectangles of the exhibit. The main function of each routine,
Chapter 31/Signalized Intersections: Supplemental Computational Engine Documentation

as well as the name given to it in the computational engine, is shown in the

exhibit. These routines are described further in the next subsection.
Exhibit 31-49 Start

Setup Module
Establish lane groups; estimate
initial group sat. flow rate, group
Initial estimate of cycle length volume, and phase duration
(InitialSetupRoutine) (InitialCapacityEstimate)
Convert input movement initial

Set demand flow = input flow rate queue to lane group initial queue
for current analysis period (InitialQueueSetup)
(PeriodVolumeSetup)
Finish
The signalized intersection module is shown in Exhibit 31-50. It consists of

nine main routines followed by a tenth and final computation routine performed
after the final phase duration equals the initial phase duration. The main function
of each routine, as well as the name given to it in the computational engine, is
shown in the exhibit. These routines are described further in the next subsection.
Exhibit 31-50
Signalized Intersection Module
The initial queue delay module is shown in Exhibit 31-51. It consists of four
main routines. The main function of each routine is shown in the exhibit.
Computational Engine Documentation Chapter 31/Signalized Intersections: Supplemental

Exhibit 31-51
Initial Queue Delay Module
The performance measures module is shown in Exhibit 31-52. It consists of

four main routines. The main function of each routine is shown in the exhibit.
Two of the routines are complicated enough to justify their development as
separate entities in the computational engine. The name given to each of these
two routines is also shown in the exhibit, and they are described further in the
next subsection.
Start Exhibit 31-52

Compute queue storage ratio Module
(QueueStorageRatio)
Establish upstream filtering
adjustment factor
Compute approach delay

Compute incremental delay and
second-term back-of-queue
(EstimateIncrementalDelay)
Finish
LINKAGE LISTS
This subsection uses linkage lists to describe the main routines that compose
the computational engine. Each list is provided in a table (an exhibit) that
identifies the routine and the various subroutines to which it refers. Conditions
for which the subroutines are used are also provided.
The lists are organized by module, as described in the previous subsection.
Four tables are provided to address the following three modules:
• Setup module (one table),
• Signalized intersection module (two tables), and
• Performance measures module (one table).
The initial queue delay module does not have a linkage list because it does
not call any specific routines.
The linkage list for the setup module is provided in Exhibit 31-53. The main
routines are listed in the far-left column of the exhibit and are identified in
Exhibit 31-49.


Setup Module Routines
InitialSetupRoutine Compute change period (Y + Rc). None
Compute initial estimate of cycle length None
C.
PeriodVolumeSetup a. Compute period volume before initial Used for multiple-period analysis
queue analysis, and
b. Restore period volume if initial queue
analysis conducted.
a. Save input volume as it will be Used for single-period analysis
overwritten if initial queue is present,
and
b. Restore input volume if initial queue
analysis conducted.
InitialCapacityEstimate getPermissiveLeftServiceTime Used if subject phase serves a
(computes gu , the duration of the left-turn movement with (a)
permitted period that is not blocked by permitted mode or (b)
an opposing queue) protected-permitted mode
getPermissiveLeftEffGreen Used if subject phase serves a
(computes gp , the duration of the left-turn movement with (a)
permitted green for permitted left-turn permitted mode or (b)
movements) protected-permitted mode
Define lane groups for each approach. None
Establish initial estimate of lane group None
volume, saturation flow rate, and
number of lanes capacity.
Establish initial estimate of proportion of Used for shared-lane lane
turns in a shared-lane lane group. groups
PermittedSatFlow Used if lane group serves a left-
(computes permitted left-turn saturation turn movement with protected-
flow rate sp) permitted mode
getParkBusSatFlowAdj Used if lane group is adjacent to
(computes combined parking and bus on-street parking or a local bus
blockage saturation flow adjustment stop
factors)
Establish initial estimate of queue None
service time gs.
InitialQueueSetup Distribute input movement initial queue Used for first analysis period
to corresponding lane groups.
Assign residual queue from last period Used for second and subsequent
to initial queue of current period, and analysis periods
distribute initial queue among affected
lane groups.
The linkage list for the signalized intersection module is provided in Exhibit
31-54. The main routines are listed in the far-left column of the exhibit and are
identified in Exhibit 31-50. The ComputeQAPolygon routine is complex enough
to justify the presentation of its subroutines in a separate linkage list. This
supplemental list is provided in Exhibit 31-55.


Signalized Intersection
InitialPortionOnGreen Compute portion arriving during None
Module: Main Routines
green P.
PedBikeEffectOnSatFlow PedBikeEffectOnLefts Used if subject phase serves a left-
turn movement with (a) permitted
mode or (b) protected-permitted
mode
PedBikeEffectOnRights Used if subject phase serves a right-
turn movement
PedBikeEffectOnLeftsUnopposed Used if subject phase serves a left-
turn movement with split phasing
ComputePermServeTime getPermissiveLeftServiceTime Used if subject phase serves a left-
(computes gu , the duration of the turn movement with (a) permitted
permitted period that is not mode or (b) protected-permitted
blocked by an opposing queue) mode
getPermissiveLeftEffGreen Used if subject phase serves a left-
(computes gp , the duration of the turn movement with (a) permitted
permitted green for permitted mode or (b) protected-permitted
left-turn movements) mode
ComputeTimeToFirstBlk getTimetoFirstBlk Used if subject phase serves a left-

(computes gf , the time before the turn movement in a shared lane with
first left-turning vehicle arrives (a) permitted mode or (b) protected-
and blocks the shared lane) permitted mode
ComputeVolumePortions PermittedSatFlow Used if lane group serves a left-turn

-AndSatFlow (computes permitted left-turn movement with protected-permitted
saturation flow rate sp) mode
PortionTurnsInSharedTRlane Used if approach has exclusive left-
(computes proportion of right- turn lane and subject lane group is a
turning vehicles in shared lane PR) shared lane serving through and
right-turning vehicles
SatFlowforPermExclLefts Used if lane group serves a left-turn
movement with a permitted mode in
an exclusive lane
PortionTurnsInSharedLTRlane Used if approach has a shared lane
(computes proportion of right- serving left-turn and through vehicles
turning vehicles in shared lane PR
and proportion of left-turning
vehicles in shared lane PL)
ComputeQAPolygon QAP_ProtPermExclLane Used if lane group serves a left-turn
movement in an exclusive lane with
the protected-permitted mode
QAP_ProtMvmtExclLane Used if lane group’s movement has
an exclusive lane and is served with
protected mode
QAP_ProtSharedLane Used if lane group has (a) a shared
lane with through and right-turning
movements or (b) a shared lane with
through and left-turning movements
served with split phasing
QAP_PermLeftExclLane Used if lane group serves a left-turn
movement in an exclusive lane with
the permitted mode
QAP_PermSharedLane Used if lane group serves a left-turn
movement in a shared lane with the
permitted mode

Exhibit 31-54 (cont’d.) Routine Subroutine Conditions for Use

Signalized Intersection VolumeComputations Determine call rate to extend green λ. None
Module: Main Routines
Determine call rate to activate a phase None
qv , qp.
MaximumAllowable- Compute maximum allowable headway Calculations vary depending on
Headway for each lane group MAH. lane group movements, lane
assignment, phase sequence,
and left-turn operational mode.
Compute equivalent maximum allowable None
headway for each phase and timer
MAH*.
ComputeAverage- Compute probability of green extension Computed for all phases except
PhaseDuration p. for the timer that serves the
protected left-turn movement in
a shared lane
Compute maximum queue service time None
for all lane groups served during the
phase.
Compute probability of phase None
termination by extension to maximum
limit (i.e., max-out).
Compute green extension time ge. None
Compute probability of a phase call pc. None
Compute unbalanced green duration Gu. None
Compute average phase duration Dp. None


QAP_ProtPermExclLane ADP_ProtPermExcl Used for lane groups with left-
Module: ComputeQAPolygon
(compute baseline first-term back-of- turn movements in exclusive
Routines
queue estimate Q1b) lane and served by protected-
permitted mode
getUniformDelay None
(compute baseline uniform delay d1b)
Compute queue service time gs. None
Compute lane group available capacity. None
Compute movement capacity. None
QAP_ProtMvmtExclLane ADP_ProtMvmt Used for lane groups with one
(compute baseline first-term back-of- service period
queue estimate Q1b)
QAP_ProtSharedLane ADP_ProtMvmt Used for lane groups with one
queue estimate Q1b)
QAP_PermLeftExclLane ADP_PermLeftExclLane Used for lane groups with left-
(compute baseline first-term back-of- turn movements in exclusive
queue estimate Q1b) lane and served by permitted
mode
QAP_PermSharedLane ADP_PermSharedMvmt Used for shared-lane lane
(compute baseline first-term back-of- groups with a permitted left-
queue estimate Q1b) turn movement
ADP_ProtMvmt Used for lane groups with one
queue estimate Q1b)
ADP_ProtPermShared Used for lane groups with left-
(compute baseline first-term back-of- turn movements in shared-lane
queue estimate Q1b) lane group and served by
protected-permitted mode

The linkage list for the performance measures module is provided in Exhibit
31-56. The main routines are listed in the far-left column and are identified in
Exhibit 31-52.

EstimateIncrementalDelay Compute incremental delay d2 and None
Module Routines
second-term back-of-queue estimate
Q2.
QueueStorageRatio Compute queue storage ratio LQ. None

8. USE OF ALTERNATIVE TOOLS
This section illustrates the use of alternative evaluation tools to evaluate the
operation of a signalized intersection. The intersection described in Example
Problem 1 of Section 9 is used for this purpose. There are no limitations in this
example that would suggest the need for alternative tools. However, it is possible
to introduce situations, such as short left-turn bays, for which an alternative tool
might provide a more realistic assessment of intersection operation.
The basic layout of the example intersection is shown in the second exhibit of
Example Problem 1 of Section 9. The left-turn movements on the north–south
street operate under protected-permitted control and lead the opposing through
movements (i.e., a lead–lead phase sequence). The left-turn movements on the
east–west street operate as permitted. To simplify the discussion, the pedestrian
and parking activity is removed. A pretimed signal operation is used.
EFFECT OF STORAGE BAY OVERFLOW

The effect of left-turn storage bay overflow is described in this subsection as
a means of illustrating the use of alternative tools. The motorized vehicle
methodology in Chapter 19 can be used to compute a queue storage ratio that
compares the back-of-queue estimate with the available storage length. This ratio
is used to identify bays that have inadequate storage. Overflow from a storage
bay can be expected to reduce approach capacity and increase the approach
delay. However, these effects of bay overflow are not addressed by the
motorized vehicle methodology.
Effect of Overflow on Approach Throughput and Delay

A simulation software product was selected as the alternative tool for this
analysis. The intersection was simulated for a range of storage bay lengths from 0
to 250 ft. All other input data remained the same. The results presented here
represent the average of 30 simulation runs for each case.
The effect of bay overflow was assessed by examining the relationship
between bay length, approach throughput, and approach delay. Exhibit 31-57
shows this effect. The throughput on each approach is equal to the demand
volume when storage is adequate but drops off when the bay length is
decreased.
A delay comparison is also presented in Exhibit 31-57. The delay on each
approach increases as bay length is reduced. The highest delay is associated with
a zero-length bay, which is effectively a shared lane. The zero-length case is
included here to establish a boundary condition. The delay value becomes
excessive when overflow occurs. This situation often degrades into
oversaturation, and a proper assessment of delay would require a multiple-
period analysis to account for the buildup of long-term queues.
Chapter 31/Signalized Intersections: Supplemental Use of Alternative Tools

500 250
Exhibit 31-57 Westbound
450
Throughput (veh/15-min)
Effect of Storage Bay Length Northbound
400 200
on Throughput and Delay 350
Delay (s/veh)
300 150
Southbound
250
200 100
150 Westbound Eastbound
100 Eastbound 50
Northbound
50
Southbound
0 0
0 50 100 150 200 250 0 50 100 150 200 250 Full
Storage Bay Length (ft) Storage Bay Length (ft)
(a) Throughput (b) Delay
For case-specific applications, parameters that could influence the evaluation

of bay overflow include the following:
• Number of lanes for each movement,
• Demand volumes for each movement,
• Impedance of left-turning vehicles by oncoming traffic during permitted
periods,
• Signal-timing plan (cycle length and phase times),
• Factors that affect the number of left-turn sneakers for left-turn
movements that have permitted operation, and
• Other factors that influence the saturation flow rates.
The example intersection described here had two through lanes in all
directions. If only one through lane had existed, the blockage effect would have
been much more severe.
Effect of Overflow on Through Movement Capacity

This subsection illustrates how an alternative tool can be used to model
congestion due to storage bay overflow. An example was set up involving
constant blockage of a through lane by left-turning vehicles. This condition arises
only under very severe oversaturation.
The following variables are used for this examination:
• Cycle length is 90 s,
• Effective green time is 41 s, and
• Saturation flow rate is 1,800 veh/h/ln.
The approach has two through lanes. Traffic volumes were sufficient to
overload both lanes, so that the number of trips processed by the simulation
model was determined to be an indication of through movement capacity. With
no storage bay overflow effect, this capacity is computed as 1,640 veh/h (= 3,600 ×
41/90). So, in a 15-min period, 410 trips were processed on average when there
was no overflow.
Exhibit 31-58 shows the effect of the storage bay length on the through
movement capacity. The percentage of the full capacity is plotted as a function of
the storage bay length over the range of 0 to 600 ft. As expected, a zero-length
bay reduces the capacity to 50% of its full value because one lane would be
Use of Alternative Tools Chapter 31/Signalized Intersections: Supplemental

constantly blocked. At the other extreme, the “no blockage” condition, achieved
by setting the left-turn volume to zero, indicates the full capacity was available.
The loss of capacity is more or less linear for storage lengths up to 600 ft, at
which point about 90% of the full capacity is achieved.
Exhibit 31-58
Effect of Storage Bay Length
on Capacity
Bay overflow is a very difficult phenomenon to deal with analytically, and a

substantial variation in its treatment is expected among alternative tools. The
main issue for modeling is the behavior of left-turning drivers denied access to
the left-turn bay because of the overflow. The animated graphics display
produced by some tools can often be used to examine this behavior and assess
the tool’s validity. Typically, some model parameters can be adjusted so that the
resulting behavior is more realistic.
EFFECT OF RIGHT-TURN-ON-RED OPERATION

The treatment of right-turn-on-red (RTOR) operation in the motorized
vehicle methodology is limited to the removal of RTOR vehicles from the right-
turn demand volume. If the right-turn movement is served by an exclusive lane,
the methodology suggests RTOR volume can be estimated as equal to the left-
turn demand of the complementary cross street left-turn movement, whenever
this movement is provided a left-turn phase. Given the simplicity of this
treatment, it may be preferable to use an alternative tool to evaluate RTOR
operation under the following conditions:
• RTOR operation occurs at the intersection,
• Right turns are a critical element of the operation,
• An acceptable LOS depends on RTOR movements, or
• Detailed phasing alternatives involving RTOR are being considered.
The remainder of this subsection examines the RTOR treatment offered in
the motorized vehicle methodology. The objective of this discussion is to
illustrate when alternative tools should be considered.
Effect of Right-Turn Lane Allocation

This subsection examines the effect of the lane allocation for the right-turn
movement. The lane-allocation scenarios considered include (a) provision of a
shared lane for the right-turn movement and (b) provision of an exclusive right-
turn lane. Exhibit 31-59 shows the results of the analysis. The intersection was
simulated with (and without) the RTOR volume.
Exhibit 31-59
Effect of Right-Turn-on-Red
and Lane Allocation on Delay
(a) Shared Lane (b) Exclusive Right-Turn Lane
The trends in Exhibit 31-59 indicate there are only minimal differences in
delay when RTOR is allowed relative to when it is not allowed. The northbound
and southbound approaches had no shadowing opportunities because the
eastbound and westbound movements did not have a protected left-turn phase.
As a result, the effect of lane allocation and RTOR operation was negligible for
the northbound and southbound right-turn movements.
In contrast, the eastbound and westbound right-turn movements were
shadowed by the protected left-turn phases for the northbound and southbound
approaches. As a result, the effect of lane allocation was more notable for the
eastbound and the westbound right-turn movements.
Effect of Right-Turn Demand Volume

This subsection examines the effect of right-turn demand volume on right-
turn delay, with and without RTOR allowed. The right-turn volumes varied from
100 to 400 veh/h on all approaches. Exclusive right-turn storage bays were
provided on each approach.
The results are shown in Exhibit 31-60. They indicate delay to the
northbound and southbound right-turn movements was fairly insensitive to
right-turn volume, with or without RTOR allowed. The available green time on
these approaches provided adequate capacity for the right turns. RTOR
operation provided about a 25% delay reduction.
The delay to the eastbound and westbound right-turn movements increased
rapidly with right-turn volume when RTOR was not allowed. At 300 veh/h and
no RTOR, the right-turn delay becomes excessive in both directions. With RTOR,
delay is less sensitive to right-turn volume. This trend indicates the additional
capacity provided by RTOR is beneficial for higher right-turn volume levels.

Right-Turn Delay (s/veh) ..

35 35 Exhibit 31-60
No RTOR
Right-Turn Delay (s/veh)

30 30
No RTOR
25 25 and Right-Turn Volume on
20
With RTOR With RTOR Delay
20
15 15
10 10
5 5
0 0
100 200 300 400 100 200 300 400
Right-Turn Volume (veh/h) Right-Turn Volume (veh/h)
(a) Northbound (b) Southbound
350 350
No RTOR
No RTOR

300 300
250 250
RTOR volume RTOR volume
200 removed per 200 removed per
methodology methodology
150 150
100 100
With RTOR
50 With RTOR 50
0 0
100 200 300 400 100 200 300 400
Right-Turn Volume (veh/h) Right-Turn Volume (veh/h)
(c) Eastbound (d) Westbound
The treatment of RTOR suggested in the motorized vehicle methodology

(i.e., removal of the RTOR vehicles from the right-turn volume) was also
examined. The simulation analysis was repeated with the right-turn volumes
reduced in this manner to explore the validity of this treatment.
The results of this analysis are shown in Exhibit 31-60 for the eastbound and
westbound approaches. The trends shown suggest the treatment yields a result
that is closer to the “with RTOR” case, as intended. However, use of the
treatment in this case could still lead to erroneous conclusions about right-turn
delay at intersections with high right-turn volumes.
Effect of a Protected Right-Turn Phase

This subsection compares the effect of adding a protected right-turn phase
without RTOR allowed relative to just allowing RTOR. The example intersection
was modified to include an exclusive right-turn storage bay and a protected
right-turn phase for both the eastbound and westbound approaches. Each phase
was timed concurrently with the complementary northbound or southbound
left-turn phase, as appropriate. The results are shown in Exhibit 31-61. The trends
in the exhibit indicate the protected phase does not improve over RTOR
operation at low volume levels. However, it does provide some delay reduction
at the high end of the volume scale.

Exhibit 31-61
and Right-Turn Protection on
Delay
(a) Eastbound (b) Westbound
This examination indicates RTOR operation can have some effect on right-
turn delay. The effect is most notable when there are no shadowing
opportunities in the phase sequence for right-turn service or the right-turn
volume is high. The use of an alternative tool to evaluate RTOR operation may
provide a more realistic estimate of delay than simply removing RTOR vehicles
from the right-turn demand volume, as suggested in Chapter 19.
EFFECT OF SHORT THROUGH LANES

One identified limitation of the motorized vehicle methodology is its
inability to evaluate short through lanes that are added or dropped at the
intersection. This subsection describes the results from an evaluation of this
geometry for the purpose of illustrating the effect of short through lanes.
Several alternative tools can address the effect of short through lanes. Each
tool will have its own unique method of representing lane drop or add geometry
and models of driver behavior. Some degree of approximation is involved with
all evaluation tools.
The question under consideration is, “How much additional through traffic
could the northbound approach accommodate if a lane were added both 150 ft
upstream and 150 ft downstream of the intersection?” The capacity of the
original two northbound lanes was computed as 1,778 veh/h (i.e., 889 veh/h/ln)
by using the motorized vehicle methodology. The simulation tool’s start-up lost
time and saturation headway parameters were then adjusted so the simulation
tool produced the same capacity. It was found in this case that a 2.3-s headway
and 3.9-s start-up lost time produced the desired capacity.
Finally, the additional through lane was added to the simulated intersection,
and the process of determining capacity was repeated. On the basis of an average
of 30 runs, the capacity of the additional lane was computed as 310 veh/h.
Theoretically, the addition of a full lane would increase the capacity by another
889 veh/h, for a total of 2,667 veh/h.
The alternative tool indicates the additional lane contributes only 0.35
equivalent lane (= 310/889). This result cannot be stated as a general conclusion
that applies to all cases because other parameters (such as the signal-timing plan
and the proportion of right turns in the lane group) will influence the results.
More important, the results are likely to vary among alternative tools given the
likely differences in their driver behavior models.

EFFECT OF CLOSELY SPACED INTERSECTIONS

The effect of closely spaced intersections is examined in this subsection. The
motorized vehicle methodology does not account for the effect of queue cyclic
spillback from a downstream signal or demand starvation from an upstream
signal. It is generally accepted that simulation of these effects is desirable when
two closely spaced signalized intersections interact with each other in this
manner.
Consider two intersections separated by 200 ft along the north–south
roadway. They operate with the same cycle length and the same northbound and
southbound green time. To keep the problem simple, only through movements
are allowed at these intersections. The northbound approach is used in this
discussion to illustrate the effect of the adjacent intersection. The layout of this
system and the resulting lane blockage are illustrated in Exhibit 31-62.
Exhibit 31-62
Closely Spaced Intersections
Exhibit 31-62 illustrates both cyclic spillback and demand starvation at one
point in the cycle. For the northbound direction, traffic queues have spilled back
from the downstream intersection to block the upstream intersection. For the
southbound direction, the traffic at the upstream intersection is prevented from
reaching the downstream intersection by the red signal at the upstream
intersection. Valuable green time is being wasted in both travel directions at the
southern intersection.
Exhibit 31-63 illustrates the relationship between signal offset and the
performance of the northbound travel direction. In terms of capacity, the exhibit
shows that under the best-case condition (i.e., zero offset), the capacity is
maintained at a value slightly above the demand volume. Under the worst-case

condition, the capacity is reduced to slightly below 1,000 veh/h. The demand
volume-to-capacity ratio under this condition is about 1.7.
Exhibit 31-63
Effect of Closely Spaced
Intersections on Capacity and
Delay
(a) Approach Capacity (b) Control Delay
The effect of signal offset time on the delay to northbound traffic

approaching the first intersection is also shown in Exhibit 31-63. As expected, the
delay is minimal under favorable offsets, but it increases rapidly as the offset
becomes less favorable. Delay is at its maximum value with a 45-s offset time.
The large value of delay suggests that approach is severely oversaturated.
The delay reported by most simulation tools represents the delay incurred by
vehicles when they depart the system during the analysis period, as opposed to
the delay incurred by vehicles that arrive during the analysis period. The latter
measure represents the delay reported by the motorized vehicle methodology.
For oversaturated conditions, the delay reported by a simulation tool may be
biased when the street system is not adequately represented. This bias occurs
when the street system represented to the tool does not physically extend beyond
the limits of the longest queue that occurs during the analysis period.
The issues highlighted in the preceding paragraphs must be considered
when an alternative tool is used. Specifically, a multiple-period analysis must be
conducted that temporally spans the period of oversaturation. Also, the spatial
boundaries of the street system must be large enough to encompass all queues
during the saturated time periods. A more detailed discussion of multiple-period
analyses is presented in Chapter 7, Interpreting HCM and Alternative Tool Results.

INTRODUCTION
This section provides base saturation flow rates for signalized intersections
that account for the presence of connected and automated vehicles (CAVs) in the
traffic stream. It also provides daily and hourly maximum service volumes for
signalized intersections for different proportions of CAVs in the traffic stream.
Although CAVs are still a developing technology, transportation agencies have
an immediate need as part of their long-range planning efforts to account for
CAVs’ potential ability to increase existing roadways’ throughput.
or both, these approaches are currently infeasible given the absence of CAVs in
the traffic stream. Instead, uncalibrated simulation modeling was conducted
using CAV logic developed for the Federal Highway Administration. Details
about this modeling are available in a paper (11) available online in HCM Volume
4 (hcmvolume4.org) in the Technical Reference Library section for Chapter 31.
All exhibits in this section assume that the CAV market penetration rate is a
global input for the entire intersection. The planning-level adjustment factors
currently do not support varying the percentage of CAVs on a per-lane or per-
approach basis.
CONCEPTS
CAV Technology
benefits. However, because a human is still driving the vehicle, car-
following and other behavior that influences saturation flow rate is not
expected to fundamentally change.
Chapter 31/Signalized Intersections: Supplemental Connected and Automated Vehicles

system (13).
Factors Influencing Signalized Intersection Capacity with CAVs in the

Traffic Stream
Protected Movements
As shown in Equation 19-16, the capacity of a lane group serving one traffic
movement, for which there are no permitted left-turn movements, depends on
three factors:
• The number of lanes in the lane group,
• The effective green–to–cycle length (g/C) ratio, and
• The saturation flow rate.
Of these factors, CAVs have the greatest influence on saturation flow rate,
due primarily to their potential ability to safely operate at closer headways than
human-driven vehicles. CAVs can also increase the effective green time slightly
(by less than 1.0 s) by eliminating the signal-change reaction component of start-
up lost time. However, this increase in effective green time only produces minor
increases in capacity. CAVs have no direct influence on the number of lanes
provided.
Permitted Left-Turn Movements

In addition to the factors listed above for protected movements, CAVs may
improve the capacity of permitted left-turn movements through their potential to
safely accept smaller gaps in traffic than human drivers. As a result, CAVs’
critical and follow-up headways would be less than the average values for
human drivers given with Equation 31-100 (Section 3, Step 3) and capacity would
increase. Further, the gap availability in the conflicting traffic stream may be
increased with CAVs organizing into platoons, resulting in larger gaps between
platoons.
Factors Influencing CAV Ability to Provide Higher Capacities

capacity benefit. A key assumption is the minimum achievable intervehicle gap
Connected and Automated Vehicles Chapter 31/Signalized Intersections: Supplemental

within a platoon, as well as the gap between platoons. Factors that could affect
the eventual intervehicle and interplatoon gaps include:
• Legal or regulatory requirements that dictate a minimum gap.
use a more conservative gap length than strictly needed for safety.
• Passenger comfort concerns on the part of vehicle manufacturers to
minimize the amount and magnitude of acceleration and deceleration that
is needed to maintain intervehicle gaps and facilitate lane changing.
• Passenger insecurity concerns on the part of vehicle owners related to
traveling close behind another vehicle.
• Need for sufficient gaps to accommodate lane-changing in preparation for
turning maneuvers at downstream intersections.
A second key assumption is that all pieces of the system will operate with a
high degree of reliability. This assumption requires, among other things, vehicle
manufacturers to build vehicles with reliable components, vehicle owners to
promptly repair components if they do break, roadway agencies to provide and
properly maintain sufficient communications infrastructure, and regulatory
agencies to provide adequate bandwidth for all the elements that need to
communicate with each other.
Finally, once CAVs become available to consumers, it may take many years
for the vehicle fleet to transition to an all-CAV fleet. In 2018, the average age of
light cars and trucks in the United States was just under 12 years (14), and it
takes even longer for the national fleet to turn over. Furthermore, based on past
airbags, and hybrid vehicles, many people will not choose a CAV the first or
prior automotive technologies. Analysts should consider all of the above factors,
incorporating the latest available information, when estimating CAV effects on
signalized intersection capacity.
others, none of which are known with great certainty. Adding assumptions
analysis inputs. Therefore, it is recommended that the saturation flow rates and
service volumes presented below be applied to the evaluation of “what if”
scenarios, rather than being taken as the final word on what will happen once
CAVs become widespread. In particular, the analyst should consider:

• What if the minimum headway permitted by technology, regulation, or policy is

longer than assumed in the research? In this case, the improvement in
saturation flow rate would be less and the corresponding capacity
increase would also be less.
CAV-capable vehicles. To the extent that roadway communications
infrastructure is inoperable due to malfunction, damage, or upgrade, a
capacity increase would not occur during those periods (and capacity
might actually decrease due to the use of adaptive cruise control, as
discussed above), which has travel time reliability implications.
• How quickly will CAV technology be available and be adopted, and how will
questions will determine the assumed volume and proportion of CAV-
capable vehicles in the traffic stream, along with the assumed saturation
flow rate.
MODIFICATIONS TO SIGNALIZED INTERSECTION CORE

METHODOLOGY INPUTS
This subsection provides replacement values for certain inputs to the
Chapter 19 core motorized vehicle methodology to account for the presence of
CAVs in the traffic stream on a signalized intersection approach. CAV
adjustment factors were derived from microsimulation using assumptions based
on current knowledge. CAV adjustment values have not been field validated due
to the current lack of commercially available CAVs.
These factors may also be used in other instances in the HCM where
saturation flow rates are calculated, including in this chapter’s planning method
(Section 5). These factors may also be used to approximate the effects of CAVs at
interchange ramp terminals and alternative intersections. Although alternative
designs change an intersection’s or interchange’s configuration, they involve the
same basic signal elements (e.g., stop bar, signal head) and timing parameters as
standard designs. As such, CAV effects on saturation flow rate also apply to
alternative intersections and interchanges. No adjustments for CAV capacities
are available for STOP-controlled movements at a signalized intersection. The
adjustments provided in Chapter 33, Roundabouts Supplemental, can be used
for YIELD-controlled movements at a signalized intersection.
Saturation Flow Rate

Exhibit 31-64 provides base saturation flow rates so for through movements at
signalized intersection approaches where CAVs are present in the traffic stream.
The base saturation flow rate is applied in Equation 19-8 along with a variety of
adjustment factors to determine an adjusted saturation flow rate. Most of these
adjustments also apply with CAVs; however, the adjustment for lane width
should not be applied when CAVs are present.

Proportion of CAVs Base Saturation Flow Rate Exhibit 31-64

in Traffic Stream (pc/h/ln) Base Saturation Flow Rates
0 1,900 for CAVs for Through
20 2,000 Movements at Signalized
40 2,150 Intersections
60 2,250
80 2,550
100 2,900
Assumes no interaction with non-motorized road users, no adverse weather impacts, and a facility without
driveways or access points impacting saturation flow rates.
Interpolate for other CAV proportions.
The saturation flow rates shown in Exhibit 31-64 assume no interaction with
non-motorized road users, no adverse weather impacts, and a facility without
driveways or access points impacting saturation flow rates. The increases in
saturation flow rate as a function of the proportion of CAVs in the traffic stream is
largely due to reduced headways between vehicles and does not consider changes
to signal timing as a result of CAV presence.
Capacity Adjustments for Protected and Permitted Left Turns

Left-turn movements at signalized intersections may also see increases in
capacity due to the presence of CAVs in the vehicle stream. The capacity benefits
are a result of reduced saturation headways for protected left turns and reduced
critical headway and follow-up headways for permitted left turns.
Exhibit 31-65 provides values of the saturation flow rate adjustment factor
for protected left turns fCAV,prot as a function of increasing proportion of CAVs in
the traffic stream. This factor should be used as an additional adjustment in
Equation 19-8 to estimate the resulting saturation flow rate for protected left
turns. Note that the factors in Exhibit 31-65 are adjustments to the base saturation
flow rate (with 0% CAVs). These factors should not be used in addition to the
values in Exhibit 31-64.
Proportion of CAVs Saturation Flow Rate Adjustment for Exhibit 31-65

in Traffic Stream Protected Left Turns, fCAV,prot Saturation Flow Rate
0 1.00 CAV Adjustment for Protected Left
20 1.01 Turns at Signalized Intersections
40 1.07
60 1.11
80 1.21
100 1.56
Assumptions: Average intervehicle gap within CAV platoons = 0.71 s, CAV interplatoon gap = 1.5 s,
maximum CAV platoon size = 8 pc, human-driven vehicles operate with through movement saturation
flow rates calibrated to 1,900, assumes no interaction with non-motorized road users, no adverse
weather impacts, and a facility without driveways or access points impacting saturation flow rates.
Exhibit 31-66 provides values of the CAV saturation flow rate adjustment
factor for permitted left turns fCAV,perm as a function of the total opposing through
volume per lane. This factor should be used as an additional adjustment in
Equation 19-8 to estimate the resulting saturation flow rate for permitted left
turns. The factors in Exhibit 31-66 are adjustments to the base saturation flow

rate (with 0% CAVs) and should not be used in addition to the values in Exhibit
31-64 or Exhibit 31-65.
Exhibit 31-66 Proportion of Saturation Flow Rate Adjustment for Permitted Left Turns fCAV,perm
Saturation Flow Rate CAV CAVs in Traffic by Opposing Through Volume Per Lane (pc/h/ln)
Adjustments for Permitted Stream 300 450 600 750
Left Turns at Signalized 0 1.00 1.00 1.00 1.00
Intersections 20 1.12 1.04 1.03 1.07
40 1.20 1.16 1.12 1.18
60 1.29 1.22 1.26 1.36
80 1.43 1.43 1.57 1.60
100 1.76 1.72 1.66 1.90
Assumptions: Average intervehicle gap within CAV platoons = 0.71 s, CAV interplatoon gap = 1.5 s,
maximum CAV platoon size = 8 pc, human-driven vehicles operate with through movement saturation
flow rates calibrated to 1,900, assumes no interaction with non-motorized road users, no adverse
weather impacts, and a facility without driveways or access points impacting saturation flow rates.
SERVICE VOLUME TABLE

Exhibit 31-67 presents illustrative service volumes for signalized intersection
approaches segments with CAVs present in the traffic stream. Assumptions used
in creating these exhibits are listed below the exhibit; see Chapter 19 for
definitions of these terms.
Exhibit 31-67 Through No. of

Illustrative Generalized Movement Through Proportion of CAVs in Traffic Stream
Service Volume LOS E g/C Ratio Lanes 0 20 40 60 80 100
Thresholds for Signalized
1 800 840 910 950 1,070 1,220
Intersections with CAV
Presence (veh/h) 0.40 2 1,550 1,630 1,750 1,840 2,080 2,370
3 2,000 2,110 2,260 2,370 2,680 3,050
1 910 960 1,030 1,080 1,220 1,390
0.45 2 1,740 1,830 1,970 2,060 2,340 2,660
3 2,250 2,370 2,550 2,660 3,020 3,430
1 1,020 1,070 1,150 1,210 1,370 1,560
0.50 2 1,930 2,030 2,180 2,290 2,590 2,950
3 2,500 2,630 2,830 2,960 3,360 3,820
Notes: LOS E threshold is defined by control delay greater than 80 s/veh or volume-to-capacity ratio >1.0.
CAV = connected and automated vehicle, defined here as a vehicle with an operating cooperative adaptive
Assumes no interaction with non-motorized road users, no adverse weather impacts, and a facility without
driveways or access points impacting saturation flow rates.
Assumed values for all entries:
Heavy vehicles: 0%
Peak hour factor: 0.92
Lane width: 12 ft
Grade: 0%
Separate left-turn lane: yes
Separate right-turn lane: no
Pretimed control
Cycle length: 90 s
Lost time: 4 s/phase
Protected left-turn phasing: yes
g/C ratio for left-turn movement: 0.10
Parking maneuvers per hour: 0
Buses stopping per hour: 0
Percentage left turns: 10%
Percentage right turns: 10%

10. EXAMPLE PROBLEMS
This section describes the application of each of the motorized vehicle,

pedestrian, and bicycle methodologies through the use of example problems.
Exhibit 31-68 provides an overview of these problems. The examples focus on the
operational analysis level. The planning and preliminary engineering analysis
level is identical to the operational analysis level in terms of the calculations,
except that default values are used when field-measured values are not available.
Problem Exhibit 31-68

Number Description Analysis Level Example Problems
1 Motorized vehicle LOS Operational
2 Pedestrian delay, LOS, and circulation area Operational
3 Bicycle LOS Operational
4 Pedestrian delay, two-stage crossing of one intersection leg Operational
5 Pedestrian delay, two-stage crossing of two intersection legs Operational
EXAMPLE PROBLEM 1: MOTORIZED VEHICLE LOS

The Intersection
The intersection of 5th Avenue and 12th Street is an intersection of two urban
arterial streets. The intersection plan view is shown in Exhibit 31-69.
Intersection Geometry Exhibit 31-69

Example Problem 1:
= Pedestrian Button
Intersection Plan View
5 Avenue
Grade = 0%
Street
= Through
th
Show North Arrow
Grade = 0% = Right
120 p/h
10' = Left
10'
40 p/h 40 p/h 10'
10' = Through + Right
10'
120 p/h
12th Street
= Left + Through
Grade = 0% Street
= Left + Right

Grade = 0%
12' 12' 12' 12' 12'
The Question
What is the motorist delay and LOS during the analysis period for each lane
group and the intersection as a whole?
The Facts
The intersection’s traffic, geometric, and signalization conditions are listed in
Exhibit 31-70, Exhibit 31-71, and Exhibit 31-72, respectively. Exhibit 31-73
presents additional data. The volume data provided represent the demand flow
rate during the 0.25-h analysis period, so a peak hour factor is not applicable to
this evaluation.
Chapter 31/Signalized Intersections: Supplemental Example Problems

Exhibit 31-70 Eastbound Westbound Northbound Southbound

Example Problem 1: Traffic
Input Data Element L T R L T R L T R L T R
Characteristics Data
Demand flow rate 71 318 106 118 600 24 133 1644 111 194 933 111
(veh/h)
RTOR flow rate (veh/h) 0 0 22 33
Percentage heavy 5 5 5 5 2 2 2 2
vehicles (%)
Platoon ratio 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Upstream filtering 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
adjustment factor
Initial queue (veh) 0 0 0 0 0 0 0 0
Base saturation flow 1,90 1,900 1,90 1,900 1,90 1,900 1,90 1,900
rate (pc/h/ln) 0 0 0 0
Pedestrian flow rate 120 120 40 40
(p/h)
Bicycle flow rate 0 0 0 0
(bicycles/h)
On-street parking 5 5
maneuver rate
(maneuvers/h)
Local bus stopping rate 0 0 0 0
(buses/h)
Note: L = left turn; T = through; R = right turn.

Example Problem 1:
Number of lanes (ln) 1 2 0 1 2 0 1 2 0 1 2 0
Average lane width
10.0 10.0 10.0 10.0 12.0 12.0 12.0 12.0
(ft)
Number of receiving
2 2 2 2
lanes (ln)
Turn bay length (ft) 200 200 200 200
Presence of on-street
No Yes No Yes No No No No
parking
Approach grade (%) 0 0 0 0
Note: L = left turn; T = through; R = right turn.
Exhibit 31-72 Input Data Element Eastbound Westbound Northbound Southbound

Type of signal control Actuated Actuated Actuated Actuated
Control Data
Phase sequence No left-turn phase No left-turn phase Leading left Lagging left
Phase number 2 6 3 8 7 4
Movement L+T+R L+T+R L T+R L T+R
Left-turn operational Prot.- Prot.-
Perm. Perm.
mode Perm. Perm.
Dallas left-turn
No No
phasing option
Passage time (s) 2.0 2.0 2.0 2.0 2.0 2.0
Maximum green (s) 30 30 25 50 25 50
Minimum green (s) 5 5 5 5 5 5
Yellow change (s) 4.0 4.0 4.0 4.0 4.0 4.0
Red clearance (s) 0 0 0 0 0 0
Walk (s) 5 5 5 5
Pedestrian clear (s) 14 14 16 16
Phase recall No No No No No No
Dual entry Yes Yes No Yes No Yes
Simultaneous gap-out Yes Yes
Note: L = left turn; T = through; R = right turn; Prot. = protected; Perm. = permitted.
Example Problems Chapter 31/Signalized Intersections: Supplemental

Eastbound Westbound Northbound Southbound Exhibit 31-73

Example Problem 1: Other
Data
Analysis period
0.25 0.25 0.25 0.25
duration (h)
Speed limit (mi/h) 35 35 35 35
Stop-line detector 40 40
40 40 40 40 40 40
length (ft)
Detection mode Pres. Presence Pres. Presence Pres. Presence Pres. Presence
Area type Central business district
Note: L = left turn; T = through; R = right turn; Pres. = presence.
The intersection is located in a central business district–type environment.

Adjacent signals are somewhat distant so the intersection is operated by using
fully actuated control. Vehicle arrivals to each approach are characterized as
“random” and are described by using a platoon ratio of 1.0.
The left-turn movements on the north–south street operate under protected-
permitted control and lead the opposing through movements (i.e., a lead–lead
phase sequence). The left-turn movements on the east–west street operate as
permitted.
All intersection approaches have a 200-ft left-turn bay, an exclusive through
lane, and a shared through and right-turn lane. The average width of the traffic
lanes on the east–west street is 10 ft. The average width of the traffic lanes on the
north–south street is 12 ft.
Crosswalks are provided on each intersection leg. A two-way flow rate of
120 p/h is estimated to use each of the east–west crosswalks and a two-way flow
rate of 40 p/h is estimated to use each of the north–south crosswalks.
On-street parking is present on the east–west street. It is estimated that
parking maneuvers on each intersection approach occur at a rate of
5 maneuvers/h during the analysis period.
The speed limit is 35 mi/h on each intersection approach. The analysis period
is 0.25 h. There is no initial queue for any movement.
As noted in the next section, none of the intersection movements have two or
more exclusive lanes. For this reason, the saturation flow rate adjustment factor
for lane utilization is not applicable. Any unequal lane use that may occur due to
the shared through and right-turn lane groups will be accounted for in the lane
group flow rate calculation, as described in the Lane Group Flow Rate on
Multiple-Lane Approaches subsection of Section 2.
Outline of Solution
The solution follows the steps listed in Exhibit 19-18 of Chapter 19.
Step 1: Determine Movement Groups and Lane Groups

The left-turn lanes are designated as separate movement groups according to
the rules described in Chapter 19. The through and shared right-turn and
through lanes are combined into one movement group on each approach. The
movement group designations are shown in Exhibit 31-74a with brackets
showing how the individual movements are combined into movement groups.

Exhibit 31-74
Example Problem 1:
Movement Groups and Lane
Groups
(a) Movement Groups (b) Lane Groups
Each lane is analyzed as a separate lane group according to the rules in

Chapter 19. The lane group designations are shown in Exhibit 31-74b with
brackets showing how the individual lanes are combined into lane groups.
Step 2: Determine Movement Group Flow Rate

Exhibit 31-75 shows the movement group flow rates, which are based on the
movement groups identified in Exhibit 31-74a. The RTOR flow rate is subtracted
from the right-turn volume for the northbound and southbound through-and-
right-turn movement groups.
Exhibit 31-75 Data Element Eastbound Westbound Northbound Southbound

Example Problem 1:
Movement group L T+R L T+R L T+R L T+R
Movement Group Flow Rates
Number of lanes (ln) 1 2 1 2 1 2 1 2
Movement group flow 318 + 106 600 + 24 1,644 + 111 933 + 111
71 118 133 194
rate (veh/h) = 424 = 624 − 22 = 1,733 − 33 = 1,011
Note: L = left turn; T+R = combined through and right turn.
Step 3: Determine Lane Group Flow Rate

There is one shared lane and two or more lanes on each intersection
approach. For this configuration, the lane group flow rates for the through-and-
right-turn movement groups are computed by the procedures in the Lane Group
Flow Rate on Multiple-Lane Approaches subsection of Section 2. The results of
these calculations are given in Exhibit 31-76. The left-turn lane group volumes
remain unchanged from Exhibit 31-75 because the movement groups and the
lane groups are the same for the left-turn lanes. The volumes shown for the
through lane group and the shared lane group represent the flow rates obtained
from the Section 2 procedure.

Example Problem 1: Lane
Lane group L T T+R L T T+R L T T+R L T T+R
Group Flow Rates
Number of lanes (ln) 1 1 1 1 1 1 1 1 1 1 1 1
Flow rate (veh/h) 71 239 185 118 337 287 133 870 863 194 513 497
Note: L = left turn; T = through; T+R = combined through and right turn.

Step 4: Determine Adjusted Saturation Flow Fate

The base saturation flow rate is 1,900 veh/h/ln for each lane group.
Adjustments made for each of the lane groups are summarized in the following
paragraphs.
The left-turn lane groups for the eastbound and westbound approaches
operate with the permitted mode. The saturation flow rate of a permitted left-
turn movement sp is determined with Equation 31-100. For example, the
saturation flow rate for the eastbound left-turn lane group is computed with the
following equation.
𝑣𝑜 𝑒 −𝑣𝑜 𝑡𝑐𝑔/3,600 624𝑒 −624(4.5)/3,600
𝑠𝑝 = = = 813 veh/h/ln
1 − 𝑒 −𝑣𝑜𝑡𝑓ℎ/3,600 1 − 𝑒 −624(2.5)/3,600
The adjustment factor for the existence of parking and parking activity fp is
applied to the shared-lane lane groups for the eastbound and westbound
approaches. This factor is computed with Equation 19-11.
The adjustment factor for area type fa is applied to all lane groups. Guidance
for determining this factor’s value is provided in Section 3 of Chapter 19 (in the
subsection titled Adjustment for Area Type).
The adjustment factor for heavy vehicles and grade fHVg is computed with
Equation 19-10. This factor is applicable to all lane groups.
The adjustment factors and the adjusted saturation flow rate for each
movement are shown in Exhibit 31-77.
Data Element Eastbound Westbound Northbound Southbound Exhibit 31-77

Example Problem 1: Adjusted
Phase number 2 2 2 6 6 6 3 8 8 7 4 4
Base saturation flow
1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900
rate so (pc/h/ln)
Permitted left turn
saturation flow rate sp 813 978
(veh/h/ln)
Adjustment factor for
left-turn vehicle 0.95 0.95
presence, fLT
heavy vehicles and 0.96 0.96 0.96 0.96 0.96 0.96 0.98 0.98 0.98 0.98 0.98 0.98
grade, fHVg
existence of parking
0.88 0.88
lane and parking
activity, fp
0.90 0.90 0.90 0.90 0.90 0.90 0.90 0.90 0.90 0.90 0.90 0.90
area type, fa
Pedestrian adjustment
factor for left-turn 1.00 0.98 1.00 1.00
groups, fLpb
Pedestrian–bicycle
adjustment factor for 0.88 0.88 0.98 0.98
right-turn groups, fRpb
Adjusted saturation
702 1,643 1,201 825 1,643 1,398 1,603 1,683 1,648 1,603 1,683 1,630
flow rate (veh/h/ln)
Notes: L = left turn; T = through; T+R = combined through and right turn.
Calculated values are based on maintaining six or more significant digits for all computed values through all
calculations. These values are shown with fewer digits for presentation purposes only.

Equation 19-8 shows all the adjustment factors that might be applied in the
calculation of saturation flow rate. However, when this equation is applied to a
given lane group, some of the factors are not applicable (or have a value of 1.0)
and can be removed from the equation. The reduced form of the saturation flow
rate equation is described in the following paragraphs for several of the lane
groups at the subject intersection.
For the eastbound and westbound left-turn lane groups, the adjusted
saturation flow rate is calculated with the following equation.
𝑠 = 𝑠𝑝 𝑓𝐻𝑉𝑔 𝑓𝑎 𝑓𝐿𝑝𝑏
The northbound and southbound left-turn lane groups operate in the
protected-permitted mode. The adjusted saturation flow rate for the protected
left-turn phase is calculated with the following equation.
𝑠 = 𝑠𝑜 𝑓𝐿𝑇 𝑓𝐻𝑉𝑔 𝑓𝑎
The adjusted saturation flow rate for the permitted left-turn period is
calculated with the same equation as for the eastbound and westbound left-turn
lane groups.
For the through lane groups on each approach, the adjusted saturation flow
rate is computed with the following equation.
𝑠 = 𝑠𝑜 𝑓𝐻𝑉𝑔 𝑓𝑎
For the shared-lane lane groups, the adjusted saturation flow rate is
computed by using Equation 31-105. This equation is reproduced below for the
eastbound shared right-turn and through lane group.
𝑠𝑡ℎ 1,438
𝑠𝑠𝑟 = = = 1,201 veh/h/ln
𝐸𝑅 106 1.18
1 + 𝑃𝑅 ( − 1) 1+( ) ( − 1)
𝑓𝑅𝑝𝑏 186 0.88
with
𝑠𝑡ℎ = 𝑠𝑜 𝑓𝐻𝑉𝑔 𝑓𝑝 𝑓𝑎 = 1,900 × 0.96 × 0.88 × 0.90 = 1,438 veh/h/ln
The calculated adjustment factors and saturation flow rates in the previous
equations are based on maintaining six or more significant digits for all
computed values through all calculations. These values are shown with fewer
digits for presentation purposes only.
Step 5: Determine Proportion Arriving During Green

The proportion arriving during green P is computed using Equation 19-15.
The results are shown in Exhibit 31-78. The effective green time g and cycle
length C are determined by using the results from the final iteration of Step 6.


Example Problem 1:
Proportion Arriving During
Phase number 2 2 2 6 6 6 3 8 8 7 4 4 Green
Effective green time g (s) 30.0 30.0 30.0 30.0 30.0 30.0 6.2 50.0 50.0 9.8 53.6 53.6
Proportion arriving on
0.29 0.29 0.29 0.29 0.29 0.29 0.06 0.49 0.49 0.10 0.53 0.53
green, P
Step 6: Determine Signal Phase Duration

The duration of each signal phase is determined by using the procedure
described in Section 2 (in the subsection titled Actuated Phase Duration). The
results of this iterative process are shown in Exhibit 31-79. The resulting cycle
length is 101.8 s.

Phase number 2 6 3 8 7 4
Phase Duration
Assigned movements L+T+R L+T+R L T+R L T+R
Phase duration Dp (s) 34.0 34.0 10.2 54.0 13.8 57.6
Maximum allowable headway MAH (s) 3.4 3.4 3.1 3.1 3.1 3.1
Maximum queue clearance time gc (s) 28.7 27.2 4.1 50.0 7.6 21.2
Green extension time ge (s) 0.0 0.4 0.2 0.0 0.3 7.8
Probability that subject phase is
1.00 1.00 0.98 1.00 1.00 1.00
called, pc
Probability of max-out, px 1.00 1.00 0.0 1.00 0.0 0.18
Duration of permitted left-turn green
not blocked by an opposing queue, gu 11.4 17.0 32.5 0.0
(s)
Notes: L = left turn; T = through; T+R = combined through and right turn; L+T+R = combined left, through, and
right turn.
Step 7: Determine Capacity and Volume-to-Capacity Ratio

The capacity of each through lane group and each shared-lane lane group is
computed with Equation 19-16. The capacity for the permitted left-turn lane
groups is computed with Equation 31-119. The latter equation is reproduced
below for the eastbound left-turn lane group.
𝑔𝑢 𝑠𝑙 + 3,600 𝑛𝑠 𝑓𝑚𝑠 𝑓𝑠𝑝
𝑐𝑙,𝑒 = 𝑁𝑙
𝐶
(11.4 × 702) + (3,600 × 2 × 1.0 × 1.0)
𝑐𝑙,𝑒 = × 1 = 149 veh/h
101.8
The capacity for the protected-permitted left-turn lane groups on the
northbound and southbound approaches is computed with Equation 31-124. The
results from the capacity and the volume-to-capacity ratio calculations are shown
in Exhibit 31-80.


Example Problem 1: Capacity
and Volume-to-Capacity Ratio
Phase number 2 2 2 6 6 6 3 8 8 7 4 4
Number of lanes N (ln) 1 1 1 1 1 1 1 1 1 1 1 1
Flow rate v (veh/h) 71 239 185 118 337 287 133 870 863 194 513 497
Adjusted saturation 1,64 1,20 1,64 1,39 1,60 1,68 1,64 1,60 1,68 1,63
702 825
flow rate s (veh/h/ln) 3 1 3 8 3 3 8 3 3 0
Effective green time g
30.0 30.0 30.0 30.0 30.0 30.0 6.2 50.0 50.0 9.8 53.6 53.6
(s)
Capacity c (veh/h) 149 484 354 208 484 412 328 827 809 225 887 859
Volume-to-capacity
0.47 0.49 0.52 0.57 0.70 0.70 0.41 1.05 1.07 0.86 0.58 0.58
ratio X
Step 8: Determine Delay

The control delay for each movement and approach, and for the intersection
as a whole, is calculated with Equation 19-18. The results of the delay
calculations are shown in Exhibit 31-81.

Example Problem 1: Control
Delay
Phase number 2 2 2 6 6 6 3 8 8 7 4 4
Uniform delay d1 (s/veh) 44.6 29.6 29.9 41.3 31.9 31.9 13.2 25.9 25.9 28.9 16.4 16.4
Incremental delay d2 (s/veh) 0.9 0.3 0.7 2.3 3.6 4.3 0.3 46.0 50.8 3.8 0.6 0.7
Initial queue delay d3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
(s/veh)
Control delay d (s/veh) 45.5 29.9 30.6 43.5 35.5 36.2 13.5 72.0 76.7 32.6 17.0 17.1
Level of service D C C D D D B F F C B B
Approach delay dA (s/veh) 32.4 37.0 70.0 19.6
Approach LOS C D E B
Intersection delay di (s/veh) 45.9
Intersection LOS D

LOS is based on the control delay. LOS values for each approach and for the
entire intersection are shown in Exhibit 31-81. The determination of LOS is based
on the LOS thresholds in Exhibit 19-8.
Step 10: Determine Queue Storage Ratio

The procedure for calculating the percentile back-of-queue size and queue
storage ratio is described in Section 4. This procedure was used to compute the
50th percentile values for both variables. The results are shown in Exhibit 31-82.


Example Problem 1: Back of
Queue and Queue Storage
Phase number 2 2 2 6 6 6 3 8 8 7 4 4 Ratio
50th percentile back of
1.8 4.8 3.8 3.0 7.6 6.6 1.4 28.9 29.4 4.9 7.7 7.5
queue Q% (veh/ln)
50th percentile queue
0.23 0.12 0.10 0.38 0.20 0.17 0.18 0.74 0.75 0.62 0.20 0.19
storage ratio RQ%

The QAP is a useful way of illustrating the signal timing and performance of
a signalized intersection. The evolution of the queue length during the cycle is
shown in the QAP. In addition, the area of the QAP is the total uniform delay
experienced by all vehicles during the cycle. The variables needed to construct
the QAP for the northbound through lane group are provided in the following
list. The QAP for this movement is shown in Exhibit 31-83.
• Flow rate: 870 veh/h,
• Adjusted saturation flow rate: 1,683 veh/h/ln,
• Cycle length: 101.8 s,
• Effective green time: 50.0 s,
• Effective red time: 51.8 s,
• Maximum queue clearance time: 50.0 s,
• Green extension time: 0.0 s, and
• Queue length at end of effective red: 13.4 veh/ln.
Exhibit 31-83
Example Problem 1: Queue
Accumulation Polygon

EXAMPLE PROBLEM 2: PEDESTRIAN LOS

The Intersection
The pedestrian crossing of interest crosses the north leg at a signalized
intersection. The north–south street is the minor street and the east–west street is
the major street. The intersection serves all north–south traffic concurrently (i.e.,
no left-turn phases) and all east–west traffic concurrently. The signal has an 80-s
cycle length. The crosswalk and intersection corners that are the subject of this
example problem are shown in Exhibit 31-84.
Exhibit 31-84 Corner 1 Corner 2

Example Problem 2:
Pedestrian Flow Rates 525 p/h
Crosswalk
490 p/h
345 p/h
530 p/h
400 p/h 540 p/h 420 p/h 480 p/h
The Question
What is the pedestrian LOS for the crossing?
The Facts
Pedestrian flow rates are shown in Exhibit 31-84. Vehicular flow rates are
Exhibit 31-85
Example Problem 2: Vehicular
Demand Flow Rates
In addition, the following facts are known about the crosswalk and the
intersection corners:
Major street: Phase duration, Dp,mj = 48 s
Yellow change interval, Ymj = 4 s
Red clearance interval, Rmj = 1 s
Walk setting, Walkmj = 7 s
Pedestrian clear setting, PCmj = 8 s
Four traffic lanes (no turn bays)

Minor street: Phase duration, Dp,mi = 32 s

Yellow change interval, Ymi = 4 s
Red clearance interval, Rmi = 1 s
Walk setting, Walkmi = 7 s
Pedestrian clear setting, PCmi = 13 s
Two traffic lanes (no turn bays)
85th percentile speed at a midsegment location, S85,mi = 35 mi/h
Corner 1: Total walkway width, Wa = Wb = 16 ft
Corner radius, R = 15 ft
Corner 2: Total walkway width, Wa = Wb = 18 ft
Corner radius, R = 15 ft
Other data: Effective crosswalk width, Wc = 16 ft
Crosswalk length, Lc = 28 ft
Walking speed, Sp = 4 ft/s
No right-turn channelizing islands are provided on any corner.
Pedestrian signal indications are provided for each crosswalk.
Rest-in-walk mode is not used for any phase.
Comments
On the basis of the variable notation in Exhibit 19-29, the subject crosswalk is
Crosswalk C because it crosses the minor street. The outbound pedestrian flow
rate vco at Corner 1 equals inbound flow rate vci at Corner 2, and the inbound flow
rate vci at Corner 1 equals the outbound flow rate vco at Corner 2.
Outline of Solution
Pedestrian delay and the pedestrian LOS score are calculated for the
crossing. Next, LOS for the crossing is determined on the basis of the computed
score and the threshold values in Exhibit 19-9.
Following the determination of LOS, the example problem continues with
optional steps 4 and 5. First, the circulation area is calculated for both corners.
Next, the circulation area is calculated for the crosswalk. The street corner and
crosswalk circulation areas are then compared with the qualitative descriptions
of pedestrian space listed in Exhibit 19-34.

Computational Steps
Step 1: Determine Pedestrian Delay

Because pedestrian signal indications are provided and rest-in-walk is not
enabled, the effective walk time for the phase serving the major street is
𝑔Walk,𝑚𝑗 = Walk 𝑚𝑗 + 4.0 = 7.0 + 4.0 = 11.0 s
The pedestrian delay is calculated by using Equation 19-54.
2
(𝐶 − 𝑔Walk,𝑚𝑗 )
𝑑𝑝 =
2𝐶
(80 − 11)2
𝑑𝑝 = = 29.8 s/p
2(80)
Step 2: Determine Pedestrian LOS Score for Intersection

The number of vehicles traveling on the minor street during a 15-min period
is computed by using Equation 19-60.
0.25
𝑛15,𝑚𝑖 = ∑ 𝑣𝑖
𝑁𝑐
0.25
𝑛15,𝑚𝑖 = (72 + 336 + 60 + 42 + 400 + 76) = 123.3 veh/ln
2
The cross-section adjustment factor is calculated by using Equation 19-56.
𝐹𝑤 = 0.681(𝑁𝑐 )0.514
𝐹𝑤 = 0.681(2)0.514 = 0.972
The motorized vehicle adjustment factor is computed with Equation 19-57.
𝑣𝑟𝑡𝑜𝑟 + 𝑣𝑙𝑡,𝑝𝑒𝑟𝑚
𝐹𝑣 = 0.00569 ( ) − 𝑁𝑟𝑡𝑐𝑖,𝑐 (0.0027𝑛15,𝑚𝑖 − 0.1946)
4
30 + 42
𝐹𝑣 = 0.00569 ( ) − (0)(0.0027(123.3) − 0.1946) = 0.102
4
The motorized vehicle speed adjustment factor is then computed with
Equation 19-58.
𝐹𝑠 = 0.00013 𝑛15,𝑚𝑖 𝑆85,𝑚𝑖
𝐹𝑠 = 0.00013(123.3)(35) = 0.561
The pedestrian delay adjustment factor is calculated with Equation 19-59.
𝐹delay = 0.0401 ln (𝑑𝑝,𝑐 )
𝐹delay = 0.0401 ln(29.8) = 0.136
The pedestrian LOS score for the intersection Ip,int is then computed with
Equation 19-55.
𝐼𝑝,𝑖𝑛𝑡 = 0.5997 + 𝐹𝑤 + 𝐹𝑣 + 𝐹𝑠 + 𝐹delay
𝐼𝑝,𝑖𝑛𝑡 = 0.5997 + 0.972 + 0.102 + 0.561 + 0.136 = 2.37


According to Exhibit 19-9, the crosswalk operates at LOS B.
Step 4: Determine Street Corner Circulation Area

A. Compute Available Time–Space
For Corner 1, the available time–space is computed with Equation 19-61.
𝑇𝑆corner = 𝐶(𝑊𝑎 𝑊𝑏 − 0.215 𝑅 2 )
𝑇𝑆corner = 80[16 × 16 − 0.215(15)2 ]
𝑇𝑆corner = 16,610 ft 2 -s
B. Compute Holding-Area Waiting Time

The number of pedestrians arriving at the corner during each cycle to cross
the minor street is computed with Equation 19-63.
𝑣𝑐𝑜
𝑁𝑐𝑜 = 𝐶
3,600
530
𝑁𝑐𝑜 = (80) = 11.8 p
3,600
The total time spent by pedestrians waiting to cross the minor street during
one cycle is then calculated with Equation 19-62. The effective walk time gWalk,mj
was determined in Step 1.
2
𝑁𝑐𝑜 (𝐶 − 𝑔Walk,mj )
𝑄𝑡𝑐𝑜 =
2𝐶
(11.8)(80 − 11)2
𝑄𝑡𝑐𝑜 = = 350.5 p-s
2(80)
By the same procedure, the total time spent by pedestrians waiting to cross
the major street during one cycle (Qtdo) is found to be 264.5 p-s.
C. Compute Circulation Time–Space

The circulation time–space is found by using Equation 19-64.
𝑇𝑆𝑐 = 𝑇𝑆corner − [5.0(𝑄𝑡𝑑𝑜 + 𝑄𝑡𝑐𝑜 )]
𝑇𝑆𝑐 = 16,610 − [5.0(350.5 + 264.5)] = 13,535 ft 2 -s
D. Compute Pedestrian Corner Circulation Area

The total number of circulating pedestrians is computed with
Equation 19-66.
𝑣𝑐𝑖 + 𝑣𝑐𝑜 + 𝑣𝑑𝑖 + 𝑣𝑑𝑜 + 𝑣𝑎,𝑏
𝑁𝑡𝑜𝑡 = 𝐶
3,600
490 + 530 + 540 + 400 + 345
𝑁𝑡𝑜𝑡 = (80) = 51.2 p
3,600
Finally, the corner circulation area per pedestrian is calculated with Equation
19-65.
𝑇𝑆𝑐
𝑀corner =
4.0 𝑁𝑡𝑜𝑡

13,535
𝑀corner = = 66.1 ft2 /p
4.0(51.2)
By following the same procedure, the corner circulation area per pedestrian
for Corner 2 is found to be 87.6 ft2/p. According to the qualitative descriptions
provided in Exhibit 19-34, pedestrians at both corners will have the ability to
move in the desired path without needing to alter their movements to avoid
conflicts.
Step 5: Determine Crosswalk Circulation Area

The analysis conducted in this step describes the circulation area for
pedestrians in the subject crosswalk.
A. Establish Walking Speed

As given in the subsection titled The Facts, the average walking speed is
determined to be 4.0 ft/s.
B. Compute Available Time–Space

Rest-in-walk is not enabled, so the pedestrian service time gped is estimated to
equal the sum of the walk and pedestrian clear settings. The time–space available
in the crosswalk is found with Equation 19-67.
𝑇𝑆𝑐𝑤 = 𝐿𝑐 𝑊𝑐 𝑔Walk,𝑚𝑗
𝑇𝑆𝑐𝑤 = (28)(16)(11) = 4,928 ft2 -s
C. Compute Effective Available Time–Space

The number of turning vehicles during the walk and pedestrian clear
intervals is calculated with Equation 19-68.
𝑣𝑙𝑡,𝑝𝑒𝑟𝑚 + 𝑣𝑟𝑡 − 𝑣𝑟𝑡𝑜𝑟
𝑁𝑡𝑣 = 𝐶
3,600
42 + 76 − 38
𝑁𝑡𝑣 = (80) = 1.8 veh
3,600
The time–space occupied by turning vehicles can then be computed with
Equation 19-69.
𝑇𝑆𝑡𝑣 = 40 𝑁𝑡𝑣 𝑊𝑐
𝑇𝑆𝑡𝑣 = 40(1.8)(16) = 1,138 ft 2 -s
The effective available crosswalk time–space TS*cw is found by subtracting the
total available crosswalk time–space TScw from the time–space occupied by
turning vehicles, as shown by Equation 19-68.
∗
𝑇𝑆𝑐𝑤 = 𝑇𝑆𝑐𝑤 − 𝑇𝑆𝑡𝑣
∗
𝑇𝑆𝑐𝑤 = 4,928 − 1,138 = 3,790 ft2 -s
D. Compute Pedestrian Service Time

The number of pedestrians exiting the curb when the WALK indication is
presented is computed by using Equation 19-73.

𝐶 − 𝑔Walk,𝑚𝑗
𝑁𝑝𝑒𝑑,𝑐𝑜 = 𝑁𝑐𝑜
𝐶
80 − 11
𝑁𝑝𝑒𝑑,𝑐𝑜 = (11.8) = 10.2 p
80
Because the crosswalk width is greater than 10 ft, the pedestrian service time
is computed by using Equation 19-71.
𝐿𝑐 𝑁𝑝𝑒𝑑,𝑐𝑜
𝑡𝑝𝑠,𝑐𝑜 = 3.2 + + 2.7
𝑆𝑝 𝑊𝑐
28 10.2
𝑡𝑝𝑠,𝑐𝑜 = 3.2 + + (2.7) = 11.9 s
4.0 16
The other travel direction in the crosswalk is analyzed next. The number of
pedestrians arriving at Corner 1 each cycle by crossing the minor street is
𝑣𝑐𝑖
𝑁𝑐𝑖 = 𝐶
3,600
490
𝑁𝑐𝑖 = (80) = 10.9 p
3,600
The sequence of calculations is repeated for this second travel direction in the
subject crosswalk to indicate that Nped,ci is equal to 9.4 p and tps,ci is 11.8.
E. Compute Crosswalk Occupancy Time

The crosswalk occupancy time for the crosswalk is computed by using
Equation 19-74.
𝑇𝑜𝑐𝑐 = 𝑡𝑝𝑠,𝑐𝑜 𝑁𝑐𝑜 + 𝑡𝑝𝑠,𝑐𝑖 𝑁𝑐𝑖
𝑇𝑜𝑐𝑐 = 11.9(11.8) + 11.8(10.9) = 268.6 p-s
F. Compute Pedestrian Crosswalk Circulation Area

Finally, the crosswalk circulation area per pedestrian for the crosswalk is
∗
𝑇𝑆𝑐𝑤
𝑀𝑐𝑤 =
𝑇𝑜𝑐𝑐
3,790
𝑀𝑐𝑤 = = 14.1 ft2 /p
268.6
The crosswalk circulation area is found to be 14.1 ft2/p. According to the
qualitative descriptions provided in Exhibit 19-28, pedestrians will find their
walking speed is restricted, with very limited ability to pass slower pedestrians.
Improvements to the crosswalk should be considered and may include a wider
crosswalk or a longer walk interval.
Discussion
The crosswalk was found to operate at LOS B in Step 3. It was determined in
Step 4 that the pedestrians at both corners have adequate space to allow freedom
of movement. However, crosswalk circulation area was found to be restricted in
Step 5 and improvements are probably justified. Moreover, the pedestrian delay

computed in Step 1 was found to be slightly less than 30 s/p. With this much
delay, some pedestrians may not comply with the signal indication.
EXAMPLE PROBLEM 3: BICYCLE LOS

The Intersection
A 5-ft-wide bicycle lane is provided at a signalized intersection.
The Question
What is the LOS of this bicycle lane?
The Facts
Saturation flow rate for bicycles = 2,000 bicycles/h
Effective green time = 48 s
Cycle length = 120 s
Bicycle flow rate = 120 bicycles/h
No on-street parking
The vehicular flow rates and street cross-section element widths are as
Exhibit 31-86
Example Problem 3: Vehicular
Demand Flow Rates and
Cross-Section Element Widths
Outline of Solution
Bicycle delay and the bicycle LOS score are computed. LOS is then
determined on the basis of the computed score and the threshold values in
Exhibit 19-9.
Computational Steps
Step 1: Determine Bicycle Delay

A. Compute Bicycle Lane Capacity
The capacity of the bicycle lane is calculated with Equation 19-106.

𝑔𝑏
𝑐𝑏 = 𝑠𝑏
𝐶
48
𝑐𝑏 = (2,000) 120 = 800 bicycles/h
B. Compute Bicycle Delay

Bicycle delay is computed with Equation 19-107.
0.5 𝐶 (1 − 𝑔𝑏 /𝐶)2
𝑑𝑏 = 𝑣 𝑔
1 − min ( 𝑏𝑖𝑐 , 1.0) 𝑏
𝑐𝑏 𝐶
0.5(120)(1−48/120)2
𝑑𝑏 = 120 48 = 23.0 s/bicycle
1−min( ,1.0)×
800 120
Step 2: Determine Bicycle LOS Score for Intersection

As shown in Exhibit 31-86, the total width of the outside through lane,
bicycle lane, and paved shoulder Wt is 17 ft (= 12 + 5 + 0 + 0). There is no on-street
parking. The cross-section adjustment factor can then be calculated with
Equation 19-109.
𝐹𝑤 = 0.0153 𝑊𝑐𝑑 − 0.2144 𝑊𝑡
𝐹𝑤 = 0.0153(70) − 0.2144(17) = −2.57
The motor-vehicle volume adjustment factor must be calculated by using
Equation 19-110.
𝐹𝑣 = 0.0066
4 𝑁𝑡ℎ
85 + 924 + 77
𝐹𝑣 = 0.0066 = 0.90
4(2)
The bicycle LOS score can then be computed with Equation 19-108.
𝐼𝑏,𝑖𝑛𝑡 = 4.1324 + 𝐹𝑤 + 𝐹𝑣
𝐼𝑏,𝑖𝑛𝑡 = 4.1324 − 2.57 + 0.90 = 2.45

According to Exhibit 19-9, the bicycle lane will operate at LOS B through the
signalized intersection.
Discussion
The bicycle lane was found to operate at LOS B. The bicycle delay was found
to be 23.0 s/bicycle, which is low enough that most bicyclists are not likely to be
impatient. However, if the signal timing at the intersection were to be changed,
the bicycle delay would need to be computed again to verify that it does not rise
above 30 s/bicycle.

EXAMPLE PROBLEM 4: PEDESTRIAN DELAY WITH TWO-STAGE

CROSSING OF ONE INTERSECTION LEG
The Intersection
The pedestrian crosswalk of interest is on the east leg of an intersection. This
intersection is shown in Exhibit 31-87(a) where north is toward the top of the
figure. The pedestrian movement of interest travels north, from corner B to
corner A. A two-stage crossing is provided for this leg of the intersection. Signal
heads are provided for all pedestrian movements.
Exhibit 31-87 N
Example Problem 4:
Intersection Geometry and
Signal Phase Sequence
Corner D 6 Corner A
4 78
34
8
2
Corner C Corner B
(a) Intersection geometry
21 57 19 43
Φ1 Φ2 Φ3 Φ4
1 2 4P 4
2P 3
Φ5 Φ6 Φ8 Φ7
6P 7 78
6
5
8 8P+78
24 54 34 28
Time
(b) Signal phase sequence
The east–west street is the major street. The major street traffic signals
provide coordination for the through movements using a 140-s cycle length. The
minor movements are actuated. Rest-in-walk is not used for any phases. The
duration of each phase is shown in Exhibit 31-87(b). The Walk intervals are set at
5.0 s and they start at the same time as the associated phase (i.e., they do not lead
or lag the phase). The distance crossed for crosswalk section 8 is 40 ft. The
median on the major street (at the location of pedestrian storage) is 16 ft wide.

The Question
The analyst desires to estimate the delay to the northbound pedestrian
movement on the east leg of the intersection.
Computational Steps
The solution follows the steps of the “Crossing One Intersection Leg in Two
Stages” procedure. The first stage of the crossing occurs in crosswalk section 8. It
is served by phase 8 so Phase X is phase 8 (i.e., X = 8). The second stage of the
crossing occurs in crosswalk section 78. It is served by phases 7 and 8. Phase 7 is
shown to occur next (after 8) so Phase Y is phase 7 (i.e., Y = 7).
Step 1. Determine the Effective Walk Time

The subject phases are actuated and rest-in-walk is not enabled so the
effective walk time is determined by Equation 19-51.
𝑔Walk,𝑖 = Walk 𝑖 + 4.0 = 5.0 + 4.0 = 9.0 s
Step 2. Determine Crossing Time during First Stage

The local pedestrian population is about 30 percent elderly so an average
pedestrian crossing speed of 3.3 ft/s is used for the analysis. Equation 19-88 is
used to compute the time for pedestrians to travel the 56-ft distance (= 40 + 16)
from corner B to the far side of the median.
𝐿𝑋 56
𝑡𝑋 = = = 17.0 s
𝑆𝑝 3.3
Step 3. Determine the Start of the Walk Intervals

The relative start time of the Walk intervals for Phase X and Phase Y are
determined by inspection of Exhibit 31-87(b). For Phase X (i.e., phase 8), the
relative start time TWalk,X is 78 s (= 21 + 57). For Phase Y (i.e., phase 7), the relative
start time TWalk,Y is 112 s (= 21 + 57 + 34).
Step 4. Compute Delay for First-Stage Crossing

The delay for the first-stage crossing is computed using Equation 19-78.
2
(𝐶 − 𝑔Walk,𝑋 ) (140 − 9.0)2
𝑑𝑝,1 = = = 61.3 s/p
2𝐶 2 × 140
Step 5. Compute Delay for Second-Stage Crossing Given Arrival is during Don’t
Walk
The time between the Walk intervals for Phases X and Y is computed using
Equation 19-79.
𝑡𝑌𝑋 = Modulo(𝑇Walk,𝑌 − 𝑇Walk,𝑋 , 𝐶) = Modulo(112 − 78, 140) = 34.0 s
The waiting time on the median (for those pedestrians that reach the median
during a DON’T WALK indication) is computed using Equation 19-81.
𝑡 = Modulo(𝑡𝑌𝑋 − 𝑡𝑋 , 𝐶) = Modulo(34.0 − 17.0, 140) = 17.0 s

The delay incurred by pedestrians waiting on the median that arrived at the
first corner during a DON’T WALK indication d2,DW1 is computed using Equation 19-80.
𝑡 if 𝑡 < 𝐶 − 𝑔Walk,𝑌
𝑑2,𝐷𝑊1 = {
0 if 𝑡 ≥ 𝐶 − 𝑔Walk,𝑌
Because the value of t is less than C – gWalk,Y (i.e., 17 < 140 – 9), the delay d2,DW1
is equal to t, or 17.0 s/p.
Step 6. Compute Delay for Second-Stage Crossing Given Arrival is during Walk
Because t is greater than gWalk,X (i.e., 17 > 9), Equation 19-84 is used to
compute the delay on the median for stage 2.
𝑡 − 0.5 𝑔Walk,𝑋 if (𝑡 + 𝑔Walk,𝑌 ) < 𝐶
2
0.5 𝑏 + 𝑏 (𝑡 − 𝑔Walk,𝑋 )
𝑑2,𝑊1 = if 𝐶 ≤ (𝑡 + 𝑔Walk,𝑌 ) ≤ (𝐶 + 𝑔Walk,𝑋 )
𝑔Walk,𝑋
{0 if (𝑡 + 𝑔Walk,𝑌 ) > (𝐶 + 𝑔Walk,𝑋 )
Because t + gWalk,Y is less than C (i.e., 17 + 9 < 140), the first part of this
equation is used to compute the desired delay value.
𝑑2,𝑊1 = 𝑡 − 0.5 𝑔Walk,𝑋 = 17 − 0.5 × 9 = 12.5 s/p
Step 7. Compute Delay for Two-Stage Crossing

The proportion of arrivals during the DON’T WALK indication at the corner
PDW1 is computed using Equation 19-87.
(𝐶 − 𝑔Walk,X ) (140 − 9)
𝑃𝐷𝑊1 = = = 0.936
𝐶 140
Equation 19-86 is then used to compute the delay for the two-stage crossing.
𝑑𝑝 = 𝑑𝑝,1 + [𝑑2,𝐷𝑊1 𝑃𝐷𝑊1 + 𝑑2,𝑊1 (1 − 𝑃𝐷𝑊1)]
𝑑𝑝 = 61.3 + [(17.0 × 0.936) + 12.5(1 − 0.936)]
𝑑𝑝 = 78 s/p
Discussion
Unlike a one-stage crossing, the delay crossing in the one direction will
generally be different than the delay crossing in the opposite direction. In this
case, using the same inputs given above, plus a crosswalk length of 52 ft for
crosswalk section 78, the average delay traveling from Corner A to Corner B is
calculated to be 147 s.
EXAMPLE PROBLEM 5: PEDESTRIAN DELAY WITH TWO-STAGE

CROSSING OF TWO INTERSECTION LEGS
The Intersection
The pedestrian crosswalks of interest are on the south and west legs of an
intersection. This intersection is shown in Exhibit 31-88(a) where north is toward
the top of the figure. The pedestrian movement of interest travels clockwise from
corner B to corner C and then to corner D. Signal heads are provided for all
pedestrian movements.

N Exhibit 31-88
Example Problem 5:
Intersection Geometry and
Signal Phase Sequence
Corner D 6 Corner A
4 8
Corner C 2 Corner B
(a) Intersection geometry
11 40 10 29
Φ1 Φ2 Φ3 Φ4
1 2 4P 4
2P 3
Φ5 Φ6 Φ7 Φ8
6P 7
6
5
8 8P
13 38 8 31
Time
(b) Signal phase sequence
The east-west street is the major street. The major street traffic signals
provide coordination for the through movements using a 90-s cycle length. The
minor movements are actuated. Rest-in-walk is not used for any phases. The
duration of each phase is shown in Exhibit 31-88(b). The Walk intervals are set at
5.0 s and they start at the same time as the associated phase (i.e., they do not lead
or lag the phase). The distance crossed for crosswalk 2 is 38 ft.
The Question
The analyst desires to estimate the delay to pedestrians traveling from corner
B to corner D by crossing the south leg and then the west leg.

Computational Steps
The solution follows the steps of the “Crossing Two Intersection Legs in Two
Stages” procedure. The first stage of the crossing occurs in crosswalk 2. It is
served by phase 2 so Phase X is phase 2 (i.e., X = 2). The second stage of the
crossing occurs in crosswalk 4. It is served by phase 4 so Phase Y is phase 4 (i.e.,
Y = 4). Finally, if the pedestrian decided to cross in the other direction around the
intersection (i.e., counterclockwise), the first phase to serve this travel direction is
phase 8 so Phase Z is phase 8 (i.e., Z = 8).
Step 1. Determine the Effective Walk Time

The subject phases are actuated and rest-in-walk is not enabled so the
effective walk time for Phase X and Phase Z is determined by Equation 19-51.
𝑔Walk,𝑖 = Walk 𝑖 + 4.0 = 5.0 + 4.0 = 9.0 s
Step 2. Determine Crossing Time during First Phase

The local pedestrian population is about 30 percent elderly so an average
pedestrian crossing speed of 3.3 ft/s is used for the analysis. Equation 19-88 is
used to compute the time for pedestrians to travel the 38-ft distance from corner
B to corner C.
𝐿𝑋 38
𝑡𝑋 = = = 11.5 s
𝑆𝑝 3.3
Step 3. Determine the Start of the Walk Intervals

The relative start time of the Walk intervals for Phase X, Phase Y, and Phase
Z are determined by inspection of Exhibit 31-88(b). For Phase X (i.e., phase 2), the
relative start time TWalk,X is 11 s. For Phase Y (i.e., phase 4), the relative start time
TWalk,Y is 61 s (= 11 + 40 + 10). For Phase Z (i.e., phase 8), the relative start time
TWalk,Z is 59 s (= 11 + 40 + 8).
Step 4. Compute Delay for First-Stage Crossing

The end of the effective walk time for Phase X is computed using Equation
19-88.
𝑇𝑋 = Modulo(𝑇Walk,𝑋 + 𝑔Walk,𝑋 , 𝐶) = Modulo(11 + 9, 90) = 20 s
Similarly, the end of the effective walk time for Phase Z is computed using
Equation 19-89.
𝑇𝑍 = Modulo(𝑇Walk,𝑍 + 𝑔Walk,𝑍 , 𝐶) = Modulo(59 + 9, 90) = 68 s
The time between the end of effective walk time for Phase Z and the start of
effective walk time for Phase X is computed using Equation 19-91.
𝑡𝑋𝑍 = Modulo(𝑇𝑋 − 𝑇𝑍 , 𝐶) = Modulo(20 − 68, 90) = 20 − 68 + 90 = 42 s
Finally, the delay for the first stage crossing is computed using Equation 19-90.
2
(𝑡𝑋𝑍 − 𝑔Walk,𝑋 ) (42 − 9)2
𝑑𝑝,1 = = = 13.0 s/p
2 𝑡𝑋𝑍 2 × 42

Step 5. Compute Delay for Entire Diagonal Crossing

Equation 19-93 is used to determine the time between arrival at corner B and
departure from the second corner.
𝑇𝑋 + 𝑇𝑍
𝑇Walk,𝑌 − if 𝑇Walk,𝑌 ≥ 𝑇𝑋 ≥ 𝑇𝑍
2
𝑇𝑋 + 𝑇𝑍 − 𝐶
𝑡𝑑 = 𝑇Walk,𝑌 − if 𝑇𝑋 < 𝑇𝑍
2
𝑇𝑋 + 𝑇𝑍
{𝑇Walk,𝑌 − 2
+𝐶 if 𝑇𝑋 ≥ 𝑇𝑍 ≥ 𝑇Walk,𝑌
Because the relative end time of the effective walk period for Phase X TX is
less than that for Phase Z TZ (i.e., 20 < 68), the second part of Equation 19-93 is
used to compute the desired time interval td.
𝑇𝑋 + 𝑇𝑍 − 𝐶 20 + 68 − 90
𝑡𝑑 = 𝑇Walk,𝑌 − = 61 − = 62 s
2 2
Finally, the diagonal crossing delay is computed using Equation 19-92.
𝑑𝑝 = 𝑡𝑑 − 𝑡𝑋 = 62.0 − 11.5 = 50.5 s/p
Step 6. Compute Delay for Second Stage Crossing

The delay for the second-stage crossing dp,2 is computed using Equation 19-94.
𝑑𝑝,2 = 𝑑𝑝 − 𝑑𝑝,1 = 50.5 − 13.0 = 37.5 s/p
Discussion
Reasonably good signal compliance can be expected for the first-stage
crossing, based on the average delay of 13.0 s. However, the second-stage delay
exceeds 30 s and some pedestrians may not comply with the signal indications.

11. REFERENCES
Many of these references are 1. Zegeer, C., K. Opiela, and M. Cynecki. Pedestrian Signalization Alternatives.
available in the Technical
Reference Library in Volume 4. Report FHWA/RD-83/102. Federal Highway Administration, Washington,
D.C., 1983.
2. Lieberman, E. B. Determining the Lateral Deployment of Traffic on an
Approach to an Intersection. In Transportation Research Record 772,
D.C., 1980, pp. 1–5.
3. Bonneson, J. Lane Volume and Saturation Flow Rate for a Multilane
Intersection Approach. Journal of Transportation Engineering, Vol. 124, No. 3,
1998, pp. 240–246.
4. Bonneson, J. A., and P. T. McCoy. NCHRP Report 395: Capacity and
Operational Effects of Midblock Left-Turn Lanes. Transportation Research Board,
National Research Council, Washington, D.C., 1997.
5. Teply, S. Accuracy of Delay Surveys at Signalized Intersections. In
Transportation Research Record 1225, Transportation Research Board, National
Research Council, Washington, D.C., 1989, pp. 24–32.
6. Transportation Research Circular 212: Interim Materials on Highway Capacity.
D.C., 1980.
7. Pline, J. L. (ed.). Traffic Engineering Handbook, 5th ed. ITE, Washington, D.C.,
1999.
8. Reilly, W. R., and C. C. Gardner. Technique for Measuring Delay at
Intersections. In Transportation Research Record 644, Transportation Research
Board, National Research Council, Washington, D.C., 1977, pp. 1–7.
9. Powell, J. L. Field Measurement of Signalized Intersection Delay for 1997
Update of the Highway Capacity Manual. In Transportation Research Record
1646, Transportation Research Board, National Research Council,
Washington, D.C., 1998, pp. 79–86.
11. Adebisi, A., Y. Guo, B. Schroeder, J. Ma, B. Cesme, A. Bibeka, and A. Morgan.
Highway Capacity Manual (HCM) Capacity Adjustment Factor (CAF)
Development for Connected and Automated Traffic at Signalized
Intersections. Presented at the Transportation Research Board 100th Annual
Meeting, 2021.
Oct. 2013.
References Chapter 31/Signalized Intersections: Supplemental

13. Krechmer, D., K. Blizzard, M. G. Cheung, R. Campbell, V. Alexiadis, J. Hyde,

June 2016.
14. Davis, S. C., and R. G. Boundy. Transportation Energy Data Book, Edition 37.
Tenn., Aug. 2019.
2019.
Chapter 31/Signalized Intersections: Supplemental References

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 32
STOP-CONTROLLED INTERSECTIONS: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION .................................................................................................. 32-1
2. TWSC POTENTIAL CAPACITY......................................................................... 32-2
3. TWSC EXAMPLE PROBLEMS ............................................................................ 32-4

TWSC Example Problem 1: TWSC at an Intersection with Three Legs ...... 32-4
TWSC Example Problem 2: Pedestrian Crossing at a
TWSC Intersection ......................................................................................32-10
TWSC Example Problem 3: Flared Approaches and Median Storage .......32-15
TWSC Example Problem 4: TWSC Intersection Within A Signalized
Urban Street Segment ................................................................................32-29
TWSC Example Problem 5: Six-Lane Street with U-Turns and
Pedestrians ..................................................................................................32-39
4. AWSC SUPPLEMENTAL ANALYSIS FOR THREE-LANE

APPROACHES ..................................................................................................... 32-48
5. AWSC EXAMPLE PROBLEMS ......................................................................... 32-57

AWSC Example Problem 1: Single-Lane, Three-Leg Intersection ..............32-57
AWSC Example Problem 2: Multilane, Four-Leg Intersection ....................32-62
Chapter 32/STOP-Controlled Intersections: Supplemental Contents

LIST OF EXHIBITS
Exhibit 32-1 Potential Capacity cp,x for Two-Lane Major Streets ..........................32-2
Exhibit 32-2 Potential Capacity cp,x for Four-Lane Major Streets ..........................32-3
Exhibit 32-3 Potential Capacity cp,x for Six-Lane Major Streets .............................32-3
Exhibit 32-4 TWSC Example Problems ....................................................................32-4
Exhibit 32-5 TWSC Example Problem 1: 15-min Volumes and Lane
Configurations .....................................................................................................32-4
Exhibit 32-6 TWSC Example Problem 1: Movement Numbers and
Calculation of Peak 15-min Flow Rates ............................................................32-5
Exhibit 32-7 TWSC Example Problem 2: Pedestrian Satisfaction Results
for Scenarios B and C ........................................................................................32-15
Exhibit 32-8 TWSC Example Problem 3: 15-min Volumes and Lane
Configurations ...................................................................................................32-15
Calculation of Peak 15-min Flow Rates ..........................................................32-16
Exhibit 32-10 TWSC Example Problem 4: TWSC Intersection Within a
Signalized Urban Street Segment ....................................................................32-29
Exhibit 32-11 TWSC Example Problem 4: 15-min Flow Rates and Lane
Configurations ...................................................................................................32-29
Exhibit 32-12 TWSC Example Problem 4: Movement-Based Access Point
Output (from Chapter 30, Example Problem 1) ............................................32-30
Exhibit 32-14 TWSC Example Problem 5: Volumes and Lane
Configurations ...................................................................................................32-40
Exhibit 32-16 Probability of Degree-of-Conflict Case: Multilane AWSC
Intersections (Three-Lane Approaches, by Lane) (Cases 1–49) ...................32-48
Exhibit 32-17 AWSC Example Problems ................................................................32-57
Exhibit 32-18 AWSC Example Problem 1: Volumes and Lane Configurations ..32-57
Exhibit 32-19 AWSC Example Problem 1: Applicable Degree-of-Conflict
Cases ....................................................................................................................32-59
Exhibit 32-20 AWSC Example Problem 1: Eastbound Saturation Headways ....32-60
Exhibit 32-21 AWSC Example Problem 1: Convergence Check ............................ 32-61
Exhibit 32-22 AWSC Example Problem 2: 15-min Volumes and Lane
Configurations ....................................................................................................32-63
Exhibit 32-23 AWSC Example Problem 2: 15-min Volumes Converted to
Hourly Flow Rates..............................................................................................32-63
Exhibit 32-24 AWSC Example Problem 2: Convergence Check ...........................32-67
Contents Chapter 32/STOP-Controlled Intersections: Supplemental

INTRODUCTION
Chapter 32 is the supplemental chapter for Chapter 20, Two-Way STOP- VOLUME 4: APPLICATIONS
GUIDE
Controlled Intersections, and Chapter 21, All-Way STOP-Controlled Intersections, 25. Freeway Facilities:
which are found in Volume 3 of the Highway Capacity Manual. This chapter Supplemental
provides supplemental material on (a) determining the potential capacity of two- Segments: Supplemental
way STOP-controlled (TWSC) intersections and (b) identifying the 512 27. Freeway Weaving:
Supplemental
combinations of degree-of-conflict cases for all-way STOP-controlled (AWSC) 28. Freeway Merges and
intersections with three-lane approaches. The chapter also provides example Diverges: Supplemental
problems demonstrating the application of the TWSC and AWSC methodologies. Supplemental
Supplemental
Supplemental
32. STOP-Controlled
Intersections:
Supplemental
33. Roundabouts:
Supplemental
Supplemental
Chapter 32/STOP-Controlled Intersections: Supplemental Introduction

TWSC POTENTIAL CAPACITY
The gap acceptance model to estimate potential capacity (presented in

Chapter 20, Equation 20-18) can be plotted for each of the non–Rank 1
movements by using values of critical headway and follow-up headway from
Chapter 20 (Exhibit 20-17 and Exhibit 20-18, respectively). These graphs are
presented in Exhibit 32-1, Exhibit 32-2, and Exhibit 32-3 for a major street with
two lanes, four lanes, and six lanes, respectively. The potential capacity is
expressed as vehicles per hour. The exhibits indicate the potential capacity is a
function of the conflicting flow rate vc,x expressed as an hourly rate, as well as the
type of minor-street movement.
Exhibit 32-1
Potential Capacity cp,x for
Two-Lane Major Streets
Note: LT = left turn, RT = right turn, and TH = through.
TWSC Potential Capacity Chapter 32/STOP-Controlled Intersections: Supplemental

Exhibit 32-2
Four-Lane Major Streets
Note: LT = left turn, U = U-turn, RT = right turn, and TH = through.
Exhibit 32-3
Six-Lane Major Streets
Note: LT = left turn, U = U-turn, RT = right turn, and TH = through.
Chapter 32/STOP-Controlled Intersections: Supplemental TWSC Potential Capacity

TWSC EXAMPLE PROBLEMS
This section provides example problems for use of the TWSC methodology.
Exhibit 32-4 provides an overview of these problems. The examples focus on the
operational analysis level. The planning and preliminary engineering analysis
level is identical to the operations analysis level in terms of the calculations,
except that default values are used when available.
Exhibit 32-4 Problem Analysis

TWSC Example Problems Number Description Level
1 TWSC at an intersection with three legs Operational
2 Pedestrian crossing at a TWSC intersection Operational
3 TWSC intersection with flared approaches and median storage Operational
4 TWSC intersection within a signalized urban street segment Operational
5 TWSC intersection on a six-lane street with U-turns and pedestrians Operational
TWSC EXAMPLE PROBLEM 1: TWSC AT AN INTERSECTION

WITH THREE LEGS
The Facts
characteristics of this location:
• T-intersection,
• Major street with one lane in each direction,
• Minor street with one lane in each direction and STOP-controlled on the
minor-street approach,
• Level grade on all approaches,
• Percentage heavy vehicles on all approaches = 10%,
• No other unique geometric considerations or upstream signal
considerations,
• No pedestrians,
• Length of analysis period = 0.25 h, and
• Volumes during the peak 15-min period and lane configurations as
Exhibit 32-5
TWSC Example Problem 1:
15-min Volumes and Lane
Configurations
TWSC Example Problems Chapter 32/STOP-Controlled Intersections: Supplemental

Comments
Steps 1 and 2: Convert Movement Demand Volumes to Flow Rates and

Label Movement Priorities
Because peak 15-min volumes have been provided, each volume is
multiplied by four to determine a peak 15-min flow rate (in vehicle per hour) for
each movement. These values, along with the associated movement numbers, are
Exhibit 32-6
Movement Numbers and
Calculation of Peak 15-min
Flow Rates
Step 3: Compute Conflicting Flow Rates

The conflicting flow rates for each minor movement at the intersection are
computed according to Equation 20-3, Exhibit 20-8, Equation 20-4, Exhibit 20-10,
Equation 20-12, Equation 20-14, and Exhibit 20-16. The conflicting flow for the
major-street left-turn vc,4 is
𝑣𝑐,4 = 𝑓𝑐,4,2 𝑣2 + 𝑓𝑐,4,3 𝑣3 + 𝑓𝑐,4,15 𝑣15
𝑣𝑐,4 = 1(240) + 1(40) + 1(0) = 280 veh/h
The conflicting flow for the minor-street right-turn movement vc,9 is
𝑣𝑐,9 = 𝑓𝑐,9,2 𝑣2 + 𝑓𝑐,9,3 𝑣3 + 𝑓𝑐,9,4𝑈 𝑣4𝑈 + 𝑓𝑐,9,14 𝑣14 + 𝑓𝑐,9,15 𝑣15
𝑣𝑐,9 = 1(240) + 0.5(40) + 0(0) + 1(0) + 1(0) = 260 veh/h
Finally, the conflicting flow for the minor-street left-turn movement vc,7 is
computed. Because two-stage gap acceptance is not present at this intersection,
the conflicting flow rates shown in Stage I (Equation 20-12) and Stage II
(Equation 20-14), with coefficients from Exhibit 20-16, are added together and
considered as one conflicting flow rate. The conflicting flow for vc,7 is computed
as follows:
𝑣𝑐,7 = [𝑓𝑐,7,1 𝑣1 + 𝑓𝑐,7,1𝑈 𝑣1𝑈 + 𝑓𝑐,7,2 𝑣2 + 𝑓𝑐,7,3 𝑣3 + 𝑓𝑐,7,15 𝑣15 ]
+ [𝑓𝑐,7,4 𝑣4 + 𝑓𝑐,7,4𝑈 𝑣4𝑈 + 𝑓𝑐,7,5 𝑣5 + 𝑓𝑐,7,6 𝑣6 + 𝑓𝑐,7,13 𝑣13 ]
𝑣𝑐,7 = [2(0) + 2(0) + 1(240) + 0.5(40) + 1(0)]
+ [2(160) + 2(0) + 1(300) + 0.5(0) + 1(0)] = 880 veh/h
Step 4: Determine Critical Headways and Follow-Up Headways

The critical headway for each minor movement is computed beginning with
the base critical headway given in Exhibit 20-17. The base critical headway for
each movement is then adjusted according to Equation 20-17. The critical
headway for the major-street left-turn movement tc,4 is computed as follows:
𝑡𝑐,4 = 𝑡𝑐,base + 𝑡𝑐,𝐻𝑉 𝑃𝐻𝑉 + 𝑡𝑐,𝐺 𝐺 − 𝑡3,𝐿𝑇
Chapter 32/STOP-Controlled Intersections: Supplemental TWSC Example Problems

𝑡𝑐,4 = 4.1 + 1.0(0.1) + 0(0) − 0 = 4.2 s

Similarly, the critical headway for the minor-street right-turn movement tc,9 is
𝑡𝑐,9 = 6.2 + 1.0(0.1) + 0.1(0) − 0 = 6.3 s
Finally, the critical headway for the minor-street left-turn movement tc,7 is
𝑡𝑐,7 = 7.1 + 1.0(0.1) + 0.2(0) − 0.7 = 6.5 s
The follow-up headway for each minor movement is computed beginning
with the base follow-up headway given in Exhibit 20-18. The base follow-up
headway for each movement is then adjusted according to Equation 20-17. The
follow-up headway for the major-street left-turn movement tf,4 is computed as
follows:
𝑡𝑓,4 = 𝑡𝑓,𝑏𝑎𝑠𝑒 + 𝑡𝑓,𝐻𝑉 𝑃𝐻𝑉
𝑡𝑓,4 = 2.2 + 0.9(0.1) = 2.29 s
Similarly, the follow-up headway for the minor-street right-turn movement
tf,9 is
𝑡𝑓,9 = 3.3 + 0.9(0.1) = 3.39 s
Finally, the follow-up headway for the minor-street left-turn movement tf,7 is
𝑡𝑓,7 = 3.5 + 0.9(0.1) = 3.59 s
Step 5: Compute Potential Capacities

The computation of a potential capacity for each movement provides the
analyst with a definition of capacity under the assumed base conditions. The
potential capacity will be adjusted in later steps to estimate the movement
capacity for each movement. The potential capacity for each movement is a
function of the conflicting flow rate, critical headway, and follow-up headway
computed in the previous steps. The potential capacity for the major-street left-
turn movement cp,4 is computed as follows from Equation 20-18:
𝑒 −𝑣𝑐,4 𝑡𝑐,4 /3,600
𝑐𝑝,4 = 𝑣𝑐,4
1 − 𝑒 −𝑣𝑐,4 𝑡𝑓,4 /3,600
𝑒 −(280)(4.2)/3,600
𝑐𝑝,4 = 280 = 1,238 veh/h
1 − 𝑒 −(280)(2.29)/3,600
Similarly, the potential capacity for the minor-street right-turn movement cp,9
is computed as follows:
𝑒 −(260)(6.3)/3,600
𝑐𝑝,9 = 260 = 760 veh/h
1 − 𝑒 −(260)(3.39)/3,600
Finally, the potential capacity for the minor-street left-turn movement cp,7 is
𝑒 −(880)(6.5)/3,600
𝑐𝑝,7 = 880 = 308 veh/h
1 − 𝑒 −(880)(3.59)/3,600
There are no upstream signals, so the adjustments for upstream signals are
ignored.

Step 6: Compute Rank 1 Movement Capacities

There are no pedestrians at the intersection; therefore, all pedestrian
impedance factors are equal to 1.0, and this step can be ignored.

The movement capacity for the major-street left-turn movement (Rank 2) cm,4
is computed as follows from Equation 20-22:
𝑐𝑚,4 = 𝑐𝑝,4 = 1,238 veh/h
Similarly, the movement capacity for the minor-street right-turn movement
(Rank 2) cm,9 is computed with Equation 20-23:
𝑐𝑚,9 = 𝑐𝑝,9 = 760 veh/h

The computation of vehicle impedance effects accounts for the reduction in
potential capacity due to the impacts of the congestion of a high-priority
movement on lower-priority movements.
Major-street movements of Rank 1 and Rank 2 are assumed to be unimpeded
by other vehicular movements. Minor-street movements of Rank 3 can be
impeded by major-street left-turn movements due to a major-street left-turning
vehicle waiting for an acceptable gap at the same time as vehicles of Rank 3. The
magnitude of this impedance depends on the probability that major-street left-
turning vehicles will be waiting for an acceptable gap at the same time as
vehicles of Rank 3. In this example, only the minor-street left-turn movement is
defined as a Rank 3 movement. Therefore, the probability of the major-street left-
turn movement operating in a queue-free state (p0,4) is computed from Equation
20-28:
𝑣4 160
𝑝0,4 = 1 − =1− = 0.871
𝑐𝑚,4 1,238
The movement capacity for the minor-street left-turn movement (Rank 3) cm,7
is found by first computing a capacity adjustment factor that accounts for the
impeding effects of higher-ranked movements. The capacity adjustment factor
for the minor-street left-turn movement f7 is computed with Equation 20-32:
𝑓7 = ∏ 𝑝0,𝑗 = 0.871
𝑗
The movement capacity for the minor-street left-turn movement (Rank 3) cm,7
is computed with Equation 20-33:
𝑐𝑚,7 = 𝑐𝑝,7 × 𝑓7 = 308(0.871) = 268 veh/h

There are no Rank 4 movements in this example problem, so this step does
not apply.

Step 10: Compute Capacity Adjustment Factors

In this example, the minor-street approach is a single lane shared by right-
turn and left-turn movements; therefore, the capacity of these two movements
must be adjusted to compute an approach capacity based on shared-lane effects.
The shared-lane capacity for the northbound minor-street approach cSH,NB is
computed from Equation 20-46:
∑𝑦 𝑣𝑦 𝑣7 + 𝑣9 40 + 120
𝑐𝑆𝐻,𝑁𝐵 = 𝑣𝑦 = 𝑣7 𝑣9 = 40 120 = 521 veh/h
∑𝑦 +
𝑐𝑚,𝑦 𝑐𝑚,7 𝑐𝑚,9 268 + 760
No other adjustments apply.
Step 11: Compute Control Delay

The control delay computation for any movement includes initial
deceleration delay, queue move-up time, stopped delay, and final acceleration
delay.
Step 11a: Compute Control Delay to Rank 2 Through Rank 4 Movements

The control delay for the major-street left-turn movement (Rank 2) d4 is
computed with Equation 20-61:
3,600 𝑣
3,600 𝑣𝑥 𝑣
2 (𝑐 ) (𝑐 𝑥 )
− 1 + √(
𝑥 𝑚,𝑥 𝑚,𝑥
𝑑= + 900𝑇 − 1) + +5
𝑐𝑚,𝑥 𝑐𝑚,𝑥 𝑐𝑚,𝑥 450𝑇
[ ]
3,600 160
3,600 160 160 2 (1,238) (1,238)
𝑑4 = + 900(0.25) √
−1+ ( − 1) + +5
1,238 1,238 1,238 450(0.25)
[ ]
𝑑4 = 8.3 s
On the basis of Exhibit 20-2, the westbound left-turn movement is assigned
level of service (LOS) A.
The control delay for the minor-street right-turn and left-turn movements is
computed by using the same formula; however, one significant difference from
the major-street left-turn computation of control delay is that these movements
share the same lane. Therefore, the control delay is computed for the approach as
a whole, and the shared-lane volume and shared-lane capacity must be used as
follows:
2
3,600 160
3,600 160 160 ( 521 ) (521)
𝑑𝑆𝐻,𝑁𝐵 = + 900(0.25) √
−1+ ( − 1) + +5
521 521 521 450(0.25)
[ ]
𝑑𝑆𝐻,𝑁𝐵 = 14.9 s
On the basis of Exhibit 20-2, the northbound approach is assigned LOS B.

Step 11b: Compute Control Delay to Rank 1 Movements

This step is not applicable as the westbound major-street through movement
v5 and westbound major-street left-turn movement v4 have exclusive lanes at this
intersection. It is assumed the eastbound through movement v2 and eastbound
major-street right-turn movement v3 do not incur any delay at this intersection.
Step 12: Compute Approach and Intersection Control Delay

The control delays to all vehicles on the eastbound approach are assumed to
be negligible as described in Step 11b. The control delay for the westbound
approach dA,WB is computed with Equation 20-64:
∑𝑖 𝑑𝑖,𝑥 𝑣𝑖,𝑥
𝑑𝐴,𝑥 =
∑𝑖 𝑣𝑖,𝑥
0(0) + 0(300) + 8.3(160)
𝑑𝐴,𝑊𝐵 = = 2.9 s
0 + 300 + 160
It is assumed the westbound through movement incurs no control delay at
this intersection. The control delay for the northbound approach was computed
in Step 11a as dSH,NB.
The intersection control delay dI is computed from Equation 20-65:
𝑑𝐴,𝐸𝐵 𝑣𝐴,𝐸𝐵 + 𝑑𝐴,𝑊𝐵 𝑣𝐴,𝑊𝐵 + 𝑑𝐴,𝑁𝐵 𝑣𝐴,𝑁𝐵
𝑑𝐼 =
𝑣𝐴,𝐸𝐵 + 𝑣𝐴,𝑊𝐵 + 𝑣𝐴,𝑁𝐵
0(280) + 2.9(460) + 14.9(160)
𝑑𝐼 = = 4.1 s
280 + 460 + 160
As noted in Chapter 20, neither major-street approach LOS nor intersection
LOS is defined.
Step 13: Compute 95th Percentile Queue Lengths

The 95th percentile queue length for the major-street westbound left-turn
movement Q95,4 is computed from Equation 20-66:
3,600 𝑣
𝑣4
2 ( )( 𝑥 ) 𝑐
√ 𝑣4 𝑐𝑚,4 𝑐𝑚,4 𝑚,4
𝑄95,4 ≈ 900𝑇 −1+ ( − 1) + ( )
𝑐𝑚,4 𝑐𝑚,4 150𝑇 3,600
[ ]
3,600 160
160 160 2 (1,238) (1,238) 1,238
𝑄95,4 ≈ 900(0.25) − 1 + √( − 1) + ( )
1,238 1,238 150(0.25) 3,600
[ ]
𝑄95,4 = 0.4 veh
The result of 0.4 vehicles for the 95th percentile queue indicates a queue of
more than one vehicle will occur very infrequently for the major-street left-turn
movement.
The 95th percentile queue length for the northbound approach is computed
by using the same formula. Similar to the control delay computation, the shared-
lane volume and shared-lane capacity must be used as shown:

2
3,600 160
160 160 ( )( )
𝑄95,𝑁𝐵 ≈ 900(0.25) √
−1+ ( − 1) + 521 521 ( 521 )
521 521 150(0.25) 3,600
[ ]
𝑄95,𝑁𝐵 = 1.3 veh
The result suggests that a queue of more than one vehicle will occur only
occasionally for the northbound approach.
Discussion
Overall, the results indicate this three-leg TWSC intersection will operate
well with brief delays and little queuing for all minor movements.
TWSC EXAMPLE PROBLEM 2: PEDESTRIAN CROSSING AT A TWSC

INTERSECTION
Calculate the pedestrian LOS of a pedestrian crossing of a major street at a
TWSC intersection under the following circumstances:
• Scenario A: unmarked crosswalk, no median refuge island;
• Scenario B: marked crosswalk, median refuge island; and
• Scenario C: marked crosswalk, median refuge island, rectangular rapid-
flashing beacons (RRFBs).
The Facts
• Four-lane major street;
• 1,700 peak hour vehicles, bidirectional;
• K-factor = 0.08;
• Crosswalk length without median = 46 ft;
• Crosswalk length with median = 20 ft each side of median;
• Observed pedestrian walking speed = 4.0 ft/s;
• Observed pedestrian start-up and end clearance time = 1.0 s; and
• No pedestrian platooning.
Comments
In addition to the input data listed above, information is required on motor
vehicle yield rates under the various scenarios. On the basis of an engineering
study of similar intersections in the vicinity, it is determined that average motor
vehicle yield rates are 0% with unmarked crosswalks, 50% with marked
crosswalks and median islands, and 80% with marked crosswalks, median
islands, and RRFBs.

Step 1: Identify Two-Stage Crossings

Scenario A does not have two-stage pedestrian crossings, as no median
refuge is available. Analysis for Scenarios B and C should assume two-stage
crossings. Thus, analysis for Scenarios B and C will combine two pedestrian
crossings of 20 ft each to determine the total delay.
Step 2: Determine Critical Headway

Because there is no pedestrian platooning, the critical headway tc is
determined with Equation 20-76:
𝐿
𝑡𝑐 = + 𝑡𝑠
𝑆𝑝
Scenario A: tc = (46 ft)/(4.0 ft/s) + 1.0 s = 12.5 s
Scenario B: tc = (20 ft)/(4.0 ft/s) + 1.0 s = 6.0 s
Scenario C: tc = (20 ft)/(4.0 ft/s) + 1.0 s = 6.0 s
Step 3: Estimate Probability of a Delayed Crossing

Equation 20-80 and Equation 20-81 are used to calculate Pb, the probability of
a blocked lane, and Pd, the probability of a delayed crossing, respectively. In the
case of Scenario A, the crossing consists of four lanes. Scenarios B and C have
only two lanes, given the two-stage crossing opportunity.
For the single-stage crossing, v is (1,700 veh/h)/(3,600 s/h) = 0.472 veh/s.
For the two-stage crossing, without any information on directional flows,
one-half the volume is used, and v is therefore (850 veh/h)/(3,600 s/h) = 0.236
veh/s.
Scenario A:
−𝑡𝑐,𝐺 𝑣
𝑃𝑏 = 1 − 𝑒 𝑁𝐿
𝑃𝑑 = 1 − (1 − 𝑃𝑏 )𝑁𝐿
−12.5(0.472)
𝑃𝑏 = 1 − 𝑒 4 = 0.771
4
𝑃𝑑 = 1 − (1 − 0.771) = 0.997
Scenarios B and C:
−6.0(0.236)
𝑃𝑏 = 1 − 𝑒 2 = 0.508
2
𝑃𝑑 = 1 − (1 − 0.508) = 0.758
Step 4: Calculate Average Delay to Wait for Adequate Gap

Average gap delay dg and average gap delay when delay is nonzero dgd are
calculated by Equation 20-82 and Equation 20-83, respectively.
Scenario A:
1 𝑣𝑡
𝑑𝑔 = (𝑒 𝑐,𝐺 − 𝑣𝑡𝑐,𝐺 − 1)
𝑣

1
𝑑𝑔 = (𝑒 0.472(12.5) − 0.472(12.5) − 1) = 761 s
0.472
𝑑𝑔 761
𝑑𝑔𝑑 = = = 763 s
𝑃𝑑 0.997
Scenarios B and C:
1
𝑑𝑔 = (𝑒 0.236(6.0) − 0.236(6.0) − 1) = 7.2 s
0.236
7.2
𝑑𝑔𝑑 = = 9.5 s
0.758
Step 5: Calculate Average Pedestrian Delay for the Crossing Stage

Under Scenario A, the motorist yielding rate is 0%. Therefore, there is no
reduction in delay due to yielding vehicles, and average delay is the same as that
shown in Step 4.
𝑑𝑝,1 = 𝑑𝑔𝑑 = 761 s
Under Scenario B, the motorist yielding rate is 50% and the reduced delay
due to yielding vehicles is determined using the process described in Step 5. To
start, the average headway of those headways less than the group critical
headway h is determined using Equation 20-85.
1/𝑣 − (𝑡𝑐,𝐺 + 1/𝑣) exp[−𝑣 𝑡𝑐,𝐺 ]
ℎ=
1 − exp[−𝑣 𝑡𝑐,𝐺 ]
1/0.236 − (6.0 + 1/0.236) exp[−(0.236)(6.0)]
ℎ=
1 − exp[−(0.236)(6.0)]
ℎ = 2.3 s
The average number of potential yielding events before an adequate gap is
available n is then
𝑑𝑔𝑑 9.5
𝑛 = int ( ) = int ( ) = 4
ℎ 2.3
The two-lane crossings require the use of Equation 20-89 to determine P(Yi).
𝑖−1
(2𝑃𝑏 [1 − 𝑃𝑏 ]𝑀𝑦 ) + (𝑃𝑏2 𝑀𝑦2 )
𝑃(𝑌𝑖 ) = [𝑃𝑑 − ∑ 𝑃(𝑌𝑗 )] [ ]
𝑃𝑑
𝑗=0
𝑃(𝑌0 ) = 0
(2[0.508][1 − 0.508][0.50]) + (0.50820.502 )
𝑃(𝑌1 ) = [0.758 − 0] [ ] = 0.314
0.758
(2[0.508][1 − 0.508][0.50]) + (0.5082 0.502 )
𝑃(𝑌2 ) = [0.758 − 0.314] [ ] = 0.184
0.758
(2[0.508][1 − 0.508][0.50]) + (0.5082 0.502 )
𝑃(𝑌3 ) = [0.758 − 0.498] [ ] = 0.108
0.758
(2[0.61][1 − 0.61][0.5]) + (0.612 0.502 )
𝑃(𝑌4 ) = [0.758 − 0.606] [ ] = 0.063
0.85

The results of Equation 20-89 are substituted into Equation 20-84 to

determine average pedestrian delay for the first crossing stage.
𝑛 𝑛
𝑑𝑝,1 = ∑ ℎ(𝑖 − 0.5)𝑃(𝑌𝑖 ) + (𝑃𝑑 − ∑ 𝑃(𝑌𝑖 )) 𝑑𝑔𝑑

𝑖=0 𝑖=0
𝑑𝑝,1 = {[(2.3)(0 − 0.5)(0)] + [(2.3)(1 − 0.5)(0.314)] + [(2.3)(2 − 0.5)(0.184)]

+ [(2.3)(3 − 0.5)(0.108)] + [(2.3)(4 − 0.5)(0.063)]}
+ {(0.758 − [0 + 0.314 + 0.184 + 0.108 + 0.063])(9.5)}
𝑑𝑝,1 = 2.12 + 0.85 = 3.0 s
The second stage of the crossing has the same characteristics as the first stage
(same conflicting flow rate and same length). Therefore, the average delay for the
second stage is the same as for the first stage:
𝑑𝑝,2 = 𝑑𝑝,1 = 3.0 s
Under Scenario C, the motorist yielding rate is 80%. Compared to Scenario B,
the different yielding rate only affects the calculation of P(Yi).
𝑃(𝑌0 ) = 0
(2[0.508][1 − 0.508][0.80]) + (0.50820.802 )
𝑃(𝑌1 ) = [0.758 − 0] [ ] = 0.565
0.758
(2[0.508][1 − 0.508][0.80]) + (0.5082 0.802 )
𝑃(𝑌2 ) = [0.758 − 0.565] [ ] = 0.144
0.758
(2[0.508][1 − 0.508][0.80]) + (0.5082 0.802 )
𝑃(𝑌3 ) = [0.758 − 0.709] [ ] = 0.037
0.758
(2[0.508][1 − 0.508][0.80]) + (0.5082 0.802 )
𝑃(𝑌4 ) = [0.758 − 0.746] [ ] = 0.009
0.758
The average pedestrian delay for the first crossing stage is then
𝑑𝑝,1 = {[(2.3)(0 − 0.5)(0)] + [(2.3)(1 − 0.5)(0.565)] + [(2.3)(2 − 0.5)(0.144)]
+ [(2.3)(3 − 0.5)(0.037)] + [(2.3)(4 − 0.5)(0.009)]}
+ {(0.758 − [0 + 0.565 + 0.144 + 0.037 + 0.009])(9.5)}
𝑑𝑝,1 = 1.4 + 0.0 = 1.5 s
The second stage of the crossing has the same characteristics as the first stage
(same conflicting flow rate and same length). Therefore, the average delay for the
second stage is the same as for the first stage:
𝑑𝑝,2 = 𝑑𝑝,1 = 1.5 s
Step 6: Calculate Average Pedestrian Delay

The average pedestrian delay for the entire crossing is the sum of the delays
for the individual crossing stages.
Scenario A = 761 s
Scenario B = 3.0 + 3.0 s = 6.0 s
Scenario C = 1.5 + 1.5 s = 3.0 s

Step 7: Calculate Pedestrian Satisfaction Probabilities and Determine LOS

Under Scenario A, the odds that pedestrians would be satisfied with their
crossing experience, relative to being dissatisfied, are determined from Equation
20-95. In this scenario, there are no pedestrian safety countermeasures at the
crossing; therefore, the indicator variables IRRFB, IMC, and IMR are all zero. The
AADT of the crossing is the peak hour volume divided by the K-factor = 1,700 /
0.08 = 21,250.
In the situation where an arriving pedestrian can cross immediately (i.e., an
adequate gap exists or all blocking vehicles yield), INY = 0. The satisfaction odds
are then:
𝑂(𝑆/𝐷) = exp(0.9951 − 0.0438𝑉𝐾𝐴𝐴𝐷𝑇 + 1.9572𝐼𝑅𝑅𝐹𝐵 + 0.9843𝐼𝑀𝐶
+ 1.5496𝐼𝑀𝑅 − 1.9059𝐼𝑁𝑌 )
𝑂(𝑆/𝐷) = exp(0.9951 − 0.0438 × 21.25 + 1.9572 × 0 + 0.9843 × 0
+ 1.5496 × 0 − 1.9059 × 0)
𝑂(𝑆/𝐷) = 1.066
The probabilities of being satisfied and dissatisfied when an arriving pedestrian
can cross immediately are then given by Equation 20-96 and Equation 20-97.
𝑂(𝑆/𝐷) 1.066
𝑃(𝑆, no delay) = = = 51.6%
𝑂(𝑆/𝐷) + 1 1.066 + 1
𝑃(𝐷, no delay) = 1 − 𝑃(𝑆, no delay) = 1 − 0.516 = 48.4%
In the situation where an arriving pedestrian is delayed crossing the street,
INY = 1. The resulting odds and probabilities are then
𝑂(𝑆/𝐷) = 0.159
𝑃(𝑆, delay) = 13.7%
𝑃(𝐷, delay) = 86.3%
The probability of a non-delayed crossing is given by Equation 20-98. The
value of Pd was determined in Step 3. The value of P(Y1) for a four-lane crossing
is determined by Equation 20-92; with a 0% yielding rate, this equation results in
P(Y1) = 0.
𝑃𝑛𝑑 = (1 − 𝑃𝑑 ) + 𝑃𝑑 𝑃(𝑌1 ) = (1 − 0.997) + 0.997 × 0 = 0.003
The average proportion of dissatisfied pedestrians is then determined from
Equation 20-99.
𝑃𝐷 = 𝑃𝑛𝑑 𝑃(𝐷, no delay) + (1 − 𝑃𝑛𝑑 )𝑃(𝐷, delay)
= (0.003)(0.484) + (0.997)(0.863) = 0.862
From Exhibit 20-3, when half or more of pedestrians would be dissatisfied,
the LOS for the crossing is F.
The calculations for Scenario B are similar to Scenario A, except that the
indicator variables IMC and IMR now have values of 1 because a marked crosswalk
and a median refuge island, respectively, are present. The value of VKAADT
remains the same even though the crossing is now performed in two stages. The
calculation for Scenario C is similar to Scenario B, except that the indicator

variable IRFFB is also 1 because RFFBs are provided. Exhibit 32-7 provides the
calculation results for Scenarios B and C.
Exhibit 32-7
Variable Scenario B Scenario C TWSC Example Problem 2:
O (S/D, no delay) 13.44 95.15 Pedestrian Satisfaction Results
P (S, no delay) 93.1% 99.0% for Scenarios B and C
P (D, no delay) 6.9% 1.0%
O (S/D, delay) 2.00 14.15
P (S, delay) 66.6% 93.4%
P (D, delay) 33.4% 6.6%
Pd 0.758 0.758
P (Y1) 0.314 0.565
Pnd 0.481 0.670
P (D) 0.207 0.029
LOS C A
Discussion
Providing a marked crosswalk and a median refuge island improves the LOS
from F to C in Scenario B, and the further addition of RRFBs improves the LOS to
A in Scenario C.
TWSC EXAMPLE PROBLEM 3: FLARED APPROACHES AND MEDIAN

STORAGE
The Facts
• Major street with two lanes in each direction, minor street with one lane
on each approach that flares with storage for one vehicle in the flare area,
and median storage for two vehicles at one time available for minor-street
through and left-turn movements;
• Level grade on all approaches;
• Percentage heavy vehicles on all approaches = 10%;
• Peak hour factor on all approaches = 0.92;
• Length of analysis period = 0.25 h; and
• Volumes and lane configurations as shown in Exhibit 32-8.
Exhibit 32-8
Configurations

Comments
All relevant input parameters are known, so no default values are needed or
used.

Because hourly volumes and a peak hour factor have been provided, each
hourly volume is divided by the peak hour factor to determine a peak 15-min
flow rate (in vehicles per hour) for each movement. These values are shown in
Exhibit 32-9.
Exhibit 32-9
Flow Rates

The conflicting flow rates for each minor movement at the intersection are
computed according to the equations in Chapter 20. The conflicting flow for the
eastbound major-street left-turn movement vc,1 is computed according to
Equation 20-2 and Exhibit 20-8 as follows:
𝑣𝑐,1 = 𝑓𝑐,1,5 𝑣5 + 𝑓𝑐,1,6 𝑣6 + 𝑓𝑐,1,16 𝑣16 = 1(300) + 1(100) + 1(0) = 400 veh/h
Similarly, the conflicting flow for the westbound major-street left-turn
movement vc,4 is computed according to Equation 20-3 and Exhibit 20-8 as
follows:
𝑣𝑐,4 = 𝑓𝑐,4,2 𝑣2 + 𝑓𝑐,4,3 𝑣3 + 𝑓𝑐,4,15 𝑣15 = 1(250) + 1(50) + 1(0) = 300 veh/h
The conflicting flows for the northbound minor-street right-turn movement
vc,9 and southbound minor-street right-turn movement vc,12 are computed with
Equation 20-4, Equation 20-5, respectively, with coefficients from Exhibit 20-10 as
follows (with pedestrians, the last two terms can be assigned zero):
𝑣𝑐,9 = 0.5(250) + 0.5(50) + 0(0) + 1(0) + 1(0) = 150 veh/h
𝑣𝑐,12 = 𝑓𝑐,12,1𝑈 𝑣1𝑈 + 𝑓𝑐,12,5 𝑣5 + 𝑓𝑐,12,6 𝑣6 + 𝑓𝑐,12,13 𝑣13 + 𝑓𝑐,12,16 𝑣16
𝑣𝑐,12 = 0(0) + 0.5(300) + 0.5(100) + 1(0) + 1(0) = 200 veh/h
Next, the conflicting flow for the northbound minor-street through
movement vc,8 is computed. Because two-stage gap acceptance is available for

this movement, the conflicting flow rates shown in Stage I and Stage II must be
computed separately. The conflicting flow for Stage I vc,I,8 is computed from
Equation 20-8 and Exhibit 20-13:
𝑣𝑐,I,8 = 𝑓𝑐,8,1 𝑣1 + 𝑓𝑐,8,1𝑈 𝑣1𝑈 + 𝑓𝑐,8,2 𝑣2 + 𝑓𝑐,8,3 𝑣3 + 𝑓𝑐,8,15 𝑣15
𝑣𝑐,I,8 = 2(33) + 2(0) + 1(250) + 0.5(50) + 1(0) = 341 veh/h
The conflicting flow for Stage II vc,II,8 is computed from Equation 20-10 and
Exhibit 20-13:
𝑣𝑐,II,8 = 𝑓𝑐,8,4 𝑣4 + 𝑓𝑐,8,4𝑈 𝑣4𝑈 + 𝑓𝑐,8,5 𝑣5 + 𝑓𝑐,8,6 𝑣6 + 𝑓𝑐,8,16 𝑣16
𝑣𝑐,II,8 = 2(66) + 2(0) + 1(300) + 1(100) + 1(0) = 532 veh/h
The total conflicting flow for the northbound through movement vc,8 is
𝑣𝑐,8 = 𝑣𝑐,I,8 + 𝑣𝑐,II,8 = 341 + 532 = 873 veh/h
Similarly, the conflicting flow for the southbound minor-street through
movement vc,11 is computed in two stages as follows:
𝑣𝑐,I,11 = 2(66) + 2(0) + 1(300) + 0.5(100) + 1(0) = 482 veh/h
𝑣𝑐,II,11 = 2(33) + 2(0) + 1(250) + 1(50) + 1(0) = 366 veh/h
𝑣𝑐,11 = 𝑣𝑐,I,11 + 𝑣𝑐,II,11 = 482 + 366 = 848 veh/h
Next, the conflicting flow for the northbound minor-street left-turn
movement vc,7 is computed. Because two-stage gap acceptance is available for
this movement, the conflicting flow rates shown in Stage I and Stage II must be
computed separately. The conflicting flow for Stage I vc,I,7 is computed with
Equation 20-12 and Exhibit 20-16 as follows:
𝑣𝑐,I,7 = 2(33) + 2(0) + 1(250) + 0.5(50) + 1(0) = 341 veh/h
The conflicting flow for Stage II vc,II,7 is computed with Equation 20-26 as
follows:
𝑣𝑐,II,7 = 2(66) + 2(0) + 0.5(300) + 0.5(110) + 1(0) = 337 veh/h
The total conflicting flow for the northbound left-turn movement vc,7 is
𝑣𝑐,7 = 𝑣𝑐,I,7 + 𝑣𝑐,II,7 = 341 + 337 = 678 veh/h
Similarly, the conflicting flow for the southbound minor-street left-turn
movement vc,10 is computed in two stages as follows:
𝑣𝑐,I,10 = 2(66) + 2(0) + 1(300) + 0.5(100) + 1(0) = 482 veh/h
𝑣𝑐,II,10 = 2(33) + 2(0) + 0.5(250) + 0.5(132) + 1(0) = 257 veh/h
𝑣𝑐,10 = 𝑣𝑐,I,10 + 𝑣𝑐,II,10 = 482 + 257 = 739 veh/h

The critical headway for each minor movement is computed beginning with
the base critical headway given in Exhibit 20-17. The base critical headway for
each movement is then adjusted according to Equation 20-16. The critical

headways for the eastbound and westbound major-street left turns tc,1 and tc,4 (in
this case, tc,1 = tc,4) are computed as follows:
𝑡𝑐,1 = 𝑡𝑐,4 = 𝑡𝑐,base + 𝑡𝑐,𝐻𝑉 𝑃𝐻𝑉 + 𝑡𝑐,𝐺 𝐺 − 𝑡3,𝐿𝑇
𝑡𝑐,1 = 𝑡𝑐,4 = 4.1 + 2.0(0.1) + 0(0) − 0 = 4.3 s
Next, the critical headways for the northbound and southbound minor-street
right-turn movements tc,9 and tc,12 (in this case, tc,9 = tc,12) are computed as follows:
𝑡𝑐,9 = 𝑡𝑐,12 = 6.9 + 2.0(0.1) + 0.1(0) − 0 = 7.1 s
Next, the critical headways for the northbound and southbound minor-street
through movements tc,8 and tc,11 (in this case, tc,8 = tc,11) are computed. Because
two-stage gap acceptance is available for these movements, the critical headways
for Stage I and Stage II must be computed, along with the critical headways for
these movements assuming single-stage gap acceptance. The critical headways
for Stage I and Stage II, tc,I,8, tc,I,11 and tc,II,8, tc,II,11, respectively (in this case, tc,I,8 = tc,II,8
= tc,I,11 = tc,II,11), are computed as follows:
𝑡𝑐,I,8 = 𝑡𝑐,II,8 = 𝑡𝑐,I,11 = 𝑡𝑐,II,11 = 5.5 + 2.0(0.1) + 0.2(0) − 0 = 5.7 s
The critical headways for tc,8 and tc,11 (in this case, tc,8 = tc,11), assuming single-
stage gap acceptance, are computed as follows:
𝑡𝑐,8 = 𝑡𝑐,11 = 6.5 + 2.0(0.1) + 0.2(0) − 0 = 6.7 s
Finally, the critical headways for the northbound and southbound minor-
street left-turn movements tc,7 and tc,10 (in this case, tc,7 = tc,10) are computed.
Because two-stage gap acceptance is available for these movements, the critical
headways for Stage I and Stage II must be computed, along with the critical
headways for these movements assuming single-stage gap acceptance. The
critical headways for Stage I and Stage II, tc,I,7, tc,I,10 and tc,II,7, tc,II,10, respectively (in
this case, tc,I,7 = tc,II,7 = tc,I,10 = tc,II,10), are computed as follows:
𝑡𝑐,I,7 = 𝑡𝑐,II,7 = 𝑡𝑐,I,10 = 𝑡𝑐,II,10 = 6.5 + 2.0(0.1) + 0.2(0) − 0 = 6.7 s
The critical headways for tc,7 and tc,10 (in this case, tc,7 = tc,10), assuming single-
stage gap acceptance, are computed as follows:
𝑡𝑐,7 = 𝑡𝑐,10 = 7.5 + 2.0(0.1) + 0.2(0) − 0 = 7.7 s
The follow-up headway for each minor movement is computed beginning
with the base follow-up headway given in Exhibit 20-18. The base follow-up
headway for each movement is then adjusted according to Equation 20-17. The
follow-up headways for the northbound and southbound major-street left-turn
movements tf,1 and tf,4 (in this case, tf,1 = tf,4) are computed as follows:
𝑡𝑓,1 = 𝑡𝑓,4 = 𝑡𝑓,𝑏𝑎𝑠𝑒 + 𝑡𝑓,𝐻𝑉 𝑃𝐻𝑉
𝑡𝑓,1 = 𝑡𝑓,4 = 2.2 + 1.0(0.1) = 2.3 s
Next, the follow-up headways for the northbound and southbound minor-
street right-turn movements tf,9 and tf,12 (in this case, tf,9 = tf,12) are computed as
follows:
𝑡𝑓,9 = 𝑡𝑓,12 = 3.3 + 1.0(0.1) = 3.4 s

Next, the follow-up headways for the northbound and southbound minor-
street through movements tf,8 and tf,11 (in this case, tf,8 = tf,11) are computed as
follows:
𝑡𝑓,8 = 𝑡𝑓,11 = 4.0 + 1.0(0.1) = 4.1 s Follow-up headways for the
minor-street through and left-
Finally, the follow-up headways for the northbound and southbound minor- turn movements are computed
for the movement as a whole.
street left-turn movements tf,7 and tf,10 (in this case, tf,7 = tf,10) are computed as Follow-up headways are not
follows: broken up by stage because
they apply only to vehicles as
𝑡𝑓,7 = 𝑡𝑓,10 = 3.5 + 1.0(0.1) = 3.6 s they exit the approach and
enter the intersection.

Because no upstream signals are present, the procedure in Step 5a is
followed.
The computation of a potential capacity for each movement provides the
analyst with a definition of capacity under the assumed base conditions. The
potential capacity will be adjusted in later steps to estimate the movement
capacity for each movement. The potential capacity for each movement is a
function of the conflicting flow rate, critical headway, and follow-up headway
computed in the previous steps. The potential capacity for the northbound
major-street left-turn movement cp,1 is computed from Equation 20-18:
𝑒 −𝑣𝑐,1 𝑡𝑐,1 /3,600
𝑐𝑝,1 = 𝑣𝑐,1
1 − 𝑒 −𝑣𝑐,1 𝑡𝑓,1 /3,600
𝑒 −(400)(4.3)/3,600
𝑐𝑝,1 = 400 = 1,100 veh/h
1 − 𝑒 −(400)(2.3)/3,600
Similarly, the potential capacities for Movements 4, 9, and 12 (cp,4, cp,9, and
cp,12, respectively) are computed as follows:
𝑒 −(300)(4.3)/3,600
𝑐𝑝,4 = 300 = 1,202 veh/h
1 − 𝑒 −(300)(2.3)/3,600
𝑒 −(150)(7.1)/3,600
𝑐𝑝,9 = 150 = 845 veh/h
1 − 𝑒 −(150)(3.4)/3,600
𝑒 −(200)(7.1)/3,600
𝑐𝑝,12 = 200 = 783 veh/h
1 − 𝑒 −(200)(3.4)/3,600
Because the two-stage gap-acceptance adjustment procedure will be
implemented for estimating the capacity of the minor-street movements, three
potential capacity values must be computed for each of Movements 7, 8, 10, and
11. First, the potential capacity must be computed for Stage I, cp,I,8, cp,I,11, cp,I,7, and
cp,I,10, for each movement as follows:
𝑒 −(341)(5.7)/3,600
𝑐𝑝,𝐼,8 = 341 = 618 veh/h
1 − 𝑒 −(341)(4.1)/3,600
𝑒 −(482)(5.7)/3,600
𝑐𝑝,𝐼,11 = 482 = 532 veh/h
1 − 𝑒 −(482)(4.1)/3,600
𝑒 −(341)(6.7)/3,600
𝑐𝑝,𝐼,7 = 341 = 626 veh/h
1 − 𝑒 −(341)(3.6)/3,600

𝑒 −(482)(6.7)/3,600
𝑐𝑝,𝐼,10 = 482 = 514 veh/h
1 − 𝑒 −(482)(3.6)/3,600
Next, the potential capacity must be computed for Stage II for each
movement, cp,II,8, cp,II,11, cp,II,7, and cp,II,10, as follows:
𝑒 −(532)(5.7)/3,600
𝑐𝑝,𝐼𝐼,8 = 532 = 504 veh/h
1 − 𝑒 −(532)(4.1)/3,600
𝑒 −(366)(5.7)/3,600
𝑐𝑝,𝐼𝐼,11 = 366 = 601 veh/h
1 − 𝑒 −(366)(4.1)/3,600
𝑒 −(337)(6.7)/3,600
𝑐𝑝,𝐼𝐼,7 = 337 = 629 veh/h
1 − 𝑒 −(337)(3.6)/3,600
𝑒 −(257)(6.7)/3,600
𝑐𝑝,𝐼,10 = 257 = 703 veh/h
1 − 𝑒 −(257)(3.6)/3,600
Finally, the potential capacity must be computed assuming single-stage gap
acceptance for each movement, cp,8, cp,11, cp,7, and cp,10, as follows:
𝑒 −(873)(6.7)/3,600
𝑐𝑝,8 = 873 = 273 veh/h
1 − 𝑒 −(873)(4.1)/3,600
𝑒 −(848)(6.7)/3,600
𝑐𝑝,11 = 848 = 283 veh/h
1 − 𝑒 −(848)(4.1)/3,600
𝑒 −(678)(7.7)/3,600
𝑐𝑝,7 = 678 = 323 veh/h
1 − 𝑒 −(678)(3.6)/3,600
𝑒 −(739)(7.7)/3,600
𝑐𝑝,10 = 739 = 291 veh/h
1 − 𝑒 −(739)(3.6)/3,600
Steps 6–9: Compute Movement Capacities

Because no pedestrians are present, the procedures given in Chapter 20 are
followed.

There is no computation for this step.

Step 7a: Movement Capacity for Major-Street Left-Turn Movements
The movement capacity of each Rank 2 major-street left-turn movement is
equal to its potential capacity:
𝑐𝑚,1 = 𝑐𝑝,1 = 1,100 veh/h
𝑐𝑚,4 = 𝑐𝑝,4 = 1,202 veh/h
Step 7b: Movement Capacity for Minor-Street Right-Turn Movements

The movement capacity of each minor-street right-turn movement is equal to
its potential capacity:
𝑐𝑚,9 = 𝑐𝑝,9 = 845 veh/h
𝑐𝑚,12 = 𝑐𝑝,12 = 783 veh/h

Step 7c: Movement Capacity for Major-Street U-Turn Movements

No U-turns are present, so this step is skipped.
Step 7d: Effect of Major-Street Shared Through and Left-Turn Lane

Separate major-street left-turn lanes are provided, so this step is skipped.

The movement capacity of each Rank 3 movement is equal to its potential
capacity, factored by any impedance due to conflicting pedestrian or vehicular
movements.
Step 8a: Rank 3 Capacity for One-Stage Movements

As there are no pedestrians assumed at this intersection, the Rank 3
movements will be impeded only by other vehicular movements. Specifically,
the Rank 3 movements will be impeded by major-street left-turning traffic, and
as a first step in determining the impact of this impedance, the probability that
these movements will operate in a queue-free state must be computed according
to Equation 20-28:
𝑣1 33
𝑝0,1 = 1 − =1− = 0.970
𝑐𝑚,1 1,100
66
𝑝0,4 = 1 − = 0.945
1,202
Next, by using the probabilities computed above, capacity adjustment factors
f8 and f11 can be computed according to Equation 20-32:
𝑓8 = 𝑓11 = 𝑝0,1 × 𝑝0,4 = (0.970)(0.945) = 0.917
Finally, under the single-stage gap-acceptance assumption, the movement
capacities cm,8 and cm,11 can be computed according to Equation 20-33:
𝑐𝑚,8 = 𝑐𝑝,8 × 𝑓8 = (273)(0.917) = 250 veh/h
𝑐𝑚,11 = 𝑐𝑝,11 × 𝑓11 = (283)(0.917) = 260 veh/h
Because Movements 8 and 11 will operate under two-stage gap acceptance,
the capacity adjustment procedure for estimating the capacity of Stage I and
Stage II of these movements must be completed.
To begin the process of estimating Stage I and Stage II movement capacities,
the probabilities of queue-free states on conflicting Rank 2 movements calculated
above are entered into Equation 20-32 as before, but this time capacity
adjustment factors are estimated for each individual stage as follows:
𝑓I,8 = 𝑝0,1 = 0.970
𝑓I,11 = 𝑝0,4 = 0.945
𝑓II,8 = 𝑝0,4 = 0.945
𝑓II,11 = 𝑝0,1 = 0.970
The Stage I movement capacities are then computed as follows:
𝑐𝑚,I,8 = 𝑐𝑝,I,8 × 𝑓𝐼,8 = (618)(0.970) = 599 veh/h
𝑐𝑚,I,11 = 𝑐𝑝,I,11 × 𝑓I,11 = (532)(0.945) = 503 veh/h

The Stage II movement capacities are then computed as follows:

𝑐𝑚,II,8 = 𝑐𝑝,II,8 × 𝑓II,8 = (504)(0.945) = 476 veh/h
Step 8b: Rank 3 Capacity for Two-Stage Movements

The two-stage gap-acceptance procedure will result in a total capacity estimate
for Movements 8 and 11. To begin the procedure, an adjustment factor a must be
computed for each movement by using Equation 20-34, under the assumption
there is storage for two vehicles in the median refuge area; thus, nm = 2.
𝑎8 = 𝑎11 = 1 − 0.32𝑒 −1.3√𝑛𝑚 = 1 − 0.32𝑒 −1.3√2 = 0.949
Next, an intermediate variable, y, must be computed for each movement by
𝑐𝑚,𝐼,8 − 𝑐𝑚,8 599 − 250
𝑦8 = = = 1.808
𝑐𝑚,𝐼𝐼,8 − 𝑣1 − 𝑐𝑚,8 476 − 33 − 250
𝑐𝑚,𝐼,11 − 𝑐𝑚,11 503 − 260
𝑦11 = = = 0.946
𝑐𝑚,𝐼𝐼,11 − 𝑣4 − 𝑐𝑚,11 583 − 66 − 260
Finally, the total capacity for each movement cT,8 and cT,11 is computed
according to Equation 20-36, because y ≠ 1:
𝑎8 𝑛
𝑐𝑚,𝑇,8 = 𝑛𝑚 +1 [𝑦8 (𝑦8 𝑚 − 1)(𝑐𝑚,𝐼𝐼,8 − 𝑣1 ) + (𝑦8 − 1)𝑐𝑚,8 ]
𝑦8 − 1
0.949
𝑐𝑚,𝑇,8 = [(1.808)(1.8082 − 1)(476 − 33) + (1.808 − 1)(250)]
1.8082+1 − 1
𝑐𝑚,𝑇,8 = 390 veh/h
𝑎11 𝑛
𝑐𝑚,𝑇,11 = 𝑛𝑚 +1 [𝑦11 (𝑦11𝑚 − 1)(𝑐𝑚,𝐼𝐼,11 − 𝑣4 ) + (𝑦11 − 1)𝑐𝑚,11 ]
𝑦11 − 1
0.949
𝑐𝑚,𝑇,11 = [(0.946)(0.9462 − 1)(583 − 66) + (0.946 − 1)(260)]
0.9462+1 − 1
𝑐𝑚,𝑇,11 = 405 veh/h

The vehicle impedance effects for Rank 4 movements are first estimated by
assuming single-stage gap acceptance. Rank 4 movements are impeded by all the
same movements impeding Rank 2 and Rank 3 movements with the addition of
impedances due to the minor-street crossing movements and minor-street right-
turn movements. The probability that these movements will operate in a queue-
free state must be incorporated into the procedure.
The probabilities that the minor-street right-turn movements will operate in
a queue-free state (p0,9 and p0,12) are computed as follows:
𝑣9 55
𝑝0,9 = 1 − =1− = 0.935
𝑐𝑚,9 845

28
𝑝0,12 = 1 − = 0.964
783
To compute the adjustment factors that account for the probability that both
the major-street left-turn movements and the minor-street crossing movements
will operate in a queue-free state simultaneously, the analyst must first compute
p0,8 and p0,11 as follows:
𝑣8 132
𝑝0,8 = 1 − = 1− = 0.662
𝑐𝑚,𝑇,8 390
110
𝑝0,11 = 1 − = 0.728
405
Next, with the probabilities computed above, capacity adjustment factors fp,7
and fp,10 can be computed according to Equation 20-38 and Equation 20-39,
respectively:
1
𝑓𝑝,7 = ( ) 𝑝0,12
1 1
+ −1
𝑝0,1 𝑝0,4 𝑝0,11
1
=( ) (0.964) = 0.658
1 1
+ −1
0.970 × 0.945 0.728
1
𝑓𝑝,10 = ( ) 𝑝0,9
1 1
+ −1
𝑝0,1 𝑝0,4 𝑝0,8
1
=( ) (0.935) = 0.584
1 1
+ 0.662 − 1
0.970 × 0.945
Finally, under the single-stage gap-acceptance assumption, the movement
capacities cm,7 and cm,10 can be computed according to Equation 20-40 and
Equation 20-41, respectively:
𝑐𝑚,7 = 𝑐𝑝,7 × 𝑓𝑝,7 = (323)(0.658) = 213 veh/h
𝑐𝑚,10 = 𝑐𝑝,10 × 𝑓𝑝,10 = (291)(0.584) = 170 veh/h

Similar to the minor-street crossing movements at this intersection,
Movements 7 and 10 will also operate under two-stage gap acceptance.
Therefore, the capacity adjustment procedure for estimating the capacity of Stage
I and Stage II of these movements must be completed.
Under the assumption of two-stage gap acceptance with a median refuge
area, the minor-street left-turn movements operate as Rank 3 movements in each
individual stage of completing the left-turn maneuver. To begin the process of
estimating two-stage movement capacities, the probabilities of queue-free states
on conflicting Rank 2 movements for Stage I of the minor-street left-turn
movement are entered into Equation 20-32, and capacity adjustment factors for
Stage I are computed as follows:

𝑓I,7 = 𝑝0,1 = 0.970

𝑓I,10 = 𝑝0,4 = 0.945
The Stage I movement capacities can then be computed as follows:
𝑐𝑚,I,7 = 𝑐𝑝,I,7 × 𝑓I,7 = (626)(0.970) = 607 veh/h
𝑐𝑚,I,10 = 𝑐𝑝,I,10 × 𝑓I,10 = (514)(0.945) = 486 veh/h
Next, the probabilities of queue-free states on conflicting Rank 2 movements
for Stage II of the minor-street left-turn movement are entered into Equation 20-
46. However, before estimating these probabilities, the probability of a queue-
free state for the first stage of the minor-street crossing movement must be
estimated as it impedes Stage II of the minor-street left-turn movement. These
probabilities are estimated with Equation 20-28:
𝑣8 132
𝑝0,I,8 = 1 − =1− = 0.780
𝑐𝑚,I,8 599
110
𝑝0,I,11 =1− = 0.781
503
The capacity adjustment factors for Stage II are then computed as follows:
𝑓II,7 = 𝑝0,4 × 𝑝0,12 × 𝑝0,I,11 = (0.945)(0.964)(0.781) = 0.711
𝑓II,10 = 𝑝0,1 × 𝑝0,9 × 𝑝0,I,8 = (0.970)(0.935)(0.780) = 0.707
Finally, the movement capacities for Stage II are computed as follows:
𝑐𝑚,II,10 = (703)(0.707) = 497 veh/h
The final result of the two-stage gap-acceptance procedure will be a total
capacity estimate for Movements 7 and 10. To begin the procedure, an
adjustment factor a must be computed for each movement by using Equation 20-
42, under the assumption there is storage for two vehicles in the median refuge
area; thus, nm = 2.
𝑎7 = 𝑎10 = 1 − 0.32𝑒 −1.3√𝑛𝑚 = 1 − 0.32𝑒 −1.3√2 = 0.949
Next, an intermediate variable y must be computed for each movement by
𝑐𝑚,I,7 − 𝑐𝑚,7 607 − 213
𝑦7 = = = 1.960
𝑐𝑚,II,7 − 𝑣1 − 𝑐𝑚,7 447 − 33 − 213
𝑐𝑚,I,10 − 𝑐𝑚,10 486 − 170
𝑦10 = = = 1.211
𝑐𝑚,II,10 − 𝑣4 − 𝑐𝑚,10 497 − 66 − 170
Finally, the total capacity for each movement, cT,7 and cT,10, is computed
according to Equation 20-44, as y ≠ 1:
𝑎7 𝑛
𝑐𝑇,7 = 𝑛𝑚+1 [𝑦7 (𝑦7 𝑚 − 1)(𝑐𝑚,𝐼𝐼,7 − 𝑣1 ) + (𝑦7 − 1)𝑐𝑚,7 ]
𝑦7 − 1
0.949
𝑐𝑇,7 = [(1.960)(1.9602 − 1)(447 − 33) + (1.960 − 1)(213)]
1.9602+1 − 1
𝑐𝑇,7 = 365 veh/h

𝑎10 𝑛
𝑐𝑇,10 = 𝑛𝑚 +1 [𝑦10 (𝑦10𝑚 − 1)(𝑐𝑚,𝐼𝐼,10 − 𝑣4 ) + (𝑦10 − 1)𝑐𝑚,10 ]
𝑦10 − 1
0.949
𝑐𝑇,10 = [(1.211)(1.2112 − 1)(497 − 66) + (1.211 − 1)(170)]
1.2112+1 − 1
𝑐𝑇,10 = 342 veh/h
Step 10: Compute Final Capacity Adjustments

In this example problem, several final capacity adjustments must be made to
account for the effect of the shared lanes and the flared lanes on the minor-street
approaches. Initially, the shared-lane capacities for each of the minor-street
approaches must be computed on the assumption of no flared lanes; after these
computations are completed, the effects of the flare can be incorporated to
compute an actual capacity for each minor-street approach.
Step 10a: Shared-Lane Capacity of Minor-Street Approaches

In this example, both minor-street approaches have single-lane entries,
meaning that all movements on the minor street share one lane. The shared-lane
capacities for the minor-street approaches are computed according to Equation
20-46:
∑𝑦 𝑣𝑦 𝑣7 + 𝑣8 + 𝑣9 44 + 132 + 55
𝑐𝑆𝐻,𝑁𝐵 = 𝑣𝑦 = 𝑣7 𝑣8 𝑣9 = 44 132 55 = 441 veh/h
∑𝑦 + +
𝑐𝑚,𝑦 𝑐𝑚,7 𝑐𝑚,8 𝑐𝑚,9 365 + 390 + 845
∑𝑦 𝑣𝑦 11 + 110 + 28
𝑐𝑆𝐻,𝑆𝐵 = 𝑣𝑦 = 11 110 28 = 439 veh/h
∑𝑦
𝑐𝑚,𝑦 342 + 405 + 783
Step 10b: Flared Minor-Street Lane Effects

In this example, the capacity of each minor-street approach will be greater
than the shared capacities computed in the previous step due to the shared-lane
condition on each approach. On each approach, it is assumed one vehicle at a
time can queue in the flared area; therefore, nR = 1.
Equation 20-47 is used to estimate the capacity of each minor-street flared
lane. From Step 7, cm,9 = 845 and cm,12 = 783. However, the capacities for a shared
left-through movement (cm,7+8 and c10+12) have not yet been calculated. These are
calculated using the same method as presented in Step 10a as follows:
∑𝑦 𝑣𝑦 𝑣7 + 𝑣8 44 + 132
𝑐𝑚,7+8 = 𝑣𝑦 = 𝑣7 𝑣8 = 44 132 = 383 veh/h
∑𝑦 +
𝑐𝑚,𝑦 𝑐𝑚,7 𝑐𝑚,8 365 + 390
∑𝑦 𝑣𝑦 11 + 110
𝑐𝑚,10+11 = 𝑣𝑦 = 11 110 = 398 veh/h
∑𝑦
𝑐𝑚,𝑦 342 + 405

𝑣𝑅 + 𝑣𝐿+𝑇𝐻
𝑐𝐹 =
(𝑛𝑅 +1)
𝑣 (𝑛𝑅+1) 𝑣𝐿+𝑇𝐻 (𝑛𝑅+1)
√( 𝑅 ) + ( )
𝑐 𝑅 𝑐 𝐿+𝑇𝐻
55 + (44 + 132)
𝑐𝐹,𝑁𝐵 = = 498 veh/h
(1+1) (1+1) (1+1)
√( 55 ) 44 + 132
+ ( 383 )
845
11 + (110 + 28)
𝑐𝐹,𝑆𝐵 = = 487 veh/h
(1+1) (1+1) (1+1)
√( 11 ) +(
110 + 28
783 398 )
Step 11: Compute Control Delay

The control delay computation for any movement includes initial
deceleration delay, queue move-up time, stopped delay, and final acceleration
delay.

The control delays for the major-street left-turn movements (Rank 2) d1 and
d4 and the minor-street approaches dNB and dSB are computed with Equation
20-61:
3,600 33
3,600 33 33 2 (1,100) (1,100)
𝑑1 = + 900(0.25) √
−1+ ( − 1) + +5
1,100 1,100 1,100 450(0.25)
[ ]
𝑑1 = 8.4 s
3,600 66
3,600 66 66 2 (1,202) (1,202)
𝑑4 = + 900(0.25) √
−1+ ( − 1) + +5
1,202 1,202 1,202 450(0.25)
[ ]
𝑑4 = 8.2 s
2
3,600 231
3,600 231 √ 231 ( 498 ) (498)
𝑑𝑁𝐵 = + 900(0.25) [ −1+ ( − 1) + ]+5
498 498 498 450(0.25)
𝑑𝑁𝐵 = 18.3 s
2
3,600 149
3,600 149 √ 149 ( 487 ) (487)
𝑑𝑆𝐵 = + 900(0.25) [ −1+ ( − 1) + ]+5
487 487 487 450(0.25)
𝑑𝑆𝐵 = 15.6 s
According to Exhibit 20-2, LOS for the major-street left-turn movements and
the minor-street approaches are as follows:

• Eastbound major-street left turn (Movement 1): LOS A,

• Westbound major-street left turn (Movement 4): LOS A,
• Northbound minor-street approach: LOS C, and
• Southbound minor-street approach: LOS C.

This step is not applicable as the major-street through movements v2 and v5
and westbound major-street left-turn movements v1 and v4 have exclusive lanes
at this intersection.

The control delay for the eastbound approach dA,EB is computed with
Equation 20-64:
𝑑𝐴,𝑥 =
0(50) + 0(250) + 8.2(33)
𝑑𝐴,𝐸𝐵 = = 0.8 s
50 + 250 + 33
The control delay for the westbound approach dA,WB is computed according
to the same equation as for the eastbound approach:
0(100) + 0(300) + 8.4(66)
𝑑𝐴,𝑊𝐵 = = 1.2 s
100 + 300 + 66
The intersection delay dI is computed from Equation 20-65:
𝑑𝐴,𝐸𝐵 𝑣𝐴,𝐸𝐵 + 𝑑𝐴,𝑊𝐵 𝑣𝐴,𝑊𝐵 + 𝑑𝐴,𝑁𝐵 𝑣𝐴,𝑁𝐵 + 𝑑𝐴,𝑆𝐵 𝑣𝐴,𝑆𝐵
𝑑𝐼 =
𝑣𝐴,𝐸𝐵 + 𝑣𝐴,𝑊𝐵 + 𝑣𝐴,𝑁𝐵 + 𝑣𝐴,𝑆𝐵
0.8(333) + 1.2(466) + 18.3(231) + 15.6(149)
𝑑𝐼 = = 6.3 s
333 + 466 + 231 + 149
LOS is not defined for the intersection as a whole or for the major-street
approaches.

The 95th percentile queue length for the major-street eastbound left-turn
movement Q95,1 is computed from Equation 20-66:
3,600 𝑣
𝑣1 𝑣1
2 (𝑐 ) (𝑐 𝑥 ) 𝑐
𝑄95,1 ≈ 900𝑇 √
−1+ ( − 1) +
𝑚,1 𝑚,1
(
𝑚,1
)
𝑐𝑚,1 𝑐𝑚,1 150𝑇 3,600
[ ]
3,600 33
33 33 2 (1,100) (1,100) 1,100
𝑄95,1 ≈ 900(0.25) √
−1+ ( − 1) + ( )
1,100 1,100 150(0.25) 3,600
[ ]
𝑄95,1 ≈ 0.1 veh

more than one vehicle will occur very infrequently for the eastbound major-street
left-turn movement.
The 95th percentile queue length for the major-street westbound left-turn
movement Q95,4 is computed as follows:
3,600 66
66 66 2 (1,202) (1,202) 1,202
𝑄95,4 ≈ 900(0.25) √
−1+ ( − 1) + ( )
1,202 1,202 150(0.25) 3,600
[ ]
𝑄95,4 ≈ 0.2 veh
more than one vehicle will occur very infrequently for the westbound major-
street left-turn movement.
The 95th percentile queue length for the northbound approach is computed
by using the same formula, but similar to the control delay computation, the
shared-lane volume and shared-lane capacity must be used.
2
3,600 231
231 √ 231 ( 498 ) (498) 498
𝑄95,𝑁𝐵 ≈ 900(0.25) [ −1+ ( − 1) + ]( )
498 498 150(0.25) 3,600
𝑄95,𝑁𝐵 ≈ 2.4 veh

more than two vehicles will occur occasionally for the northbound approach.
The 95th percentile queue length for the southbound approach is computed
by using the same formula, but similar to the control delay computation, the
shared-lane volume and shared-lane capacity must be used.
2
3,600 149
149 √ 149 ( 487 ) (487) 487
𝑄95,𝑆𝐵 ≈ 900(0.25) [ −1+ ( − 1) + ]( )
487 487 150(0.25) 3,600
𝑄95,𝑆𝐵 ≈ 1.3 veh

more than one vehicle will occur occasionally for the southbound approach.
Discussion
Overall, the results indicate the four-leg TWSC intersection with two-stage
gap acceptance and flared minor-street approaches will operate satisfactorily
with low delays for major-street movements and average delays for the minor-
street approaches.

TWSC EXAMPLE PROBLEM 4: TWSC INTERSECTION WITHIN A

SIGNALIZED URBAN STREET SEGMENT
The Facts
This problem analyzes the performance of the TWSC intersection at Access
Point 1 (AP1) from Example Problem 1 in Chapter 30, Urban Street Segments:
Supplemental, which looks at the motor vehicle performance of the urban street
segment bounded by two signalized intersections, as shown in Exhibit 32-10. The
street has a four-lane cross section with two lanes in each direction.
Exhibit 32-10
1800 ft N TWSC Example Problem 4:
600 ft 600 ft TWSC Intersection Within a
Signalized Urban Street
Segment
1 2
AP1 AP2
From Example Problem 1 in Chapter 30, the following data are relevant:
• Major street with two lanes in each direction,
• Minor street with separate left-turn and right-turn lanes in each direction
(through movements considered negligible) and STOP control on minor-
street approach,
• Flow rates and lane configurations as shown in Exhibit 32-11.
Exhibit 32-11
15-min Flow Rates and Lane
Configurations
The proportion time blocked and delay to through vehicles from the
methodology of Chapter 18, Urban Street Segments, are as shown in Exhibit
32-12.

Exhibit 32-12 Access Point Data

Segment 1
EB
L
EB
T
EB
R
WB
L
WB
T
WB
R
NB
L
NB
T
NB
R
SB
L
SB
T
SB
R
TWSC Example Problem 4: Movement:
1 2 3 4 5 6 7 8 9 10 11 12
Movement-Based Access Point 1: Volume, veh/h 74.80 981.71 93.50 75.56 991.70 94.45 80.00 0.00 100.00 80.00 0.00 100.00
Output (from Chapter 30,

1: Lanes 0 2 0 0 2 0 1 0 1 1 0 1
Example Problem 1) 1: Delay to through vehicles, s/veh
1: Prob. inside lane blocked by left
0.163
0.101
0.164
0.101
2: Volume, veh/h 75.56 991.70 94.45 74.80 981.71 93.50 80.00 0.00 100.00 80.00 0.00 100.00
2: Lanes 0 2 0 0 2 0 1 0 1 1 0 1
Comments
Default values are needed for the saturation flow rates of the major-street
through and right-turn movements for the analysis of shared or short major-
street left-turn lanes:
• Major-street through movement, si1 = 1,800 veh/h; and
• Major-street right-turn movement, si2 = 1,500 veh/h.
All other input parameters are known.

Flow rates for each turning movement have been provided from the
methodology of Chapter 17, Urban Street Reliability and ATDM. They are
assigned movement numbers as shown in Exhibit 32-13.
Exhibit 32-13
Flow Rates

Major-Street Left-Turn Movements (Rank 2, Movements 1 and 4)
The conflicting flows for the major-street left-turn movements are computed
as follows:
𝑣𝑐,1 = 𝑓𝑐,1,5 𝑣5 + 𝑓𝑐,1,6 𝑣6 + 𝑓𝑐,1,16 𝑣16 = 1(992) + 1(94) + 1(0) = 1,086 veh/h
𝑣𝑐,4 = 𝑓𝑐,4,2 𝑣2 + 𝑓𝑐,4,3 𝑣3 + 𝑓𝑐,4,15 𝑣15 = 1(982) + 1(94) + 1(0) = 1,076 veh/h
Minor-Street Right-Turn Movements (Rank 2, Movements 9 and 12)

The conflicting flows for minor-street right-turn movements are computed as
follows:


𝑣𝑐,9 = 0.5(982) + 0.5(94) + 0(0) + 1(0) + 1(0) = 538 veh/h
𝑣𝑐,12 = 𝑓𝑐,12,1𝑈 𝑣1𝑈 + 𝑓𝑐,12,5 𝑣5 + 𝑓𝑐,12,6 𝑣6 + 𝑓𝑐,12,13 𝑣13 + 𝑓𝑐,12,16 𝑣16
𝑣𝑐,12 = 0(0) + 0.5(992) + 0.5(94) + 1(0) + 1(0) = 543 veh/h
Major-Street U-Turn Movements (Rank 2, Movements 1U and 4U)

U-turns are assumed to be negligible.
Minor-Street Pedestrian Movements (Rank 2, Movements 13 and 14)

Minor-street pedestrian movements are assumed to be negligible.
Minor-Street Through Movements (Rank 3, Movements 8 and 11)

Because there are no minor-street through movements, this step can be
skipped.
Minor-Street Left-Turn Movements (Rank 4, Movements 7 and 10)

Because the major street has four lanes without left-turn lanes or other
possible median storage, the minor-street left-turn movement is assumed to be
conducted in one stage. As a result, the conflicting flows for Stages I and II can be
combined.
𝑣𝑐,7 = [2(75) + 2(0) + 1(982) + 0.5(94) + 1(0)]
+ [2(76) + 2(0) + 0.5(992) + 0.5(0) + 1(0)]
𝑣𝑐,7 = 1,827 veh/h
𝑣𝑐,10 = [2(76) + 2(0) + 1(992) + 0.5(94) + 1(0)]
+ [2(75) + 2(0) + 0.5(982) + 0.5(0) + 1(0)]
𝑣𝑐,10 = 1,832 veh/h

Critical headways for each movement are computed from Equation 20-16:
𝑡𝑐,𝑥 = 𝑡𝑐,base + 𝑡𝑐,𝐻𝑉 𝑃𝐻𝑉 + 𝑡𝑐,𝐺 𝐺 − 𝑡3,𝐿𝑇
𝑡𝑐,1 = 𝑡𝑐,4 = 4.1 + (2.0)(0.01) + 0 − 0 = 4.12 s
𝑡𝑐,9 = 𝑡𝑐,12 = 6.9 + (2.0)(0.01) + 0.1(0) − 0 = 6.92 s
𝑡𝑐,7 = 𝑡𝑐,10 = 7.5 + (2.0)(0.01) + 0.2(0) − 0 = 7.52 s
Follow-up headways for each movement are computed from Equation 20-17:
𝑡𝑓,𝑥 = 𝑡𝑓,𝑏𝑎𝑠𝑒 + 𝑡𝑓,𝐻𝑉 𝑃𝐻𝑉
𝑡𝑓,1 = 𝑡𝑓,4 = 2.2 + (1.0)(0.01) = 2.21 s
𝑡𝑓,9 = 𝑡𝑓,12 = 3.3 + (1.0)(0.01) = 3.31 s
𝑡𝑓,7 = 𝑡𝑓,10 = 3.5 + (1.0)(0.01) = 3.51 s


Because upstream signals are present, Step 5b is used. The proportion time
blocked for each movement x is given as pb,x and has been computed by the
Chapter 30 procedure.
The flow for the unblocked period (no platoons) is determined by first
computing the conflicting flow for each movement during the unblocked period
(Equation 20-19). The minimum platooned flow rate vc,min over two lanes is
assumed to be equal to 1,000N = 1,000(2) = 2,000. The flow rate assumed to occur
during the blocked period is calculated as follows:
𝑣𝑐,𝑥 − 1.5𝑣𝑐,𝑚𝑖𝑛 𝑝𝑏,𝑥
if 𝑣𝑐,𝑥 > 1.5𝑣𝑐,𝑚𝑖𝑛 𝑝𝑏,𝑥
𝑣𝑐,𝑢,𝑥 ={ 1 − 𝑝𝑏,𝑥
0 otherwise
1.5𝑣𝑐,𝑚𝑖𝑛 𝑝𝑏,1 = 1.5(2,000)(0.170) = 510 veh/h
The value for vc,1 = 1,086 exceeds this value, which indicates some of the
conflicting flow occurs in the unblocked period. Therefore, vc,u,1 is calculated as
follows:
𝑣𝑐,1 − 1.5𝑣𝑐,𝑚𝑖𝑛 𝑝𝑏,1 1,086 − 1.5(2,000)(0.170)
𝑣𝑐,𝑢,1 = = = 694 veh/h
1 − 𝑝𝑏,1 1 − 0.170
Similar calculations are made for the other movements:
1,076 − 1.5(2,000)(0.170)
𝑣𝑐,𝑢,4 = = 682 veh/h
1 − 0.170
538 − 1.5(2,000)(0.170)
𝑣𝑐,𝑢,9 = = 34 veh/h
1 − 0.170
543 − 1.5(2,000)(0.170)
𝑣𝑐,𝑢,12 = = 40 veh/h
1 − 0.170
1,827 − 1.5(2,000)(0.260)
𝑣𝑐,𝑢,7 = = 1,415 veh/h
1 − 0.260
1,832 − 1.5(2,000)(0.260)
𝑣𝑐,𝑢,10 = = 1,422 veh/h
1 − 0.260
The potential capacity for each movement is then calculated with Equation
20-20 and Equation 20-21 (combined) as follows:
𝑒 −𝑣𝑐,𝑢,1 𝑡𝑐,1 /3,600
𝑐𝑝,1 = (1 − 𝑝𝑏,1 )(𝑣𝑐,𝑢,1 )
1 − 𝑒 −𝑣𝑐,𝑢,1 𝑡𝑓,1 /3,600
𝑒 −(694)(4.12)/3,600
𝑐𝑝,1 = (1 − 0.170)(694) = 750 veh/h
1 − 𝑒 −(694)(2.21)/3,600
𝑒 −(682)(4.12)/3,600
𝑐𝑝,4 = (1 − 0.170)(682) = 758 veh/h
1 − 𝑒 −(682)(2.21)/3,600
𝑒 −(34)(6.92)/3,600
𝑐𝑝,9 = (1 − 0.170)(34) = 859 veh/h
1 − 𝑒 −(34)(3.31)/3,600

𝑒 −(40)(6.92)/3,600
𝑐𝑝,12 = (1 − 0.170)(40) = 852 veh/h
1 − 𝑒 −(40)(3.31)/3,600
𝑒 −(1,415)(7.52)/3,600
𝑐𝑝,7 = (1 − 0.260)(1,415) = 73 veh/h
1 − 𝑒 −(1,415)(3.51)/3,600
𝑒 −(1,422)(7.52)/3,600
𝑐𝑝,10 = (1 − 0.260)(1,422) = 72 veh/h
1 − 𝑒 −(1,422)(3.51)/3,600

Because no pedestrians are present, the procedures given in Chapter 20 are
followed.

There is no computation for this step. The adjustment for the delay to
through movements caused by left-turn movements in the shared left–through
lane is accounted for by using adjustments provided later in this procedure.

Step 7a: Movement Capacity for Major-Street Left-Turn Movements
The movement capacity of each Rank 2 major-street left-turn movement is
equal to its potential capacity as follows:
𝑐𝑚,1 = 𝑐𝑝,1 = 750 veh/h
𝑐𝑚,4 = 𝑐𝑝,4 = 758 veh/h
Step 7b: Movement Capacity for Minor-Street Right-Turn Movements

The movement capacity of each minor-street right-turn movement is equal to
its potential capacity:
𝑐𝑚,9 = 𝑐𝑝,9 = 859 veh/h
𝑐𝑚,12 = 𝑐𝑝,12 = 852 veh/h
Step 7c: Movement Capacity for Major-Street U-Turn Movements

No U-turns are present, so this step is skipped.
Step 7d: Effect of Major-Street Shared Through and Left-Turn Lane

The probability that the major-street left-turning traffic will operate in a
queue-free state, assuming the left-turn movement occupies its own lane, is
calculated with Equation 20-28 as follows:
𝑣1 75
𝑝0,1 = 1 − =1− = 0.900
𝑐𝑚,1 750
𝑣4 76
𝑝0,4 =1− =1− = 0.900
𝑐𝑚,4 758
However, for this problem the major-street left-turn movement shares a lane
with the through movement. First, the combined degree of saturation for the

major-street through and right-turn movements is calculated as follows (using

default values for s and fLL):
𝑣2 𝑣3 982 94
𝑥2+3 = 𝑓𝐿𝐿,2+3 ( + ) = 0.5 ( + ) = 0.304
𝑠2 𝑠3 1,800 1,500
𝑣5 𝑣6 992 94
𝑥5+6 = 𝑓𝐿𝐿,5+6 ( + ) = 0.5 ( + ) = 0.307
𝑠5 𝑠6 1,800 1,500
Next, the probability that there will be no queue in the major-street shared
lane p*0,j is calculated according to the special case (nL = 0) given in Equation 20-33
and Equation 20-34:
∗
1 − 𝑝0,1+1𝑈 1 − 0.900
𝑝0,1+1𝑈 =1− =1− = 0.856
1 − 𝑥2+3 1 − 0.304
∗
1 − 𝑝0,4+4𝑈 1 − 0.900
𝑝0,4+4𝑈 =1− =1− = 0.856
1 − 𝑥5+6 1 − 0.307
These values of p*0,1+1U and p*0,4+4U are used in lieu of p0,1+1U and p0,4+4U for the
remaining calculations.

No minor-street through movements are present, so this step is skipped.

No two-stage movements are present, so this step is skipped.

The probabilities that the minor-street right-turn movements will operate in
the queue-free state p0,9 and p0,12 are computed as follows:
𝑣9 100
𝑝0,9 = 1 − =1− = 0.884
𝑐𝑚,9 859
𝑣12 100
𝑝0,12 =1− =1− = 0.883
𝑐𝑚,12 852
To compute pʹ, the probability that both the major-street left-turn movements
and the minor-street crossing movements will operate in a queue-free state
simultaneously, the analyst must first compute p0,k, which is done in the same
manner as the computation of p0,j, except k represents Rank 3 movements. The
values for p0,k are computed as follows:
𝑣8
𝑝0,8 = 1 − =1−0=1
𝑐𝑚,8
𝑣11
𝑝0,11 =1− =1−0=1
𝑐𝑚,11
Next, by using the probabilities computed above and substituting p*0,j and p*0,j
for p0,j and p0,j, capacity adjustment factors f p,7 and fp,10 can be computed as
follows:

1
𝑓𝑝,7 = ( ) 𝑝0,12
1 1
∗ ∗ +𝑝 −1
𝑝0,1 𝑝0,4 0,11
1
=( ) (0.883) = 0.647
1 1
+1−1
(0.856)(0.856)
1
𝑓𝑝,10 = ( ) 𝑝0,9
1 1
∗ ∗ +𝑝 −1
𝑝0,1 𝑝0,4 0,8
1
=( ) (0.884) = 0.648
1 1
+1−1
(0.856)(0.856)
Finally, the movement capacities cm,7 and cm,10 can be computed as follows:
𝑐𝑚,7 = 𝑐𝑝,7 × 𝑓7 = (73)(0.647) = 47 veh/h
𝑐𝑚,10 = 𝑐𝑝,10 × 𝑓10 = (72)(0.648) = 47 veh/h

No two-stage movements are present, so this step is skipped.
Step 10: Final Capacity Adjustments

Step 10a: Shared-Lane Capacity of Minor-Street Approaches
No shared lanes are present on the side street, so this step is skipped.
Step 10b: Flared Minor-Street Lane Effects

No flared lanes are present, so this step is skipped.
Step 10c: Shared Major-Street Lane Effects

Because the left-turn movement shares a lane with the through movement
(e.g., nL = 0), this step is needed.
𝑣1+1𝑈 75
𝑥1+1𝑈 = = = 0.100
𝑐𝑚,1+1𝑈 750
𝑣2 𝑣3 982 94
𝑥2+3 = 𝑓𝐿𝐿,2+3 ( + ) = 0.5 ( + ) = 0.304
𝑠2 𝑠3 1,800 1,500
(𝑛𝐿 +1) (𝑛 +1)
𝐿
𝑥2+3 0.304
𝑥1+1𝑈+2+3 = 𝑥1+1𝑈 [ √1 + ] = 0.100 [1 + ] = 0.144
1 − 𝑥2+3 1 − 0.304
𝑣2 + 𝑣3 982 + 94
𝑠2+3 = 𝑣 𝑣 = = 3,208
2
+ 3 982 94
+
𝑁𝑠2 𝑠3 (2)(1,800) 1,500

75 + 982 + 94
𝑐𝑆𝑆,1+1𝑈+2+3 = min ( , 3,208) = 3,208
0.144
and
𝑣4+4𝑈 76
𝑥4+4𝑈 = = = 0.100
𝑐𝑚,4+4𝑈 758
𝑣5 𝑣6 992 94
𝑥5+6 = 𝑓𝐿𝐿,5+6 ( + ) = 0.5 ( + ) = 0.307
𝑠5 𝑠6 1,800 1,500
(𝑛𝐿 +1) (𝑛 +1)
𝐿
𝑥5+6 0.307
𝑥4+4𝑈+5+6 = 𝑥4+4𝑈 [ √1 + ] = 0.100 [1 + ] = 0.144
1 − 𝑥5+6 1 − 0.307
𝑣5 + 𝑣6 992 + 94
𝑠5+6 = 𝑣 𝑣 = = 3,211
5
+ 6 992 94
+
𝑁𝑠5 𝑠6 (2)(1,800) 1,500
76 + 992 + 94
𝑐𝑆𝑆,4+4𝑈+5+6 = min ( , 3,211) = 3,211
0.144
Step 11: Compute Movement Control Delay

The delay for each minor-street movement is calculated from Equation 20-61:
2 3,600 75
3,600 75 75 ( 750 ) (750)
𝑑1 = + 900(0.25) √
−1+ ( − 1) + +5
750 750 750 450(0.25)
[ ]
𝑑1 = 10.3 s
2
3,600 76
3,600 76 √ 76 ( 758 ) (758)
𝑑4 = + 900(0.25) [ −1+ ( − 1) + ]+5
758 758 758 450(0.25)
𝑑4 = 10.3 s
2
3,600 100
3,600 100 √ 100 ( 859 ) (859)
𝑑9 = + 900(0.25) [ −1+ ( − 1) + ]+5
859 859 859 450(0.25)
𝑑9 = 9.7 s
2
3,600 100
3,600 100 √ 100 ( )( )
𝑑12 = + 900(0.25) [ −1+ ( − 1) + 851 851 ] + 5
851 851 851 450(0.25)
𝑑12 = 9.8 s
2
3,600 80
3,600 80 √ 80 ( 47 ) (47)
𝑑7 = + 900(0.25) [ − 1 + ( − 1) + ]+5
47 47 47 450(0.25)
𝑑7 = 529 s

2
3,600 80
3,600 80 √ 80 ( 47 ) (47)
𝑑10 = + 900(0.25) [ − 1 + ( − 1) + ]+5
47 47 47 450(0.25)
𝑑10 = 529 s
According to Exhibit 20-2, the LOS for the major-street left-turn movements
and the minor-street approaches are as follows:
• Eastbound major-street left turn (Movement 1): LOS B,
• Westbound major-street left turn (Movement 4): LOS B,
• Northbound minor-street right turn (Movement 9): LOS A,
• Southbound minor-street right turn (Movement 12): LOS A,
• Northbound minor-street left turn (Movement 7): LOS F, and
• Southbound minor-street left turn (Movement 10): LOS F.

The presence of a shared left–through lane on the major street creates delay
for Rank 1 movements (major-street through movements). Because there are two
through lanes, the delay to Rank 1 movements only affects Movements 2 and 5.
Assuming that major-street through vehicles distribute equally across both lanes
(i.e., fLL,2+3 = 0.5 and fLL,5+6 = 0.5), the average delay to Rank 1 vehicles is computed
as follows:
∗
(1 − 𝑝0,1+1𝑈 )𝑓𝐿𝐿,2+3 (𝑣2 )
𝑑1+1𝑈 𝑁>1
𝑑2+3 = { 𝑣1+1𝑈 + 𝑓𝐿𝐿,2+3 (𝑣2 )
∗
(1 − 𝑝0,1+1𝑈 )𝑑1+1𝑈 𝑁=1
(1 − 𝑝∗0,1+1𝑈 ) 𝑓𝐿𝐿,2+3(𝑣2 + 𝑣3 )
𝑑2+3 = 𝑑1+1𝑈
𝑣1+1𝑈 + 𝑓𝐿𝐿,2+3 (𝑣2 + 𝑣3 )
(1 − 0.856)(0.5)(982 + 94)
𝑑2+3 = (10.3)
75 + 0.5(982 + 94)
𝑑2+3 = 1.3 s
(1 − 𝑝∗0,4+4𝑈 ) 𝑓𝐿𝐿,5+6(𝑣5 + 𝑣6 )
𝑑5+6 = 𝑑4+4𝑈
𝑣4+4𝑈 + 𝑓𝐿𝐿,5+6 (𝑣5 + 𝑣6 )
(1 − 0.856)(0.5)(992 + 94)
𝑑5+6 = (10.3)
76 + 0.5(992 + 94)
𝑑5+6 = 1.3 s
The procedures in Chapter 18 provide a better estimate of delay to major-
street through vehicles: d2 = 0.2 and d5 = 0.2. These values account for the
likelihood of major-street through vehicles shifting out of the shared left–through
lane to avoid being delayed by major-street left-turning vehicles.


The control delay for each approach is computed as follows:
𝑑𝐴,𝑥 =
1.3(94 + 982) + 10.3(75)
𝑑𝐴,𝐸𝐵 = = 1.9 s
94 + 982 + 75
1.3(94 + 992) + 10.3(76)

𝑑𝐴,𝑊𝐵 = = 1.9 s
94 + 992 + 76
9.7(100) + 0 + 529(80)
𝑑𝐴,𝑁𝐵 = = 241 s
100 + 0 + 80
9.8(100) + 0 + 529(80)
𝑑𝐴,𝑆𝐵 = = 241 s
100 + 0 + 80
The intersection control delay dI is computed as follows:
𝑑𝐴,𝐸𝐵 𝑣𝐴,𝐸𝐵 + 𝑑𝐴,𝑊𝐵 𝑣𝐴,𝑊𝐵 + 𝑑𝐴,𝑁𝐵 𝑣𝐴,𝑁𝐵 + 𝑑𝐴,𝑆𝐵 𝑣𝐴,𝑆𝐵
𝑑𝐼 =
𝑣𝐴,𝐸𝐵 + 𝑣𝐴,𝑊𝐵 + 𝑣𝐴,𝑁𝐵 + 𝑣𝐴,𝑆𝐵
1.9(1,151) + 1.9(1,162) + 241(180) + 241(180)
𝑑𝐼 = = 34.1 s
1,151 + 1,162 + 180 + 180
LOS is not defined for the intersection as a whole or for the major-street
approaches. This fact is particularly important for this problem, as the
assignment of LOS to the intersection as a whole would mask the severe LOS F
condition on the minor-street left-turn movement. However, this control delay
value may be useful for intersection control evaluation comparisons across a
range of intersection forms and controls.

The 95th percentile queue length for each minor movement is computed by
3,600 𝑣
𝑣1
2 (𝑐 ) (𝑐 1 ) 𝑐
√ 𝑣1 𝑚,1 𝑚,1 𝑚,1
𝑄95,1 ≈ 900𝑇 −1+ ( − 1) + ( )
𝑐𝑚,1 𝑐𝑚,1 150𝑇 3,600
[ ]
2 3,600 75
75 75 ( 750 ) (750) 750
𝑄95,1 ≈ 900(0.25) √
−1+ ( − 1) + ( )
750 750 150(0.25) 3,600
[ ]
𝑄95,1 ≈ 0.3 veh
2
3,600 76
76 √ 76 ( 758 ) (758) 758
𝑄95,4 ≈ 900(0.25) [ −1+ ( − 1) + ]( )
758 758 150(0.25) 3,600
𝑄95,4 ≈ 0.3 veh

2
3,600 100
100 √ 100 ( 859 ) (859) 859
𝑄95,9 ≈ 900(0.25) [ −1+ ( − 1) + ]( )
859 859 150(0.25) 3,600
𝑄95,9 ≈ 0.4 veh
2
3,600 100
100 100 ( )( )
𝑄95,12 ≈ 900(0.25) [ √
−1+ ( − 1) + 851 851 ] ( 851 )
851 851 150(0.25) 3,600
𝑄95,12 ≈ 0.4 veh
2
3,600 80
80 √ 80 ( 47 ) (47) 47
𝑄95,7 ≈ 900(0.25) [ − 1 + ( − 1) + ]( )
47 47 150(0.25) 3,600
𝑄95,7 ≈ 7.9 veh
2
3,600 80
80 √ 80 ( 47 ) (47) 47
𝑄95,10 ≈ 900(0.25) [ − 1 + ( − 1) + ]( )
47 47 150(0.25) 3,600
𝑄95,10 ≈ 7.9 veh

The results indicate that queues of more than one vehicle will rarely occur
for the major-street left-turn and minor-street right-turn movements. Longer
queues are expected for the minor-street left-turn movements, and these queues
are likely to be unstable under the significantly oversaturated conditions.
Discussion
The results indicate that Access Point 1 will operate over capacity (LOS F) for
the minor-street left-turn movements. All other movements are expected to
operate at LOS B or better, with low average delays and short queue lengths.
TWSC EXAMPLE PROBLEM 5: SIX-LANE STREET WITH U-TURNS

AND PEDESTRIANS
The Facts
• T-intersection,
• Major street with three lanes in each direction,
• Minor street with separate left-turn and right-turn lanes and STOP control
on the minor-street approach (minor-street left turns operate in two stages
with room for storage of one vehicle),

• No other unique geometric considerations or upstream signal

considerations,
• 20 p/h crossing both the west and south legs [each pedestrian is assumed
to cross in his or her own group (i.e., independently)],
• Peak hour factor = 1.00,
• Hourly volumes and lane configurations as shown in Exhibit 32-14.
Exhibit 32-14
Volumes and Lane
Configurations
Comments
The assumed walking speed of pedestrians is 3.5 ft/s.

Flow rates for each turning movement are the same as the peak hour
volumes because the peak hour factor equals 1.0. These movements are assigned
numbers as shown in Exhibit 32-15.
Exhibit 32-15
Flow Rates

Major-Street Left-Turn Movement (Rank 2, Movement 4)
The conflicting flow rate for the major-street left-turn movement is computed
as follows:
𝑣𝑐,4 = 𝑓𝑐,4,2 𝑣2 + 𝑓𝑐,4,3 𝑣3 + 𝑓𝑐,4,15 𝑣15 = 1(1,000) + 1(100) + 1(20)
= 1,120 veh/h

Minor-Street Right-Turn Movement (Rank 2, Movement 9)

The conflicting flow rate for the minor-street right-turn movement is
computed as follows (the v3 term has a conflicting flow factor of 0 due to a
separate major-street right-turn lane, and the v4U term has a coefficient of zero as
a default):
𝑣𝑐,9 = 0.5(1,000) + 0(100) + 0(25) + 1(0) + 1(20) = 520 veh/h
Major-Street U-Turn Movements (Rank 2, Movements 1U and 4U)

The conflicting flow rates for the major-street U-turns are computed as
follows (again the v3 term has a conflicting flow factor of 0):
𝑣𝑐,1𝑈 = 𝑓𝑐,1𝑈,5 𝑣5 + 𝑓𝑐,1𝑈,6 𝑣6 + 𝑓𝑐,1𝑈,12 𝑣12 = 0.73(1,200) + 0.73(0) + 0(0)
= 876 veh/h
𝑣𝑐,4𝑈 = 𝑓𝑐,4𝑈,2 𝑣2 + 𝑓𝑐,4𝑈,3 𝑣3 + 𝑓𝑐,4𝑈,9 𝑣9 = 0.73(1,000) + 0(100) + 0(100)
= 730 veh/h
Minor-Street Left-Turn Movements (Rank 3, Movement 7)

The conflicting flow rate for Stage I of the minor-street left-turn movement is
computed as follows (the v3 term in these equations is assumed to be zero
because of the right-turn lane on the major street):
𝑣𝑐,I,7 = 2(0) + 2(50) + 1(1,000) + 0.5(0) + 1(20) = 1,120 veh/h
The conflicting flow rate for Stage II of the minor-street left-turn movement
is computed as follows:
𝑣𝑐,II,7 = 2(100) + 2(25) + 0.4(1,200) + 0.5(0) + 1(20) = 750 veh/h
𝑣𝑐,7 = 𝑣𝑐,I,7 + 𝑣𝑐,II,7 = 1,120 + 750 = 1,870 veh/h

Critical headways for each minor movement are computed as follows:
𝑡𝑐,𝑥 = 𝑡𝑐,base + 𝑡𝑐,𝐻𝑉 𝑃𝐻𝑉 + 𝑡𝑐,𝐺 𝐺 − 𝑡3,𝐿𝑇
𝑡𝑐,1𝑈 = 5.6 + 0 + 0 − 0 = 5.6 s
𝑡𝑐,4 = 5.3 + 0 + 0 − 0 = 5.3 s
𝑡𝑐,4𝑈 = 5.6 + 0 + 0 − 0 = 5.6 s
𝑡𝑐,9 = 7.1 + 0 + 0 − 0 = 7.1 s
𝑡𝑐,7 = 6.4 + 0 + 0 − 0.7 = 5.7 s
𝑡𝑐,𝐼,7 = 7.3 + 0 + 0 − 0.7 = 6.6 s
𝑡𝑐,𝐼𝐼,7 = 6.7 + 0 + 0 − 0.7 = 6.0 s
Follow-up headways for each minor movement are computed as follows:
𝑡𝑓,𝑥 = 𝑡𝑓,base + 𝑡𝑓,𝐻𝑉 𝑃𝐻𝑉
𝑡𝑓,1𝑈 = 2.3 + 0 = 2.3 s

𝑡𝑓,4 = 3.1 + 0 = 3.1 s

𝑡𝑓,4𝑈 = 2.3 + 0 = 2.3 s
𝑡𝑓,9 = 3.9 + 0 = 3.9 s
𝑡𝑓,7 = 3.8 + 0 = 3.8 s

Because no upstream signals are present, Step 5a is used. The potential
capacity cp,x for each movement is computed as follows:
𝑒 −𝑣𝑐,𝑥𝑡𝑐,𝑥/3,600
𝑐𝑝,𝑥 = 𝑣𝑐,𝑥
1 − 𝑒 −𝑣𝑐,𝑥𝑡𝑓,𝑥 /3,600
𝑒 −𝑣𝑐,1𝑈 𝑡𝑐,1𝑈/3,600 𝑒 −(876)(5,6)/3,600

𝑐𝑝,1𝑈 = 𝑣𝑐,1𝑈 = 876 = 523 veh/h
1 − 𝑒 −𝑣𝑐,1𝑈 𝑡𝑓,1𝑈/3,600 1 − 𝑒 −(876)(2.3)/3,600
𝑒 −(1,120)(5,3)/3,600
𝑐𝑝,4 = 1,120 = 348 veh/h
1 − 𝑒 −(1,120)(3.1)/3,600
𝑒 −(730)(5,6)/3,600
𝑐𝑝,4𝑈 = 730 = 629 veh/h
1 − 𝑒 −(730)(2.3)/3,600
𝑒 −(520)(7.1)/3,600
𝑐𝑝,9 = 520 = 433 veh/h
1 − 𝑒 −(520)(3.9)/3,600
𝑒 −(1,870)(5,7)/3,600
𝑐𝑝,7 = 1,870 = 112 veh/h
1 − 𝑒 −(1,870)(3.8)/3,600
𝑒 −(1,120)(6.6)/3,600
𝑐𝑝,𝐼,7 = 1,120 = 207 veh/h
1 − 𝑒 −(1,120)(3.8)/3,600
𝑒 −(750)(6.0)/3,600
𝑐𝑝,𝐼𝐼,7 = 750 = 393 veh/h
1 − 𝑒 −(750)(3.8)/3,600

Because of the presence of pedestrians, the computation steps provided in
Section 4 of Chapter 20 should be used.

The methodology assumes Rank 1 vehicles are unimpeded by pedestrians.

Step 7a: Pedestrian Impedance
The factor accounting for pedestrian blockage is computed by Equation 20-67
as follows:
𝑤
𝑣𝑥 × 𝑆
𝑝
𝑓𝑝𝑏 =
3,600

𝑤 12
𝑣13 ×
𝑆𝑝 20 × 3.5
𝑓𝑝𝑏,13 = = = 0.019
3,600 3,600
12
20 ×
𝑓𝑝𝑏,15 = 3.5 = 0.019
3,600
The pedestrian impedance factor for each pedestrian movement x, pp,x is
computed by Equation 20-68 as follows:
𝑝𝑝,13 = 1 − 𝑓𝑝𝑏,13 = 1 − 0.019 = 0.981
𝑝𝑝,15 = 1 − 𝑓𝑝𝑏,15 = 1 − 0.019 = 0.981
Step 7b: Movement Capacity for Major-Street Left-Turn Movements

On the basis of Exhibit 20-7, vehicular Movement 4 is impeded by pedestrian
Movement 15. Therefore, the movement capacity for Rank 2 major-street left-turn
movements is computed as follows:
𝑐𝑚,4 = 𝑐𝑝,4 × 𝑝𝑝,15 = (348)(0.981) = 341 veh/h
Step 7c: Movement Capacity for Minor-Street Right-Turn Movements

The northbound minor-street right-turn movement (Movement 9) is
impeded by one conflicting pedestrian movement: Movement 15.
𝑓9 = 𝑝𝑝,15 = 0.981
The movement capacity is then computed as follows:
𝑐𝑚,9 = 𝑐𝑝,9 × 𝑓9 = (433)(0.981) = 425 veh/h
Step 7d: Movement Capacity for Major-Street U-Turn Movements

The eastbound U-turn is unimpeded by queues from any other movement.
Therefore, f1U = 1, and the movement capacity is computed as follows:
𝑐𝑚,1𝑈 = 𝑐𝑝,1𝑈 × 𝑓1𝑈 = (523)(0.981) = 523 veh/h
For the westbound U-turn, the movement capacity is found by first
computing a capacity adjustment factor that accounts for the impeding effects of
minor-street right turns as follows:
𝑣9 100
𝑓4𝑈 = 𝑝0,9 = 1 − =1− = 0.765
𝑐𝑚,9 425
The movement capacity is therefore computed as follows:
𝑐𝑚,4𝑈 = 𝑐𝑝,4𝑈 × 𝑓4𝑈 = (629)(0.765) = 481 veh/h
Because the westbound left-turn and U-turn movements are conducted from
the same lane, their shared-lane capacity is computed as follows:
𝑣4 + 𝑣4𝑈 100 + 25
𝑐𝑚,4+4𝑈 = 𝑣 𝑣 = = 362 veh/h
4
+ 4𝑈 100 25
+
𝑐𝑚,4 𝑐𝑚,4𝑈 341 481
Step 7e: Effect of Major-Street Shared Through and Left-Turn Lane

This step is skipped.


There are no minor-street through movements, so the minor-street left-turn
movement is treated as a Rank 3 movement.
Step 8a: Pedestrian Impedance

The northbound minor-street left turn (Movement 7) must yield to
pedestrian Movements 13 and 15. Therefore, the impedance factor for
pedestrians is as follows:
𝑝𝑝,7 = 𝑝𝑝,15 × 𝑝𝑝,13 = (0.981)(0.981) = 0.962
Step 8b: Rank 3 Capacity for One-Stage Movements

The movement capacity cm,k for all Rank 3 movements is found by first
computing a capacity adjustment factor that accounts for the impeding effects of
higher-ranked movements, assuming the movement operates in one stage. This
value is computed as follows:
𝑣1𝑈 𝑣4+4𝑈
𝑓7 = 𝑝0,1𝑈 × 𝑝0,4+4𝑈 × 𝑝𝑝,7 = (1 − ) (1 − ) (𝑝𝑝,7 )
𝑐𝑚,1𝑈 𝑐𝑚,4+4𝑈
50 100 + 25
𝑓7 = (1 − ) (1 − ) (0.962) = 0.570
523 362
𝑐𝑚,7 = 𝑐𝑝,7 × 𝑓7 = (112)(0.570) = 64 veh/h
Step 8c: Rank 3 Capacity for Two-Stage Movements

Because the minor-street left-turn movement operates in two stages, the
procedure for computing the total movement capacity for the subject movement
considering the two-stage gap-acceptance process is followed.
First, the movement capacities for each stage of the left-turn movement are
computed on the basis of the impeding movements for each stage. For Stage I,
the left-turn movement is impeded by the major-street left and U-turns and by
pedestrian Movement 15. Therefore,
𝑣1𝑈 𝑣4+4𝑈
𝑓I,7 = 𝑝0,1𝑈 × 𝑝0,4+4𝑈 × 𝑝𝑝,15 = (1 − ) (1 − ) (𝑝𝑝,15 )
𝑐𝑚,1𝑈 𝑐𝑚,4+4𝑈
50 100 + 25
𝑓I,7 = (1 − ) (1 − ) (0.981) = 0.581
523 362
𝑐𝑚,I,7 = 𝑐𝑝,I,7 × 𝑓I,7 = (207)(0.581) = 120 veh/h
For Stage II, the left-turn movement is impeded only by pedestrian
Movement 13. Therefore,
𝑓II,7 = 𝑝𝑝,13 = 0.981
Next, an adjustment factor a and an intermediate variable y are computed for
Movement 7 as follows:
𝑎7 = 1 − 0.32𝑒 −1.3√𝑛𝑚 = 1 − 0.32𝑒 −1.3√1 = 0.913
𝑐𝑚,𝐼,7 − 𝑐𝑚,7 120 − 64
𝑦7 = = = 0.284
𝑐𝑚,𝐼𝐼,7 − 𝑣4+4𝑈 − 𝑐𝑚,7 386 − 125 − 64

Therefore, the total capacity cT is computed as follows:

𝑎7 𝑛
𝑐𝑚,𝑇,7 = 𝑛𝑚 +1 [𝑦7 (𝑦7 𝑚 − 1)(𝑐𝑚,𝐼𝐼,7 − 𝑣4+4𝑈 ) + (𝑦7 − 1)𝑐𝑚,7 ]
𝑦7 − 1
0.913
𝑐𝑚,𝑇,7 = [(0.284)(0.2841 − 1)(386 − 125) + (0.284 − 1)(64)]
0.2841+1 − 1
𝑐𝑚,𝑇,7 = 98 veh/h

Because there are no Rank 4 movements, this step is skipped.

There are no shared or flared lanes on the minor street, so this step is
skipped.

The control delay for each minor movement is computed as follows:
2 3,600 50
3,600 50 50 ( 523 ) (523)
𝑑1𝑈 = + 900(0.25) √
−1+ ( − 1) + +5
523 523 523 450(0.25)
[ ]
𝑑1 = 12.6 s
This movement would be assigned LOS B.
2 3,600 125
3,600 125 125 ( 362 ) (362)
𝑑4+4𝑈 = + 900(0.25) √
−1+ ( − 1) + +5
362 362 362 450(0.25)
[ ]
𝑑4+4𝑈 = 20.1 s
This movement would be assigned LOS C.
2
3,600 100
3,600 100 √ 100 ( 425 ) (425)
𝑑9 = + 900(0.25) [ −1+ ( − 1) + ]+5
425 425 425 450(0.25)
𝑑9 = 16.1 s
This movement would be assigned LOS C.
2 3,600 75
3,600 75 75 ( 98 ) (98)
𝑑7 = + 900(0.25) √
− 1 + ( − 1) + +5
98 98 98 450(0.25)
[ ]
𝑑1 = 113 s
This movement would be assigned LOS F.


No shared lanes are present on the major street, so this step is skipped.

The control delay for each approach is computed as follows:
0(100) + 0(1,000) + 12.6(50)
𝑑𝐴,𝐸𝐵 = = 0.5 s
100 + 1,000 + 50
0(1,200) + 20.1(125)
𝑑𝐴,𝑊𝐵 = = 1.9 s
1,200 + 125
16.1(100) + 113(75)
𝑑𝐴,𝑁𝐵 = = 57.6 s
100 + 75
The northbound approach is assigned LOS F. No LOS is assigned to the

major-street approaches.
The intersection delay dI is computed as follows:
𝑑𝐴,𝐸𝐵 𝑣𝐴,𝐸𝐵 + 𝑑𝐴,𝑊𝐵 𝑣𝐴,𝑊𝐵 + 𝑑𝐴,𝑁𝐵 𝑣𝐴,𝑁𝐵
𝑑𝐼 =
𝑣𝐴,𝐸𝐵 + 𝑣𝐴,𝑊𝐵 + 𝑣𝐴,𝑁𝐵
0.5(1,150) + 1.9(1,325) + 57.6(175)

𝑑𝐼 = = 5.0 s
1,150 + 1,325 + 175
LOS is not defined for the intersection as a whole.

The 95th percentile queue length for each movement is computed from
Equation 20-66:
2 3,600 50
50 50 ( 523 ) (523) 523
𝑄95,1𝑈 ≈ 900(0.25) √
−1+ ( − 1) + ( )
523 523 150(0.25) 3,600
[ ]
𝑄95,1 ≈ 0.3 veh
2 3,600 125
125 125 ( 362 ) (362) 362
𝑄95,4+4𝑈 ≈ 900(0.25) √
−1+ ( − 1) + ( )
362 362 150(0.25) 3,600
[ ]
𝑄95,4+4𝑈 ≈ 1.5 veh
2
3,600 100
100 √ 100 ( 425 ) (425) 425
𝑄95,9 ≈ 900(0.25) [ −1+ ( − 1) + ]( )
425 425 150(0.25) 3,600
𝑄95,9 ≈ 0.9 veh

2 3,600 75
75 75 ( 98 ) (98) 98
𝑄95,7 ≈ 900(0.25) √
− 1 + ( − 1) + ( )
98 98 150(0.25) 3,600
[ ]
𝑄95,7 ≈ 4.1 veh
Discussion
Overall, the results indicate that although most minor movements are
operating at low to moderate delays and at LOS C or better, the minor-street left
turn experiences high delays and operates at LOS F.

AWSC SUPPLEMENTAL ANALYSIS FOR THREE-LANE

APPROACHES
Exhibit 32-16 provides the 512 possible combinations of probability of

degree-of-conflict cases when alternative lane occupancies are considered for
three-lane approaches. A 1 indicates a vehicle is in the lane; a 0 indicates a
vehicle is not in the lane.
Exhibit 32-16 Opposing Conflicting Left Conflicting Right

Probability of Degree-of- DOC No. of Approach Approach Approach
Conflict Case: Multilane AWSC i Case Vehicles L1 L2 L3 L1 L2 L3 L1 L2 L3
Intersections (Three-Lane 1 1 0 0 0 0 0 0 0 0 0 0
Approaches, by Lane) 2 2 1 1 0 0 0 0 0 0 0 0
(Cases 1–49) 3 0 1 0 0 0 0 0 0 0
4 0 0 1 0 0 0 0 0 0
5 2 1 1 0 0 0 0 0 0 0
6 0 1 1 0 0 0 0 0 0
7 1 0 1 0 0 0 0 0 0
8 3 1 1 1 0 0 0 0 0 0
9 3 1 0 0 0 1 0 0 0 0 0
10 0 0 0 0 1 0 0 0 0
11 0 0 0 0 0 1 0 0 0
12 0 0 0 0 0 0 1 0 0
13 0 0 0 0 0 0 0 1 0
14 0 0 0 0 0 0 0 0 1
15 2 0 0 0 1 1 0 0 0 0
16 0 0 0 0 1 1 0 0 0
17 0 0 0 1 0 1 0 0 0
18 0 0 0 0 0 0 1 1 0
19 0 0 0 0 0 0 0 1 1
20 0 0 0 0 0 0 1 0 1
21 3 0 0 0 1 1 1 0 0 0
22 0 0 0 0 0 0 1 1 1
23 4 2 1 0 0 1 0 0 0 0 0
24 1 0 0 0 1 0 0 0 0
25 1 0 0 0 0 1 0 0 0
26 0 1 0 1 0 0 0 0 0
27 0 1 0 0 1 0 0 0 0
28 0 1 0 0 0 1 0 0 0
29 0 0 1 1 0 0 0 0 0
30 0 0 1 0 1 0 0 0 0
31 0 0 1 0 0 1 0 0 0
32 1 0 0 0 0 0 1 0 0
33 1 0 0 0 0 0 0 1 0
34 1 0 0 0 0 0 0 0 1
35 0 1 0 0 0 0 1 0 0
36 0 1 0 0 0 0 0 1 0
37 0 1 0 0 0 0 0 0 1
38 0 0 1 0 0 0 1 0 0
39 0 0 1 0 0 0 0 1 0
40 0 0 1 0 0 0 0 0 1
41 0 0 0 1 0 0 1 0 0
42 0 0 0 1 0 0 0 1 0
43 0 0 0 1 0 0 0 0 1
44 0 0 0 0 1 0 1 0 0
45 0 0 0 0 1 0 0 1 0
46 0 0 0 0 1 0 0 0 1
47 0 0 0 0 0 1 1 0 0
48 0 0 0 0 0 1 0 1 0
49 0 0 0 0 0 1 0 0 1
AWSC Supplemental Analysis for Three-Lane Approaches Chapter 32/STOP-Controlled Intersections: Supplemental
Opposing Conflicting Left Conflicting Right Exhibit 32-16 (cont’d.)

DOC No. of Approach Approach Approach Probability of Degree-of-
i Case Vehicles L1 L2 L3 L1 L2 L3 L1 L2 L3 Conflict Case: Multilane AWSC
50 4 3 1 1 0 1 0 0 0 0 0 Intersections (Three-Lane
51 (cont’d.) 1 1 0 0 1 0 0 0 0 Approaches, by Lane)
52 1 1 0 0 0 1 0 0 0 (Cases 50–112)
53 0 1 1 1 0 0 0 0 0
54 0 1 1 0 1 0 0 0 0
55 0 1 1 0 0 1 0 0 0
56 1 0 1 1 0 0 0 0 0
57 1 0 1 0 1 0 0 0 0
58 1 0 1 0 0 1 0 0 0
59 1 1 0 0 0 0 1 0 0
60 1 1 0 0 0 0 0 1 0
61 1 1 0 0 0 0 0 0 1
62 0 1 1 0 0 0 1 0 0
63 0 1 1 0 0 0 0 1 0
64 0 1 1 0 0 0 0 0 1
65 1 0 1 0 0 0 1 0 0
66 1 0 1 0 0 0 0 1 0
67 1 0 1 0 0 0 0 0 1
68 1 0 0 1 1 0 0 0 0
69 1 0 0 0 1 1 0 0 0
70 1 0 0 1 0 1 0 0 0
71 0 1 0 1 1 0 0 0 0
72 0 1 0 0 1 1 0 0 0
73 0 1 0 1 0 1 0 0 0
74 0 0 1 1 1 0 0 0 0
75 0 0 1 0 1 1 0 0 0
76 0 0 1 1 0 1 0 0 0
77 0 0 0 1 1 0 1 0 0
78 0 0 0 1 1 0 0 1 0
79 0 0 0 1 1 0 0 0 1
80 0 0 0 0 1 1 1 0 0
81 0 0 0 0 1 1 0 1 0
82 0 0 0 0 1 1 0 0 1
83 0 0 0 1 0 1 1 0 0
84 0 0 0 1 0 1 0 1 0
85 0 0 0 1 0 1 0 0 1
86 1 0 0 0 0 0 1 1 0
87 1 0 0 0 0 0 0 1 1
88 1 0 0 0 0 0 1 0 1
89 0 1 0 0 0 0 1 1 0
90 0 1 0 0 0 0 0 1 1
91 0 1 0 0 0 0 1 0 1
92 0 0 1 0 0 0 1 1 0
93 0 0 1 0 0 0 0 1 1
94 0 0 1 0 0 0 1 0 1
95 0 0 0 1 0 0 1 1 0
96 0 0 0 1 0 0 0 1 1
97 0 0 0 1 0 0 1 0 1
98 0 0 0 0 1 0 1 1 0
99 0 0 0 0 1 0 0 1 1
100 0 0 0 0 1 0 1 0 1
101 0 0 0 0 0 1 1 1 0
102 0 0 0 0 0 1 0 1 1
103 0 0 0 0 0 1 1 0 1
104 4 1 1 0 1 1 0 0 0 0
105 1 1 0 0 1 1 0 0 0
106 1 1 0 1 0 1 0 0 0
107 0 1 1 1 1 0 0 0 0
108 0 1 1 0 1 1 0 0 0
109 0 1 1 1 0 1 0 0 0
110 1 0 1 1 1 0 0 0 0
111 1 0 1 0 1 1 0 0 0
112 1 0 1 1 0 1 0 0 0
Chapter 32/STOP-Controlled Intersections: Supplemental AWSC Supplemental Analysis for Three-Lane Approaches
Exhibit 32-16 (cont’d.) Opposing Conflicting Left Conflicting Right

Intersections (Three-Lane 113 4 4 (cont’d.) 1 1 0 0 0 0 1 1 0
Approaches, by Lane) 114 (cont’d.) 1 1 0 0 0 0 0 1 1
(Cases 113–175) 115 1 1 0 0 0 0 1 0 1
116 0 1 1 0 0 0 1 1 0
117 0 1 1 0 0 0 0 1 1
118 0 1 1 0 0 0 1 0 1
119 1 0 1 0 0 0 1 1 0
120 1 0 1 0 0 0 0 1 1
121 1 0 1 0 0 0 1 0 1
122 0 0 0 1 1 0 1 1 0
123 0 0 0 1 1 0 0 1 1
124 0 0 0 1 1 0 1 0 1
125 0 0 0 0 1 1 1 1 0
126 0 0 0 0 1 1 0 1 1
127 0 0 0 0 1 1 1 0 1
128 0 0 0 1 0 1 1 1 0
129 0 0 0 1 0 1 0 1 1
130 0 0 0 1 0 1 1 0 1
131 1 1 1 1 0 0 0 0 0
132 1 1 1 0 1 0 0 0 0
133 1 1 1 0 0 1 0 0 0
134 1 1 1 0 0 0 1 0 0
135 1 1 1 0 0 0 0 1 0
136 1 1 1 0 0 0 0 0 1
137 1 0 0 1 1 1 0 0 0
138 0 1 0 1 1 1 0 0 0
139 0 0 1 1 1 1 0 0 0
140 0 0 0 1 1 1 1 0 0
141 0 0 0 1 1 1 0 1 0
142 0 0 0 1 1 1 0 0 1
143 1 0 0 0 0 0 1 1 1
144 0 1 0 0 0 0 1 1 1
145 0 0 1 0 0 0 1 1 1
146 0 0 0 1 0 0 1 1 1
147 0 0 0 0 1 0 1 1 1
148 0 0 0 0 0 1 1 1 1
149 5 1 1 1 1 1 0 0 0 0
150 1 1 1 0 1 1 0 0 0
151 1 1 1 1 0 1 0 0 0
152 1 1 1 0 0 0 1 1 0
153 1 1 1 0 0 0 0 1 1
154 1 1 1 0 0 0 1 0 1
155 1 1 0 1 1 1 0 0 0
156 0 1 1 1 1 1 0 0 0
157 1 0 1 1 1 1 0 0 0
158 0 0 0 1 1 1 1 1 0
159 0 0 0 1 1 1 0 1 1
160 0 0 0 1 1 1 1 0 1
161 1 1 0 0 0 0 1 1 1
162 0 1 1 0 0 0 1 1 1
163 1 0 1 0 0 0 1 1 1
164 0 0 0 1 1 0 1 1 1
165 0 0 0 0 1 1 1 1 1
166 0 0 0 1 0 1 1 1 1
167 6 1 1 1 1 1 1 0 0 0
168 1 1 1 0 0 0 1 1 1
169 0 0 0 1 1 1 1 1 1
170 5 3 1 0 0 1 0 0 1 0 0
171 1 0 0 1 0 0 0 1 0
172 1 0 0 1 0 0 0 0 1
173 1 0 0 0 1 0 1 0 0
174 1 0 0 0 1 0 0 1 0
175 1 0 0 0 1 0 0 0 1

176 5 3 (cont’d.) 1 0 0 0 0 1 1 0 0 Intersections (Three-Lane
178 1 0 0 0 0 1 0 0 1 (Cases 176–238)
179 0 1 0 1 0 0 1 0 0
180 0 1 0 1 0 0 0 1 0
181 0 1 0 1 0 0 0 0 1
182 0 1 0 0 1 0 1 0 0
183 0 1 0 0 1 0 0 1 0
184 0 1 0 0 1 0 0 0 1
185 0 1 0 0 0 1 1 0 0
186 0 1 0 0 0 1 0 1 0
187 0 1 0 0 0 1 0 0 1
188 0 0 1 1 0 0 1 0 0
189 0 0 1 1 0 0 0 1 0
190 0 0 1 1 0 0 0 0 1
191 0 0 1 0 1 0 1 0 0
192 0 0 1 0 1 0 0 1 0
193 0 0 1 0 1 0 0 0 1
194 0 0 1 0 0 1 1 0 0
195 0 0 1 0 0 1 0 1 0
196 0 0 1 0 0 1 0 0 1
197 4 1 1 0 1 0 0 1 0 0
198 1 1 0 1 0 0 0 1 0
199 1 1 0 1 0 0 0 0 1
200 1 1 0 0 1 0 1 0 0
201 1 1 0 0 1 0 0 1 0
202 1 1 0 0 1 0 0 0 1
203 1 1 0 0 0 1 1 0 0
204 1 1 0 0 0 1 0 1 0
205 1 1 0 0 0 1 0 0 1
206 0 1 1 1 0 0 1 0 0
207 0 1 1 1 0 0 0 1 0
208 0 1 1 1 0 0 0 0 1
209 0 1 1 0 1 0 1 0 0
210 0 1 1 0 1 0 0 1 0
211 0 1 1 0 1 0 0 0 1
212 0 1 1 0 0 1 1 0 0
213 0 1 1 0 0 1 0 1 0
214 0 1 1 0 0 1 0 0 1
215 1 0 1 1 0 0 1 0 0
216 1 0 1 1 0 0 0 1 0
217 1 0 1 1 0 0 0 0 1
218 1 0 1 0 1 0 1 0 0
219 1 0 1 0 1 0 0 1 0
220 1 0 1 0 1 0 0 0 1
221 1 0 1 0 0 1 1 0 0
222 1 0 1 0 0 1 0 1 0
223 1 0 1 0 0 1 0 0 1
224 1 0 0 1 1 0 1 0 0
225 1 0 0 1 1 0 0 1 0
226 1 0 0 1 1 0 0 0 1
227 1 0 0 0 1 1 1 0 0
228 1 0 0 0 1 1 0 1 0
229 1 0 0 0 1 1 0 0 1
230 1 0 0 1 0 1 1 0 0
231 1 0 0 1 0 1 0 1 0
232 1 0 0 1 0 1 0 0 1
233 0 1 0 1 1 0 1 0 0
234 0 1 0 1 1 0 0 1 0
235 0 1 0 1 1 0 0 0 1
236 0 1 0 0 1 1 1 0 0
237 0 1 0 0 1 1 0 1 0
238 0 1 0 0 1 1 0 0 1

Approaches, by Lane) 240 (cont’d.) 0 1 0 1 0 1 0 1 0
(Cases 239–301) 241 0 1 0 1 0 1 0 0 1
242 0 0 1 1 1 0 1 0 0
243 0 0 1 1 1 0 0 1 0
244 0 0 1 1 1 0 0 0 1
245 0 0 1 0 1 1 1 0 0
246 0 0 1 0 1 1 0 1 0
247 0 0 1 0 1 1 0 0 1
248 0 0 1 1 0 1 1 0 0
249 0 0 1 1 0 1 0 1 0
250 0 0 1 1 0 1 0 0 1
251 1 0 0 1 0 0 1 1 0
252 1 0 0 1 0 0 0 1 1
253 1 0 0 1 0 0 1 0 1
254 1 0 0 0 1 0 1 1 0
255 1 0 0 0 1 0 0 1 1
256 1 0 0 0 1 0 1 0 1
257 1 0 0 0 0 1 1 1 0
258 1 0 0 0 0 1 0 1 1
259 1 0 0 0 0 1 1 0 1
260 0 1 0 1 0 0 1 1 0
261 0 1 0 1 0 0 0 1 1
262 0 1 0 1 0 0 1 0 1
263 0 1 0 0 1 0 1 1 0
264 0 1 0 0 1 0 0 1 1
265 0 1 0 0 1 0 1 0 1
266 0 1 0 0 0 1 1 1 0
267 0 1 0 0 0 1 0 1 1
268 0 1 0 0 0 1 1 0 1
269 0 0 1 1 0 0 1 1 0
270 0 0 1 1 0 0 0 1 1
271 0 0 1 1 0 0 1 0 1
272 0 0 1 0 1 0 1 1 0
273 0 0 1 0 1 0 0 1 1
274 0 0 1 0 1 0 1 0 1
275 0 0 1 0 0 1 1 1 0
276 0 0 1 0 0 1 0 1 1
277 0 0 1 0 0 1 1 0 1
278 5 1 1 0 1 1 0 1 0 0
279 1 1 0 1 1 0 0 1 0
280 1 1 0 1 1 0 0 0 1
281 1 1 0 0 1 1 1 0 0
282 1 1 0 0 1 1 0 1 0
283 1 1 0 0 1 1 0 0 1
284 1 1 0 1 0 1 1 0 0
285 1 1 0 1 0 1 0 1 0
286 1 1 0 1 0 1 0 0 1
287 0 1 1 1 1 0 1 0 0
288 0 1 1 1 1 0 0 1 0
289 0 1 1 1 1 0 0 0 1
290 0 1 1 0 1 1 1 0 0
291 0 1 1 0 1 1 0 1 0
292 0 1 1 0 1 1 0 0 1
293 0 1 1 1 0 1 1 0 0
294 0 1 1 1 0 1 0 1 0
295 0 1 1 1 0 1 0 0 1
296 1 0 1 1 1 0 1 0 0
297 1 0 1 1 1 0 0 1 0
298 1 0 1 1 1 0 0 0 1
299 1 0 1 0 1 1 1 0 0
300 1 0 1 0 1 1 0 1 0
301 1 0 1 0 1 1 0 0 1

304 1 0 1 1 0 1 0 0 1 (Cases 302–364)
305 1 1 0 1 0 0 1 1 0
306 1 1 0 0 1 0 1 1 0
307 1 1 0 0 0 1 1 1 0
308 1 1 0 1 0 0 0 1 1
309 1 1 0 0 1 0 0 1 1
310 1 1 0 0 0 1 0 1 1
311 1 1 0 1 0 0 1 0 1
312 1 1 0 0 1 0 1 0 1
313 1 1 0 0 0 1 1 0 1
314 0 1 1 1 0 0 1 1 0
315 0 1 1 0 1 0 1 1 0
316 0 1 1 0 0 1 1 1 0
317 0 1 1 1 0 0 0 1 1
318 0 1 1 0 1 0 0 1 1
319 0 1 1 0 0 1 0 1 1
320 0 1 1 1 0 0 1 0 1
321 0 1 1 0 1 0 1 0 1
322 0 1 1 0 0 1 1 0 1
323 1 0 1 1 0 0 1 1 0
324 1 0 1 0 1 0 1 1 0
325 1 0 1 0 0 1 1 1 0
326 1 0 1 1 0 0 0 1 1
327 1 0 1 0 1 0 0 1 1
328 1 0 1 0 0 1 0 1 1
329 1 0 1 1 0 0 1 0 1
330 1 0 1 0 1 0 1 0 1
331 1 0 1 0 0 1 1 0 1
332 1 0 0 1 1 0 1 1 0
333 1 0 0 1 1 0 0 1 1
334 1 0 0 1 1 0 1 0 1
335 1 0 0 0 1 1 1 1 0
336 1 0 0 0 1 1 0 1 1
337 1 0 0 0 1 1 1 0 1
338 1 0 0 1 0 1 1 1 0
339 1 0 0 1 0 1 0 1 1
340 1 0 0 1 0 1 1 0 1
341 0 1 0 1 1 0 1 1 0
342 0 1 0 1 1 0 0 1 1
343 0 1 0 1 1 0 1 0 1
344 0 1 0 0 1 1 1 1 0
345 0 1 0 0 1 1 0 1 1
346 0 1 0 0 1 1 1 0 1
347 0 1 0 1 0 1 1 1 0
348 0 1 0 1 0 1 0 1 1
349 0 1 0 1 0 1 1 0 1
350 0 0 1 1 1 0 1 1 0
351 0 0 1 1 1 0 0 1 1
352 0 0 1 1 1 0 1 0 1
353 0 0 1 0 1 1 1 1 0
354 0 0 1 0 1 1 0 1 1
355 0 0 1 0 1 1 1 0 1
356 0 0 1 1 0 1 1 1 0
357 0 0 1 1 0 1 0 1 1
358 0 0 1 1 0 1 1 0 1
359 1 1 1 1 0 0 1 0 0
360 1 1 1 1 0 0 0 1 0
361 1 1 1 1 0 0 0 0 1
362 1 1 1 0 1 0 1 0 0
363 1 1 1 0 1 0 0 1 0
364 1 1 1 0 1 0 0 0 1

Approaches, by Lane) 366 (cont’d. 1 1 1 0 0 1 0 1 0
(Cases 365–427) 367 1 1 1 0 0 1 0 0 1
368 1 0 0 1 1 1 1 0 0
369 1 0 0 1 1 1 0 1 0
370 1 0 0 1 1 1 0 0 1
371 0 1 0 1 1 1 1 0 0
372 0 1 0 1 1 1 0 1 0
373 0 1 0 1 1 1 0 0 1
374 0 0 1 1 1 1 1 0 0
375 0 0 1 1 1 1 0 1 0
376 0 0 1 1 1 1 0 0 1
377 1 0 0 1 0 0 1 1 1
378 1 0 0 0 1 0 1 1 1
379 1 0 0 0 0 1 1 1 1
380 0 1 0 1 0 0 1 1 1
381 0 1 0 0 1 0 1 1 1
382 0 1 0 0 0 1 1 1 1
383 0 0 1 1 0 0 1 1 1
384 0 0 1 0 1 0 1 1 1
385 0 0 1 0 0 1 1 1 1
386 6 1 1 0 1 1 0 1 1 0
387 1 1 0 1 1 0 0 1 1
388 1 1 0 1 1 0 1 0 1
389 1 1 0 0 1 1 1 1 0
390 1 1 0 0 1 1 0 1 1
391 1 1 0 0 1 1 1 0 1
392 1 1 0 1 0 1 1 1 0
393 1 1 0 1 0 1 0 1 1
394 1 1 0 1 0 1 1 0 1
395 0 1 1 1 1 0 1 1 0
396 0 1 1 1 1 0 0 1 1
397 0 1 1 1 1 0 1 0 1
398 0 1 1 0 1 1 1 1 0
399 0 1 1 0 1 1 0 1 1
400 0 1 1 0 1 1 1 0 1
401 0 1 1 1 0 1 1 1 0
402 0 1 1 1 0 1 0 1 1
403 0 1 1 1 0 1 1 0 1
404 1 0 1 1 1 0 1 1 0
405 1 0 1 1 1 0 0 1 1
406 1 0 1 1 1 0 1 0 1
407 1 0 1 0 1 1 1 1 0
408 1 0 1 0 1 1 0 1 1
409 1 0 1 0 1 1 1 0 1
410 1 0 1 1 0 1 1 1 0
411 1 0 1 1 0 1 0 1 1
412 1 0 1 1 0 1 1 0 1
413 1 1 1 1 1 0 1 0 0
414 1 1 1 1 1 0 0 1 0
415 1 1 1 1 1 0 0 0 1
416 1 1 1 0 1 1 1 0 0
417 1 1 1 0 1 1 0 1 0
418 1 1 1 0 1 1 0 0 1
419 1 1 1 1 0 1 1 0 0
420 1 1 1 1 0 1 0 1 0
421 1 1 1 1 0 1 0 0 1
422 1 1 1 1 0 0 1 1 0
423 1 1 1 1 0 0 0 1 1
424 1 1 1 1 0 0 1 0 1
425 1 1 1 0 1 0 1 1 0
426 1 1 1 0 1 0 0 1 1
427 1 1 1 0 1 0 1 0 1

430 1 1 1 0 0 1 1 0 1 (Cases 428–490)
431 1 1 0 1 1 1 1 0 0
432 1 1 0 1 1 1 0 1 0
433 1 1 0 1 1 1 0 0 1
434 0 1 1 1 1 1 1 0 0
435 0 1 1 1 1 1 0 1 0
436 0 1 1 1 1 1 0 0 1
437 1 0 1 1 1 1 1 0 0
438 1 0 1 1 1 1 0 1 0
439 1 0 1 1 1 1 0 0 1
440 1 0 0 1 1 1 1 1 0
441 1 0 0 1 1 1 0 1 1
442 1 0 0 1 1 1 1 0 1
443 0 1 0 1 1 1 1 1 0
444 0 1 0 1 1 1 0 1 1
445 0 1 0 1 1 1 1 0 1
446 0 0 1 1 1 1 1 1 0
447 0 0 1 1 1 1 0 1 1
448 0 0 1 1 1 1 1 0 1
449 1 1 0 1 0 0 1 1 1
450 1 1 0 0 1 0 1 1 1
451 1 1 0 0 0 1 1 1 1
452 0 1 1 1 0 0 1 1 1
453 0 1 1 0 1 0 1 1 1
454 0 1 1 0 0 1 1 1 1
455 1 0 1 1 0 0 1 1 1
456 1 0 1 0 1 0 1 1 1
457 1 0 1 0 0 1 1 1 1
458 1 0 0 1 1 0 1 1 1
459 1 0 0 0 1 1 1 1 1
460 1 0 0 1 0 1 1 1 1
461 0 1 0 1 1 0 1 1 1
462 0 1 0 0 1 1 1 1 1
463 0 1 0 1 0 1 1 1 1
464 0 0 1 1 1 0 1 1 1
465 0 0 1 0 1 1 1 1 1
466 0 0 1 1 0 1 1 1 1
467 7 1 1 1 1 1 0 1 1 0
468 1 1 1 1 1 0 0 1 1
469 1 1 1 1 1 0 1 0 1
470 1 1 1 0 1 1 1 1 0
471 1 1 1 0 1 1 0 1 1
472 1 1 1 0 1 1 1 0 1
473 1 1 1 1 0 1 1 1 0
474 1 1 1 1 0 1 0 1 1
475 1 1 1 1 0 1 1 0 1
476 1 1 0 1 1 1 1 1 0
477 1 1 0 1 1 1 0 1 1
478 1 1 0 1 1 1 1 0 1
479 0 1 1 1 1 1 1 1 0
480 0 1 1 1 1 1 0 1 1
481 0 1 1 1 1 1 1 0 1
482 1 0 1 1 1 1 1 1 0
483 1 0 1 1 1 1 0 1 1
484 1 0 1 1 1 1 1 0 1
485 1 1 0 1 1 0 1 1 1
486 1 1 0 0 1 1 1 1 1
487 1 1 0 1 0 1 1 1 1
488 0 1 1 1 1 0 1 1 1
489 0 1 1 0 1 1 1 1 1
490 0 1 1 1 0 1 1 1 1

Intersections (Three-Lane 491 5 7 1 0 1 1 1 0 1 1 1
Approaches, by Lane) 492 (cont’d.) (cont’d.) 1 0 1 0 1 1 1 1 1
(Cases 491–512) 493 1 0 1 1 0 1 1 1 1
494 1 1 1 1 1 1 1 0 0
495 1 1 1 1 1 1 0 1 0
496 1 1 1 1 1 1 0 0 1
497 1 1 1 1 0 0 1 1 1
498 1 1 1 0 1 0 1 1 1
499 1 1 1 0 0 1 1 1 1
500 1 0 0 1 1 1 1 1 1
501 0 1 0 1 1 1 1 1 1
502 0 0 1 1 1 1 1 1 1
503 8 1 1 1 1 1 1 1 1 0
504 1 1 1 1 1 1 0 1 1
505 1 1 1 1 1 1 1 0 1
506 1 1 1 1 1 0 1 1 1
507 1 1 1 0 1 1 1 1 1
508 1 1 1 1 0 1 1 1 1
509 1 1 0 1 1 1 1 1 1
510 0 1 1 1 1 1 1 1 1
511 1 0 1 1 1 1 1 1 1
512 9 1 1 1 1 1 1 1 1 1
Note: DOC = degree-of-conflict; No. of vehicles = total number of vehicles on the opposing and conflicting
approaches; L1, L2, and L3 = Lane 1, 2, and 3, respectively.
AWSC EXAMPLE PROBLEMS
This part of the chapter provides example problems for use of the AWSC
methodology. Exhibit 32-17 provides an overview of these problems. The
examples focus on the operational analysis level. The planning and preliminary
engineering analysis level is identical to the operations analysis level in terms of
the calculations, except default values are used when available.
Problem Analysis Exhibit 32-17

Number Description Level AWSC Example Problems
1 Single-lane, three-leg AWSC intersection Operational
2 Multilane, four-leg AWSC intersection Operational
AWSC EXAMPLE PROBLEM 1: SINGLE-LANE, THREE-LEG

INTERSECTION
The Facts
The following describes this location’s traffic and geometric characteristics:
• Three legs (T-intersection),
• One-lane entries on each leg,
• Peak hour factor = 0.95, and
• Volumes and lane configurations are as shown in Exhibit 32-18.
Exhibit 32-18
AWSC Example Problem 1:
Volumes and Lane
Configurations
Comments
All input parameters are known, so no default values are needed or used. The use of a spreadsheet or
software for AWSC intersection
The use of a spreadsheet or software is recommended because of the repetitive analysis is recommended
computations required. Slight differences in reported values may result from because of the repetitive and
iterative computations
rounding differences between manual and software computations. Because required.
showing all the individual computations is not practical, this example problem
shows how one or more computations are made. All computational results can
be found in the spreadsheet output located in the Volume 4 Technical Reference
Library section for Chapter 32.
Chapter 32/STOP-Controlled Intersections: Supplemental AWSC Example Problems

Step 1: Convert Movement Demand Volumes to Flow Rates

Peak 15-min flow rates for each turning movement at the intersection are
equal to the hourly volumes divided by the peak hour factor (Equation 21-12).
For example, the peak 15-min flow rate for the eastbound through movement is
as follows:
𝑉𝐸𝐵𝑇𝐻 300
𝑣𝐸𝐵𝑇𝐻 = = = 316 veh/h
𝑃𝐻𝐹 0.95
Step 2: Determine Lane Flow Rates

This step does not apply because the intersection has one-lane approaches on
all legs.
Step 3: Determine Geometry Group for Each Approach

Exhibit 21-11 shows each approach should be assigned to Geometry Group 1.
Step 4: Determine Saturation Headway Adjustments

Exhibit 21-12 shows the headway adjustments for left turns, right turns, and
heavy vehicles are 0.2, –0.6, and 1.7, respectively. These values apply to all
approaches because all are assigned to Geometry Group 1. The saturation
headway adjustment for the eastbound approach is calculated from Equation 21-
13 as follows:
ℎ𝑎𝑑𝑗 = ℎ𝐿𝑇,𝑎𝑑𝑗 𝑃𝐿𝑇 + ℎ𝑅𝑇,𝑎𝑑𝑗 𝑃𝑅𝑇 + ℎ𝐻𝑉,𝑎𝑑𝑗 𝑃𝐻𝑉
53
ℎ𝑎𝑑𝑗,𝐸𝐵 = 0.2 − 0.6(0) + 1.7(0.02) = 0.063
53 + 316
Similarly, the saturation headway adjustments for the westbound and
northbound approaches are as follows:
105
ℎ𝑎𝑑𝑗,𝑊𝐵 = 0.2(0) − 0.6 ( ) + 1.7(0.02) = −0.116
105 + 316
105 53
ℎ𝑎𝑑𝑗,𝑁𝐵 = 0.2 − 0.6 ( ) + 1.7(0.02) = −0.034
105 + 53 105 + 53
Steps 5–11: Determine Departure Headways

These steps are iterative. The following narrative highlights some of the key
calculations using the eastbound approach for Iteration 1.
Step 6: Calculate Initial Degree of Utilization

By using the lane flow rates from Step 2 and the assumed initial departure
headway from Step 5, the initial degree of utilization x is computed as follows
from Equation 21-14:
𝑣ℎ𝑑 (368)(3.2)
𝑥𝐸𝐵 = = = 0.327
3,600 3,600
(421)(3.2)
𝑥𝑊𝐵 = = 0.374
3,600
(158)(3.2)
𝑥𝑁𝐵 = = 0.140
3,600
AWSC Example Problems Chapter 32/STOP-Controlled Intersections: Supplemental

Step 7: Compute Probability States

The probability state of each combination i is determined with Equation
21-15.
𝑃(𝑖) = ∏ 𝑃(𝑎𝑗 ) = 𝑃(𝑎𝑂 ) × 𝑃(𝑎𝐶𝐿 ) × 𝑃(𝑎𝐶𝑅 )

𝑗
For an intersection with single-lane approaches, only eight cases from
Exhibit 21-14 apply, as shown in Exhibit 32-19:
Conflicting Conflicting Exhibit 32-19

DOC No. of Opposing Left Right AWSC Example Problem 1:
i Case Vehicles Approach Approach Approach Applicable Degree-of-Conflict
1 1 0 0 0 0 Cases
2 2 1 1 0 0
5 3 1 0 1 0
7 3 1 0 0 1
13 4 2 0 1 1
16 4 2 1 1 0
21 4 2 1 0 1
45 5 3 1 1 1
For example, the probability state for the eastbound leg under the condition
of no opposing vehicles on the other approaches (degree-of-conflict Case 1, i = 1)
is as follows:
𝑃(𝑎𝑂 ) = 1 − 𝑥𝑂 = 1 − 0.374 = 0.626 (no opposing vehicle present)
𝑃(𝑎𝐶𝐿 ) = 1 − 𝑥𝐶𝐿 = 1 − 0.140 = 0.860 (no conflicting vehicle from left)
𝑃(𝑎𝐶𝑅 ) = 1 (no approach conflicting from right)
Therefore,
𝑃(1) = 𝑃(𝑎𝑂 ) × 𝑃(𝑎𝐶𝐿 ) × 𝑃(𝑎𝐶𝑅 ) = (0.626)(0.860)(1) = 0.538
Similarly,
𝑃(2) = (0.374)(0.860)(1) = 0.322
𝑃(5) = (0.626)(0.140)(1) = 0.088
𝑃(7) = (0.626)(0.860)(0) = 0
𝑃(13) = (0.626)(0.140)(0) = 0
𝑃(16) = (0.374)(0.140)(1) = 0.052
𝑃(21) = (0.374)(0.860)(0) = 0
𝑃(45) = (0.374)(0.140)(0) = 0
Step 8: Compute Probability Adjustment Factors

The probability adjustment is computed as follows, using Equation 21-16
through Equation 21-20:
𝑃(𝐶1 ) = 𝑃(1) = 0.538
𝑃(𝐶2 ) = 𝑃(2) = 0.322
𝑃(𝐶3 ) = 𝑃(5) + 𝑃(7) = 0.088 + 0 = 0.088
𝑃(𝐶4 ) = 𝑃(13) + 𝑃(16) + 𝑃(21) = 0 + 0.052 + 0 = 0.052
𝑃(𝐶5) = 𝑃(45) = 0

The probability adjustment factors for the nonzero cases are calculated from
𝐴𝑑𝑗𝑃(1) = 0.01[0.322 + 2(0.088) + 3(0.052) + 0]/1 = 0.0065
𝐴𝑑𝑗𝑃(2) = 0.01[0.088 + 2(0.052) + 0 − 0.322]/3 = −0.0004
𝐴𝑑𝑗𝑃(5) = 0.01[0.052 + 2(0) − 3(0.088)]/6 = −0.0004
𝐴𝑑𝑗𝑃(16) = 0.01[0 − 6(0.052)]/27 = −0.0001
Therefore, the adjusted probability for Combination 1, for example, is as
follows from Equation 21-16:
𝑃′ (1) = 𝑃(1) + 𝐴𝑑𝑗𝑃(1) = 0.538 + 0.0065 = 0.5445
Step 9: Compute Saturation Headways

The base saturation headways for each combination can be determined with
Exhibit 21-15. They are adjusted by using the adjustment factors calculated in
Step 4 and added to the base saturation headways to determine saturation
headways as shown in Exhibit 32-20 (eastbound illustrated):
Exhibit 32-20 i hbase hadj hsi

AWSC Example Problem 1: 1 3.9 0.063 3.963
Eastbound Saturation
2 4.7 0.063 4.763
Headways
5 5.8 0.063 5.863
7 7.0 0.063 7.063
Step 10: Compute Departure Headways

The departure headway of the lane is the sum of the products of the adjusted
probabilities and the saturation headways as follows (eastbound illustrated):
64
ℎ𝑑 = ∑ 𝑃′ (𝑖)ℎ𝑠𝑖
𝑖=1
ℎ𝑑,𝐸𝐵 = (0.5445)(3.963) + (0.3213)(4.763) + (0.0875)(5.863) + (0.0524)(7.063)
ℎ𝑑,𝐸𝐵 = 4.57 s
Step 11: Check for Convergence

The calculated values of hd are checked against the initial values assumed for
hd. After one iteration, each calculated headway differs from the initial value by
more than 0.1 s. Therefore, the new calculated headway values are used as initial
values in a second iteration. For this problem, four iterations are required for
convergence, as shown in Exhibit 32-21.

EB EB WB WB NB NB SB SB Exhibit 32-21
L1 L2 L1 L2 L1 L2 L1 L2 AWSC Example Problem 1:
Total Lane Flow Rate 368 421 158 Convergence Check
hd, initial value, iteration 1 3.2 3.2 3.2
x, initial, iteration 1 0.327 0.374 0.140
hd, computed value, iteration 1 4.57 4.35 5.14
Convergence? N N N

Convergence? N N N

Convergence? Y Y N

Convergence? Y Y Y
Step 12: Compute Capacities

The capacity of each lane in a subject approach is computed by increasing the
given flow rate on the subject lane (assuming the flows on the opposing and
conflicting approaches are constant) until the degree of utilization for the subject
lane reaches 1. This level of calculation requires running an iterative procedure
many times, which is practical for a spreadsheet or software implementation.
Here, the eastbound lane capacity is approximately 720 veh/h, which is lower
than the value that could be estimated by dividing the lane volume by the degree
of utilization (368/0.492 = 748 veh/h). The difference is due to the interaction
effects among the approaches: increases in eastbound traffic volume increase the
departure headways of the lanes on the other approaches, which in turn
increases the departure headways of the lane(s) on the subject approach.
Step 13: Compute Service Times

The service time required to calculate control delay is computed on the basis
of the final calculated departure headway and the move-up time by using
Equation 21-29. For the eastbound lane (using a value for m of 2.0 for Geometry
Group 1), the calculation is as follows:
𝑡𝑠,𝐸𝐵 = ℎ𝑑,𝐸𝐵 − 𝑚 = 4.97 − 2.0 = 2.97 s
Step 14: Compute Control Delay and Determine LOS for Each Lane
The control delay for each lane is computed with Equation 21-30 as follows
(eastbound illustrated):
ℎ𝑑,𝐸𝐵 𝑥𝐸𝐵
𝑑𝐸𝐵 = 𝑡𝑠,𝐸𝐵 + 900𝑇 [(𝑥𝐸𝐵 − 1) + √(𝑥𝐵 − 1)2 + ]+5
450𝑇
4.97(0.508)
𝑑𝐸𝐵 = 2.97 + 900(0.25) [(0.508 − 1) + √(0.508 − 1)2 + ]+5
450(0.25)
𝑑𝐸𝐵 = 13.0 s

By using Exhibit 21-8, the eastbound lane (and thus approach) is assigned
LOS B. A similar calculation for the westbound and southbound lanes (and thus
approaches) yields 13.5 and 10.6 s, respectively.
Step 15: Compute Control Delay and Determine LOS for the
Intersection
The control delays for the approaches can be combined into an intersection
control delay by using a weighted average as follows:
∑ 𝑑𝑎 𝑣𝑎
𝑑intersection =
∑ 𝑣𝑎
(13.0)(368) + (13.5)(421) + (10.6)(158)
𝑑intersection = = 12.8 s
368 + 421 + 158
This value of delay is assigned LOS B.
Step 16: Compute Queue Lengths

The 95th percentile queue for each lane is computed with Equation 21-33 as
follows (eastbound approach illustrated):
900𝑇 ℎ𝑑,𝐸𝐵 𝑥𝐸𝐵

𝑄95,𝐸𝐵 ≈ [(𝑥𝐸𝐵 − 1) + √(𝑥𝐸𝐵 − 1)2 + ]
ℎ𝑑,𝐸𝐵 150𝑇
900(0.25) 4.97(0.508)
𝑄95,𝐸𝐵 ≈ [(0.508 − 1) + √(0.508 − 1)2 + ] = 2.9 veh
4.97 150(0.25)
This queue length would be reported as three vehicles.
Discussion
The results indicate the intersection operates well with brief delays.
AWSC EXAMPLE PROBLEM 2: MULTILANE, FOUR-LEG INTERSECTION

The Facts
• Four legs;
• Two-lane approaches on the east and west legs;
• Three-lane approaches on the north and south legs;
• Percentage heavy vehicles on all approaches = 2%;
• Demand volumes are provided in 15-min intervals (therefore, a peak hour
factor is not required), and the analysis period length is 0.25 h; and
• Volumes and lane configurations are as shown in Exhibit 32-22.

Exhibit 32-22
Configurations
Comments
The use of a spreadsheet or software is required because of the several thousand
repetitive computations needed. Slight differences in reported values may result
from rounding differences between manual and software computations. Because
showing all the individual computations is not practical, this example problem
shows how one or more computations are made. All computational results can
be found in the spreadsheet output located in the Volume 4 Technical Reference
Library section for Chapter 32.

To convert the peak 15-min demand volumes to hourly flow rates, the
individual movement volumes are simply multiplied by four, as shown in
Exhibit 32-23:
Exhibit 32-23
15-min Volumes Converted to
Hourly Flow Rates
Step 2: Determine Lane Flow Rates

This step simply involves assigning the turning movement volume to each of
the approach lanes. The left-turn volume is assigned to the separate left-turn lane
on each approach. For the east and west approaches, the through and right-turn
volumes are assigned to the shared through and right lanes. For the north and

south approaches, the through volumes are assigned to the through lanes and
the right-turn volumes are assigned to the right-turn lanes.
Step 3: Determine Geometry Group for Each Approach

Exhibit 21-11 shows each approach should be assigned to Geometry Group 6.
Step 4: Determine Saturation Headway Adjustments

Exhibit 21-12 shows the headway adjustments for left turns, right turns, and
heavy vehicles are 0.5, –0.7, and 1.7, respectively. These values apply to all
approaches as all are assigned Geometry Group 6. The saturation headway
adjustment for the eastbound approach is as follows for Lane 1 (the left-turn
lane):
ℎ𝑎𝑑𝑗 = ℎ𝐿𝑇,𝑎𝑑𝑗 𝑃𝐿𝑇 + ℎ𝑅𝑇,𝑎𝑑𝑗 𝑃𝑅𝑇 + ℎ𝐻𝑉,𝑎𝑑𝑗 𝑃𝐻𝑉
ℎ𝑎𝑑𝑗,𝐸𝐵,1 = 0.5(1.0) − 0.7(0) + 1.7(0.02) = 0.534
Similarly, the saturation headway adjustment for Lane 2 of the eastbound
approach is as follows:
64
ℎ𝑎𝑑𝑗,𝐸𝐵,2 = 0.5(0) − 0.7 ( ) + 1.7(0.02) = −0.173
64 + 152
The saturation headway adjustment for all the remaining lanes by approach
is similarly calculated. The full computational results can be seen in the
“HdwyAdj” spreadsheet tab.
Steps 5–11: Determine Departure Headways

These steps are iterative and, for this example, involve several thousand
calculations. The following narrative highlights some of the key calculations
using the eastbound approach for Iteration 1, but it does not attempt to
reproduce all calculations for all iterations. The full computational results for
each of the iterative computations can be seen in the “DepHdwyIterX”
spreadsheet tab, where “X” is the iteration.
Step 6: Calculate Initial Degree of Utilization

The remainder of this example illustrates the calculations needed to evaluate
Lane 1 on the eastbound approach (eastbound left turn). Step 6 requires
calculating the initial degree of utilization for all the opposing and conflicting
lanes. They are computed as follows:
𝑣ℎ𝑑 (156)(3.2)
𝑥𝑊𝐵,1 = = = 0.1387
3,600 3,600
𝑣ℎ𝑑 (164)(3.2)
𝑥𝑊𝐵,2 = = = 0.1458
3,600 3,600
𝑣ℎ𝑑 (76)(3.2)
𝑥𝑁𝐵,1 = = = 0.0676
3,600 3,600
𝑣ℎ𝑑 (164)(3.2)
𝑥𝑁𝐵,2 = = = 0.1458
3,600 3,600
𝑣ℎ𝑑 (116)(3.2)
𝑥𝑁𝐵,3 = = = 0.1031
3,600 3,600

𝑣ℎ𝑑 (48)(3.2)
𝑥𝑆𝐵,1 = = = 0.0427
3,600 3,600
𝑣ℎ𝑑 (124)(3.2)
𝑥𝑆𝐵,2 = = = 0.1102
3,600 3,600
𝑣ℎ𝑑 (88)(3.2)
𝑥𝑆𝐵,3 = = = 0.0782
3,600 3,600
Step 7: Compute Probability States

Because three-lane approaches are involved, the modified methodology
presented in Section 4 of Chapter 21 is used.
The probability state of each combination i is determined with Equation
21-34:
𝑃(𝑖) = ∏ 𝑃(𝑎𝑗 ) = 𝑃( 𝑂) × 𝑃(𝑎𝐶𝐿 ) × 𝑃(𝑎𝐶𝑅 )

𝑗
For example, the probability state for the eastbound leg under the condition
of no opposing vehicles on the other approaches (Degree-of-Conflict Case 1, i = 1)
is as follows (using Exhibit 21-16):
𝑃(𝑎𝑂1 ) = 1 − 𝑥𝑂1 = 1 − 0.1387 = 0.8613 (opposing westbound Lane 1)
𝑃(𝑎𝑂2 ) = 1 − 𝑥𝑂2 = 1 − 0.1458 = 0.8542 (opposing westbound Lane 2)
𝑃(𝑎𝐶𝐿1 ) = 1 − 𝑥𝐶𝐿1 = 1 − 0.0427 = 0.9573 (conflicting from left Lane 1)
𝑃(𝑎𝐶𝑅1 ) = 1 − 𝑥𝐶𝑅1 = 1 − 0.0676 = 0.9324 (conflicting from right Lane 1)
Therefore,
𝑃(1) = 𝑃(𝑎𝑂1 ) × 𝑃(𝑎𝑂2 ) × 𝑃(𝑎𝐶𝐿1 ) × 𝑃(𝑎𝐶𝐿2 ) × 𝑃(𝑎𝐶𝐿3 ) × 𝑃(𝑎𝐶𝑅1 ) × 𝑃(𝑎𝐶𝑅2 )
× 𝑃(𝑎𝐶𝑅3 )
𝑃(1) = (0.8613)(0.8542)(0.9573)(0.8898)(0.9218)(0.9324)(0.8542)(0.8969)
𝑃(1) = 0.4127
To complete the calculations for Step 7, the computations are completed for
the remaining 511 possible combinations. The full computational results for the
eastbound leg (Lane 1) can be seen in the “DepHdwyIter1” spreadsheet tab,
Rows 3118–3629 (Columns C–K).
Step 8: Compute Probability Adjustment Factors

The probability adjustment is computed with Equation 21-35 through
Equation 21-39 to account for the serial correlation in the previous probability
computation. First, the probability of each degree-of-conflict case must be
determined. For the example of eastbound Lane 1, these computations are made
by summing Rows 3118–3629 in the spreadsheet for each of the five cases
(Columns R–V). The resulting computations are shown in Row 3630 (Columns
R–V), where

𝑃(𝐶1 ) = 𝑃(1) = 0.4127

8
𝑃(𝐶2 ) = ∑ 𝑃(𝑖) = 0.1482

𝑖=2
22
𝑃(𝐶3 ) = ∑ 𝑃(𝑖) = 0.2779

𝑖=9
169
𝑃(𝐶4 ) = ∑ 𝑃(𝑖) = 0.1450

𝑖=23
512
𝑃(𝐶5) = ∑ 𝑃(𝑖) = 0.0162

𝑖=170
The probability adjustment factors are then computed with Equation 21-40
through Equation 21-44, where  equals 0.01 (or 0.00 if correlation among
saturation headways is not taken into account).
For example, by using Equation 21-35, AdjP(1) is calculated as follows:
𝐴𝑑𝑗𝑃(1) = 0.01[0.1482 + 2(0.2779) + 3(0.1450) + 4(0.0162)]/1 = 0.01204
The results of the remaining computations for eastbound Lane 1 are located
in Row 3632 of the spreadsheet (Columns S–V).
Step 9: Compute Saturation Headways

The base saturation headways for each of the 512 combinations can be
determined with Exhibit 21-15. They are adjusted by using the adjustment factors
calculated in Step 4 and added to the base saturation headways to determine
saturation headways.
For the example of eastbound Lane 1, these computations are shown in Rows
3118–3629 of the spreadsheet (Columns M–O).
Step 10: Compute Departure Headways

The departure headway of the lane is the sum of the products of the adjusted
probabilities and the saturation headways. For the example of eastbound Lane 1,
these computations are made by summing the product of Columns O and Y for
Rows 3118–3629 in the example spreadsheet.
Step 11: Check for Convergence

The calculated values of hd are checked against the assumed initial values for
hd. After one iteration, each calculated headway differs from the initial value by
more than 0.1 s. Therefore, the new calculated headway values are used as initial
values in a second iteration. For this problem, five iterations were required for
convergence, as shown in Exhibit 32-24.

EB EB EB WB WB WB NB NB NB SB SB SB Exhibit 32-24
L1 L2 L3 L1 L2 L3 L1 L2 L3 L1 L2 L3
Total lane flow rate 56 216 156 164 76 164 116 48 124 88 AWSC Example Problem 2:
hd, initial value, Iteration 1 3.2 3.2 3.2 3.2 3.2 3.2 3.2 3.2 3.2 3.2 Convergence Check
x, initial, Iteration 1 0.0498 0.192 0.1387 0.1458 0.0676 0.1458 0.1031 0.0427 0.1102 0.0782
hd, computed value, Iteration 1 6.463 5.755 6.405 5.597 6.440 5.935 5.228 6.560 6.055 5.347
Convergence? N N N N N N N N N N
hd, initial value, Iteration 2 6.463 5.755 6.405 5.597 6.440 5.935 5.228 6.560 6.055 5.347
Convergence? Y Y Y Y Y Y Y Y Y Y
Step 12: Compute Capacity

As noted in the procedure, the capacity of each lane in a subject approach is
computed by increasing the given flow rate on the subject lane (assuming the
flows on the opposing and conflicting approaches are constant) until the degree
of utilization for the subject lane reaches 1. This level of calculation requires
running an iterative procedure many times, which is practical only for a
spreadsheet or software implementation.
For this example, the capacity of eastbound Lane 1 can be found to be
approximately 420 veh/h. This value is lower than the value that could be
estimated by dividing the lane volume by the degree of utilization (56/0.1265 =
443 veh/h). The difference is due to the interaction effects among the approaches:
increases in eastbound traffic volume increase the departure headways of the
lanes on the other approaches, which increases the departure headways of the
lanes on the subject approach.
Step 13: Compute Service Times

The service time required to calculate control delay is computed on the basis
of the final calculated departure headway and the move-up time by using
Equation 21-29. For the eastbound Lane 1 (using a value for m of 2.3 for
Geometry Group 6), the calculation is as follows:
𝑡𝑠,𝐸𝐵,1 = ℎ𝑑,𝐸𝐵,1 − 𝑚 = 8.19 − 2.3 = 5.89 s
Step 14: Compute Control Delay and Determine LOS for Each Lane
The control delay for each lane is computed with Equation 21-30 as follows
(eastbound Lane 1 illustrated):
ℎ𝑑,𝐸𝐵,1 𝑥𝐸𝐵,1
𝑑𝐸𝐵,1 = 𝑡𝑠,𝐸𝐵,1 + 900𝑇 [(𝑥𝐸𝐵,1 − 1) + √(𝑥𝐸𝐵,1 − 1)2 + ]+5
450𝑇
8.19(0.1274)
𝑑𝐸𝐵,1 = 5.89 + 900(0.25) [(0.1274 − 1) + √(0.1274 − 1)2 + ]+5
450(0.25)

𝑑𝐸𝐵,1 = 12.1 s
On the basis of Exhibit 20-2, eastbound Lane 1 is assigned LOS B.
Step 15: Compute Control Delay and Determine LOS for Each Approach
and the Intersection
The control delay for each approach is calculated using Equation 21-31 as
follows (eastbound approach illustrated):
(12.1)(272) + (16.1)(216)
𝑑EB = = 15.3 s
56 + 216
This value of delay is assigned LOS C.
Similarly, the control delay for the intersection is calculated as follows:
(15.3)(272) + (14.3)(320) + (13.1)(356) + (12.6)(260)
𝑑intersection = = 14.0 s
272 + 320 + 356 + 260
This value of delay is assigned LOS B.
Step 16: Compute Queue Lengths

The 95th percentile queue for each lane is computed with Equation 21-33 as
follows for eastbound Lane 1:
900(0.25) 8.19(0.1274)
𝑄95,𝐸𝐵1 ≈ [(0.1274 − 1) + √(0.1274 − 1)2 + ]
8.19 150(0.25)
𝑄95,𝐸𝐵1 ≈ 0.4 veh

This queue length commonly would be rounded up to one vehicle.
Discussion
The overall results can be found in the “DelayLOS” spreadsheet tab. As
indicated in the output, all movements at the intersection are operating well with
small delays. The worst-performing movement is eastbound Lane 2, which is
operating with a volume-to-capacity ratio of 0.45 and a control delay of 16.1
s/veh, which results in LOS C. The intersection as a whole operates at LOS B, so
the reporting of individual movements is important to avoid masking results
caused by aggregating delays.

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 33
ROUNDABOUTS: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 33-1
2. SUPPLEMENTAL GUIDANCE .......................................................................... 33-2

Variability and Uncertainty............................................................................... 33-2
Lane-Use Assignment ........................................................................................ 33-4
Capacity Model Calibration .............................................................................. 33-6

Example Problem 1: Single-Lane Roundabout with Bypass Lanes ............. 33-8
Example Problem 2: Multilane Roundabout..................................................33-13
4. CONNECTED AND AUTOMATED VEHICLES .......................................... 33-19

Introduction ........................................................................................................33-19
Concepts .............................................................................................................33-19
Capacity Adjustment Factors ...........................................................................33-22
5. REFERENCES ....................................................................................................... 33-24
Chapter 33/Roundabouts: Supplemental Contents

LIST OF EXHIBITS
Exhibit 33-1 Observed Combinations of Entry Flow and Conflicting Flow

During 1-min Periods of Continuous Queuing: One-Lane Entry
Opposed by One Circulating Lane ................................................................... 33-2
During 1-min Periods of Continuous Queuing: Both Lanes of Two-
Lane Entry Opposed by One Circulating Lane .............................................. 33-3
During 1-min Periods of Continuous Queuing: Left Lane of Two-Lane
Entry Opposed by Two Circulating Lanes ...................................................... 33-3
During 1-min Periods of Continuous Queuing: Right Lane of Two-
Lane Entry Opposed by Two Circulating Lanes ............................................ 33-4
Exhibit 33-5 Roundabout Example Problems ........................................................ 33-8
Exhibit 33-6 Example Problem 1: Demand Volumes and Lane
Configurations .................................................................................................... 33-8
Exhibit 33-7 Example Problem 1: Adjusted Flow Rates........................................ 33-9
Exhibit 33-8 Example Problem 1: LOS by Lane ................................................... 33-12
Exhibit 33-9 Example Problem 2: Demand Volumes and Lane
Configurations .................................................................................................. 33-13
Exhibit 33-10 Example Problem 2: Adjusted Flow Rates .................................... 33-14
Exhibit 33-11 Example Problem 2: LOS by Lane ................................................. 33-17
Exhibit 33-12 Roundabout Entry Lane Capacity Model Parameters ................ 33-23
Exhibit 33-13 Capacity Adjustment Factors for CAVs for Roundabouts ......... 33-23
Contents Chapter 33/Roundabouts: Supplemental

1. INTRODUCTION
Chapter 33 is the supplemental chapter for Chapter 22, Roundabouts, which VOLUME 4: APPLICATIONS
GUIDE
is found in Volume 3 of the Highway Capacity Manual (HCM). This chapter 25. Freeway Facilities:
presents detailed information about the following aspects of the Chapter 22 Supplemental
motorized vehicle methodology: Segments: Supplemental
• Information about the large variability in U.S. driver behavior at Supplemental
roundabouts, 28. Freeway Merges and
• Guidance on making an appropriate selection of a lane utilization factor, 29. Urban Street Facilities:
Supplemental
and 30. Urban Street Segments:
Supplemental
• Guidance on calibrating the capacity model to reflect local conditions. 31. Signalized Intersections:
Supplemental
This chapter also provides two example problems that demonstrate the 32. STOP-Controlled
application of the Chapter 22 methodology to single-lane and multilane Intersections:
Supplemental
roundabouts, and provides guidance on estimating roundabout capacity when 33. Roundabouts:
connected and automated vehicles are present in the traffic stream. Supplemental
Supplemental
Chapter 33/Roundabouts: Supplemental Introduction

2. SUPPLEMENTAL GUIDANCE
This section presents supplemental guidance on the methodology provided

in Chapter 22, Roundabouts.
VARIABILITY AND UNCERTAINTY

The analyst should be aware of the large observed variation in driver
behavior at roundabouts. Exhibit 33-1 through Exhibit 33-4 show observed
combinations of entry flow and conflicting flow at different roundabout
configurations, along with the capacity models for the respective configuration
as presented in Chapter 22. The bulk of this variation is attributable to variations
in driver behavior, truck percentage, and exiting vehicles. As there is no external
control device regulating flow interactions at roundabouts, driver interactions
govern the operation, and they are highly variable by nature.
This variability should be considered by the analyst when evaluating a
roundabout approach.
Exhibit 33-1
Observed Combinations of
Entry Flow and Conflicting
Flow During 1-min Periods of
Continuous Queuing: One-
Lane Entry Opposed by One
Circulating Lane
Source: Rodegerdts et al. (1 ).
Supplemental Guidance Chapter 33/Roundabouts: Supplemental

Exhibit 33-2
Continuous Queuing: Both
Lanes of Two-Lane Entry
Opposed by One Circulating
Lane
Exhibit 33-3
Continuous Queuing: Left
Lane of Two-Lane Entry
Opposed by Two Circulating
Lanes
Chapter 33/Roundabouts: Supplemental Supplemental Guidance

Exhibit 33-4
Continuous Queuing: Right
Lane of Two-Lane Entry
Opposed by Two Circulating
Lanes
LANE-USE ASSIGNMENT
Lane-use assignment is best determined by measuring lane use in the field
under the conditions being analyzed. In the absence of this information, default
values or estimates can be used. This section provides background on the process
by which an analyst can make an appropriate selection of a lane utilization
factor.
In general, several factors contribute to the assignment of traffic flow to each
lane:
Turning movement patterns 1. The assignment of turning movements to each lane (either as exclusive
greatly influence lane
assignments. lanes or as shared lanes) directly influences the assignment of traffic
volumes to each lane. Lane assignment is generally accomplished through
the use of signs and pavement markings that designate the lane use for
each lane. Multilane entries with no lane-use signing or pavement
markings may be assumed to operate with a shared left–through lane in
the left lane and a shared through–right lane in the right lane, although
field observations should be made to confirm the lane-use pattern of an
existing roundabout.
Dominant turning movements 2. Dominant turning movements may create de facto lane assignments for
may create de facto lanes. A
de facto lane is one designated which there is no advantage for drivers in using both lanes assigned to a
for multiple movements but given turning movement. For example, at an entry with left–through and
that may operate as an
exclusive lane because of a through–right lanes and a dominant left-turn movement, there may be no
dominant movement demand. advantage for through drivers in using the left lane. In addition, a lack of
A common example is a left–
through lane with a left-turn lane balance through the roundabout (e.g., two entry lanes but only one
flow rate that greatly exceeds downstream circulating lane or one downstream exit lane) can create de
the through flow rate.
facto lane-use assignments for a particular entry.

3. Destinations downstream of a roundabout may influence the lane choice Downstream destinations may
influence lane assignment.
at the roundabout entry. A downstream destination such as a freeway on-
ramp may increase use of the right entry lane, for example, even though
both lanes could be used.
4. The alignment of the lane relative to the circulatory roadway seems to Poor geometric alignment of
the entry may cause drivers to
influence the use of entry lanes where drivers can choose between lanes. avoid the left lane.
Some roundabouts have been designed with rather perpendicular entries
that have a natural alignment of the right entry lane into the left lane of
the circulatory roadway. Under this design, the left entry lane is naturally
aimed at the central island and is thus less comfortable and less desirable
for drivers. This phenomenon of poor path alignment, documented
elsewhere (2), may result in poor use of the left entry lane. Similarly,
poorly aligned multilane exits, where vehicles exiting in the inside lane
cross the path of vehicles exiting in the outside lane, may influence lane
use on upstream entries. In either case, the effect is most readily measured
in the field at existing roundabouts, and it should be avoided in the
design of new roundabouts.
5. Drivers may be uncertain about lane use when they use the roundabout, Unfamiliar drivers may
incorrectly select lanes for their
particularly at roundabouts without designated lane assignments intended movements.
approaching or circulating through the roundabout. This uncertainty may
contribute to the generally incorrect use of the right entry lane for left
turns, for example, because of a perceived or real difficulty in exiting from
the inside lane of the circulatory roadway. Proper signing and striping of
lane use on the approach and through the roundabout may reduce this
uncertainty, although it is likely to be present to some extent at multilane
roundabouts.
The first three factors described above are common to all intersections and
are accounted for in the assignment of turning-movement patterns to individual
lanes; the remaining two factors are unique to roundabouts. The fourth factor
should be addressed through proper alignment of the entry relative to the
circulatory roadway and thus may not need to be considered in the analysis of
new facilities. However, existing roundabouts may exhibit poor path alignment,
resulting in poor lane utilization. It may be possible to reduce the fifth factor
through proper design, particularly through lane-use arrows and striping. These
factors collectively make accurate estimation of lane utilization difficult, but it
can be measured at existing roundabouts.
For entries with two through lanes, limited field data suggest drivers Multilane roundabouts
generally exhibit a bias to the
generally have a bias for the right lane. For entries with two left-turn lanes (e.g., right lane except where a
left-turn-only and shared left–through–right lanes), limited field data suggest double left-turn movement is
present.
drivers have a bias for the left lane. Although no field observations have been
documented for entries with two right-turn lanes, experience at other types of
intersections with two right-turn lanes suggests drivers have a bias for the right
lane.

CAPACITY MODEL CALIBRATION

As discussed in Chapter 22, Roundabouts, the capacity model can be
calibrated by using one of two methods: using two parameters, the critical
headway tc and the follow-up headway tf, or using only the follow-up headway tf.
An example of calibration using two parameters was performed for
roundabouts in California (3). Field-measured values for critical headway and
follow-up headway were determined as follows:
• Critical headway:
o Single-lane roundabouts: 4.8 s;
o Multilane roundabouts, left lane: 4.7 s; and
o Multilane roundabouts, right lane: 4.4 s.
• Follow-up headway:
o Single-lane roundabouts: 2.5 s;
o Multilane roundabouts, left lane: 2.2 s; and
o Multilane roundabouts, right lane: 2.2 s.
By using these values and the expressions in Equation 22-21 through
Equation 22-23, the capacity equation for single-lane roundabouts can be
expressed as follows:
3,600 3,600
𝐴= = = 1,440
𝑡𝑓 2.5
𝑡𝑐 − (𝑡𝑓 ⁄2) 4.8 − (2.5⁄2)
𝐵= = = 1.0 × 10−3
3,600 3,600
−3 𝑣 )
𝑐𝑝𝑐𝑒 = 𝐴𝑒 (−𝐵𝑣𝑐 ) = 1,440𝑒 (−1.0×10 𝑐
Therefore, the model resulting from the use of California-specific data for
critical headway and follow-up time has a higher intercept, and thus higher
capacity, over its entire range than does the model based on the national study.
These equations replace the equations in Step 5 of the Chapter 22 methodology.
An example of calibration using only follow-up headway can be
demonstrated using data collected as part of a national study for the US-9/
Warren Street/Hudson Avenue/Glen Street intersection in Glen Falls, New York
(1). Field-measured values for follow-up headway for the five-legged
roundabout were determined as follows (rounded to the nearest 0.1 s):
• East leg: 2.9 s,
• Northwest leg: 2.8 s,
• South leg: 2.9 s,
• West leg: 2.7 s, and
• North leg: 2.8 s.
The mean value using unrounded values for follow-up time for the
intersection is 2.85 s. The intercept can therefore be calculated as follows:

3,600 3,600
𝐴= = = 1,260
𝑡𝑓 2.85
With this value for the intercept, the resulting capacity model is
−3 𝑣 )
𝑐𝑝𝑐𝑒 = 𝐴𝑒 (−𝐵𝑣𝑐 ) = 1,260𝑒 (−1.02×10 𝑐
The resulting model has a lower intercept than the national model. Based on
the observations of each approach of this intersection under queued conditions
from the national study, this site-specific model has a better goodness of fit than
the national model (an improvement in the root mean squared error from 164 to
126 pc/h). Variation in driver behavior between individual drivers or from
minute to minute makes eliminating prediction error impossible, but calibration
can improve the accuracy of the prediction.

3. EXAMPLE PROBLEMS
This section illustrates the application of the roundabout methodology

through the two example problems listed in Exhibit 33-5.

Roundabout Example Problem Description Application
Problems 1 Single-lane roundabout with bypass lanes Operational analysis
2 Multilane roundabout Operational analysis
EXAMPLE PROBLEM 1: SINGLE-LANE ROUNDABOUT WITH BYPASS

LANES
This is an example of an
The Facts
operational analysis. It uses The following data are available to describe the traffic and geometric
traffic data and geometric
characteristics to determine characteristics of this location:
capacities, control delay, and
LOS. • Four legs,
• One-lane entries on each leg,
• A westbound right-turn bypass lane that yields to exiting vehicles,
• A southbound right-turn bypass lane that forms its own lane adjacent to
exiting vehicles,
• Percentage heavy vehicles for all movements = 2%,
• Demand volumes and lane configurations as shown in Exhibit 33-6, and
• 50 p/h across the south leg and negligible pedestrian activity across the
other three legs.
Exhibit 33-6
Example Problem 1: Demand
Volumes and Lane
Configurations
Comments
Example Problems Chapter 33/Roundabouts: Supplemental


Each turning-movement volume given in the problem is converted to a
demand flow rate by dividing by the peak hour factor. As an example, the
northbound left-turn volume is converted to a flow rate as follows by using
Equation 22-8:
𝑉𝑁𝐵𝐿 105
𝑣𝑁𝐵𝐿 = = = 112 pc/h
𝑃𝐻𝐹 0.94
Step 2: Adjust Flow Rates for Heavy Vehicles

The flow rate for each movement may be adjusted to account for vehicle
stream characteristics by using Equations 22-9 and 22-10 as follows (northbound
left turn illustrated):
1 1
𝑓𝐻𝑉 = = = 0.980
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.02(2 − 1)
𝑣𝑁𝐵𝐿 112
𝑣𝑁𝐵𝐿,𝑝𝑐𝑒 = = = 114 pc/h
𝑓𝐻𝑉 0.980
The resulting adjusted flow rates for all movements, accounting for Steps 1
and 2, are therefore computed as shown in Exhibit 33-7:
Exhibit 33-7
Example Problem 1:
Adjusted Flow Rates
Step 3: Determine Circulating and Exiting Flow Rates

The circulating and exiting flows are calculated for each leg. For the south
leg (northbound entry), the circulating flow is calculated by using the process
illustrated by Equation 22-11 as follows:
𝑣𝑐,𝑁𝐵,𝑝𝑐𝑒 = 𝑣𝑊𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝑈,𝑝𝑐𝑒
𝑣𝑐,𝑁𝐵,𝑝𝑐𝑒 = 21 + 190 + 21 + 304 + 206 + 54 = 796 pc/h
Similarly, vc,SB,pce = 769 pc/h; vc,EB,pce = 487 pc/h; and vc,WB,pce = 655 pc/h.
For this problem, one exit flow rate is needed: the northbound exit flow rate,
which serves as the conflicting flow for the westbound bypass lane. Because all
westbound right turns are assumed to use the bypass lane, they are excluded
from the conflicting exit flow by using the process illustrated by Equation 22-12
as follows:
Chapter 33/Roundabouts: Supplemental Example Problems

𝑣𝑒𝑥,𝑁𝐵,𝑝𝑐𝑒 = 𝑣𝑆𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝑁𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝑊𝐵𝑅,𝑝𝑐𝑒 − 𝑣𝑊𝐵𝑅,𝑏𝑦𝑝𝑎𝑠𝑠,𝑝𝑐𝑒

𝑣𝑒𝑥,𝑁𝐵,𝑝𝑐𝑒 = 21 + 206 + 227 + 662 − 662 = 454 pc/h
Step 4: Determine Entry Flow Rates by Lane

The entry flow rate is calculated by summing the movement flow rates that
enter the roundabout (without using a bypass lane). Because this is a single-lane
roundabout, no lane-use calculations are needed.
The entry flow rates are calculated as follows, assuming all right-turn
volumes on the westbound and southbound approaches use the bypass lane
provided and not the entry:
𝑣𝑒,𝑁𝐵,𝑝𝑐𝑒 = 𝑣𝑁𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝑁𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝑁𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝑁𝐵𝑅,𝑒,𝑝𝑐𝑒
𝑣𝑒,𝑁𝐵,𝑝𝑐𝑒 = 33 + 114 + 227 + 54 = 428 pc/h
𝑣𝑒,𝑆𝐵,𝑝𝑐𝑒 = 𝑣𝑆𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝑅,𝑒,𝑝𝑐𝑒

𝑣𝑒,𝑆𝐵,𝑝𝑐𝑒 = 21 + 190 + 103 + 0 = 314 pc/h
𝑣𝑒,𝐸𝐵,𝑝𝑐𝑒 = 𝑣𝐸𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝑅,𝑒,𝑝𝑐𝑒

𝑣𝑒,𝐸𝐵,𝑝𝑐𝑒 = 54 + 206 + 304 + 92 = 656 pc/h
𝑣𝑒,𝑊𝐵,𝑝𝑐𝑒 = 𝑣𝑊𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝑊𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝑊𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝑊𝐵𝑅,𝑒,𝑝𝑐𝑒

𝑣𝑒,𝑊𝐵,𝑝𝑐𝑒 = 21 + 119 + 428 + 0 = 568 pc/h
Step 5: Determine the Capacity of Each Entry Lane and Bypass Lane as
Appropriate in Passenger Car Equivalents
By using the single-lane capacity equation (Equation 22-1), the capacity for
each entry lane is given as follows:
−3 )𝑣 −3 )(796)
𝑐𝑝𝑐𝑒,𝑁𝐵 = 1,380𝑒 (−1.02×10 𝑐,𝑝𝑐𝑒,𝑁𝐵 = 1,380𝑒 (−1.02×10 = 613 pc/h
(−1.02×10−3 )𝑣𝑐,𝑝𝑐𝑒,𝑆𝐵 (−1.02×10−3 )(769)
𝑐𝑝𝑐𝑒,𝑆𝐵 = 1,380𝑒 = 1,380𝑒 = 630 pc/h
(−1.02×10−3 )𝑣𝑐,𝑝𝑐𝑒,𝐸𝐵 (−1.02×10−3 )(487)
𝑐𝑝𝑐𝑒,𝐸𝐵 = 1,380𝑒 = 1,380𝑒 = 840 pc/h
(−1.02×10−3 )𝑣𝑐,𝑝𝑐𝑒,𝑊𝐵 (−1.02×10−3 )(655)
𝑐𝑝𝑐𝑒,𝑊𝐵 = 1,380𝑒 = 1,380𝑒 = 708 pc/h
By using the equation for a bypass lane opposed by a single exit lane
(Equation 22-6), the capacity for the westbound bypass lane is given as follows:
−3 )𝑣 −3 )(454)
𝑐bypass,𝑝𝑐𝑒,𝑊𝐵 = 1,380𝑒 (−1.02×10 𝑒𝑥,𝑝𝑐𝑒,𝑁𝐵 = 1,380𝑒 (−1.02×10 = 868 pc/h
Step 6: Determine Pedestrian Impedance to Vehicles

The south leg (northbound entry) has a conflicting pedestrian flow rate, nped,
of 50 p/h. The pedestrian impedance factor is calculated by using Exhibit 22-18 as
follows:
𝑓𝑝𝑒𝑑 = 1 − 0.000137𝑛𝑝𝑒𝑑 = 1 − 0.000137(50) = 0.993
Because the other legs and bypass lanes have negligible pedestrian activity
(nped = 0), they have fped = 1.

Step 7: Convert Lane Flow Rates and Capacities into Vehicles per Hour
The capacity for a given lane is converted back to vehicles by first
determining the heavy-vehicle adjustment factor for the lane and then
multiplying it by the capacity in passenger car equivalents (Equation 22-14). For
this example, because all turning movements on each entry have the same fHV,
each entry will also have the same fHV, 0.980. The capacities for each of the entries
are also adjusted by the pedestrian impedance factor.
𝑐𝑁𝐵 = 𝑐𝑝𝑐𝑒,𝑁𝐵 𝑓𝐻𝑉,𝑒,𝑁𝐵 𝑓𝑝𝑒𝑑 = (613)(0.980)(0.993) = 597 veh/h
𝑐𝑆𝐵 = 𝑐𝑝𝑐𝑒,𝑆𝐵 𝑓𝐻𝑉,𝑒,𝑆𝐵 𝑓𝑝𝑒𝑑 = (630)(0.980)(1) = 618 veh/h
𝑐𝐸𝐵 = 𝑐𝑝𝑐𝑒,𝐸𝐵 𝑓𝐻𝑉,𝑒,𝐸𝐵 𝑓𝑝𝑒𝑑 = (840)(0.980)(1) = 824 veh/h
𝑐𝑊𝐵 = 𝑐𝑝𝑐𝑒,𝑊𝐵 𝑓𝐻𝑉,𝑒,𝑊𝐵 𝑓𝑝𝑒𝑑 = (708)(0.980)(1) = 694 veh/h
𝑐bypass,𝑊𝐵 = 𝑐bypass,𝑝𝑐𝑒,𝑊𝐵 𝑓𝐻𝑉,𝑒,𝑊𝐵 𝑓𝑝𝑒𝑑 = (868)(0.980)(1) = 851 veh/h
Calculations for the entry flow rates are as follows (Equation 22-13):
𝑣𝑁𝐵 = 𝑣𝑝𝑐𝑒,𝑁𝐵 𝑓𝐻𝑉,𝑒,𝑁𝐵 = (428)(0.980) = 420 veh/h
𝑣𝑆𝐵 = 𝑣𝑝𝑐𝑒,𝑆𝐵 𝑓𝐻𝑉,𝑒,𝑆𝐵 = (314)(0.980) = 308 veh/h
𝑣𝐸𝐵 = 𝑣𝑝𝑐𝑒,𝐸𝐵 𝑓𝐻𝑉,𝑒,𝐸𝐵 = (656)(0.980) = 643 veh/h
𝑣𝑊𝐵 = 𝑣𝑝𝑐𝑒,𝑊𝐵 𝑓𝐻𝑉,𝑒,𝑊𝐵 = (568)(0.980) = 557 veh/h
𝑣bypass,𝑊𝐵 = 𝑣bypass,𝑝𝑐𝑒,𝑊𝐵 𝑓𝐻𝑉,𝑒,𝑊𝐵 = (662)(0.980) = 649 veh/h
Step 8: Compute the Volume-to-Capacity Ratio for Each Lane

The volume-to-capacity ratios for each entry lane are calculated from
420
𝑥𝑁𝐵 = = 0.70
597
308
𝑥𝑆𝐵 = = 0.50
618
643
𝑥𝐸𝐵 = = 0.78
824
557
𝑥𝑊𝐵 = = 0.80
694
649
𝑥bypass,𝑊𝐵 = = 0.76
851
Step 9: Compute the Average Control Delay for Each Lane

The control delay for the northbound entry lane is computed from Equation
22-17 as follows:
3,600
3,600 √ ( ) 0.70
𝑑𝑁𝐵 = + 900(0.25) [0.70 − 1 + (0.70 − 1)2 + 597 ]
597 450(0.25)
+5(min[0.70,1])
𝑑𝑁𝐵 = 22.6 s/veh

Similarly, dSB = 14.0 s; dbypass,SB = 0 s (assumed); dEB = 22.0 s; dWB = 26.8 s; and
dbypass,WB = 20.2 s.
Step 10: Determine LOS for Each Lane on Each Approach
From Exhibit 22-8, the level of service (LOS) for each lane is determined as
shown in Exhibit 33-8:
Exhibit 33-8 Lane Control Delay (s/veh) LOS

Example Problem 1: Northbound entry 22.6 C
LOS by Lane Southbound entry 14.0 B
Southbound bypass lane 0 (assumed) A
Eastbound entry 22.0 C
Westbound entry 26.8 D
Westbound bypass lane 20.2 C
Step 11: Compute the Average Control Delay and Determine LOS for
Each Approach and the Roundabout as a Whole
The control delays for the northbound and eastbound approaches are equal
to the control delay for the entry lanes, as both of these approaches have only one
lane. On the basis of Exhibit 22-8, these approaches are both assigned LOS C.
The control delay calculations for the westbound and southbound
approaches include the effects of their bypass lanes as follows (Equation 22-18):
(26.8)(557) + (20.2)(649)
𝑑𝑊𝐵 = = 23.3 s/veh
557 + 649
(14.0)(308) + (0.0)(617)
𝑑𝑆𝐵 = = 4.7 s/veh
308 + 617
On the basis of Exhibit 22-8, these approaches are respectively assigned LOS
C and LOS A.
Similarly, intersection control delay is computed as follows (Equation 22-19):
(22.6)(420) + (4.7)(925) + (22.0)(643) + (23.3)(1,206)
𝑑intersection =
420 + 925 + 643 + 1,206
𝑑intersection = 17.5 s/veh
On the basis of Exhibit 22-8, the intersection is assigned LOS C.
Step 12: Compute 95th Percentile Queues for Each Lane

The 95th percentile queue is computed for each lane. An example calculation
for the northbound entry is given as follows (Equation 22-20):
3,600
√ ( 597 ) 0.70 597
𝑄95,𝑁𝐵 = 900(0.25) [0.70 − 1 + (1 − 0.70)2 + ]( )
150(0.25) 3,600
𝑄95,𝑁𝐵 = 5.7 veh

For design purposes, this value is typically rounded up to the nearest
vehicle, which for this case would be six vehicles.
Similarly, Q95,SB = 2.8 veh; Q95,EB = 7.9 veh; Q95,WB = 8.2 veh; and Q95,bypass,WB = 7.4
veh.

Discussion
The results indicate the overall roundabout is operating at LOS C. However, The analyst should be careful
not to mask key operational
one lane (the westbound entry) is operating at LOS D. If, for example, the performance issues by
performance standard for this intersection was LOS C, this entry would not meet reporting overall intersection
performance without also
the standard, even though the overall intersection meets the standard. For these reporting the performance of
reasons, the analyst should consider reporting volume-to-capacity ratios, control each lane, or at least the
worst-performing lane.
delay, and queue lengths for each lane, in addition to the aggregated measures,
for a more complete picture of operational performance.
EXAMPLE PROBLEM 2: MULTILANE ROUNDABOUT

Example Problem 2 is also an
The Facts example of an operational
The following data are available to describe the traffic and geometric analysis, despite the fact that
lane utilization data are
characteristics of this location: unknown and must be
assumed.
• Percentage heavy vehicles for eastbound and westbound movements =
5%,
• Percentage heavy vehicles for northbound and southbound movements =
2%,
• Negligible pedestrian activity, and
• Volumes and lane configurations as shown in Exhibit 33-9.
Exhibit 33-9
Example Problem 2: Demand
Volumes and Lane
Configurations
Comments
Lane use is not specified for the eastbound and westbound approaches;
therefore, the percentage flow in the right lane is assumed to be 53%, as specified
in Exhibit 22-9.

Each turning-movement demand volume given in the problem is converted
to a demand flow rate by dividing by the peak hour factor. As an example, the
eastbound-left demand volume is converted to a demand flow rate by using

𝑉𝐸𝐵𝐿 230
𝑣𝐸𝐵𝐿 = = = 242 veh/h
𝑃𝐻𝐹 0.95
Step 2: Adjust Flow Rates for Heavy Vehicles

The heavy-vehicle adjustment factor for the eastbound and westbound
movements is calculated by using Equation 22-10 as follows:
1 1
𝑓𝐻𝑉 = = = 0.952
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.05(2 − 1)
Similarly, the heavy-vehicle adjustment factor for the northbound and
southbound movements is calculated as follows:
1 1
𝑓𝐻𝑉 = = = 0.980
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.02(2 − 1)
This factor is applied to each movement by using Equation 22-9 as follows
(eastbound left turn illustrated):
𝑣𝐸𝐵𝐿 242
𝑣𝐸𝐵𝐿,𝑝𝑐𝑒 = = = 254 pc/h
𝑓𝐻𝑉 0.952
The resulting adjusted flow rates for all movements, accounting for Steps 1
and 2, are therefore as shown in Exhibit 33-10:
Exhibit 33-10
Example Problem 2:
Adjusted Flow Rates
Step 3: Determine Circulating and Exiting Flow Rates

For this problem, only circulating flows need to be calculated for each leg.
For the west leg (eastbound entry), the circulating flow is calculated by using the
process illustrated by Equation 22-11 as follows:
𝑣𝑐,𝐸𝐵,𝑝𝑐𝑒 = 𝑣𝑁𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝑊𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝑊𝐵𝑈,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝑈,𝑝𝑐𝑒
𝑣𝑐,𝐸𝐵,𝑝𝑐𝑒 = 0 + 442 + 0 + 64 + 258 + 0 = 764 pc/h
Similarly, vc,WB,pce = 372 pc/h; vc,NB,pce = 976 pc/h; and vc,SB,pce = 772 pc/h.
Step 4: Determine Entry Flow Rates by Lane

The entry flow rate is calculated by summing up the movement flow rates
that enter the roundabout. This problem presents four unique cases.
• Northbound: The northbound entry has only one lane. Therefore, the entry
flow is simply the sum of the movements, or 54 + 64 + 129 = 247 pc/h.

• Southbound: The southbound entry has two lanes: a shared through–left

lane and a right-turn-only lane. Therefore, the flow rate in the right lane is
simply the right-turn movement flow, or 429 pc/h, and the flow rate in the
left lane is the sum of the left-turn and through movements, or 258 + 64 =
322 pc/h.
• Eastbound: The eastbound entry has shared left–through and through–
right lanes. A check is needed to determine whether any de facto lanes are
in effect. These checks are as follows:
o Left lane: The left-turn flow rate, 254 pc/h, is less than the sum of the
through and right-turn flow rates, 464 + 88 = 552 pc/h. Therefore,
some of the through volume is assumed to use the left lane, and no de
facto left-turn lane condition is present.
o Right lane: The right-turn flow rate, 88 pc/h, is less than the sum of the
left-turn and through flow rates, 254 + 464 = 718 pc/h. Therefore, some
of the through volume is assumed to use the right lane, and no de
facto right-turn lane condition is present.
The total entry flow (254 + 464 + 88 = 806 pc/h) is therefore distributed
over the two lanes, with flow biased to the right lane by using the
assumed lane-use factor identified previously:
o Right lane: (806)(0.53) = 427 pc/h, and
o Left lane: 806 – 427 = 379 pc/h.
• Westbound: The westbound entry also has shared left–through and
through–right lanes, and so a similar check is needed for de facto lanes.
The left-turn flow rate, 442 pc/h, is greater than the sum of the through
and right-turn flow rates, 276 + 100 = 376 pc/h. Therefore, the left lane is
assumed to operate as a de facto left-turn lane. Therefore, the left-lane
flow rate is equal to the left-turn flow rate, or 442 pc/h, and the right-lane
flow rate is equal to the sum of the through- and right-turn-movement
flow rates, or 376 pc/h.
Step 5: Determine the Capacity of Each Entry Lane and Bypass Lane as
Appropriate in Passenger Car Equivalents
The capacity calculations for each approach are calculated as follows:
• Northbound: The northbound entry is a single-lane entry opposed by two
circulating lanes. Therefore, Equation 22-3 is used as follows:
−3 )(976)
𝑐𝑝𝑐𝑒,𝑁𝐵 = 1,420𝑒 (−0.85×10 = 619 pc/h
• Southbound: The southbound entry is a two-lane entry opposed by two
circulating lanes. Therefore, Equation 22-4 is used for the right lane, and
Equation 22-5 is used for the left lane:
−3 )(772)
𝑐𝑝𝑐𝑒,𝑆𝐵,𝑅 = 1,420𝑒 (−0.85×10 = 737 pc/h
(−0.92×10−3 )(772)
𝑐𝑝𝑐𝑒,𝑆𝐵,𝐿 = 1,350𝑒 = 664 pc/h

• Eastbound: The eastbound entry is a two-lane entry opposed by one

circulating lane. Therefore, the capacity for each lane is calculated by
using Equation 22-2 as follows:
−3 )(764)
𝑐𝑝𝑐𝑒,𝐸𝐵,𝑅 = 𝑐𝑝𝑐𝑒,𝐸𝐵,𝐿 = 1,420𝑒 (−0.91×10 = 709 pc/h
• Westbound: The westbound entry is also a two-lane entry opposed by one
circulating lane, so its capacity calculation is similar to that for the
eastbound entry:
−3 )(372)
𝑐𝑝𝑐𝑒,𝑊𝐵 = 1,420𝑒 (−0.91×10 = 1,012 pc/h
There are no bypass lanes in this example problem.
Step 6: Determine Pedestrian Impedance to Vehicles

For this problem pedestrians have been assumed to be negligible, so no
impedance calculations are performed.
Step 7: Convert Lane Flow Rates and Capacities into Vehicles per Hour
The capacity for a given lane is converted back to vehicles by first
determining the heavy-vehicle adjustment factor for the lane and then
multiplying it by the capacity in passenger car equivalents (Equation 22-14). For
this example, because all turning movements on the eastbound and westbound
entries have the same fHV, each of the lanes on the eastbound and westbound
entries can be assumed to have the same fHV, 0.952.
𝑐𝐸𝐵,𝑅 = 𝑐𝑝𝑐𝑒,𝐸𝐵,𝑅 𝑓𝐻𝑉,𝑒,𝐸𝐵 = (709)(0.952) = 675 veh/h
Similarly, cEB,L = 675 veh/h; cWB,L = 964 veh/h; and cWB,R = 964 veh/h.
Because all turning movements on the northbound and southbound entries
have the same fHV, each of the lanes on those entries can be assumed to have the
same fHV, 0.980.
𝑐𝑁𝐵 = 𝑐𝑝𝑐𝑒,𝑁𝐵 𝑓𝐻𝑉,𝑒,𝑁𝐵 = (619)(0.980) = 607 veh/h
Similarly, cSB,L = 651 veh/h, and cSB,R = 723 veh/h.
Calculations for the entry flow rates are as follows (Equation 22-13):
𝑣𝐸𝐵,𝑅 = 𝑣𝑝𝑐𝑒,𝐸𝐵,𝑅 𝑓𝐻𝑉,𝑒,𝐸𝐵 = (427)(0.952) = 407 veh/h
𝑣𝑁𝐵 = 𝑣𝑝𝑐𝑒,𝑁𝐵 𝑓𝐻𝑉,𝑒,𝑁𝐵 = (247)(0.980) = 242 veh/h
Similarly, vEB,L = 361 veh/h; vWB,L = 421 veh/h; vWB,R = 358 veh/h; vSB,L = 316
veh/h; and vSB,R = 421 veh/h.


The volume-to-capacity ratio for each lane is calculated from Equation 22-16
as follows:
242
𝑥𝑁𝐵 = = 0.40
607
316
𝑥𝑆𝐵,𝐿 = = 0.48
651
421
𝑥𝑆𝐵,𝑅 = = 0.58
723
361
𝑥𝐸𝐵,𝐿 = = 0.53
675
407
𝑥𝐸𝐵,𝑅 = = 0.60
675
421
𝑥𝑊𝐵,𝐿 = = 0.44
964
358
𝑥𝑊𝐵,𝑅 = = 0.37
964
Step 9: Compute the Average Control Delay for Each Lane

The control delay for the northbound entry lane is computed from Equation
22-17 as follows:
2
3,600 242
3,600 242 √ 242 ( 607 ) 607
𝑑𝑁𝐵 = + 900(0.25) [ −1+ ( − 1) + ]
607 607 607 450(0.25)
242
+5 (min [ , 1])
607
𝑑𝑁𝐵 = 11.8 s/veh
Similarly, dSB,L = 13.0 s/veh; dSB,R = 14.6 s/veh; dEB,L = 14.0 s/veh; dEB,R = 16.1
s/veh; dWB,L = 8.8 s/veh; and dWB,R = 7.8 s/veh.
Step 10: Determine LOS for Each Lane on Each Approach

On the basis of Exhibit 22-8, the LOS for each lane is determined as shown in
Exhibit 33-11:
Critical Lane Control Delay (s/veh) LOS Exhibit 33-11

Northbound entry 11.8 B Example Problem 2:
Southbound left lane 13.0 B LOS by Lane
Southbound right lane 14.6 B
Eastbound left lane 14.0 B
Eastbound right lane 16.1 C
Westbound left lane 8.8 A
Westbound right lane 7.8 A
Step 11: Compute the Average Control Delay and Determine LOS for
Each Approach and the Roundabout as a Whole
The control delay for the northbound approaches is equal to the control
delay for the entry lane, 11.8 s, as the approach has only one lane. The control
delays for the other approaches are as follows (Equation 22-18):

(13.0)(316) + (14.6)(421)
𝑑𝑆𝐵 = = 13.9 s/veh
316 + 421
(14.0)(361) + (16.1)(407)
𝑑𝐸𝐵 = = 15.1 s/veh
361 + 407
(8.8)(421) + (7.8)(358)
𝑑𝑊𝐵 = = 8.3 s/veh
421 + 358
On the basis of Exhibit 22-8, these approaches are respectively assigned LOS
B, LOS B, LOS C, and LOS A.
Similarly, control delay for the intersection is computed as follows (Equation
22-19):
(11.8)(242) + (13.9)(736) + (15.1)(768) + (8.3)(779)
𝑑intersection =
242 + 736 + 768 + 779
𝑑intersection = 12.3 s/veh
On the basis of Exhibit 22-8, the intersection is assigned LOS B.

The 95th percentile queue is computed for each lane. An example calculation
for the northbound entry is given as follows (Equation 22-20):
3,600 242
242 √ 242 2 ( 607 ) (607) 607
𝑄95,𝑁𝐵 = 900(0.25) [ − 1 + (1 − ) + ]( )
607 607 150(0.25) 3,600
𝑄95,𝑁𝐵 = 1.9 veh

For design purposes, this value is typically rounded up to the nearest
vehicle, in this case two vehicles.
Discussion
The results indicate the intersection as a whole operates at LOS B on the basis
of control delay during the peak 15 min of the analysis hour. However, the
eastbound approach operates at LOS C, as does the right lane of the eastbound
approach. The analyst should consider reporting both the overall performance
and those of the individual lanes to provide a more complete picture of
operational performance.

INTRODUCTION
This section provides capacity adjustment factors (CAFs) for roundabout
approaches to account for the presence of connected and automated vehicles
(CAVs) in the traffic stream. Although CAVs are still a developing technology,
transportation agencies have an immediate need as part of their long-range
planning efforts to account for CAVs’ potential ability to increase existing
roadways’ throughput.
or both, these approaches are currently infeasible given the absence of level 4 and
5 CAVs in the traffic stream. Instead, uncalibrated simulation modeling was
conducted using CAV logic developed for the Federal Highway Administration.
Details about this modeling are available in a paper (5) available online in HCM
Volume 4 (hcmvolume4.org) in the Technical Reference Library section for
Chapter 33.
All exhibits in this section assume that the CAV market penetration rate is a
global input for the entire intersection. The planning-level adjustment factors
currently do not support varying the percentage of CAVs on a per-lane or per-
approach basis.
This chapter’s adjustments for CAVs were developed for roundabout
intersections. No specific simulation was completed for other YIELD-controlled
movements, including those that may be present at alternative intersections or
interchanges. The adjustments provided in this section may be used to
approximate these effects in the absence of other data. There are no CAV
adjustments at present for STOP-controlled intersection approaches.
CONCEPTS
CAV Technology
Chapter 33/Roundabouts: Supplemental Connected and Automated Vehicles

benefits. However, it is not clear how connectivity affects car following

and driver behavior and subsequently intersection movement capacity.
system (7). In particular, the cooperative adaptive cruise control (CACC)
feature enabled by vehicle-to-vehicle communication allows CAVs to
safely accept shorter gaps than possible by either human-driven vehicles
or automated vehicles using adaptive cruise control only.
Factors Influencing Roundabout Capacity with CAVs in the Traffic

Stream
The capacity of a given roundabout approach reflects how many vehicles can
reasonably be expected to enter the roundabout in a given 15-min period.
Capacity is a function of the distribution of gaps in the circulating traffic stream,
driver judgment in selecting gaps to enter the roundabout, and the follow-up
headways required by each driver in a queue. CAVs may improve the capacity
of roundabout approaches through their potential to safely accept smaller gaps
in traffic than human drivers.
The proportion of the traffic stream that is composed of CAVs will influence
the achievable capacity increase. The greater the proportion of CAVs in the traffic
stream, the more frequently the benefits of connectivity can be realized, because
more vehicles will be able to accept smaller gaps safely.
Assumptions Affecting CAV Ability to Provide Higher Capacities

Critical Headway and Follow-up Headway
capacity benefit. A key assumption used in developing this section’s CAFs was
the yielding logic used to determine whether a CAV accepted or rejected gaps to
(a) conflicting human-driven vehicles and (b) conflicting CAVs, which in turn
determines the resulting critical and follow-up headways for both the two conflict
Connected and Automated Vehicles Chapter 33/Roundabouts: Supplemental

scenarios. Factors that could affect CAVs’ eventual gap-acceptance behavior

include:
• Legal or regulatory requirements dictating a minimum headway.
use more conservative gap-acceptance behavior than strictly needed for
safety.
• Passenger lack of trust concerns on the part of vehicle owners related to
accepting a short gap to a conflicting vehicle.
The simulation modeling that developed this section’s CAFs incorporated
the following assumptions related to gap acceptance (5):
• CAV capability. The modeled CAVs had vehicle-to-vehicle communication
abilities and a working CACC system.
• Human-driven vehicle capability. The operation of human-driven vehicles
was calibrated by comparing the capacity of each entry lane under
different conflicting flow rates with the values calculated by the Chapter
22 method. Separate networks were calibrated for single- and double-lane
approaches.
• CAV gap-acceptance behavior. When the conflicting vehicle was a human-
driven vehicle, a subject CAV relied on adaptive cruise control to regulate
the driving behavior. When the conflicting vehicle was a CAV, the subject
CAV used CACC to receive the conflicting vehicle’s location and speed
and to determine whether to accept or reject the gap.
System Reliability
The ability of CAVs to safely operate with short intervehicle gaps and critical
headways requires, among other things, low communications latency (i.e.,
information can be quickly exchanged between vehicles and acted upon), vehicle
manufacturers to build vehicles with reliable components, vehicle owners to
promptly repair components if they do break, and regulatory agencies to provide
adequate bandwidth for vehicle-to-vehicle communication. Consistent with other
base-condition assumptions in the roundabout methodology (e.g., good weather),
a base assumption for CAV analysis is that all necessary communication
elements are in place and working with a high degree of reliability.
Traffic Stream Composition

A key assumption an analyst will need to make when performing a CAV
analysis is the percentage of CAVs that will be in the traffic stream during the
analysis year(s). Once CAVs become available to consumers, it may take many
years for the vehicle fleet to transition to an all-CAV fleet. In 2018, the average
age of light cars and trucks in the United States was just under 12 years (8), and it
takes even longer for the national fleet to turn over. Furthermore, based on past
airbags, and hybrid vehicles, many people may not choose a CAV the first or

prior automotive technologies.
The simulation modeling that developed this section’s CAFs assumed a
traffic stream consisting of 100% passenger cars. The percentage of CAVs in the
traffic stream was varied from 0% to 100% in 20% increments. Analysts should
consider the latest available information about CAV adoption rates and the
effects of CAV usage on travel demand when performing an analysis of CAV
effects on roundabout capacity.
Addressing Uncertain Assumptions in a CAV Analysis

other factors, none of which are known with great certainty. Adding assumptions
analysis inputs.
Because of this uncertainty, it is recommended that the CAV CAFs presented
in this section be applied to the evaluation of “what if” scenarios, rather than being
taken as the final word on what will happen once CAVs become widespread. In
particular, the analyst should consider:
• What if the minimum gap-acceptance value permitted by technology, regulation,
or policy, or the average value produced by different vehicles’ user settings, is
longer than the modeling assumed? In this case, the capacity increase would
be less than predicted by the CAV CAFs.
CAV-capable vehicles.
• How quickly will CAV technology become available and adopted, and how will
questions will determine the assumed volume and proportion of CAVs in
the traffic stream, along with the assumed CAF.
CAPACITY ADJUSTMENT FACTORS

This subsection provides adjustment factors for the capacity relationships for
single-lane and multilane roundabouts to account for the presence of CAVs in the
traffic stream. CAV adjustment values were derived from microsimulation using
assumptions based on current knowledge; CAV adjustment values were not field
validated due to the current lack of CAV market penetration in the field.
Equation 33-1 shows the general form of the roundabout entry capacity
models presented in Equations 22-1 through 22-5 in Chapter 22, where A is a
parameter that controls the intercept of the capacity curve and B is a parameter
that controls the slope of the curve. The pair of adjustment factors, fA and fB, used
Connected and Automated Vehicles Chapter 33/Roundabouts: Supplemental

in Equation 33-2 adjust A and B and thereby determine an adjusted entry lane
capacity ce,adj,pce reflecting the presence of CAVs. All else being equal, an increase
in A or a decrease in B will increase entry lane capacity.
𝑐𝑒,𝑝𝑐𝑒 = 𝐴𝑒 −𝐵𝑣𝑐,𝑝𝑐𝑒 Equation 33-1
𝑐𝑒,𝑎𝑑𝑗,𝑝𝑐𝑒 = 𝑓𝐴 𝐴𝑒 −𝑓𝐵𝐵𝑣𝑐,𝑝𝑐𝑒 Equation 33-2
where
ce,pce = entry lane capacity, adjusted for heavy vehicles (pc/h);
A = intercept parameter, from Exhibit 33-12;
B = slope parameter, from Exhibit 33-12;
vc,pce = conflicting flow rate (pc/h);
ce,adj,pce = entry lane capacity, adjusted for CAVs and heavy vehicles (pc/h);
fA = adjustment factor for the intercept parameter, from Exhibit 33-13; and
fB = adjustment factor for the intercept parameter, from Exhibit 33-13.
To determine the CAV-adjusted capacity, first identify the values of A and B

from Exhibit 33-12 for the appropriate combination of number of entry lanes and
number of conflicting circulating lanes. Next, identify the values of fA and fB from
Exhibit 33-13 for the combination of subject entry lane type and proportion of
CAVs in the traffic stream. Finally, apply these values in Equation 33-2 to
determine the subject entry lane’s adjusted capacity.
Entry Lane Type A B Exhibit 33-12

One-lane entry conflicted by one circulating lane 1,380 1.02×10−3 Roundabout Entry Lane
Two-lane entry conflicted by one circulating lane (both entry lanes) 1,420 0.91×10−3 Capacity Model Parameters
(without CAVs)
One-lane entry conflicted by two circulating lanes 1,420 0.85×10−3
Two-lane entry conflicting by two circulating lanes (right entry lane) 1,420 0.85×10−3
Two-lane entry conflicting by two circulating lanes (left entry lane) 1,350 0.92×10−3
Source: Equations 22-1 through 22-5.
1-Lane Entry 2-Lane Entry Exhibit 33-13

1 Circulating 2 Circulating 2 Circulating Capacity Adjustment Factors
Proportion of 1 Circulating 2 Circulating Lane, Lanes, Lanes, for CAVs for Roundabouts
CAVs in Lane Lanesa Both Lanesa Left Lane Right Lane
Traffic Stream fA fB fA fB fA fB fA fB fA fB
0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
20 1.05 0.99 1.03 0.99 1.05 0.99 1.03 0.99 1.05 0.96
40 1.12 0.97 1.08 0.96 1.12 0.97 1.08 0.96 1.12 0.93
60 1.22 0.94 1.18 0.92 1.22 0.94 1.18 0.92 1.20 0.87
80 1.29 0.90 1.28 0.89 1.29 0.90 1.28 0.89 1.27 0.84
100 1.35 0.85 1.38 0.85 1.35 0.85 1.38 0.85 1.34 0.80
a
Notes: These cases were not specifically analyzed in the research and thus are suggested approximations.
CAV = connected and automated vehicle, defined here as a vehicle with an operating cooperative adaptive
Assumptions: Human-driven vehicles operate with average gaps calibrated to the entry lane capacity given
by Chapter 22.

5. REFERENCES
Some of these references can 1. Rodegerdts, L. A., A. Malinge, P. S. Marnell, S. G. Beaird, M. J. Kittelson, and
Reference Library in Volume 4. Y. S. Mereszczak. Assessment of Roundabout Capacity Models for the Highway
Capacity Manual: Volume 2 of Accelerating Roundabout Implementation in the
United States. Report FHWA-SA-15-070. Federal Highway Administration,
Washington, D.C., Sept. 2015.
3. Tian, Z. Z., F. Xu, L. A. Rodegerdts, W. E. Scarbrough, B. L. Ray, W. E.
Bishop, T. C. Ferrara, and S. Mam. Roundabout Geometric Design Guidance.
Report No. F/CA/RI-2006/13. Division of Research and Innovation, California
Department of Transportation, Sacramento, Calif., June 2007.
5. Jang, Q., B. Schroeder, J. Ma, L. Rodegerdts, B. Cesme, A. Bibeka, and A.
Morgan. Developing Highway Capacity Manual Capacity Adjustment Factors
for Connected and Automated Traffic on Roundabouts. Working paper.
2020.
Oct. 2013.
7. Krechmer, D., K. Blizzard, M.G. Cheung, R. Campbell, V. Alexiadis, J. Hyde,
June 2016.
8. Davis, S.C., and R.G. Boundy. Transportation Energy Data Book, Edition 37.
Tenn., Aug. 2019.
2019.
References Chapter 33/Roundabouts: Supplemental

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 34
INTERCHANGE RAMP TERMINALS: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 34-1

Introduction .......................................................................................................... 34-2
Intersection Traffic Movements ......................................................................... 34-2
Example Problem 1: Diamond Interchange ..................................................... 34-3
Example Problem 2: Parclo A-2Q Interchange ................................................ 34-9
Example Problem 3: Diamond Interchange with Queue Spillback ............ 34-16
Example Problem 4: Diamond Interchange with Demand Starvation ....... 34-23
Example Problem 5: Diverging Diamond Interchange with Signal
Control ......................................................................................................... 34-30
Example Problem 6: Diverging Diamond Interchange with Yield
Control ......................................................................................................... 34-34
Example Problem 7: Single-Point Urban Interchange .................................. 34-37
Example Problem 8: Diamond Interchange with Adjacent Intersection .... 34-43
Example Problem 9: Diamond Interchange with Roundabouts ................. 34-51
Example Problem 10: Operational Analysis for Type Selection .................... 34-53
Example Problem 11: Alternative Analysis Tool........................................... 34-58
Example Problem 12: Four-Legged Restricted Crossing U-Turn
Intersection with Merges ........................................................................... 34-64
Example Problem 13: Three-Legged Restricted Crossing U-Turn
Intersection with Stop Signs...................................................................... 34-67
Example Problem 14: Four-Legged Restricted Crossing U-Turn
Intersection with Signals ........................................................................... 34-71
Example Problem 15: Four-Legged Median U-Turn Intersection with
Stop Signs .................................................................................................... 34-75
Example Problem 16: Partial Displaced Left-Turn Intersection .................. 34-79
Example Problem 17: Full Displaced Left-Turn Intersection....................... 34-84
3. OPERATIONAL ANALYSIS FOR INTERCHANGE TYPE

SELECTION .............................................................................................................. 34-91
Introduction ........................................................................................................ 34-91
Inputs and Applications ................................................................................... 34-92
Saturation Flow Rates ....................................................................................... 34-92
Computational Steps ......................................................................................... 34-93
Chapter 34/Interchange Ramp Terminals: Supplemental Contents

4. O-D AND TURNING MOVEMENTS ...........................................................34-100

O-D and Turning Movements for Interchanges with Roundabouts ........34-100
O-D and Turning Movements for Conventional Interchanges .................34-102
5. REFERENCES .....................................................................................................34-109
Contents Chapter 34/Interchange Ramp Terminals: Supplemental

LIST OF EXHIBITS
Exhibit 34-1 Example Problem Descriptions ........................................................... 34-2

Exhibit 34-2 Intersection Traffic Movements and Numbering Scheme .............. 34-2
Exhibit 34-3 Example Problem 1: Interchange Volumes and
Channelization ..................................................................................................... 34-3
Exhibit 34-4 Example Problem 1: Signalization Information ................................ 34-3
Exhibit 34-5 Example Problem 1: Adjusted O-D Table.......................................... 34-4
Exhibit 34-6 Example Problem 1: Lane Utilization Adjustment
Calculations .......................................................................................................... 34-4
Exhibit 34-7 Example Problem 1: Saturation Flow Rate Calculation for
Eastbound and Westbound Approaches .......................................................... 34-5
Northbound and Southbound Approaches ..................................................... 34-5
Exhibit 34-9 Example Problem 1: Common Green Calculations .......................... 34-6
Exhibit 34-10 Example Problem 1: Lost Time due to Downstream Queues ....... 34-6
Exhibit 34-11 Example Problem 1: Lost Time due to Demand Starvation .......... 34-7
Exhibit 34-12 Example Problem 1: Queue Storage Ratio for Eastbound
and Westbound Movements .............................................................................. 34-7
Exhibit 34-13 Example Problem 1: Queue Storage Ratio for Northbound
and Southbound Movements ............................................................................. 34-8
Exhibit 34-14 Example Problem 1: Control Delay for Eastbound and
Westbound Movements ...................................................................................... 34-8
Exhibit 34-15 Example Problem 1: Control Delay for Northbound and
Southbound Movements .................................................................................... 34-9
Exhibit 34-16 Example Problem 1: O-D Movement LOS ....................................... 34-9
Exhibit 34-17 Example Problem 2: Intersection Plan View ................................. 34-10
Exhibit 34-18 Example Problem 2: Signalization Information ............................ 34-10
Exhibit 34-19 Example Problem 2: Adjusted O-D Table ...................................... 34-11
Calculations ........................................................................................................ 34-11
Northbound and Southbound Approaches ................................................... 34-11
Eastbound and Westbound Approaches ........................................................ 34-12
Exhibit 34-23 Example Problem 2: Common Green Calculations ...................... 34-12
Exhibit 34-24 Example Problem 2: Lost Time due to Downstream Queues ..... 34-13
and Westbound Movements ............................................................................ 34-13


and Southbound Movements ...........................................................................34-14
Westbound Movements ....................................................................................34-14
Southbound Movements...................................................................................34-15
Exhibit 34-29 Example Problem 2: O-D Movement LOS .....................................34-15
Exhibit 34-30 Example Problem 3: Intersection Plan View .................................34-16
Exhibit 34-31 Example Problem 3: Signalization Information ............................34-16
Exhibit 34-32 Example Problem 3: Adjusted O-D Table ......................................34-17
Calculations ........................................................................................................34-17
Eastbound and Westbound Approaches ........................................................34-18
Northbound and Southbound Approaches ...................................................34-18
Exhibit 34-36 Example Problem 3: Common Green Calculations ......................34-19
Exhibit 34-37 Example Problem 3: Lost Time due to Downstream Queues .....34-19
Exhibit 34-38 Example Problem 3: Lost Time due to Demand Starvation
Calculations ........................................................................................................34-20
and Westbound Movements ............................................................................34-20
Southbound Movements ..................................................................................34-22
Exhibit 34-46 Example Problem 4: Adjusted O-D Table ......................................34-24
Calculations ........................................................................................................34-24
Eastbound and Westbound Approaches ........................................................34-25
Northbound and Southbound Approaches ...................................................34-25

Exhibit 34-52 Example Problem 4: Lost Time due to Demand Starvation

Calculations ........................................................................................................ 34-27
and Westbound Movements ............................................................................ 34-27
and Southbound Movements ........................................................................... 34-28
Westbound Movements .................................................................................... 34-28
Southbound Movements .................................................................................. 34-29
Exhibit 34-57 Example Problem 4: O-D Movement LOS ..................................... 34-29
Exhibit 34-58 Example Problem 5: DDI Geometry, Lane, and Volume
Inputs .................................................................................................................. 34-30
Exhibit 34-59 Example Problem 5: Signal Timing and Volume Inputs ............. 34-31
Calculations ........................................................................................................ 34-32
All Approaches .................................................................................................. 34-32
Exhibit 34-63 Example Problem 5: Lost Time and Effective Green
Calculations ........................................................................................................ 34-33
Exhibit 34-64 Example Problem 5: Performance Results ..................................... 34-33
Exhibit 34-65 Example Problem 5: ETT and LOS Results .................................... 34-34
Exhibit 34-66 Example Problem 6: Geometry, Lane, and Volume Inputs ......... 34-34
Exhibit 34-67 Example Problem 6: Capacity of Blocked Regime ....................... 34-35
Exhibit 34-68 Example Problem 6: Capacity of Gap Acceptance Regime ......... 34-36
Exhibit 34-69 Example Problem 6: Capacity of No-Opposing-Flow
Regime ................................................................................................................ 34-36
Exhibit 34-70 Example Problem 6: Performance Results ..................................... 34-36
Exhibit 34-71 Example Problem 6: ETT and LOS Results .................................... 34-37
Exhibit 34-72 Example Problem 7: Intersection Plan View ................................. 34-37
Exhibit 34-73 Example Problem 7: Signalization Information ............................ 34-37
Eastbound and Westbound Approaches ........................................................ 34-39
Northbound and Southbound Approaches ................................................... 34-39
Exhibit 34-77 Example Problem 7: Uniform Delay Calculations for Left
Turns Featuring Both Permissive and Protected Phasing ............................ 34-40


and Westbound Movements ............................................................................34-41
Southbound Movements ..................................................................................34-42
Calculations ........................................................................................................34-44
Interchange Eastbound and Westbound Approaches ..................................34-44
Interchange Northbound and Southbound Approaches .............................34-45
Adjacent Intersection.........................................................................................34-45
Exhibit 34-91 Example Problem 8: Queue Storage Ratio for Interchange
Eastbound and Westbound Movements ........................................................34-48
Exhibit 34-92 Example Problem 8: Queue Storage Ratio for Interchange
Northbound and Southbound Movements ....................................................34-48
Exhibit 34-93 Example Problem 8: Queue Storage Ratio for Adjacent
Intersection Movements....................................................................................34-49
Exhibit 34-94 Example Problem 8: Control Delay for Interchange
Eastbound and Westbound Movements ........................................................34-49
Exhibit 34-95 Example Problem 8: Control Delay for Interchange
Northbound and Southbound Movements ....................................................34-50
Exhibit 34-96 Example Problem 8: Control Delay for Adjacent
Intersection Movements....................................................................................34-50
Exhibit 34-97 Example Problem 8: Interchange O-D Movement LOS ...............34-51
Exhibit 34-98 Example Problem 8: Adjacent Intersection Movement LOS .......34-51
Exhibit 34-100 Example Problem 9: Adjusted O-D Table ....................................34-52
Exhibit 34-101 Example Problem 9: Approach Capacity and Delay
Calculations ........................................................................................................34-52

Exhibit 34-102 Example Problem 9: Control Delay and LOS for Each O-D
Movement ........................................................................................................... 34-53
Exhibit 34-103 Example Problem 10: O-D Demand Information for the
Interchange ......................................................................................................... 34-54
Exhibit 34-104 Example Problem 10: NEMA Flows (veh/h) for the
Interchange ......................................................................................................... 34-54
Exhibit 34-105 Example Problem 10: NEMA Flows for the Interchange
Without Channelized Right Turns .................................................................. 34-55
Exhibit 34-106 Example Problem 10: SPUI Critical Flow Ratio
Calculations ........................................................................................................ 34-55
Exhibit 34-107 Example Problem 10: TUDI Critical Flow Ratio
Calculations ........................................................................................................ 34-55
Exhibit 34-108 Example Problem 10: CUDI Critical Flow Ratio
Calculations ........................................................................................................ 34-55
Exhibit 34-109 Example Problem 10: CDI Critical Flow Ratio Calculations ..... 34-56
Exhibit 34-110 Example Problem 10: Parclo A-4Q Critical Flow Ratio
Calculations ........................................................................................................ 34-56
Exhibit 34-111 Example Problem 10: Parclo A-2Q Critical Flow Ratio
Calculations ........................................................................................................ 34-57
Exhibit 34-112 Example Problem 10: Parclo B-4Q Critical Flow Ratio
Calculations ........................................................................................................ 34-57
Exhibit 34-113 Example Problem 10: Parclo B-2Q Critical Flow Ratio
Calculations ........................................................................................................ 34-57
Exhibit 34-114 Example Problem 10: Interchange Delay for the Eight
Interchange Types ............................................................................................. 34-58
Exhibit 34-115 Example Problem 11: Interchange Configuration and
Demand Volumes .............................................................................................. 34-59
Exhibit 34-116 Example Problem 11: Signal Timing Plan ................................... 34-59
Exhibit 34-117 Example Problem 11: Physical Configurations Examined ........ 34-60
Exhibit 34-118 Example Problem 11: Congested Approaches to Diamond
Interchange ......................................................................................................... 34-60
Exhibit 34-119 Example Problem 11: Discharge from the Diamond
Interchange Under the Full Range of Arterial Demand ............................... 34-61
Exhibit 34-120 Example Problem 11: Discharge from the Southbound Exit
Ramp Under the Full Range of Ramp Demand ............................................ 34-62
Exhibit 34-121 Example Problem 11: Congested Approaches to the TWSC
Intersection ......................................................................................................... 34-62
Exhibit 34-122 Example Problem 11: Effect of Arterial Demand on Minor-
Street Discharge at the TWSC Intersection .................................................... 34-63
Exhibit 34-123 Example Problem 12: Turning Movement Demands ................. 34-64

Exhibit 34-124 Example Problem 12: Demands Converted to the RCUT

Geometry ............................................................................................................34-64
Exhibit 34-125 Example Problem 12: Flow Rates in the RCUT Geometry ........34-65
Exhibit 34-126 Example Problem 13: Turning Movement Demands and
Intersection Diagram .........................................................................................34-67
Exhibit 34-127 Example Problem 13: Demands and Flow Rates in the
RCUT Geometry ................................................................................................34-68
Exhibit 34-128 Example Problem 13: Control Delay Calculations from the
Two-Way STOP-Controlled Intersections Methodology ...............................34-69
Exhibit 34-129 Example Problem 13: Experienced Travel Time
Calculations and Level of Service ....................................................................34-70
Exhibit 34-130 Example Problem 14: Turning Movement Demands .................34-71
Exhibit 34-131 Example Problem 14: Demands and Flow Rates in the
RCUT Geometry ................................................................................................34-72
Exhibit 34-132 Example Problem 14: Control Delay for Each Junction .............34-73
Exhibit 34-133 Example Problem 14: ETT and LOS Results ................................34-73
Exhibit 34-134 Example Problem 15: Turning Movement Demands and
Average Interval Durations ..............................................................................34-75
Exhibit 34-135 Example Problem 15: Demands Converted to the MUT
Geometry ............................................................................................................34-76
Exhibit 34-136 Example Problem 15: Flow Rates in the MUT Geometry ..........34-76
Exhibit 34-137 Example Problem 15: Control Delay for Each Junction .............34-77
Exhibit 34-138 Example Problem 15: ETT and LOS Results ................................34-78
Exhibit 34-139 Example Problem 16: Intersection Volumes and
Channelization ...................................................................................................34-79
Exhibit 34-140 Example Problem 16: Intersection Signalization ........................34-79
Exhibit 34-141 Example Problem 16: Flow Rates at the Supplemental and
Main Intersections .............................................................................................34-80
Exhibit 34-142 Example Problem 16: Lane Geometries at the
Supplemental and Main Intersections ............................................................34-80
Exhibit 34-143 Example Problem 16: Signalization at the DLT
Intersections........................................................................................................34-81
Exhibit 34-144 Example Problem 16: Maximum Phase Times at the Main
Intersection .........................................................................................................34-82
Exhibit 34-145 Example Problem 16: Weighted Average Control Delays.........34-83
Exhibit 34-146 Example Problem 17: Flow Rates at the Supplemental and
Main Intersections .............................................................................................34-85
Exhibit 34-147 Example Problem 17: Lane Geometries at the
Supplemental and Main Intersections ............................................................34-85

Page 34-viii Version 7.0
Exhibit 34-148 Example Problem 17: East–West Signalization at the DLT

Intersections ....................................................................................................... 34-86
Exhibit 34-149 Example Problem 17: North–South Signalization at the
DLT Intersections............................................................................................... 34-88
Exhibit 34-150 Example Problem 17: Weighted Average Control Delays ........ 34-89
Exhibit 34-151 Default Values of Saturation Flow Rate for Use with the
Operational Analysis for Interchange Type Selection .................................. 34-93
Exhibit 34-152 Mapping of Interchange Origins and Destinations into
Phase Movements for Operational Interchange Type Selection
Analysis............................................................................................................... 34-94
Exhibit 34-153 Phase Movements in a SPUI .......................................................... 34-94
Exhibit 34-154 Phase Movements in a Tight Urban or Compressed Urban
Diamond Interchange ....................................................................................... 34-95
Exhibit 34-155 Default Values for yt ....................................................................... 34-95
Exhibit 34-156 Phase Movements in a CDI ........................................................... 34-96
Exhibit 34-157 Phase Movements in Parclo A-2Q and A-4Q Interchanges ...... 34-97
Exhibit 34-158 Phase Movements in Parclo B-2Q and B-4Q Interchanges ........ 34-97
Exhibit 34-159 Estimation of Interchange Delay dI for Eight Basic
Interchange Types ............................................................................................. 34-99
Exhibit 34-160 Illustration and Notation of O-D Demands at an
Interchange with Roundabouts ..................................................................... 34-100
Exhibit 34-161 Notation of O-D Demands at Interchanges with
Roundabouts .................................................................................................... 34-101
Exhibit 34-162 O-D Flows for Each Interchange Configuration ....................... 34-102
Exhibit 34-163 Worksheet for Obtaining O-D Movements from Turning
Movements for Parclo A-2Q Interchanges ................................................... 34-103
Movements for Parclo A-4Q Interchanges ................................................... 34-103
Movements for Parclo AB-2Q Interchanges................................................. 34-104
Movements for Parclo AB-4Q Interchanges................................................. 34-104
Movements for Parclo B-2Q Interchanges.................................................... 34-105
Movements for Parclo B-4Q Interchanges.................................................... 34-105
Movements for Diamond Interchanges ........................................................ 34-106
Movements for SPUIs ..................................................................................... 34-106

Version 7.0 Page 34-ix
Exhibit 34-171 Worksheet for Obtaining Turning Movements from O-D

Movements for Parclo A-2Q and Parclo A-4Q Interchanges .....................34-106
Movements for Parclo AB-2Q Interchanges .................................................34-107
Movements for Parclo AB-4Q Interchanges .................................................34-107
Movements for Parclo B-2Q Interchanges ....................................................34-107
Movements for Parclo B-4Q Interchanges ....................................................34-108
Movements for Diamond Interchanges ........................................................34-108
Movements for SPUIs ......................................................................................34-108

Page 34-x Version 7.0
1. INTRODUCTION
Chapter 34 is the supplemental chapter for Chapter 23, Ramp Terminals and VOLUME 4: APPLICATIONS
GUIDE
Alternative Intersections, which is found in Volume 3 of the Highway Capacity 25. Freeway Facilities:
Manual (HCM). This chapter provides 17 example problems demonstrating the Supplemental
application of the Chapter 23 methodologies for evaluating the performance of Segments: Supplemental
distributed intersections, including restricted crossing U-turn (RCUT), median 27. Freeway Weaving:
Supplemental
U-turn (MUT), and displaced left-turn (DLT) intersections. It also presents a 28. Freeway Merges and
procedure for interchange type selection, which can be used to evaluate the Diverges: Supplemental
operational performance of various interchange types. Finally, this chapter Supplemental
provides worksheets for converting origin–destination (O-D) flows to turn 30. Urban Street Segments:
Supplemental
movement flows, and vice versa, for various interchange types. 31. Signalized Intersections:
Supplemental
Methodologies for the analysis of interchanges involving freeways and 32. STOP-Controlled
surface streets (i.e., service interchanges) were developed primarily on the basis Intersections:
Supplemental
of research conducted through the National Cooperative Highway Research 33. Roundabouts:
Program (1–3) and elsewhere (4). Development of HCM analysis procedures for Supplemental
alternative intersection and interchange designs was conducted through the Terminals:
Federal Highway Administration (5). Supplemental
Supplemental
Chapter 34/Interchange Ramp Terminals: Supplemental Introduction

2. EXAMPLE PROBLEMS
INTRODUCTION
This section describes the application of each of the final design, operational
analysis for interchange type selection, and roundabouts analysis methods
through the use of example problems. Exhibit 34-1 describes each of the example
problems included in this chapter and indicates the methodology applied.

Example Problem Descriptions Problem Description Application
1 Diamond interchange Operational
2 Parclo A-2Q interchange Operational
3 Diamond interchange with four-phase signalization and queue spillback Operational
4 Diamond interchange with demand starvation Operational
5 Diverging diamond interchange with signalized control Operational
6 Diverging diamond interchange with YIELD-controlled turns Operational
7 Single-point urban interchange Operational
8 Diamond interchange with closely spaced intersections Operational
9 Diamond interchange with roundabouts Operational
Interchange type
10 Compare eight types of signalized interchanges
selection
11 Diamond interchange analysis using simulation Alternative tools
12 Four-legged RCUT with merges Operational
13 Three-legged RCUT with STOP signs Operational
14 Four-legged RCUT with signals Operational
15 Four-legged MUT with STOP signs Operational
16 Partial DLT intersection Operational
17 Full DLT intersection Operational
Note: Parclo = partial cloverleaf, RCUT = restricted crossing U-turn, MUT = median U-turn, DLT = displaced left
turn.
INTERSECTION TRAFFIC MOVEMENTS

Exhibit 34-2 illustrates typical vehicle and pedestrian traffic movements for
the intersections in this chapter. Three vehicular traffic movements and one
pedestrian traffic movement are shown for each intersection approach. Each
movement is assigned a unique number or a number and letter combination. The
letter P denotes a pedestrian movement. The number assigned to each left-turn
and through movement is the same as the number assigned to each phase by
National Electrical Manufacturers Association (NEMA) specification.
Exhibit 34-2 Minor Street

Intersection Traffic Vehicle Movements 14 4 7
Movements and
Numbering Scheme Pedestrian Movements
6P
Major Street 16
5 8P 6
4P 1
2
12
2P
3 8 18
Example Problems Chapter 34/Interchange Ramp Terminals: Supplemental

Intersection traffic movements are assigned the right-of-way by the signal

controller. Each movement is assigned to one or more signal phases. A phase is
defined as the green, yellow change, and red clearance intervals in a cycle that
are assigned to a specified traffic movement (or movements) (6). The assignment
of movements to phases varies in practice with the desired phase sequence and
the movements present at the intersection.
EXAMPLE PROBLEM 1: DIAMOND INTERCHANGE

The Interchange
The interchange of I-99 (northbound/southbound, NB/SB) and University
Drive (eastbound/westbound, EB/WB) is a diamond interchange. Exhibit 34-3
provides the interchange volumes and channelization, and Exhibit 34-4 provides
the signalization information. The offset is referenced to the beginning of green
on the EB direction of the arterial.
2%
grade = _________ = Pedestrian Button Exhibit 34-3
Example Problem 1:
= Lane Width
Interchange Volumes and
400 ft
= Through Channelization
= Right
600 ft 0%
grade = _________
156 185 135 = Left
795 797
200 ft
212 200 ft 96 = Through + Right
781 870
80 210 204 600 ft = Left + Through
0%
grade = _________ University Drive
______________
Street
= Left + Right
I-99
_____________
400 ft
400 ft
Freeway
2%
grade = _________
D= 500 ft
Intersection I Intersection II Exhibit 34-4

Phase 1 2 3 1 2 3 Example Problem 1:
NEMA Φ (2+6) Φ (1+6) Φ (4+7) Φ (2+6) Φ (3+8) Φ (2+5) Signalization Information
Green time (s) 63 43 39 63 53 29
Yellow + all red (s) 5 5 5 5 5 5
Offset (s) 19 9
The Question
What are the control delay, queue storage ratio, and level of service (LOS) for
this interchange?
The Facts
There are no closely spaced intersections to this interchange, and it operates
as a pretimed signal with no right turns on red allowed. Travel path radii are 50
ft for all right-turning movements and 75 ft for all left-turning movements.
Arrival Type 4 is assumed for all arterial movements and Arrival Type 3 for all
other movements. Extra distance traveled along each freeway ramp is 100 ft.
Chapter 34/Interchange Ramp Terminals: Supplemental Example Problems

Heavy vehicles account for 6.1% of both the external and the internal
through movements, and the peak hour factor (PHF) for the interchange is
estimated to be 0.90. Start-up lost time and extension of effective green are both
2 s for all approaches. During the analysis period, there is no parking, and no
buses, bicycles, or pedestrians utilize the interchange. The grade is 2% on the NB
and SB approaches.
Solution
Calculation of Origin–Destination Movements
O-D movements through this diamond interchange are calculated on the
basis of the worksheet provided in Exhibit 34-169 in Section 4. Since all
movements utilize the signal, O-Ds can be calculated directly from the turning
movements at the two intersections. The results of these calculations and the
PHF-adjusted values are presented in Exhibit 34-5.
Exhibit 34-5 O-D Movement Demand (veh/h) PHF-Adjusted Demand (veh/h)

Example Problem 1: Adjusted A 210 233
O-D Table B 204 227
C 156 173
D 185 206
E 96 107
F 80 89
G 135 150
H 212 236
I 685 761
J 585 650
K 0 0
L 0 0
M 0 0
N 0 0
Lane Utilization and Saturation Flow Rate Calculations

Both external approaches to this interchange consist of a two-lane shared
right and through lane group. Lane utilization factors for the external through
approaches are presented in Exhibit 34-6.
Exhibit 34-6 Maximum Lane Lane Utilization

Example Problem 1: Lane Approach V1 V2 Utilization Factor
Utilization Adjustment Eastbound external 0.5056 0.4944 0.5056 0.9890
Calculations Westbound external 0.5181 0.4819 0.5181 0.9651
Saturation flow rates are calculated on the basis of reductions in the base
saturation flow rate of 1,900 pc/hg/ln by using Equation 23-14. The lane
utilization of the approaches external to the interchange is obtained as shown
above in Exhibit 34-6. Traffic pressure is calculated by using Equation 23-15. The
left- and right-turn adjustment factors are estimated by using Equations 23-20
through 23-23. These equations use an adjustment factor for travel path radius
calculated by Equation 23-19. The remaining adjustment factors are calculated as
indicated in Chapter 19, Signalized Intersections. The estimated saturation flow
rates for all approaches are shown in Exhibit 34-7 and Exhibit 34-8.

Eastbound Westbound Exhibit 34-7

Value EXT-TH&R INT-TH INT-L EXT-TH&R INT-TH INT-L Example Problem 1:
Base saturation flow Saturation Flow Rate
1,900 1,900 1,900 1,900 1,900 1,900 Calculation for Eastbound and
(s0, pc/hg/ln)
Number of lanes (N) 2 2 1 2 2 1 Westbound Approaches
Lane width adjustment (fw) 1.000 1.000 1.000 1.000 1.000 1.000
Heavy vehicle and grade
0.952 0.952 1.000 0.952 0.952 1.000
adjustment (fHVg)
Parking adjustment (fp) 1.000 1.000 1.000 1.000 1.000 1.000
Bus blockage adjustment (fbb) 1.000 1.000 1.000 1.000 1.000 1.000
Area type adjustment (fa) 1.000 1.000 1.000 1.000 1.000 1.000
Lane utilization adjustment (fLU) 0.989 0.952 1.000 0.965 0.952 1.000
Left-turn adjustment (fLT) 1.000 1.000 0.930 1.000 1.000 0.930
Right-turn adjustment (fRT) 0.999 1.000 1.000 0.998 1.000 1.000
Left-turn pedestrian–bicycle
1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fLpb)
Right-turn pedestrian–bicycle
1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fRpb)
Turn radius adjustment for lane
0.991 1.000 0.930 0.985 1.000 0.930
group (fR)
Traffic pressure adjustment for
1.034 1.036 0.963 1.044 1.026 1.000
lane group (fv)
Adjusted saturation flow
3,700 3,568 1,703 3,637 3,535 1,767
(s, veh/hg/ln)
Notes: EXT = external, INT = internal, TH = through, R = right, L = left.
Northbound Southbound Exhibit 34-8

Value Left Right Left Right Example Problem 1:
Base saturation flow (s0, pc/hg/ln) 1,900 1,900 1,900 1,900 Saturation Flow Rate
Number of lanes (N) 1 1 1 1 Calculation for Northbound
Lane width adjustment (fw) 1.000 1.000 1.000 1.000 and Southbound Approaches
Heavy vehicle and grade adjustment (fHVg) 0.990 0.990 0.990 0.990
Parking adjustment (fp) 1.000 1.000 1.000 1.000
Bus blockage adjustment (fbb) 1.000 1.000 1.000 1.000
Area type adjustment (fa) 1.000 1.000 1.000 1.000
Lane utilization adjustment (fLU) 1.000 1.000 1.000 1.000
Left-turn adjustment (fLT) 0.930 1.000 0.930 1.000
Right-turn adjustment (fRT) 1.000 0.899 1.000 0.899
Left-turn pedestrian–bicycle adjustment (fLpb) 1.000 1.000 1.000 1.000
Right-turn pedestrian–bicycle adjustment (fRpb) 1.000 1.000 1.000 1.000
Turn radius adjustment for lane group (fR) 0.930 0.899 0.930 0.899
Traffic pressure adjustment for lane group (fv) 1.000 0.979 0.991 0.968
Adjusted saturation flow (s, veh/hg/ln) 1,749 1,656 1,734 1,638
Common Green and Lost Time due to Downstream Queue and Demand
Starvation Calculations
Exhibit 34-9 first provides the beginning and end times of the green for each
phase at the two intersections on the assumption that Phase 1 of the first
intersection begins at time zero. On the basis of the information provided in
Exhibit 34-9, the relative offset between the two intersections is Offset 2 – Offset 1
+ n × cycle length = 9 – 19 + 160 = 150 s. Next, the exhibit provides the beginning
and end of green for the six pairs of movements between the two intersections
and the respective common green time for each pair of movements. For example,
the EB external through movement has the green between 0 and 63 s, while the
EB internal through movement has the green twice during the cycle, between 150
and 53 s and between 116 and 150 s. The common green time when both
movements have the green is between 0 and 53 s, for a duration of 53 s.

Exhibit 34-9 Intersection I Intersection II

Example Problem 1: Common Phase Green Begin Green End Green Begin Green End
Green Calculations Phase 1 0 63 150 53
Phase 2 68 111 58 111
Phase 3 116 155 116 145
First Green Time Second Green Time Common
Within Cycle Within Cycle Green
Movement Begin End Begin End Time
EB EXT THRU 0 63
53
EB INT THRU 150 53 116 150
WB EXT THRU 150 53
53
WB INT THRU 0 111
SB RAMP 116 155
34
EB INT THRU 150 53 116 150
NB RAMP 58 111
53
WB INT THRU 0 111
WB INT LEFT 68 111
0
EB INT THRU 150 53
EB INT LEFT 116 145
0
WB INT THRU 0 111
Notes: EXT = external, INT = internal, THRU = through, EB = eastbound, WB = westbound, SB = southbound,
NB = northbound.
The next step involves the calculation of lost time due to downstream
queues. First, the queues at the beginning of the upstream arterial phase and at
the beginning of the upstream ramp phase must be calculated by using Equation
23-33 and Equation 23-34, respectively. Exhibit 34-10 presents the calculation of
these downstream queues followed by the calculation of the respective lost time
due to those queues.
Exhibit 34-10 Movement

Example Problem 1: Lost Time Value EB EXT-TH SB-L WB EXT-TH NB-L
due to Downstream Queues Downstream Queue Calculations
VR or VA (veh/h) 206 868 233 886
NR or NA 1 2 1 2
GR or GA (s) 39 63 53 63
GD (s) 97 97 111 111
C (s) 160 160 160 160
CGUD or CGRD (s) 53 34 53 53
Queue length (QA or QR) (ft) 0.0 4.1 0.0 0.0
Lost Time Calculations
GR or GA (s) 63 39 63 53
C (s) 160 160 160 160
DQA or DQR (ft) 500 496 500 500
CGUD or CGRD (s) 53 34 53 53
Additional lost time, LD-A or LD-R (s) 0.0 0.0 0.0 0.0
Total lost time, t'L (s) 5.0 5.0 5.0 5.0
Effective green time, g' (s) 63.0 39.0 63.0 53.0
Notes: EXT = external, TH = through, L = left, EB = eastbound, WB = westbound, NB = northbound, SB =
southbound.
The lost time due to demand starvation is calculated by using Equation 23-
38. The respective calculations are presented in Exhibit 34-11. As shown, in this
case there is no lost time due to demand starvation (LDS = 0).

Movement Exhibit 34-11

Value EB-INT-TH WB-INT-TH Example Problem 1: Lost Time
vRamp-L (veh/h) 206 233 due to Demand Starvation
vArterial (veh/h) 868 886
C (s) 160 160
NRamp-L 1 1
NArterial 2 2
CGRD (s) 34 53
CGUD (s) 53 53
HI 2.02 2.04
Qinitial (ft) 0 0
CGDS (s) 0 0
LDS (s) 0 0
t”L (s) 5 5
Effective green time, g'' (s) 97 111
Notes: EB-INT-TH = eastbound internal through, WB-INT-TH = westbound internal through.
Queue Storage and Control Delay

The queue storage ratio is estimated as the ratio of the average maximum
queue to the available queue storage by using Equation 31-154. Exhibit 34-12 and
Exhibit 34-13 present the calculations of the queue storage ratio for all
movements in Example 1. Those exhibits also show the volume-to-capacity (v/c)
ratio for each movement. Control delay for each movement is calculated
according to Equation 19-18. Exhibit 34-14 and Exhibit 34-15 provide the control
delay for each movement of the interchange.
Eastbound Movements Westbound Movements Exhibit 34-12

Value EXT-TH&R INT-L INT-TH EXT-TH&R INT-L INT-TH Example Problem 1: Queue
QbL (ft) 0.0 0.0 0.0 0.0 0.0 0.0 Storage Ratio for Eastbound
v (veh/h/ln group) 957 107 967 1,036 236 883 and Westbound Movements
s (veh/h/ln) 1,850 1,703 1,784 1,819 1,768 1,768
g (s) 63 29 97 63 43 111
g/C 0.39 0.18 0.61 0.39 0.27 0.69
I 1.00 0.71 0.71 1.00 0.62 0.62
c (veh/h/ln group) 1,459 309 2,163 1,437 475 2,452
X = v/c 0.66 0.35 0.45 0.72 0.50 0.36
ra (ft/s2) 3.5 3.5 3.5 3.5 3.5 3.5
rd (ft/s2) 4.0 4.0 4.0 4.0 4.0 4.0
Ss (mi/h) 5 5 5 5 5 5
Spl (mi/h) 40 40 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04 12.04 12.04
Rp 1.000 1.333 1.333 1.000 1.333 1.333
P 0.39 0.24 0.81 0.39 0.36 0.92
r (s) 97 131 63 97 117 49
tf (s) 0.01 0.00 0.00 0.01 0.00 0.00
q (veh/s) 0.27 0.03 0.27 0.27 0.07 0.25
qg (veh/s) 0.27 0.04 0.36 0.28 0.13 0.25
qr (veh/s) 0.27 0.03 0.13 0.72 0.50 0.36
Q1 (veh) 15.2 3.5 3.8 13.9 6.9 1.2
Q2 (veh) 0.9 0.2 0.1 1.2 0.3 0.1
T 0.25 0.25 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00 0.00 0.00
tA 0 0 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00 0.00 0.00
Qb (veh) 0.00 0.00 0.00 0.00 0.00 0.00
Q3 (veh) 0.0 0.0 0.0 0.0 0.0 0.0
Q (veh) 16.2 3.7 4.0 15.2 7.2 1.3
Lh (ft) 25.01 25.00 25.01 25 25 25
La (ft) 600 200 500 600 200 500
RQ 0.67 0.46 0.20 0.63 0.90 0.06
Notes: EXT = external, INT = internal, TH = through, R = right, L= left.

Exhibit 34-13 Northbound Movements Southbound Movements

Example Problem 1: Queue Value Left Right Left Right
Storage Ratio for Northbound QbL (ft) 0.0 0.0 0.0 0.0
and Southbound Movements v (veh/h/ln group) 233 227 206 173
s (veh/h/ln) 1,749 1,656 1,734 1,638
g (s) 53 53 39 39
g/C 0.33 0.33 0.24 0.24
I 1.00 1.00 1.00 1.00
c (veh/h/ln group) 580 549 423 399
X = v/c 0.40 0.41 0.49 0.43
ra (ft/s2) 3.5 3.5 3.5 3.5
rd (ft/s2) 4.0 4.0 4.0 4.0
Ss (mi/h) 5 5 5 5
Spl (mi/h) 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04
Rp 1.000 1.000 1.000 1.000
P 0.33 0.33 0.24 0.24
r (s) 107.00 107.00 121.00 121.00
tf (s) 0.00 0.00 0.00 0.00
q (veh/s) 0.06 0.06 0.06 0.05
qg (veh/s) 0.06 0.06 0.06 0.05
qr (veh/s) 0.06 0.06 0.06 0.05
Q1 (veh) 7.1 6.9 7.1 5.9
Q2 (veh) 0.3 0.3 0.5 0.4
T 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00
tA 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00
Qb (veh) 0.00 0.00 0.00 0.00
Q3 (veh) 0.0 0.0 0.0 0.0
Q (veh) 7.4 7.3 7.5 6.2
Lh (ft) 25 25 25 25
La (ft) 400 400 400 400
RQ 0.46 0.45 0.47 0.39
Exhibit 34-14 Eastbound Movements Westbound Movements

Example Problem 1: Control EXT-
Delay for Eastbound and Value INT-L INT-TH EXT-TH&R INT-L INT-TH
TH&R
Westbound Movements g (s) - 29 97 - 43 111
g' (s) 63 - - 63 - -
g/C or g'/C 0.39 0.18 0.61 0.39 0.27 0.69
c (veh/h) 1,459 309 2,163 1,437 475 2,452
X = v/c 0.66 0.35 0.45 0.72 0.50 0.36
d1 (s/veh) 39.6 52.8 7.3 31.3 42.9 2.0
k 0.5 0.5 0.5 0.5 0.5 0.5
d2 (s/veh) 4.6 2.2 0.5 6.2 2.3 0.3
d3 (s/veh) 0.0 0.0 0.0 0.0 0.0 0.0
PF 1.000 1.000 0.560 1.000 1.000 0.283
kmin 0.04 0.04 0.04 0.04 0.04 0.04
u 0 0 0 0 0 0
t 0 0 0 0 0 0
d (s/veh) 44.1 55.0 7.8 37.5 45.2 2.3
Notes: EXT = external, INT = internal, TH = through, R = right, L= left.

Northbound Movements Southbound Movements Exhibit 34-15

Value Left Right Left Right Example Problem 1: Control
g (s) - 53 - 39 Delay for Northbound and
g' (s) 53 - 39 - Southbound Movements
g/C or g'/C 0.33 0.33 0.24 0.24
c (veh/h) 580 549 423 399
X = v/c 0.42 0.41 0.49 0.43
d1 (s/veh) 41.3 41.5 51.9 51.2
k 0.5 0.5 0.5 0.5
d2 (s/veh) 2.1 2.1 4.0 3.4
d3 (s/veh) 0.0 0.0 0.0 0.0
PF 1.000 1.000 1.000 1.000
kmin 0.04 0.04 0.04 0.04
u 0 0 0 0
t 0 0 0 0
d (s/veh) 43.4 43.4 55.9 54.6
Results
Delay for each O-D is estimated as the sum of the movement delays for each
movement utilized by the O-D, as indicated in Equation 23-2. Next, the v/c and
queue storage ratios are checked. If either of these parameters exceeds 1, the LOS
for all O-Ds that utilize that movement is F. Exhibit 34-16 summarizes the results
for all O-D movements at this interchange. As shown, all the movements have v/c
and queue storage ratios less than 1; for these O-D movements, the LOS is
determined by using Exhibit 23-10. After extra distances are measured according
to the Exhibit 23-8 discussion, EDTT can be obtained from Equation 23-50 [i.e.,
EDTT = 100 / (1.47 × 35) + 0 = 1.9 s/veh]. Interchangewide ETT is calculated by a
weighted average of O-D movement flow rates. Although certain individual O-D
movements perform at a worse LOS, this interchange operates at LOS C overall.
PHF-Adjusted Control Exhibit 34-16

O-D Demand Delay EDTT ETT Demand Example Problem 1: O-D
Movement (veh/h) (s/veh) (s/veh) (s/veh) v/c > 1? RQ > 1? LOS × ETT Movement LOS
A 233 45.6 1.9 47.5 No No C 11,067.5
B 227 43.7 −1.9 41.8 No No C 9,488.6
C 173 54.6 −1.9 52.7 No No C 9,117.1
D 206 63.6 1.9 65.5 No No D 13,493.0
E 107 99.2 1.9 101.1 No No E 10,817.7
F 89 44.2 −1.9 42.3 No No C 3,764.7
G 150 37.5 −1.9 35.6 No No C 5,340.0
H 236 82.7 1.9 84.6 No No D 19,965.6
I 761 52.0 0.0 52.0 No No C 39,572.0
J 650 39.8 0.0 39.8 No No C 25,870.0
Totals 2,832 148,496.2
Interchange ETT (s/veh) and LOS 52.4 C
EXAMPLE PROBLEM 2: PARCLO A-2Q INTERCHANGE

The Interchange
The interchange of I-75 (NB/SB) and Newberry Avenue (EB/WB) is a Parclo
A-2Q interchange. Exhibit 34-17 provides the interchange volumes and
channelization, while Exhibit 34-18 provides the signalization information. The
offset is referenced to the beginning of green on the EB direction of the arterial.

Exhibit 34-17 2%
grade = _________
= Pedestrian Button
Example Problem 2:
Intersection Plan View = Lane Width
400 ft
= Through
= Right
0%
grade = _________
800 ft 350
120 275 = Left
1187
200 ft 1055
300 165 = Through + Right
200 ft
1100
1013 188
218 250 800 ft
= Left + Through
0%
grade = _________ Newberry Avenue
______________
I-75 = Left + Right

_____________
Freeway
400 ft
2%
grade = _________
D= 800 ft

Example Problem 2: Phase 1 2 3 1 2 3
Signalization Information NEMA Φ (2+5) Φ (2+6) Φ (4+7) Φ (1+6) Φ (3+8) Φ (2+6)
Green time (s) 25 60 40 25 35 65
Offset (s) 0 0
The Question
What are the control delay, queue storage ratio, and LOS for this interchange?
The Facts
as a pretimed signal with no right turns on red allowed. The eastbound and
westbound left-turn radii are 80 ft, while all remaining turning movements have
radii of 50 ft. The arrival type is assumed to be 4 for all arterial movements and 3
for all other movements. Extra distance traveled along each freeway loop ramp is
1,600 ft. The grade is 2% on the NB and SB approaches.
There are 11.7% heavy vehicles on both the external and the internal through
movements, and the PHF for the interchange is estimated to be 0.95. Start-up lost
time is 3 s for all approaches, while the extension of effective green is 2 s for all
approaches. During the analysis period, there is no parking, and no buses,
bicycles, or pedestrians utilize the interchange.
Solution
O-Ds through this parclo interchange are calculated on the basis of the
worksheet provided in Exhibit 34-163 in Section 4. Since all movements utilize
the signal, O-Ds can be calculated directly from the turning movements at the
two intersections. The results of these calculations and the PHF-adjusted values
are presented in Exhibit 34-19.

O-D Movement Demand (veh/h) PHF-Adjusted Demand (veh/h) Exhibit 34-19

A 218 229 Example Problem 2: Adjusted
B 250 263 O-D Table
C 120 126
D 275 289
E 188 198
F 300 316
G 165 174
H 350 368
I 825 868
J 837 881
K 0 0
L 0 0
M 0 0
N 0 0

The external approaches to this interchange consist of a three-lane through
lane group. Use of the three-lane model from Exhibit 23-24 results in the
predicted lane utilization percentages for the external through approaches that
Exhibit 34-20
Maximum Lane Lane Utilization
Approach V1 V2 V3 Utilization Factor
Utilization Adjustment
Eastbound external 0.2660 0.2791 0.4549 0.4549 0.7328 Calculations
Westbound external 0.2263 0.2472 0.5265 0.5265 0.6332
above in Exhibit 34-20. Traffic pressure is calculated by using Equation 23-15.
The left- and right-turn adjustment factors are estimated by using Equations 23-
20 through 23-23. These equations use an adjustment factor for travel path radius
calculated by Equation 23-19. The remaining adjustment factors are calculated
according to Chapter 19, Signalized Intersections. The results of these
calculations for all approaches are presented in Exhibit 34-21 and Exhibit 34-22.
Exhibit 34-21
Northbound Southbound
Example Problem 2:
Value Left Right Left Right
Base saturation flow (s0, pc/hg/ln) 1,900 1,900 1,900 1,900 Calculation for Northbound
Number of lanes (N) 1 1 1 1 and Southbound Approaches
Lane width adjustment (fw) 1.000 1.000 1.000 1.000

Exhibit 34-22 Eastbound Westbound

Example Problem 2: Value EXT-TH EXT-L INT-TH&R EXT-TH EXT-L INT-TH&R
Saturation Flow Rate Base saturation flow (s0, pc/hg/ln) 1,900 1,900 1,900 1,900 1,900 1,900
Calculation for Eastbound and
Number of lanes (N) 3 1 3 3 1 3
Westbound Approaches
Heavy vehicle and grade adj. (fHVg) 0.909 1.000 0.909 0.909 1.000 0.909
LT adjustment (fLT) 1.000 0.934 1.000 1.000 0.934 1.000
RT adjustment (fRT) 1.000 1.000 0.998 1.000 1.000 0.994
LT pedestrian–bicycle adj. (fLpb) 1.000 1.000 1.000 1.000 1.000 1.000
RT pedestrian–bicycle adj. (fRpb) 1.000 1.000 1.000 1.000 1.000 1.000
Turn radius adj. for lane group (fR) 1.000 0.934 0.985 1.000 0.934 0.975
Traffic pressure adj. for lane group
0.997 1.013 1.016 1.009 0.976 1.024
(fv)
3,786 1,798 5,253 3,310 1,733 5,271
(s, veh/hg/ln)
Notes: EXT = external, INT = internal, TH = through, R = right, L = left, RT = right turn, LT = left turn,
adj. = adjustment.
Exhibit 34-23 provides the beginning and end times of the green for each
phase followed by the beginning and end of green for the four pairs of
movements at the two intersections. Phase 1 of the first intersection is assumed to
begin at time zero (in this case the offset for both intersections is zero, and
therefore the beginning of Phase 1 for the second intersection is also zero).

Phase 2 30 90 30 65
Phase 3 95 135 70 135
EB EXT THRU 0 90
20
EB INT THRU 70 135
WB EXT THRU 0 25 70 135
20
WB INT THRU 30 90
SB RAMP 95 135
40
EB INT THRU 70 135
NB RAMP 30 65
35
WB INT THRU 30 90
Notes: EXT = external, INT = internal, EB = eastbound, WB = westbound, NB = northbound, SB = southbound,
THRU = through.
due to those queues.


Value EB EXT-TH SB-L WB EXT-TH NB-L Example Problem 2: Lost Time
Downstream Queue Calculations due to Downstream Queues
VR or VA (veh/h) 289 1,066 229 1,249
NR or NA 1 3 1 3
GR or GA (s) 40 90 35 95
GD (s) 65 65 60 60
C (s) 140 140 140 140
CGUD or CGRD (s) 20 40 20 35
GR or GA (s) 90 40 95 35
C (s) 140 140 140 140
DQA or DQR (ft) 799 751 800 711
CGUD or CGRD (s) 20 40 20 35
Additional lost time, LD-A or LD-R (s) 0 0 0 0
Total lost time, t'L (s) 6 6 6 6
Effective green time, g' (s) 89 39 94 34
Notes: EXT = external, TH = through, L = left, EB = eastbound, WB = westbound, NB = northbound, SB =
southbound.

The queue storage ratio is estimated using Equation 31-154 as the ratio of the
average maximum queue to the available queue storage. Exhibit 34-25 and Exhibit
34-26 present the calculation of the queue storage ratio for all movements in Example
Problem 2. The exhibit also shows the v/c ratio for each movement. Control delay
for each movement is calculated according to Equation 19-18. Exhibit 34-27 and
Exhibit 34-28 provide the control delay for each movement of this interchange.
Exhibit 34-25
Eastbound Movements Westbound Movements
Example Problem 2: Queue
Value EXT-TH EXT-L INT-TH&R EXT-TH EXT-L INT-TH&R Storage Ratio for Eastbound
QbL (ft) 0.0 0.0 0.0 0.0 0.0 0.0 and Westbound Movements
v (veh/h/ln group) 1,066 316 1,282 1,249 174 1,479
s (veh/h/ln) 1,262 1,798 1,751 1,103 1,733 1,757
g (s) 89 24 64 94 24 59
g/C 0.64 0.17 0.46 0.67 0.17 0.42
I 1.00 1.00 0.90 1.00 1.00 0.81
c (veh/h/ln group) 2,407 308 2,401 2,222 297 2,221
X = v/c 0.44 1.02 0.54 0.56 0.58 0.67
ra (ft/s2) 3.5 3.5 3.5 3.5 3.5 3.5
rd (ft/s2) 4.0 4.0 4.0 4.0 4.0 4.0
Ss (mi/h) 5 5 5 5 5 5
Spl (mi/h) 40 40 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04 12.04 12.04
Rp 1.000 1.000 1.333 1.000 1.000 1.333
P 0.636 0.171 0.609 0.671 0.171 0.562
r (s) 51 116 76 46 116 81
tf (s) 0.00 0.01 0.00 0.01 0.00 0.01
q (veh/s) 0.30 0.09 0.38 0.35 0.05 0.41
qg (veh/s) 0.30 0.09 0.50 0.35 0.05 0.55
qr (veh/s) 0.30 0.09 0.27 0.35 0.05 0.31
Q1 (veh) 5.4 10.7 6.9 6.3 5.6 10.4
Q2 (veh) 0.1 4.9 0.3 0.2 0.7 0.5
T 0.25 0.25 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00 0.00 0.00
tA 0 0 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00 0.00 0.00
Qb (veh) 0.00 0.00 0.00 0.00 0.00 0.00
Q3 (veh) 0.0 0.0 0.0 0.0 0.0 0.0
Q (veh) 5.5 15.7 7.2 6.5 6.3 10.9
Lh (ft) 25.02 25.00 25.02 25.02 25.00 25.02
La (ft) 800 200 800 800 200 800
RQ 0.17 1.96 0.23 0.20 0.78 0.34


s (veh/h/ln) 1,674 1,658 1,701 1,617
g (s) 34 34 39 39
g/C 0.24 0.24 0.28 0.28
I 1.00 1.00 1.00 1.00
c (veh/h/ln group) 407 403 474 450
X = v/c 0.56 0.65 0.61 0.28
ra (ft/s2) 3.5 3.5 3.5 3.5
rd (ft/s2) 4.0 4.0 4.0 4.0
Ss (mi/h) 5 5 5 5
Spl (mi/h) 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04
Rp 1.000 1.000 1.000 1.000
P 0.243 0.243 0.279 0.279
r (s) 106 106 101 101
tf (s) 0.00 0.00 0.01 0.00
q (veh/s) 0.06 0.07 0.08 0.04
qg (veh/s) 0.06 0.07 0.08 0.04
qr (veh/s) 0.06 0.07 0.08 0.04
Q1 (veh) 7.8 9.2 9.8 3.4
Q2 (veh) 0.6 0.9 0.8 0.2
T 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00
tA 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00
Qb (veh) 0.00 0.00 0.00 0.00
Q3 (veh) 0.0 0.0 0.0 0.0
Q (veh) 8.5 10.1 10.5 3.6
Lh (ft) 25 25 25 25
La (ft) 400 400 400 400
RQ 0.53 0.63 0.66 0.22

Example Problem 2: Control Value EXT-TH EXT-L INT-TH&R EXT-TH EXT-L INT-TH&R
Delay for Eastbound and g (s) - 24 64 - 24 59
Westbound Movements g' (s) 89 - - 94 - -
g/C or g'/C 0.64 0.17 0.46 0.67 0.17 0.42
c (veh/h) 2,407 308 2,401 2,222 297 2,221
X = v/c 0.44 1.02 0.56 0.56 0.58 0.67
d1 (s/veh) 12.9 58.0 18.8 12.1 53.4 24.1
k 0.5 0.5 0.5 0.5 0.5 0.5
d2 (s/veh) 0.6 57.7 1.5 1.0 8.2 2.6
d3 (s/veh) 0.0 0.0 0.0 0.0 0.0 0.0
PF 1.000 1.000 0.827 1.000 1.000 0.871
kmin 0.04 0.04 0.04 0.04 0.04 0.04
u 0 0 0 0 0 0
t 0 0 0 0 0 0
d (s/veh) 13.5 115.7 20.3 13.2 61.6 26.8


g (s) - 34 39 - Delay for Northbound and
g' (s) 34 - - 39 Southbound Movements
g/C or g'/C 0.24 0.24 0.28 0.28
c (veh/h) 407 403 474 450
X = v/c 0.56 0.65 0.61 0.28
d1 (s/veh) 46.5 47.7 43.9 39.5
k 0.5 0.5 0.5 0.5
d2 (s/veh) 5.6 8.0 5.8 1.6
d3 (s/veh) 0.0 0.0 0.0 0.0
PF 1.000 1.000 1.000 1.000
kmin 0.04 0.04 0.04 0.04
u 0 0 0 0
t 0 0 0 0
d (s/veh) 52.1 55.7 49.7 41.1
Results
for all O-Ds that utilize that movement is F. Exhibit 34-29 presents the resulting
delay, v/c ratio, and RQ for each O-D movement. As shown, O-D Movement F
(which consists of the EB external left movement) has v/c and RQ ratios greater
than 1, resulting in LOS F. For the remaining movements, the LOS is determined
by using Exhibit 23-10. After extra distances are measured according to the
Exhibit 23-9 discussion, EDTT can be obtained from Equation 23-50 [i.e., EDTT =
1,200 / (1.47 × 25) + 5 = 37.7 s/veh]. Interchangewide ETT is calculated by a
movements perform at a worse LOS, this interchange operates at LOS D overall.

A 229 78.9 20.6 99.5 No No E 22,785.5
B 263 55.7 -15.6 40.1 No No C 10,546.3
C 126 41.1 -15.6 25.5 No No B 3,213.0
D 289 70.0 20.6 90.6 No No E 26,183.4
E 198 33.8 37.7 71.5 No No D 14,157.0
F 316 115.7 20.6 136.3 Yes Yes F 43,070.8
G 174 61.6 20.6 82.2 No No D 14,302.8
H 368 40.0 37.7 77.7 No No D 28,593.6
I 868 33.8 0.0 33.8 No No C 29,338.4
J 881 40.0 0.0 40.0 No No C 35,240.0
Totals 3,712 227,430.8
Interchange ETT (s/veh) and LOS 61.3 D

EXAMPLE PROBLEM 3: DIAMOND INTERCHANGE WITH QUEUE

SPILLBACK
The Interchange
The interchange of I-95 (NB/SB) and 22nd Avenue (EB/WB) is a diamond
interchange. The traffic, geometric, and signalization conditions for this study
site are provided in Exhibit 34-30 and Exhibit 34-31. The offset is referenced to
the beginning of green on the EB direction of the arterial.
Exhibit 34-30 2%
grade = _________ = Pedestrian Button
Example Problem 3: = Lane Width
400 ft
= Through
= Right
600 ft
0%
grade = _________
104 56 68 = Left
860 1020
2000 801
1991
300 135 460 600 ft = Left + Through
0%
grade = _________ 22nd Avenue
______________
= Left + Right
I-95
_____________
400 ft
Freeway
2%
grade = _________
D= 300 ft

Example Problem 3: Phase 1 2 3 1 2 3
Signalization Information NEMA Φ (4+7) Φ (2+6) Φ (1+6) Φ (2+5) Φ (2+6) Φ (3+8)
Green time (s) 27 59 19 27 39 39
Offset (s) 0 0
The Question
The Facts
ft for all turning movements except the eastbound and westbound left
movements, which have radii of 75 ft. Extra distance traveled along each freeway
ramp is 60 ft. The grade is 2% on the NB and SB approaches.
There are 6.1% heavy vehicles on both the external and the internal through
movements, and the PHF for the interchange is 0.97. Start-up lost time and
extension of effective green are both 2 s for all approaches. During the analysis
period, there is no parking, and no buses, bicycles, or pedestrians utilize the
interchange.

Solution
O-Ds through this diamond interchange are calculated on the basis of the
worksheet provided in Exhibit 34-169 in Section 4. Since all movements utilize
the signal, O-Ds can be calculated directly from the turning movements at the
two intersections. The results of these calculations and the PHF-adjusted values

B 460 474 O-D Table
C 104 107
D 56 58
E 1,255 1,294
F 300 309
G 68 70
H 295 304
I 745 768
J 725 747
K 0 0
L 0 0
M 0 0
N 0 0

This interchange consists of external approaches with three through lanes
and an exclusive right-turn lane. The lane utilization for Lane 1 is predicted by
using the three-lane model of Exhibit 23-24. Since there is an exclusive right-turn
lane for both external approaches, according to the first note of Exhibit 23-24 the
lane utilization for Lane 3 should be estimated by assuming that the right-
turning O-D (vF, vG) is zero. Exhibit 34-33 presents the calculation results and the
lane utilization factor for each approach.
Maximum Exhibit 34-33

Lane Lane Utilization Example Problem 3: Lane
Approach V1 V2 V3 Utilization Factor Utilization Adjustment
3-lane EB 0.5551 0.2224 0.2224 0.5551 0.6005 Calculations
3-lane WB 0.4441 0.2779 0.2779 0.4441 0.7506
Notes: EB = eastbound, WB = westbound.
indicated in Chapter 19, Signalized Intersections. The results of these calculations
for all approaches are presented in Exhibit 34-34 and Exhibit 34-35.


Example Problem 3: Value EXT-TH EXT-R INT-TH INT-L EXT-TH EXT-R INT-TH INT-L
Saturation Flow Rate Base saturation flow
Calculation for Eastbound and 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900
(s0, pc/hg/ln)
Westbound Approaches Number of lanes (N) 3 1 3 1 3 1 3 1
Lane width adjustment (fw) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.952 1.000 0.952 1.000 0.952 1.000 0.952 1.000
adjustment (fHVg)
Parking adjustment (fp) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Bus blockage adjustment (fbb) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Area type adjustment (fa) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Lane utilization adjustment
0.600 1.000 0.908 1.000 0.751 1.000 0.908 1.000
(fLU)
Left-turn adjustment (fLT) 1.000 1.000 1.000 0.930 1.000 1.000 1.000 0.930
Right-turn adjustment (fRT) 1.000 0.899 1.000 1.000 1.000 0.899 1.000 1.000
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fLpb)
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fRpb)
Turn radius adjustment for
1.000 0.899 1.000 0.930 1.000 0.899 1.000 0.930
lane group (fR)
Traffic pressure adjustment
1.043 0.980 0.975 0.948 0.987 0.945 0.978 0.998
for lane group (fv)
3,400 1,675 4,807 1,676 4,021 1,614 4,822 1,764
(s, veh/hg/ln)
Exhibit 34-35 Northbound Southbound

Example Problem 3: Value Left Right Left Right
Saturation Flow Rate Base saturation flow (s0, pc/hg/ln) 1,900 1,900 1,900 1,900
Calculation for Northbound Number of lanes (N) 1 1 1 1
and Southbound Approaches Lane width adjustment (fw) 1.000 1.000 1.000 1.000
Heavy vehicle adjustment (fHV) 1.000 1.000 1.000 1.000
Grade adjustment (fg) 0.990 0.990 0.990 0.990
Exhibit 34-36 first provides the beginning and ending of the green time for
each phase at the two intersections, on the assumption that Phase 1 of the first
intersection begins at time zero. In this case, the offset for both intersections is
zero; therefore, the beginning of Phase 1 for the second intersection is also zero.


Green Example Problem 3: Common
Phase Begin Green End Green Begin Green End Green Calculations
Phase 1 0 27 0 27
Phase 2 32 91 32 71
Phase 3 96 115 76 115
EB EXT THRU 32.0 91.0
39
EB INT THRU 0.0 71.0
WB EXT THRU 32.0 71.0
39
WB INT THRU 32.0 115.0
SB RAMP 0.0 27.0
27
NB RAMP 76.0 115.0
39
WB INT LEFT 96.0 115.0
0
EB INT LEFT 0.0 27.0
0
THRU = through.
due to those queues. As shown, the SB-L movement has additional lost time of
5.5 s due to the downstream queue.
The lost time due to demand starvation is calculated by using Equation 23-
38. The respective calculations are presented in Exhibit 34-38. As shown, in this
case there is no lost time due to demand starvation.

Value EB EXT-TH SB-L WB EXT-TH NB-L Example Problem 3: Lost Time
Downstream Queue Calculations due to Downstream Queues
VR or VA (veh/h) 58 2,062 139 1,052
NR or NA 1 3 1 3
GR or GA (s) 27 59 39 39
GD (s) 71 71 83 83
C (s) 120 120 120 120
CGUD or CGRD (s) 39.0 27.0 39.0 39.0
GR or GA (s) 59 27 39 39
C (s) 120 120 120 120
DQA or DQR (ft) 300 191 300 300
CGUD or CGRD (s) 39.0 27 39 39
Effective green time, g' (s) 59.0 21.5 39.0 39.0
Notes: EXT = external, TH = through, L = left, EB = eastbound, WB = westbound, NB = northbound,
SB = southbound.


Example Problem 3: Lost Time Value EB-INT-TH WB-INT-TH
due to Demand Starvation vRamp-L (veh/h) 58 139
Calculations vArterial (veh/h) 2,062 1,052
C (s) 120 120
NRamp-L 1 1
NArterial 3 3
CGRD (s) 27 39
CGUD (s) 39 39
HI 2.25 2.24
Qinitial (ft) 0 0
CGDS (s) 0 0
LDS (s) 0 0
t”L (s) 5 5
Effective green time, g'' (s) 71 83

movements. Those exhibits also provide the v/c ratio for each movement. Control
delay for each movement is calculated according to Equation 19-18. Exhibit 34-41
and Exhibit 34-42 provide the control delay for each movement of the interchange.

Example Problem 3: Queue Value EXT-TH EXT-R INT-L INT-TH EXT-TH EXT-R INT-L INT-TH
Storage Ratio for Eastbound QbL (ft) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
and Westbound Movements v (veh/h/ln group) 2,062 309 67 826 1,052 70 304 887
s (veh/h/ln) 1,133 1,675 1,676 1,602 1,340 1,614 1,764 1,607
g (s) 59.0 59.0 27.0 71.0 39.0 39.0 19.0 83.0
g/C 0.49 0.49 0.23 0.59 0.33 0.33 0.16 0.69
I 1.00 1.00 0.09 0.09 1.00 1.00 0.49 0.49
c (veh/h/ln group) 1,672 824 377 2,844 1,307 524 279 3,336
X = v/c 1.23 0.38 0.18 0.29 0.80 0.13 1.09 0.27
ra (ft/s2) 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4 4 4 4 4
Ss (mi/h) 5 5 5 5 5 5 5 5
Spl (mi/h) 40 40 40 40 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04 12.04 12.04 12.04 12.04
Rp 1 1 1 1.333 1 1 1 1.333
P 0.49 0.49 0.23 0.79 0.33 0.33 0.16 0.92
r (s) 61.00 61.00 93.00 49.00 81.00 81.00 101.00 37.00
tf (s) 0.02 0.00 0.00 0.00 0.01 0.00 0.01 0.00
q (veh/s) 0.57 0.09 0.02 0.23 0.29 0.02 0.08 0.25
qg (veh/s) 0.57 0.09 0.02 0.31 0.29 0.02 0.08 0.33
qr (veh/s) 0.57 0.09 0.02 0.12 0.29 0.02 0.08 0.06
Q1 (veh) 14.9 5.2 1.6 1.6 9.1 1.4 8.2 0.5
Q2 (veh) 17.1 0.3 0.0 0.0 0.6 0.1 5.0 0.1
T 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25
Qeo (veh) 97.50 0.00 0.00 0.00 0.00 0.00 6.22 0.00
tA 0.25 0 0 0 0 0 0.25 0
Qe (veh) 97.50 0.00 0.00 0.00 0.00 0.00 6.22 0.00
Qb (veh) 0 0 0 0 0 0 0 0
Q3 (veh) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Q (veh) 32.0 5.5 1.6 1.6 9.7 1.5 13.2 0.6
Lh (ft) 25 25 25 25 25 25 25 25
La (ft) 600 600 200 300 600 600 200 300
RQ 1.33 0.23 0.20 0.13 0.41 0.06 1.65 0.12


Value Left Right Left Right Example Problem 3: Queue
QbL (ft) 0.0 0.0 0.0 0.0 Storage Ratio for Northbound
v (veh/h/ln group) 139 474 58 107 and Southbound Movements
s (veh/h/ln) 1,628 1,703 1,600 1,607
g (s) 39.0 39.0 22.0 27.0
g/C 0.33 0.33 0.18 0.23
I 1.00 1.00 1.00 1.00
c (veh/h/ln group) 529 553 287 362
X = v/c 0.26 0.86 0.20 0.30
ra (ft/s2) 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4
Ss (mi/h) 5 5 5 5
Spl (mi/h) 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04
Rp 1 1 1 1
P 0.33 0.33 0.18 0.23
r (s) 81.00 81.00 98.50 93.00
tf (s) 0.00 0.01 0.00 0.00
q (veh/s) 0.04 0.13 0.02 0.03
qg (veh/s) 0.04 0.13 0.02 0.03
qr (veh/s) 0.04 0.13 0.02 0.03
Q1 (veh) 2.9 12.6 1.4 2.6
Q2 (veh) 0.2 2.4 0.1 0.2
T 0.25 0.25 0.25 0.25
Qeo (veh) 0.0 0.0 0.0 0.0
tA 0.0 0.0 0.0 0.0
Qe (veh) 0.0 0.0 0.0 0.0
Qb (veh) 0.0 0.0 0.0 0.0
Q3 (veh) 0.0 0.0 0.0 0.0
Q (veh) 3.1 15.0 1.6 2.8
Lh (ft) 25 25 25 25
La (ft) 400 400 400 400
RQ 0.19 0.94 0.10 0.17

Value EXT-TH EXT-R INT-L INT-TH EXT-TH EXT-R INT-L INT-TH Example Problem 3: Control
g (s) - 59 27 71 - 39 19 83 Delay for Eastbound and
g' (s) 59 - - - 39 - - - Westbound Movements
g/C or g'/C 0.49 0.49 0.23 0.59 0.33 0.33 0.16 0.69
c (veh/h) 1,672 824 377 2,844 1,307 524 279 3,336
X = v/c 1.23 0.38 0.18 0.29 0.80 0.13 1.09 0.27
d1 (s/veh) 30.5 19.0 37.5 5.8 37.0 28.6 50.5 1.5
k 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5
d2 (s/veh) 110.5 1.3 0.1 0.0 5.4 0.5 64.1 0.1
d3 (s/veh) 0 0 0 0 0 0 0 0
PF 1.000 1.000 1.000 0.595 1.000 1.000 1.000 0.291
kmin 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04
u 0 0 0 0 0 0 0 0
t 0 0 0 0 0 0 0 0
d (s/veh) 141.0 20.3 37.6 5.8 42.4 29.1 114.6 1.6


Example Problem 3: Control Value Left Right Left Right
Delay for Northbound and g (s) - 39 - 27
Southbound Movements g' (s) 39 - 21.5 -
g/C or g'/C 0.33 0.33 0.18 0.23
c (veh/h) 529 553 287 361
X = v/c 0.26 0.86 0.20 0.30
d1 (s/veh) 29.9 37.9 41.9 38.6
k 0.5 0.5 0.5 0.5
d2 (s/veh) 1.2 15.7 1.6 2.1
d3 (s/veh) 0.0 0.0 0.0 0.0
PF 1.000 1.000 1.000 1.000
kmin 0.04 0.04 0.04 0.04
u 0 0 0 0
t 0 0 0 0
d (s/veh) 31.1 53.6 43.5 40.7
Results
movement utilized by the O-D, as indicated in Equation 23-2. Next, the v/c ratio
and queue storage ratio are checked. If either of these parameters exceeds 1, the
LOS for all O-Ds that utilize that movement is F. Exhibit 34-43 presents a summary
of the results for all O-D movements at this interchange. As shown, v/c and RQ for
parts of O-Ds E, H, I, and M exceed 1; therefore, these O-Ds operate in LOS F.
O-D E and O-D I include the EB external through movement, while O-D H and
O-D M include the WB internal left. These movements have v/c ratios exceeding
1. The remaining movements have v/c and queue storage ratios less than 1; the
LOS for these O-D movements is determined by using Exhibit 23-10. After extra
distances are measured according to the Exhibit 23-8 discussion, EDTT can be
obtained from Equation 23-50 [i.e., EDTT = 60 / (1.47 × 35) + 0 = 1.2 s/veh].
Interchangewide ETT is calculated by a weighted average of O-D movement flow
rates. It is important to note that although certain individual movements
experience a v/c ratio or RQ greater than 1.00, the interchange is still considered to
be under capacity, operating at LOS E overall.
Exhibit 34-43 PHF-Adjusted Control

Example Problem 3: O-D O-D Demand Delay EDTT ETT Demand
Movement LOS Movement (veh/h) (s/veh) (s/veh) (s/veh) v/c > 1? RQ > 1? LOS × ETT
A 139 32.7 1.2 33.9 No No C 4,712.1
B 474 53.6 -1.2 52.4 No No C 24,837.6
C 107 40.7 -1.2 39.5 No No C 4,226.5
D 58 49.3 1.2 50.5 No No C 2,929.0
E 1,294 178.6 1.2 179.8 Yes Yes F 232,661.2
F 309 20.3 -1.2 19.1 No No B 5,901.9
G 70 29.1 -1.2 27.9 No No B 1,953.0
H 304 157.0 1.2 158.2 Yes Yes F 48,092.8
I 768 146.8 0.0 146.8 Yes Yes F 112,742.4
J 747 44.0 0.0 44.0 No No C 32,868.0
Totals 4,270 470,924.5
Interchange ETT (s/veh) and LOS 110.3 E

EXAMPLE PROBLEM 4: DIAMOND INTERCHANGE WITH DEMAND

STARVATION
The Interchange
The interchange of I-75 (NB/SB) and Archer Road (EB/WB) is a diamond
interchange. The traffic, geometric, and signalization conditions for this
interchange are provided in Exhibit 34-44 and Exhibit 34-45. The offset is
referenced to the beginning of green on the EB direction of the arterial.
2%
= Lane Width
Example Problem 4:
400 ft
= Through
= Right
600 ft 0%
grade = _________
120 185 180 = Left
925 1085
1100
1085
110 125 210 600 ft = Left + Through
0%
grade = _________ Archer Road
______________
Street
= Left + Right
I-75
_____________
400 ft
Freeway
2%
grade = _________
D= 400 ft

Green time (s) 30 25 30 30 25 30
Offset (s) 0 0
The Question
What are the control delay, queue storage ratio, and LOS for this
interchange?
The Facts
ft for all turning movements except the eastbound and westbound left, which are
75 ft. Extra distance traveled along each freeway ramp is 100 ft.
There are 6.1% heavy vehicles on both external and internal through
movements, and the PHF for the interchange is estimated to be 0.97. Start-up lost
time and extension of effective green are both 2 s for all approaches. During the
analysis interval, there is no parking, and no buses, bicycles, or pedestrians
utilize the interchange. The grade is 2% on the NB and SB approaches.

Solution
O-Ds through this diamond interchange are calculated by using the
worksheet given in Exhibit 34-169 in Section 4. Since all movements utilize the
signal, O-Ds can be calculated directly from the turning movements at the two
intersections. The results of these O-D calculations and the PHF-adjusted values

O-D Table B 210 216
C 120 124
D 185 191
E 200 206
F 110 113
G 180 186
H 285 294
I 900 928
J 800 825
K 0 0
L 0 0
M 0 0
N 0 0

This interchange consists of a three-lane shared right and through lane group
for the external approaches. Use of the three-lane model from Exhibit 23-24
results in the predicted lane utilization percentages for the external through
approaches that are presented in Exhibit 34-47.
Exhibit 34-47 Maximum Lane Lane Utilization

Example Problem 4: Lane Approach V1 V2 V3 Utilization Factor
Utilization Adjustment 3-lane EB 0.3879 0.2773 0.3348 0.3879 0.8593
Calculations 3-lane WB 0.4032 0.2502 0.3465 0.4032 0.8266
Notes: EB = eastbound, WB = westbound.
indicated in Chapter 19, Signalized Intersections. The results of the saturation
flow rate calculations for all approaches are presented in Exhibit 34-48 and
Exhibit 34-49.


Value EXT-TH&R INT-TH INT-L EXT-TH&R INT-TH INT-L Example Problem 4:
Base saturation flow Saturation Flow Rate
1,900 1,900 1,900 1,900 1,900 1,900 Calculation for Eastbound and
(s0, pc/hg/ln)
Number of lanes (N) 3 3 1 3 3 1 Westbound Approaches
0.952 0.952 1.000 0.952 0.952 1.000
adjustment (fHVg)
1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fLpb)
1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fRpb)
0.991 1.000 0.930 0.986 1.000 0.930
group (fR)
0.986 0.981 0.969 0.989 0.974 0.985
lane group (fv)
4,597 4,834 1,714 4,428 4,799 1,741
(s, veh/hg/ln)

Number of lanes (N) 1 1 1 1 Calculation for Northbound
Lane width adjustment (fw) 1.000 1.000 1.000 1.000 and Southbound Approaches
Exhibit 34-50 presents the beginning and end times of the green for each
phase at the two intersections. Phase 1 of the first intersection is assumed to
begin at time zero. In this case the offset for both intersections is zero; therefore
the beginning of Phase 1 for the second intersection is also zero.


Phase 2 35 60 35 60
Phase 3 65 95 65 95
EB EXT THRU 35 60
25
EB INT THRU 0 60
WB EXT THRU 0 30
30
WB INT THRU 0 60
SB RAMP 65 95
0
EB INT THRU 35 60
NB RAMP 65 95
0
WB INT THRU 0 60
WB INT LEFT 0 30
30
EB INT THRU 0 60
EB INT LEFT 35 60
25
WB INT THRU 0 60
THRU = through.
due to those queues. As shown, there is no additional lost time due to
downstream queues.

Example Problem 4: Lost Time Value EB EXT-TH SB-L WB EXT-TH NB-L
due to Downstream Queues Downstream Queue Calculations
VR or VA (veh/h) 191 1,134 129 1,119
NR or NA 1 3 1 3
GR or GA (s) 30 25 30 30
GD (s) 60 60 60 60
C (s) 100 100 100 100
CGUD or CGRD (s) 25 0 30 0
Queue length (QA or QR) (ft) 0.0 31.5 0.0 40
GR or GA (s) 25 30 30 30
C (s) 100 100 100 100
DQA or DQR (ft) 400 369 400 360
CGUD or CGRD (s) 25 0 30 0
Additional lost time, LD-A or LD-R (s) 0 0 0 0
Total lost time, t'L (s) 5 5 5 5
Notes: EXT = external, EB = eastbound, WB = westbound, NB = northbound, SB = southbound, TH = through,
L = left.
The lost time due to demand starvation is calculated by using Equation

23-38. The respective calculations are presented in Exhibit 34-52. As shown, both
internal through movements experience lost time due to demand starvation.


Value EB-INT-TH WB-INT-TH Example Problem 4: Lost Time
vRamp-L (veh/h) 191 129 due to Demand Starvation
vArterial (veh/h) 1,134 1,119 Calculations
C (s) 100 100
NRamp-L 1 1
NArterial 3 3
CGRD (s) 5 5
CGUD (s) 25 30
HI 2.23 2.25
Qinitial (ft) 6.8 2.8
CGDS (s) 30 25
LDS (s) 14.7 18.6
t”L (s) 19.7 23.6
Effective green time, g'' (s) 45.3 41.4

Exhibit 34-54 present the calculations of the queue storage ratio for all movements.
These exhibits also provide the v/c ratios for all movements. Control delay for
each movement is calculated according to Equation 19-18. Exhibit 34-55 and
Exhibit 34-56 provide the control delay for each movement of the interchange.

Value EXT-TH&R INT-L INT-TH EXT-TH&R INT-L INT-TH Example Problem 4: Queue
v (veh/h/ln group) 1,247 206 1,119 1,304 294 954 and Westbound Movements
s (veh/h/ln) 1,532 1,714 1,611 1,476 1,741 1,600
g (s) 25 25 45 30 30 41
g/C 0.25 0.25 0.45 0.30 0.30 0.41
I 1.00 0.09 0.09 1.00 0.13 0.13
c (veh/h/ln group) 1,198 428 2,190 1,383 522 1,987
X = v/c 1.04 0.48 0.51 0.94 0.56 0.48
ra (ft/s2) 3.5 3.5 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4 4 4
Ss (mi/h) 5 5 5 5 5 5
Spl (mi/h) 40 40 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04 12.04 12.04
Rp 1.000 1.000 1.333 1.000 1.000 1.333
P 0.25 0.25 0.60 0.30 0.30 0.55
r (s) 75.00 75.00 54.71 70.00 70.00 58.64
tf (s) 0.01 0.00 0.00 0.01 0.00 0.00
q (veh/s) 0.35 0.06 0.31 0.36 0.08 0.26
qg (veh/s) 0.35 0.06 0.41 0.36 0.08 0.35
qr (veh/s) 0.35 0.06 0.23 0.36 0.08 0.20
Q1 (veh) 9.2 4.1 3.8 9.8 5.7 3.7
Q2 (veh) 5.5 0.0 0.0 3.0 0.1 0.0
T 0.25 0.25 0.25 0.25 0.25 0.25
Qeo (veh) 24.54 0.00 0.00 0.00 0.00 0.00
tA 0.25 0 0 0 0 0
Qe (veh) 24.54 0.00 0.00 0.00 0.00 0.00
Qb (veh) 0 0 0 0 0 0
Q3 (veh) 0.0 0.0 0.0 0.0 0.0 0.0
Q (veh) 14.7 4.1 3.9 12.8 5.8 3.7
Lh (ft) 25 25 25 25 25 25
La (ft) 600 200 400 600 200 400
RQ 0.61 0.52 0.24 0.53 0.72 0.23


s (veh/h/ln) 1,617 1,625 1,635 1,606
g (s) 30 30 30 30
g/C 0.30 0.30 0.30 0.30
I 1.00 1.00 1.00 1.00
c (veh/h/ln group) 485 487 491 482
X = v/c 0.27 0.44 0.39 0.26
ra (ft/s2) 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4
Ss (mi/h) 5 5 5 5
Spl (mi/h) 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04
Rp 1.00 1.00 1.00 1.00
P 0.30 0.30 0.30 0.30
r (s) 70.00 70.00 70.00 70.00
tf (s) 0.00 0.00 0.00 0.00
q (veh/s) 0.04 0.06 0.05 0.03
qg (veh/s) 0.04 0.06 0.05 0.03
qr (veh/s) 0.04 0.06 0.05 0.03
Q1 (veh) 2.3 4.0 3.5 2.2
Q2 (veh) 0.2 0.4 0.3 0.2
T 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00
tA 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00
Qb (veh) 0 0 0 0
Q3 (veh) 0.0 0.0 0.0 0.0
Q (veh) 2.4 4.4 3.8 2.3
Lh (ft) 25 25 25 25
La (ft) 400 400 400 400
RQ 0.15 0.28 0.24 0.15

Example Problem 4: Control Value EXT-TH&R INT-L INT-TH EXT-TH&R INT-L INT-TH
g/C or g'/C 0.25 0.25 0.45 0.30 0.30 0.41
c (veh/h) 1,198 428 2,190 1,385 522 1,985
X = v/c 1.04 0.48 0.51 0.94 0.56 0.48
d1 (s/veh) 37.4 32.0 13.5 34.2 29.5 15.8
k 0.5 0.5 0.5 0.5 0.5 0.5
d2 (s/veh) 50.3 0.3 0.1 23.7 0.6 0.1
d3 (s/veh) 0.0 0.0 0.0 0.0 0.0 0.0
PF 1.000 1.000 0.863 1.000 1.000 0.902
kmin 0.04 0.04 0.04 0.04 0.04 0.04
u 0 0 0 0 0 0
t 0 0 0 0 0 0
d (s/veh) 87.6 32.3 13.5 57.9 30.1 16.0


g (s) - 30 - 30 Delay for Northbound and
g' (s) 30 - 30 - Southbound Movements
g/C or g'/C 0.30 0.30 0.30 0.30
c (veh/h) 485 487 490 482
X = v/c 0.27 0.44 0.39 0.26
d1 (s/veh) 26.6 28.3 27.7 26.5
k 0.5 0.5 0.5 0.5
d2 (s/veh) 1.3 2.9 2.3 1.3
d3 (s/veh) 0.0 0.0 0.0 0.0
PF 1.000 1.000 1.000 1.000
kmin 0.04 0.04 0.04 0.04
u 0 0 0 0
t 0 0 0 0
d (s/veh) 28.0 31.2 30.1 27.8
Results
for all O-Ds that utilize that movement is F. Exhibit 34-57 summarizes the results
for all O-D movements at this interchange. As shown, the v/c ratio exceeds 1 for
O-D Movements E, F, and I, all of which include the EB external through and
right movements. Therefore, these O-D movements operate in LOS F. The
remaining movements have v/c and queue storage ratios less than 1; the LOS is
determined by using Exhibit 23-10 for these movements. After extra distances are
measured according to the Exhibit 23-8 discussion, EDTT can be obtained from
Equation 23-50 [i.e., EDTT = 80 / (1.47 × 35) + 0 = 1.6 s/veh]. Interchangewide ETT
is calculated by a weighted average of O-D movement flow rates. It is important
to note that although certain individual movements experience a v/c ratio or RQ
greater than 1.00, the interchange is still considered to be under capacity, operating
at LOS D overall.

A 129 43.9 1.6 45.5 No No C 5,869.5
B 216 31.2 -1.6 29.6 No No B 6,393.6
C 124 27.8 -1.6 26.2 No No B 3,248.8
D 191 43.6 1.6 45.2 No No C 8,633.2
E 206 119.9 1.6 121.5 Yes No F 25,029.0
F 113 87.6 -1.6 86.0 Yes No F 9,718.0
G 186 57.9 -1.6 56.3 No No D 10,471.8
H 294 88.0 1.6 89.6 No No E 26,342.4
I 928 101.1 0.0 101.1 Yes No F 93,820.8
J 825 73.9 0.0 73.9 No No D 60,967.5
Totals 3,212 250,494.6

EXAMPLE PROBLEM 5: DIVERGING DIAMOND INTERCHANGE WITH

SIGNAL CONTROL
The Interchange
The interchange of Main Street at Interstate I-40 is a diverging diamond
interchange (DDI) with signalized right turns and left turns controlling
movements from the freeway onto the Main Street arterial. The turning
movements onto the freeway from Main Street are not signalized. The traffic,
geometric, and signalization conditions of the interchange are provided in
Exhibit 34-58 and Exhibit 34-59.
Exhibit 34-58
Example Problem 5: DDI
Geometry, Lane, and Volume
Inputs
Exhibit 34-58 shows movement numbers M1 through M8, their associated

volume levels (in vehicles per hour), and the number of lanes for each movement
approach. Note that the eastbound movement has an exclusive left-turn lane
onto the freeway between crossovers, which is carried through the external
crossover at Movement M6. For the westbound movement, the left turn onto the
freeway is made from a shared lane, which is expected to affect the lane
utilization of Movement M2.
The Question
What are the control delays, experienced travel time, and LOS for this
interchange?
The Facts
There are no closely spaced intersections to this interchange, and it operates as
a pretimed signal with no right turns on red allowed. Travel path radii are 75 ft
for right-turn movements and 150 ft for left turns.
There are 6.1% heavy vehicles for all movements, and the PHF for the
interchange is 0.95. Start-up lost time and extension of effective green are both 2 s
for all approaches. During the analysis period, there is no parking, and no buses,
bicycles, or pedestrians utilize the interchange.

Exhibit 34-59 provides basic signal timing information for the DDI. The cycle
length is set at 70 s for this pretimed signal. The arterial street free-flow speed is
35 mi/h.
West Crossover East Crossover Exhibit 34-59

Movement M5 M6 M7 M8 M1 M2 M3 M4 Example Problem 5: Signal
Green time (s) 25 35 25 35 35 25 35 25 Timing and Volume Inputs
Yellow time (s) 4 4 4 4 4 4 4 4
All-red time (s) 1 1 1 1 1 1 1 1
Phase split (s) 30 40 30 40 40 30 40 30
Turn radius (ft) 150 75 150 75
Width of clear zone (ft) 200 100 200 100
Shortest distance, stop bar
20 60 20 60
to conflict point (ft)
Volume (veh/h) 500 1,300 300 200 1,000 450 350 200
The DDI is timed with two critical phases to allow the northbound and
southbound through movements to be processed through the interchange
sequentially. The signalized right-turn movements from the freeway move
concurrently with the inbound through movement into the interchange at each
crossover, and the left turns move concurrently with the outbound through
movements. Overlap phasing is used to reduce the lost time for the through
movement while providing adequate clearance times for the turning traffic. In
the methodology, this results in additional lost time applied to the ramp
movements (Step 4 of DDI methodology in Chapter 23).
Solution
O-D movements through this diamond interchange are calculated by using
the worksheet in Exhibit 34-169 in Section 4. Because all movements utilize the
signal, O-Ds can be calculated directly from the turning movements at the two
intersections. The results of these calculations and the PHF-adjusted values are
presented in Exhibit 34-60.

B 200 211 O-D Table
C 200 211
D 300 316
E 600 632
F 200 211
G 300 316
H 300 316
I 700 737
J 150 158
K 0 0
L 0 0
M 0 0
N 0 0


Lane utilization for DDIs is calculated by using Exhibit 23-26 for the two
external approaches to the DDI. The eastbound movement has an exclusive left-
turn lane onto the freeway between crossovers, which is carried through the
external crossover at Movement M6. For the westbound movement, the left turn
onto the freeway is made from a shared lane, which is expected to affect lane
utilization at Movement M2. The calculated maximum lane utilization and
associated lane utilization factors are shown in Exhibit 34-61.
Exhibit 34-61 Left-Turn Maximum Lane

Example Problem 5: Lane Demand Lane Utilization
Utilization Adjustment Approach Lane Configuration Ratio Utilization Factor
Calculations Eastbound external 3-lane exclusive 0.46 0.45 0.74
Westbound external 2-lane shared 0.67 0.77 0.65
above. Traffic pressure is calculated by using Equation 23-15. The left- and right-
turn adjustment factors are estimated by using Equation 23-20 through Equation
23-23. These equations use an adjustment factor for travel path radius calculated
by Equation 23-19. The DDI adjustment factor is applied to the internal and
external through movements at both crossovers. The remaining adjustment
factors are calculated as indicated in Chapter 19, Signalized Intersections. The
estimated saturation flow rates for all approaches are shown in Exhibit 34-62.
Exhibit 34-62 West Crossover East Crossover

Example Problem 5: Value M5 M6 M7 M8 M1 M2 M3 M4
Saturation Flow Rate Base saturation flow
Calculation for All Approaches 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900
(s0, pc/hg/ln)
Number of lanes (N) 2 3 1 1 2 2 1 1
0.952 0.952 0.952 0.952 0.952 0.952 0.952 0.952
adjustment (fHVg)
Bus blockage adjustment
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
(fbb)
Lane utilization adjustment
1.000 0.740 1.000 1.000 1.000 0.649 1.000 1.000
(fLU)
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fLpb)
Right-turn pedestrian–
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
bicycle adjustment (fRpb)
Turn radius adjustment for
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
lane group (fR)
Traffic pressure adjustment
0.956 0.972 0.960 0.951 0.978 0.954 0.964 0.951
for lane group (fv)
DDI adjustment factor (fDDI) 0.913 0.913 1.000 1.000 0.913 0.913 1.000 1.000
Adjusted saturation flow per
1,578 1,188 1,674 1,601 1,615 1,022 1,682 1,601
lane (s, veh/hg/ln)
Adjusted approach
3,156 3,563 1,674 1,601 3,229 2,045 1,682 1,601
saturation flow (s, veh/hg)

Effective Green and Lost Time Calculations

Next, effective green time adjustments for the DDI movements are calculated
according to Step 4 of the DDI methodology, as shown in Exhibit 34-63. The lost
time adjustment due to internal queues was illustrated in previous examples and
is assumed to be 4 s/veh for this example. Lost time due to demand starvation
does not apply to DDIs and is set at zero. Lost time due to overlap phasing for
the DDI ramp movements is calculated from Equation 23-37.

Value M5 M6 M7 M8 M1 M2 M3 M4 Example Problem 5:
Lost time due to internal queues (s) 0 4 4 0 0 4 4 0 Lost Time and Effective
Lost time due to demand starvation (s) 0 0 0 0 0 0 0 0 Green Calculations
Lost time on DDI ramps from overlap
0 0 6.5 4.9 0 0 6.5 4.9
phasing (s)
Start-up lost time (s) 2 2 2 2 2 2 2 2
Extension of effective green (s) 2 2 2 2 2 2 2 2
Adjusted lost time, external (s) 8 15 9 9 15 9
Adjusted lost time, internal (s) 4 4
Effective green time (s) 25 31 14 30 35 20 24 20
Results
With the effective green time and saturation flow adjustments complete, the
volume-to-capacity ratios for each lane group are calculated from Equation 23-
48. Because this is an isolated DDI, no adjustments due to closely spaced
intersections apply. Because all turning movements from the freeway are
signalized, Step 6 for estimating performance of YIELD-controlled turns also does
not apply. The results are shown in Exhibit 34-64.
Control delay and its various components (uniform delay, incremental delay,
and initial queue delay) are calculated by using the procedures in Chapter 19,
and the results are shown in Exhibit 34-64.

Value M5 M6 M7 M8 M1 M2 M3 M4 Example Problem 5:
Demand flow rate, lane group (veh/h) 500 1,300 300 200 1,000 450 350 200 Performance Results
Saturation flow rate, lane group (veh/h) 3,156 3,563 1,674 1,601 3,229 2,045 1,682 1,601
Effective green time (s) 25 31 14 30 35 20 24 20
Cycle length (s) 70 70 70 70 70 70 70 70
g/C ratio 0.36 0.44 0.21 0.43 0.50 0.29 0.35 0.29
v/c ratio for lane group 0.44 0.82 0.87 0.29 0.62 0.77 0.60 0.44
Uniform delay (s/veh) 16.0 17.6 26.8 13.2 21.7 25.4 19.1 22.3
Incremental delay (s/veh) 1.2 5.2 25.7 0.1 0.2 23 1.9 0.6
Initial queue delay (s/veh) 0 0 0 0 0 0 0 0
Control delay (s/veh) 17.2 22.8 52.5 13.3 21.9 48.4 21.0 22.9
From these results, the performance measures are aggregated for each O-D
movement. The naming convention for converting turning movements to O-Ds is
followed. Furthermore, for each O-D movement, the EDTT is calculated with
Equation 23-50. The LOS for each lane group can then be determined from
Exhibit 23-10. The results of all steps are shown in Exhibit 34-65.
In the exhibit, the extra distance traveled is 100 ft for the left turn from the
freeway (Movements A and D), reflecting some out-of-direction travel distance at
the interchange. Similarly, 40 ft of added travel distance is applied to the arterial
through movements (I and J) to account for the two crossover shifts. For an

actual site, these distances should be measured from design drawings or aerial
images. The EDTT is then calculated on the assumption of a travel speed of 35
mi/h for that added distance. Note that the methodology does not consider
delays for the free-flow right-turn bypass movements onto the freeway, which
are therefore assumed to be zero. Interchangewide ETT is calculated by a
movements perform at a worse LOS, this interchange operates at LOS C overall.
Exhibit 34-65 PHF- Control Total Extra

Example Problem 5: ETT and Adjusted Delay Control Dis-
LOS Results Demand Move- Compo- Delay tance EDTT ETT Demand
O-D (veh/h) ment nents (s/veh) (ft) (s/veh) (s/veh) LOS × ETT
A 368 NB L M3 + M5 38.2 100 1.9 40.1 C 14,756.8
B 211 NB R M4 22.9 −100 −1.9 21.0 B 4,431.0
C 211 SB R M8 13.3 −100 −1.9 11.4 A 2,405.4
D 316 SB L M7 + M1 74.4 100 1.9 76.3 D 24,110.8
E 632 EB L M6 22.8 100 1.9 24.7 B 15,610.4
F 211 EB R N/A 0.0 0 0.0 0.0 A 0.0
G 316 WB R N/A 0.0 0 0.0 0.0 A 0.0
H 316 WB L M2 48.4 100 1.9 50.3 C 15,894.8
I 737 EB T M6 + M1 44.7 40 0.8 45.5 C 33,533.5
J 158 WB T M2 + M5 65.6 40 0.8 66.4 D 10,491.2
Totals 3,476 121,233.9
Note: NB = northbound, SB = southbound, EB = eastbound, WB = westbound, L = left, R = right, T = through,
N/A = not applicable.
EXAMPLE PROBLEM 6: DIVERGING DIAMOND INTERCHANGE WITH

YIELD CONTROL
The Interchange
In this example, the same DDI is used that was introduced in Example
Problem 5. The only difference is that the left turns (M3 and M7) and right turns
(M4 and M8) from the freeway off-ramps are now YIELD-controlled movements.
The estimation of control delays for Movements M1, M2, M5, and M6 is
unchanged from the previous example. The geometry is shown in Exhibit 34-66.
Exhibit 34-66
Example Problem 6:
Geometry, Lane, and Volume
Inputs

The Question
What are the control delays, experienced travel time, and LOS for the turning
movements off the freeway for this interchange if they are controlled by YIELD
signs?
The Facts
The basic assumptions for this freeway are the same as for Example Problem
5. Similarly, Steps 1 through 5 are unchanged for the signalized movements.
Solution
Capacity of YIELD-Controlled Movement
Step 6 of the interchange methodology evaluates the capacity of the YIELD-
controlled movement in three regimes: (a) Regime 1⎯blocked by conflicting
platoon when the conflicting signal has just turned green, with zero capacity for
turning movement; (b) Regime 2⎯gap acceptance in conflicting traffic after the
initial platoon has cleared, with gap acceptance controlled by the critical gap,
follow-up time, and conflicting flow rate; and (c) Regime 3⎯no conflicting flow
when the conflicting signal is red, with full capacity, controlled by the follow-up
time of the YIELD-controlled approach.
For each regime, the methodology computes the proportion of time the
regime is active, as well as the capacity that applies over that period of time. The
evaluation is performed for the two right-turn movements (M4 and M8) and the
two left-turn movements (M3 and M7).
In Regime 1, the capacity is equal to zero, since no YIELD-controlled
movements can enter the interchange while the opposing queue clears. The
duration of the blocked period is estimated from Equation 23-53. For an isolated
interchange, Equation 23-54 and Equation 23-56 are used to estimate the time to
clear the opposing queue and the time for the last queued vehicle to clear the
conflict point, respectively. The calculation results are shown in Exhibit 34-67.
Value M7 M8 M3 M4 Exhibit 34-67

Green time for opposing movement (s) 31 25 20 35 Example Problem 6: Capacity
Red time for opposing movement (s) 39 45 50 35 of Blocked Regime
Volume of opposing movement per lane (veh/h/ln) 433 250 225 500
Saturation flow rate for opposing movement (veh/h) 1,188 1,578 1,022 1,615
Time to clear queue, tCQ (s) 22.4 8.5 14.1 15.7
Distance to clear, xclear (ft) 200.0 100.0 200.0 100.0
Speed of opposing movement (mi/h) 25.0 25.0 25.0 25.0
Time to clear last vehicle, tclear (s) 5.5 2.7 5.5 2.7
Proportion of time blocked, pb 0.40 0.16 0.28 0.26
Capacity of blocked period, cb (veh/h) 0 0 0 0
In Regime 2, the capacity of the YIELD-controlled movement when gaps are

accepted in opposing traffic is estimated by using Equation 23-42. The proportion
of time for that gap acceptance regime is estimated from Equation 23-43. The
computation results are shown in Exhibit 34-68. Note that in the exhibit, the pGA
time calculated for M3 was originally negative and therefore was set to zero.

Exhibit 34-68 Value M7 M8 M3 M4

Example Problem 6: Capacity Critical gap, tc (s) 3.9 1.8 3.9 1.8
of Gap Acceptance Regime Follow-up time, tf (s) 2.6 2.4 2.6 2.4
Conflicting flow rate, qc (veh/h) 1,300 500 450 1,200
Capacity of gap acceptance regime, cGA (veh/h) 541 1,380 1,000 1,228
Proportion of time of gap acceptance, pGA 0.04 0.20 0.00a 0.24
a
Note: Set to zero to avoid negative numbers.
In Regime 3, conflicting flow is stopped at the crossover signal, and the

capacity is estimated from Equation 23-44. The proportion of time for this regime
is estimated from Equation 23-45. The results are shown in Exhibit 34-69.

Example Problem 6: Capacity Capacity of no-opposing-flow regime, cNOF (veh/h) 1,385 1,500 1,385 1,500
of No-Opposing-Flow Regime Proportion of time with no opposing flow, pNOF 0.56 0.64 0.71 0.50
Results
The combined capacity of the YIELD-controlled movement is estimated from
Equation 23-46 or Equation 23-47. With that capacity and the movement demand,
a volume-to-capacity ratio can be estimated. The control delay for the movement
is then estimated by using the control delay procedure for roundabouts given in
Equation 22-17. The computations of other terms contributing to the experienced
travel time service measure are consistent with Example Problem 5. The results
are shown in Exhibit 34-70.

Example Problem 6: Demand flow rate for lane group (veh/h) 300 200 350 200
Performance Results v/c ratio for lane group (decimal) 0.38 0.16 0.35 0.19
Control delay (s/veh) 34.7 13.4 31.0 16.3
The results suggest that under these assumptions, YIELD-controlled left-turn

Movements M7 and M4 perform better than the signalized alternatives evaluated
in Example Problem 5, while unsignalized right-turn Movements M8 and M3
show slightly higher delay than with the signal.
From these results, the performance measures are aggregated for each O-D
movement. The naming convention for converting turning movements to O-Ds is
followed. Furthermore, for each O-D movement, the EDTT is calculated with
Equation 23-50. From the O-D ETT, the LOS for each lane group is estimated
from Exhibit 23-10. The results of all steps are shown in Exhibit 34-71.
In the exhibit, the extra distance traveled is 100 ft for the left turn from the
freeway (Movements A and D), reflecting some out-of-direction travel distance at
the interchange. For right turns from the freeway (Movements B and C), an
equivalent negative extra travel distance is applied. Similarly, 40 ft of added
travel distance is applied to the arterial through movements (I and J) to account
for the two crossover shifts. For an actual site, these distances should be
measured from design drawings or aerial images. The EDTT is then calculated
on the assumption of a travel speed of 35 mi/h for that added distance. Note that
the methodology does not consider delays for the free-flow right-turn bypass
movements onto the freeway, which are therefore assumed to be zero.

Interchangewide ETT is calculated by a weighted average of O-D movement

flow rates. Although certain individual O-D movements perform at a worse LOS,
this interchange operates at LOS C overall.
PHF- Control Total Extra Exhibit 34-71

Adjusted Delay Control Dis- Example Problem 6: ETT and
Demand Move- Compo- Delay tance EDTT ETT Demand LOS Results
O-D (veh/h) ment nents (s/veh) (ft) (s/veh) (s/veh) LOS × ETT
A 368 NB L M3 + M5 38.2 100 1.9 40.1 C 18,436.8
B 211 NB R M4 22.9 −100 −1.9 21.0 B 3,038.4
C 211 SB R M8 13.3 −100 −1.9 11.4 A 2,426.5
D 316 SB L M7 + M1 74.4 100 1.9 76.3 D 18,486.0
E 632 EB L M6 22.8 100 1.9 24.7 B 15,610.4
F 211 EB R N/A 0.0 0 0.0 0.0 A 0.0
G 316 WB R N/A 0.0 0 0.0 0.0 A 0.0
H 316 WB L M2 48.4 100 1.9 50.3 C 15,894.8
I 737 EB T M6 + M1 44.7 40 0.8 45.5 C 33,533.5
J 158 WB T M2 + M5 65.6 40 0.8 66.4 D 10,491.2
Totals 3,476 117,917.6
Note: NB = northbound, SB = southbound, EB = eastbound, WB = westbound, L = left, R = right, T = through,
N/A = not applicable.
EXAMPLE PROBLEM 7: SINGLE-POINT URBAN INTERCHANGE

The Interchange
The interchange of I-95 (NB/SB) and University Drive (EB/WB) is a single-
point urban interchange (SPUI). The traffic, geometric, and signalization
conditions of the interchange are provided in Exhibit 34-72 and Exhibit 34-73.
0%
grade = _________ 0%
grade = _________ Exhibit 34-72
Example Problem 7:
600 ft = Pedestrian Button
= Lane Width
= Through
520 2%
grade = _________
= Right
210 600 ft
= Left
120
200 ft 837
184 = Through + Right
168
200 ft
865 = Left + Through
160
600 ft 80 165 = Left + Right
2%
_____________
Street
I-95
_____________ 600 ft
Freeway
0%
grade = _________ 0%
grade = _________
SPUI Interchange Exhibit 34-73

Phase 1 2 3 Example Problem 7:
NEMA Φ (1+5+4R+8R) Φ (2+6) Φ (3+8+2R+6R) Signalization Information
Green time (s) 16 32 38
Yellow + all red (s) 8 8 8
The Question

The Facts
ft and 50 ft for all left-turn and right-turn movements, respectively. Lane widths
are 10.3 ft for all lanes. There is no extra distance traveled along the freeway
ramps. The grade is 2% on the eastbound and westbound approaches.
There are 3.4% heavy vehicles on all eastbound and westbound movements.
There are 5% heavy vehicles on all northbound and southbound movements. The
PHF for the interchange is 0.95. Start-up lost time and extension of effective
green are both 2 s for all approaches. During the analysis period, there is no
parking, and no buses, bicycles, or pedestrians utilize the interchange.
Solution
O-Ds through this SPUI are calculated on the basis of the worksheet
provided in Exhibit 34-170. O-Ds can be calculated directly from the turning
movements at a SPUI because it has only one intersection. The O-Ds and the
corresponding PHF-adjusted values are presented in Exhibit 34-74.

O-D Table B 160 168
C 120 126
D 520 547
E 168 177
F 80 84
G 210 221
H 184 194
I 865 911
J 837 881
K 0 0
L 0 0
M 0 0
N 0 0
Saturation Flow Rate Calculations

saturation flow rate of 1,900 pc/hg/ln by using Equation 23-14. Traffic pressure is
calculated by using Equation 23-15. The left- and right-turn adjustment factors
are estimated by using Equation 23-20 through Equation 23-23. These equations
use an adjustment factor for travel path radius calculated by Equation 23-19. The
remaining adjustment factors are calculated as indicated in Chapter 19,
Signalized Intersections. The results of the saturation flow rate calculations for all
approaches are presented in Exhibit 34-75 and Exhibit 34-76.


Left Left Left Left Example Problem 7:
Value Prot. Perm. Through Right Prot. Perm. Through Right Saturation Flow Rate
Calculation for Eastbound and
1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 Westbound Approaches
(s0, pc/hg/ln)
Number of lanes (N) 1 1 2 1 1 1 2 1
0.961 0.961 0.961 0.961 0.961 0.961 0.961 0.961
adjustment (fHVg)
Bus blockage adjustment (fbb) 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Lane utilization adjustment (fLU) 1.000 1.000 0.952 1.000 1.000 1.000 0.952 1.000
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fLpb)
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fRpb)
0.930 0.930 1.000 0.899 0.930 0.930 1.000 0.899
group (fR)
0.950 0.951 0.998 0.946 0.950 0.954 0.995 0.964
lane group (fv)
1,560 228 3,353 1,659 1,561 211 3,346 1,673
(s, veh/hg/ln)
Note: Prot. = protected, Perm. = permitted.

Value Left Through Right Left Through Right Example Problem 7:
Base saturation flow (s0, pc/hg/ln) 1,900 1,900 1,900 1,900 1,900 1,900
Calculation for Northbound
Number of lanes (N) 1 1 1 1 1 1 and Southbound Approaches
Heavy vehicle and grade adjustment (fHVg) 1.000 1.000 1.000 1.000 1.000 1.000
Left-turn pedestrian–bicycle adjustment
1.000 1.000 1.000 1.000 1.000 1.000
(fLpb)
Right-turn pedestrian–bicycle adjustment
1.000 1.000 1.000 1.000 1.000 1.000
(fRpb)
Turn radius adjustment for lane group (fR) 0.899 1.000 0.899 0.899 1.000 0.899
Traffic pressure adjustment for lane group
0.967 0.935 0.957 1.044 0.935 0.951
(fv)
Adjusted saturation flow (s, veh/hg/ln) 1,597 1,717 1,580 1,724 1,717 1,571

Supplemental Uniform Delay Worksheet for Left Turns from Exclusive Lanes with
Protected and Permitted Phases
Uniform delay for the eastbound and westbound left-turn movements must
be calculated with a supplemental worksheet since both of these exclusive left-
turn lanes have both protected and permitted movements. The intermediate
calculations and uniform delay for the eastbound and westbound left turns are
completed according to the methodology of Chapter 19, Signalized Intersections,
and are shown in Exhibit 34-77.
Exhibit 34-77 Value Eastbound Left Westbound Left

Example Problem 7: Uniform C (s) 110 110
Delay Calculations for Left Leading left? Yes Yes
Turns Featuring Both g (s) 16 16
Permissive and Protected gq (s) 17 20
Phasing gu (s) 13.01 11.78
r (s) 62.00 62.00
X = v/c 0.60 0.67
qa (veh/s) 0.05 0.05
sp (veh/s) 0.43 0.43
ss (veh/s) 0.16 0.16
Xperm 0.78 0.92
Xprot 0.55 0.60
Case 1 1
QA (ft) 3.0 3.3
Qu (ft) 0.9 1.1
Qr (ft) 0.0 0.0
d1 (s/veh) 22.1 22.7

movements. These exhibits also show the v/c ratio for each movement. Control
delay for each movement is calculated according to Equation 19-18. Exhibit 34-80
and Exhibit 34-81 provide the control delay for each movement of the
interchange. The eastbound left turns for the permissive and protected phases
are treated in combination in these calculations.


Value Left Through Right Left Through Right Example Problem 7: Queue
v (veh/h/ln group) 177 911 84 194 881 221 and Westbound Movements
s (veh/h/ln) 672 1,676 1,659 661 1,673 1,673
g (s) 48.0 32.0 38.0 48.0 32.0 38.0
g/C 0.44 0.29 0.35 0.44 0.29 0.35
I 1.0 1.0 1.0 1.0 1.0 1.0
c (veh/h/ln group) 293 975 573 288 973 578
X = v/c 0.60 0.93 0.15 0.67 0.91 0.38
ra (ft/s2) 3.5 3.5 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4 4 4
Ss (mi/h) 5 5 5 5 5 5
Spl (mi/h) 40 40 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04 12.04 12.04
Rp 1 1 1 1 1 1
P 0.44 0.29 0.35 0.44 0.29 0.35
r (s) 62.0 78.0 72.0 62.0 78.0 72.0
tf (s) 0.00 0.01 0.00 0.01 0.01 0.00
q (veh/s) 0.05 0.25 0.02 0.05 0.24 0.06
qg (veh/s) 0.05 0.25 0.02 0.05 0.24 0.06
qr (veh/s) 0.05 0.25 0.02 0.05 0.24 0.06
Q1 (veh) 4.1 14.2 1.8 4.7 13.6 5.1
Q2 (veh) 0.7 2.3 0.1 0.9 1.9 0.3
T 0.25 0.25 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00 0.00 0.00
tA 0 0 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00 0.00 0.00
Qb (veh) 0 0 0 0 0 0
Q3 (veh) 0 0 0 0 0 0
Q (veh) 4.9 16.5 1.9 5.7 15.4 5.4
Lh (ft) 25.006 25.006 25.006 25.006 25.006 25.006
La (ft) 200 600 600 200 600 600
RQ 0.61 0.69 0.08 0.71 0.64 0.23

Throug Example Problem 7: Queue
Value Left h Right Left Through Right Storage Ratio for Northbound
QbL (ft) 0.0 0.0 0.0 0.0 0.0 0.0 and Southbound Movements
v (veh/h/ln group) 174 0 168 547 0 126
s (veh/h/ln) 1,597 1,717 1,580 1,724 1,717 1,571
g (s) 38.0 38.0 16.0 38.0 38.0 16.0
g/C 0.35 0.35 0.15 0.35 0.35 0.15
I 1.0 1.0 1.0 1.0 1.0 1.0
c (veh/h/ln group) 552 593 230 596 593 228
X = v/c 0.31 0.00 0.73 0.92 0.00 0.55
ra (ft/s2) 3.5 3.5 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4 4 4
Ss (mi/h) 5 5 5 5 5 5
Spl (mi/h) 40 40 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04 12.04 12.04
Rp 1 1 1 1 1 1
P 0.35 0.35 0.15 0.35 0.35 0.15
r (s) 72.0 72.0 94.0 72.0 72.0 94.0
tf (s) 0.00 0.00 0.00 0.01 0.00 0.00
q (veh/s) 0.05 0.00 0.05 0.15 0.00 0.04
qg (veh/s) 0.05 0.00 0.05 0.15 0.00 0.04
qr (veh/s) 0.05 0.00 0.05 0.15 0.00 0.04
Q1 (veh) 3.9 0.0 4.9 16.0 0.0 3.6
Q2 (veh) 0.2 0.0 1.2 3.6 0.0 0.6
T 0.25 0.25 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00 0.00 0.00
tA 0 0 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00 0.00 0.00
Qb (veh) 0 0 0 0 0 0
Q3 (veh) 0 0 0 0 0 0
Q (veh) 4.1 0.0 6.1 19.6 0.0 4.2
Lh (ft) 25 25 25 25 25 25
La (ft) 600 600 600 600 600 600
RQ 0.17 0.00 0.25 0.82 0.00 0.17


Example Problem 7: Control Value Left Through Right Left Through Right
g/C or g'/C 0.44 0.29 0.35 0.44 0.29 0.35
c (veh/h) 293 975 573 288 973 578
X = v/c 0.60 0.93 0.15 0.67 0.91 0.38
d1 (s/veh) 22.1 38.0 24.8 22.8 37.5 27.2
k 0.5 0.5 0.5 0.5 0.5 0.5
d2 (s/veh) 8.9 16.6 0.5 11.8 13.4 1.9
d3 (s/veh) 0.0 0.0 0.0 0.0 0.0 0.0
PF 1.000 1.000 1.000 1.000 1.000 1.000
kmin 0.04 0.04 0.04 0.04 0.04 0.04
u 0 0 0 0 0 0
t 0 0 0 0 0 0
d (s/veh) 31.0 54.6 25.4 34.6 51.0 29.1

Example Problem 7: Control Value Left Through Right Left Through Right
Delay for Northbound and g (s) - 38 16 - 38 16
Southbound Movements g' (s) 38 - - 38 - -
g/C or g'/C 0.35 0.35 0.15 0.35 0.35 0.15
c (veh/h) 552 593 230 596 593 228
X = v/c 0.31 0.00 0.73 0.92 0.00 0.55
d1 (s/veh) 26.4 23.6 45.0 34.5 23.6 43.7
k 0.5 0.5 0.5 0.5 0.5 0.5
d2 (s/veh) 1.5 0.0 18.6 21.5 0.0 9.3
d3 (s/veh) 0.0 0.0 0.0 0.0 0.0 0.0
PF 1.000 1.000 1.000 1.000 1.000 1.000
kmin 0.04 0.04 0.04 0.04 0.04 0.04
u 0 0 0 0 0 0
t 0 0 0 0 0 0
d (s/veh) 27.9 23.6 63.6 56.0 23.6 53.0
Results
Delay for each O-D is estimated as the movement delay for the corresponding
movement, as shown in Exhibit 34-82. Next, the v/c and queue storage ratios are
checked. If either of these parameters exceeds 1, the LOS for the respective O-D is
F. As shown, no movements have a v/c ratio or RQ exceeding 1, and therefore the
LOS result is based on the second column of Exhibit 23-10. Interchangewide ETT
is calculated by a weighted average of O-D movement flow rates. Although
certain individual O-D movements perform at a worse LOS, this interchange
operates at LOS C overall.
Exhibit 34-82 O-D PHF-Adjusted ETT

Example Problem 7: O-D Movement Demand (veh/h) (s/veh) v/c > 1? RQ > 1? LOS Demand × ETT
Movement LOS A 174 27.9 No No B 4,854.6
B 168 63.6 No No D 10,684.8
C 126 53.0 No No C 6,678.0
D 547 56.0 No No D 30,632.0
E 177 31.0 No No C 5,487.0
F 84 25.4 No No B 2,133.6
G 221 29.1 No No B 6,431.1
H 194 34.6 No No C 6,712.4
I 911 54.6 No No C 49,740.6
J 881 51.0 No No C 44,931.0
Totals 3,483 168,285.1

EXAMPLE PROBLEM 8: DIAMOND INTERCHANGE WITH ADJACENT

INTERSECTION
The Interchange
At the diamond interchange described in Example Problem 1 (I-99 and
University Drive), an adjacent intersection was built 300 ft to the west of the
interchange (Spring Street, NB/SB, and University Drive, EB/WB). The traffic,
geometric, and signalization conditions are shown in Exhibit 34-83 and Exhibit
34-84. The offset is referenced to the beginning of the green for the respective EB
arterial approach.
0%
grade = _________ 0%
= Lane Width Example Problem 8:
400 ft = Through Intersection Plan View
200 ft
98 = Right
800 ft
600 ft 0%
grade = _________
165 280160 185 135
156 = Left
400 795 797
200 ft 200 ft
220 150 200 ft
212 200 ft 96 = Through + Right
780 781 870
200
60 105 180 80 210 204 600 ft = Left + Through
0%
______________
Street
University Drive
200 ft
______________ = Left + Right

EB/WB I-99
_____________
400 ft
400 ft
Freeway
Spring Street
_____________
NB/SB
2%
grade = _________
D= 300 ft D= 500 ft

Green time (s) 63 43 39 63 53 29
Offset (s) 19 9
Adjacent Intersection
Phase 1 2 3 4
NEMA Φ (1+5) Φ (2+6) Φ (4+7) Φ (3+8)
Green time (s) 33 59 24 24
Yellow + all red (s) 5 5 5 5
Offset (s) 19
The Question
What are the control delay, queue storage ratio, and LOS for this interchange
and the adjacent intersection?
The Facts
The closely spaced intersection operates as a pretimed signal with no right
turns on red allowed. Travel path radii at the interchange are 50 ft for all right-
turning movements and 75 ft for all left-turning movements. Extra distance
traveled along each freeway ramp is 100 ft.
There are 6.1% heavy vehicles on eastbound and westbound through
movements of the interchange and all movements of the adjacent intersection.
The PHF for the interchange–intersection system is 0.97. Start-up lost time and
extension of effective green are both 2 s for all approaches. During the analysis
period, there is no parking, and no buses, bicycles, or pedestrians utilize the
interchange. The grade is 2% on the northbound approach.

Solution
The O-Ds for the interchange are obtained as explained in Example Problem
1 and were presented in Exhibit 34-5.

The adjacent intersection has a two-lane shared right and through lane group
for both the inbound (arriving at the interchange) and the outbound (leaving the
interchange) approaches. The lane utilization factors for the inbound and
outbound approaches of the closely spaced intersection are estimated by
obtaining the respective lane utilization values (through or shared) from Exhibit
19-15 and subtracting 0.05. The resulting lane utilization factors are shown in
Exhibit 34-85.
Exhibit 34-85 Lane Group Lane Utilization Factor

Example Problem 8: Lane 2-lane group eastbound (inbound) 0.902
Utilization Adjustment 2-lane group westbound (outbound) 0.902
Calculations
saturation flow rate of 1,900 pc/h/ln by using Equation 23-14. The saturation
flows for each lane group of the adjacent intersection are estimated according to
Chapter 19, Signalized Intersections. The results of the saturation flow rate
calculations for all movements of the adjacent intersection and the interchange
are presented in Exhibit 34-86 through Exhibit 34-88. Note that turn radius and
traffic pressure adjustments are not considered in the adjacent intersection.

Example Problem 8: Value EXT-TH&R INT-TH INT-L EXT-TH&R INT-TH INT-L
Calculation for Interchange 1,900 1,900 1,900 1,900 1,900 1,900
(s0, pc/hg/ln)
Eastbound and Westbound
Approaches Number of lanes (N) 2 2 1 2 2 1
0.952 0.952 1.000 0.952 0.952 1.000
adjustment (fHVg)
1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fLpb)
1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fRpb)
0.991 1.000 0.930 0.985 1.000 0.930
group (fR)
1.027 1.028 0.961 1.044 1.019 0.995
lane group (fv)
3,670 3,540 1,698 3,637 3,510 1,759
(s, veh/hg/ln)


Number of lanes (N) 1 1 1 1 Calculation for Interchange
Lane width adjustment (fw) 1.000 1.000 1.000 1.000 Northbound and Southbound
Heavy vehicle and grade adjustment (fHVg) 0.990 0.990 0.990 0.990 Approaches

Value TH&R L TH&R L TH R L TH&R L Example Problem 8:
1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 1,900 Calculation for Adjacent
(s0, pc/hg/ln)
Intersection
Number of lanes (N) 2 1 2 1 1 1 1 2 1
Lane width adjustment
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
(fw)
Heavy vehicle and
grade adjustment 0.952 0.952 0.952 0.952 0.952 0.952 0.952 0.952 0.952
(fHVg)
Parking adjustment
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
(fp)
Bus blockage
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fbb)
Area type adjustment
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
(fa)
Lane utilization
0.902 1.000 0.902 1.000 1.000 1.000 1.000 1.000 1.000
adjustment (fLU)
Left-turn adjustment
1.000 0.930 1.000 0.930 1.000 1.000 0.899 1.000 0.899
(fLT)
Right-turn adjustment
1.000 1.000 1.000 1.000 1.000 0.899 1.000 1.000 1.000
(fRT)
Left-turn pedestrian–
bicycle adjustment 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
(fLpb)
Right-turn pedestrian–
bicycle adjustment 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
(fRpb)
Adjusted saturation
3,359 1,680 3,251 1,645 1,765 1,580 1,568 3,434 1,654
flow (s, veh/hg/ln)
Notes: TH = through, R = right, L = left.
Common Green and Lost Time due to Downstream Queue Calculations

Common green is calculated between certain movements that can contribute
to excessive downstream queues or demand starvation, depending on the signal
phasing sequence. The adjacent intersection is offset by 10 s from Intersection 2
and by 0 s from Intersection 1. Exhibit 34-89 presents the beginning and end of
each phase at the three intersections and the calculations of common green
between the relevant movements at the three intersections.


Phase 2 68 111 58 111
Phase 3 116 155 116 145
Adjacent Intersection
Phase Phase Begin Phase End
Phase 1 0 33
Phase 2 38 62
Phase 3 67 96
Phase 4 96 155
EB EXT THRU 0 63
53
EB INT THRU 150 53 116 150
WB EXT THRU 150 53
53
WB INT THRU 0 111
SB RAMP 116 155
34
EB INT THRU 150 53 116 150
NB RAMP 58 111
53
WB INT THRU 0 111
WB INT LEFT 68 111
0
EB INT THRU 150 53
EB INT LEFT 116 145
0
WB INT THRU 0 111
EB EXT THRU 0 63
25
ADJ EB THRU 38 97
EB EXT THRU 0 63
0
ADJ SB LEFT 102 126
EB EXT THRU 0 63
0
ADJ NB RIGHT 131 155
ADJ WB THRU 38 97
59
WB INT THRU 0 111
ADJ WB THRU 38 97
0
SB RAMP 116 155
Notes: ADJ = adjacent, EXT = external, INT = internal, THRU = through, EB = eastbound, WB = westbound,
NB = northbound, SB = southbound.
The next step is the calculation of lost time due to downstream queues. At an
adjacent intersection, additional lost time due to interchange operations may
occur at the intersection’s eastbound, southbound left-turn, and northbound
right-turn approaches. Furthermore, the interchange westbound internal link
and southbound ramp may experience additional lost time due to operations at
the adjacent closely spaced intersection.
To estimate whether these approaches experience additional lost time, the
procedure determines the queue at the beginning of the intersection’s eastbound
through arterial phase, southbound left-turn phase, and northbound right-turn
phase. They are calculated by using Equation 23-24 and Equation 23-25. The
resulting queues are subtracted from the downstream link length (link between
the closely spaced intersection and the interchange) to determine the storage at
the beginning of each phase. Exhibit 34-90 presents the calculation of lost time
due to downstream queues. The results indicate that the southbound left-turn
and northbound right-turn movements of the adjacent intersection experience
additional lost time of 2.10 and 3.07 s, respectively.

Interchange Exhibit 34-90

Movement EB EXT-TH SB-L WB EXT-TH NB-L Example Problem 8: Lost Time
VR or VA (veh/h) 191 805 216 822 due to Downstream Queues
NR or NA 1 2 1 2
GR or GA (s) 39 63 53 63
GD (s) 97 97 111 111
C (s) 160 160 160 160
CGUD or CGRD (s) 53 34 53 53
Lost Time due to Downstream Queue
Interchange
Effective Green Adjustment EB EXT-TH SB-L WB EXT-TH NB-L
GR or GA (s) 63 39 63 53
C (s) 160 160 160 160
DQA or DQR (ft) 500 500 500 500
CGUD or CGRD (s) 53 34 53 53
Adjacent Intersection Interchange
EB- WB INT-
Movement TH SB-L NB-R TH SB-R
VR or VA (veh/h) 474 804 804 156 795
NR or NA 1 2 2 1 2
GR or GA (s) 48 59 59 39 111
GD (s) 63 63 63 59 59
C (s) 160 160 160 160 160
CGUD or CGRD (s) 25.0 0.0 0.0 15 39
Queue length (QA or QR) (ft) 56.9 102.6 102.6 0.0 91.1
Lost Time due to Downstream Queue
Adjacent Intersection Interchange
EB- WB INT-
Effective Green Adjustment TH SB-L NB-R TH SB-R
GR or GA (s) 59 24 24 119 39
C (s) 160 160 160 160 160
DQA or DQR (ft) 243 197 197 300 209
CGUD or CGRD (s) 25.0 29 0 15 39
Additional lost time, LD-A or LD-R (s) 0.00 2.10 3.07 0.0 0.0
Total lost time, t'L (s) 5.00 7.10 8.07 5.0 5.0
Effective green time, g' (s) 59.0 21.9 20.9 119 39
Notes: EXT = external, INT = internal, TH = through, L = left, R = right, EB = eastbound, WB = westbound,
NB = northbound, SB = southbound.

The queue storage ratio is estimated as the average maximum queue divided
by the available queue storage by using Equation 31-154. Exhibit 34-91 and
approaches of the interchange, while Exhibit 34-93 gives the results of all
approaches of the adjacent intersection. The v/c ratio for the respective
movements is also provided in these exhibits.
Control delay for each movement is calculated according to Equation 19-18.
Exhibit 34-94 through Exhibit 34-96 summarize the control delay estimates for all
approaches of the interchange and adjacent signalized intersection.


Example Problem 8: Queue Value EXT-TH&R INT-L INT-TH EXT-TH&R INT-L INT-TH
Storage Ratio for Interchange QbL (ft) 0.0 0.0 0.0 0.0 0.0 0.0
Eastbound and Westbound v (veh/h/ln group) 888 99 897 961 219 820
Movements s (veh/h/ln) 1,835 1,699 1,770 1,819 1,759 1,755
g (s) 63 29 97 63 43 111
g/C 0.39 0.18 0.61 0.39 0.27 0.69
I 1.00 0.75 0.75 1.00 0.68 0.68
c (veh/h/ln group) 1,448 308 2,146 1,448 473 2,435
X = v/c 0.61 0.32 0.42 0.66 0.46 0.34
ra (ft/s2) 3.5 3.5 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4 4 4
Ss (mi/h) 5 5 5 5 5 5
Spl (mi/h) 40 40 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04 12.04 12.04
Rp 1.000 1.000 1.333 1.000 1.000 1.333
P 0.39 0.18 0.81 0.39 0.27 0.92
r (s) 97.00 131.00 63.00 97.00 117.00 49.00
tf (s) 0.01 0.00 0.00 0.01 0.00 0.00
q (veh/s) 0.25 0.03 0.25 0.27 0.06 0.23
qg (veh/s) 0.25 0.03 0.33 0.27 0.06 0.30
qr (veh/s) 0.25 0.03 0.12 0.27 0.06 0.06
Q1 (veh) 13.8 3.5 8.5 13.0 7.3 1.1
Q2 (veh) 0.8 0.2 0.1 1.1 0.3 0.1
T 0.25 0.25 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00 0.00 0.00
tA 0 0 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00 0.00 0.00
Qb (veh) 0 0 0 0 0 0
Q3 (veh) 0.0 0.0 0.0 0.0 0.0 0.0
Q (veh) 14.6 3.7 8.6 14.1 7.6 1.2
Lh (ft) 25 25 25 25 25 25
La (ft) 600 200 500 600 200 500
RQ 0.61 0.46 0.43 0.59 0.95 0.06
Notes: EXT = external, INT = internal, TH = through, L = left, R = right.

Storage Ratio for Interchange QbL (ft) 0.0 0.0 0.0 0.0
Northbound and Southbound v (veh/h/ln group) 216 210 191 161
Movements s (veh/h/ln) 1,682 1,651 1,669 1,634
g (s) 53 53 39 39
g/C 0.33 0.33 0.24 0.24
I 1.00 1.00 1.00 1.00
c (veh/h/ln group) 557 547 407 398
X = v/c 0.39 0.38 0.47 0.40
ra (ft/s2) 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4
Ss (mi/h) 5 5 5 5
Spl (mi/h) 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04
Rp 1.000 1.000 1.000 1.000
P 0.33 0.33 0.24 0.24
r (s) 107.00 107.00 121.00 121.00
tf (s) 0.00 0.00 0.00 0.00
q (veh/s) 0.06 0.06 0.05 0.04
qg (veh/s) 0.06 0.06 0.05 0.04
qr (veh/s) 0.06 0.06 0.05 0.04
Q1 (veh) 6.6 6.4 6.5 5.4
Q2 (veh) 0.3 0.3 0.4 0.3
T 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00
tA 0 0 0 0
Qe (veh) 0.00 0.00 0.00 0.00
Qb (veh) 0 0 0 0
Q3 (veh) 0.0 0.0 0.0 0.0
Q (veh) 6.9 6.7 7.0 5.7
Lh (ft) 25 25 25 25
La (ft) 400 400 400 400
RQ 0.43 0.42 0.43 0.36

Eastbound Westbound Southbound Exhibit 34-93

Through Through Northbound Through Example Problem 8: Queue
Value & Right Left & Right Left Through Right Left & Right Left Storage Ratio for Adjacent
QbL (ft) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Intersection Movements
v (veh/h/ln group) 866 227 577 309 206 186 108 542 289
s (veh/h/ln) 1,679 1,680 1,650 1,722 1,765 1,580 1,568 1,717 1,654
g (s) 59.0 33 59 33 24.0 20.9 24.0 24 21.9
g/C 0.37 0.21 0.37 0.21 0.15 0.13 0.15 0.15 0.14
I 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
c (veh/h/ln group) 1,288 346 1,218 355 265 237 235 515 248
X = v/c 0.67 0.65 0.47 0.46 0.78 0.90 0.46 1.05 1.28
ra (ft/s2) 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5
rd (ft/s2) 4 4 4 4 4 4 4 4 4
Ss (mi/h) 5 5 5 5 5 5 5 5 5
Spl (mi/h) 40 40 40 40 40 40 40 40 40
Sa (mi/h) 39.96 39.96 39.96 39.96 39.96 39.96 39.96 39.96 39.96
da (s) 12.04 12.04 12.04 12.04 12.04 12.04 12.04 12.04 12.04
Rp 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
P 0.37 0.21 0.37 0.21 0.15 0.13 0.15 0.15 0.14
r (s) 101.00 127.00 101.00 127.00 136.00 139.07 136.00 136.00 138.10
tf (s) 0.01 0.00 0.01 0.00 0.00 0.00 0.00 0.01 0.01
q (veh/s) 0.24 0.06 0.16 0.04 0.06 0.05 0.03 0.08 0.08
qg (veh/s) 0.24 0.06 0.16 0.04 0.06 0.05 0.03 0.08 0.08
qr (veh/s) 0.24 0.06 0.16 0.04 0.06 0.05 0.03 0.08 0.08
Q1 (veh) 14.3 8.4 8.7 5.5 8.0 7.4 4.0 10.4 9.2
Q2 (veh) 1.1 0.9 0.5 0.4 1.5 2.3 0.4 5.0 9.7
T 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25
Qeo (veh) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.39 15.6
tA 0 0 0 0 0 0 0 0.25 0.25
Qe (veh) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.39 15.6
Qb (veh) 0 0 0 0 0 0 0 0 0
Q3 (veh) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Q (veh) 15.4 9.3 9.1 5.9 9.5 9.8 4.4 15.5 18.8
Lh (ft) 25 25 25 25 25 25 25 25 25
La (ft) 800 200 300 200 800 800 200 800 200
RQ 0.48 1.16 0.76 0.73 0.30 0.30 0.55 0.48 2.36

Value EXT-TH&R INT-L INT-TH EXT-TH&R INT-L INT-TH Example Problem 8: Control
g (s) - 29 97 - 43 111 Delay for Interchange
g' (s) 63 - - 63 - - Eastbound and Westbound
g/C or g'/C 0.39 0.18 0.61 0.39 0.27 0.69 Movements
c (veh/h) 1,448 308 2,146 1,448 473 2,435
X = v/c 0.61 0.32 0.42 0.68 0.46 0.34
d1 (s/veh) 38.8 56.9 16.6 30.6 48.8 2.0
k 0.5 0.5 0.5 0.5 0.5 0.5
d2 (s/veh) 3.9 2.1 0.5 5.4 2.2 0.3
d3 (s/veh) 0.0 0.0 0.0 0.0 0.0 0.0
PF 1.000 1.000 0.560 1.000 1.000 0.283
kmin 0.04 0.04 0.04 0.04 0.04 0.04
u 0 0 0 0 0 0
t 0 0 0 0 0 0
d (s/veh) 42.6 59.0 17.1 36.0 51.0 2.2
Notes: EXT = external, INT = internal, TH = through, L = left, R = right.


Example Problem 8: Control Value Left Right Left Right
Delay for Interchange g (s) - 53 - 39
Northbound and Southbound g' (s) 53 - 39 -
Movements g/C or g'/C 0.33 0.33 0.24 0.24
c (veh/h) 557 547 407 398
X = v/c 0.39 0.38 0.47 0.40
d1 (s/veh) 41.1 41.0 51.7 50.7
k 0.5 0.5 0.5 0.5
d2 (s/veh) 2.0 2.0 3.8 3.0
d3 (s/veh) 0.0 0.0 0.0 0.0
PF 1 1 1 1
kmin 0.04 0.04 0.04 0.04
u 0 0 0 0
t 0 0 0 0
d (s/veh) 43.1 43.0 55.5 53.8
Exhibit 34-96 Eastbound Westbound Southbound

Example Problem 8: Control Through Through Northbound Through
Delay for Adjacent Value & Right Left & Right Left Through Right Left & Right Left
Intersection Movements
g (s) - 33.0 59.0 33.0 24.0 - 24.0 24.0 -
g' (s) 59.0 - - - - 20.9 - - 21.9
g/C or g'/C 0.37 0.21 0.37 0.21 0.15 0.13 0.15 0.15 0.14
c (veh/h) 1,288 346 1,218 355 265 237 235 258 248
X = v/c 0.67 0.65 0.47 0.87 0.78 0.78 0.46 1.05 1.28
d1 (s/veh) 42.5 58.3 38.7 55.6 65.4 68.5 62.1 68.0 69.0
k 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5
d2 (s/veh) 6.0 9.3 2.7 4.4 20.0 40.7 6.4 70.6 153.6
d3 (s/veh) 0 0 0 0 0 0 0 0 0
PF 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
kmin 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04
u 0 0 0 0 0 0 0 0 0
t 0 0 0 0 0 0 0 0 0
d (s/veh) 48.5 67.6 41.4 60.0 85.4 109.1 68.4 138.6 226.6
Results
movement utilized by the O-D, as indicated in Equation 23-2. The v/c and queue
storage ratios are checked next. If either of these parameters exceeds 1, the LOS
for all O-Ds that utilize that movement is F. The final delay calculations and
resulting LOS for each O-D and each lane group are presented in Exhibit 34-97
and Exhibit 34-98. As shown, the v/c ratio and RQ for all O-Ds are all below 1, and
therefore the LOS for all O-Ds is determined by using the second column of
Exhibit 23-10. The LOS for each lane group at the adjacent intersection is
assigned on the basis of Chapter 19, Signalized Intersections. After extra
distances are measured according to the Exhibit 23-8 discussion, EDTT can be
obtained from Equation 23-50 [i.e., EDTT = 100 / (1.47 × 35) + 0 = 1.9 s].
Interchangewide ETT is calculated by a weighted average of O-D movement flow
rates. Although certain individual O-D movements perform at a worse LOS, this
interchange operates at LOS C overall.


O-D Demand Delay EDTT ETT Demand Example Problem 8:
Movement (veh/h) (s/veh) (s/veh) (s/veh) v/c > 1? RQ > 1? LOS × ETT Interchange O-D Movement
A 233 45.3 1.9 47.2 No No C 10,997.6 LOS
B 227 43.0 −1.9 41.1 No No C 9,329.7
C 173 53.8 −1.9 51.9 No No C 8,978.7
D 206 72.6 1.9 74.5 No No D 15,347.0
E 107 98.1 1.9 100.0 No No E 10,700.0
F 89 39.1 −1.9 37.2 No No C 3,310.8
G 150 36.0 −1.9 34.1 No No C 5,115.0
H 236 87.0 1.9 88.9 No No E 20,980.4
I 761 56.2 0.0 56.2 No No D 42,768.2
J 650 38.2 0.0 38.2 No No C 24,830.0
Totals 2,832 152,357.4
Approach Lane Group Control Delay (s) LOS Exhibit 34-98

Through and right 48.5 C Example Problem 8: Adjacent
EB Intersection Movement LOS
Left 67.6 D
Through and right 41.4 C
WB
Left 60.0 D
Through 85.4 E
NB Right 109.1 E
Left 68.4 D
Through and right 138.6 F
SB
Left 226.6 F
Notes: EB = eastbound, WB = westbound, NB = northbound, SB = southbound.
EXAMPLE PROBLEM 9: DIAMOND INTERCHANGE WITH ROUNDABOUTS

The Interchange
The interchange of I-99 (NB/SB) and University Drive (EB/WB) is a diamond
interchange featuring roundabouts. The traffic conditions of the interchange are
2%
grade = _________ = Pedestrian Button
Exhibit 34-99
Example Problem 9:
= Lane Width
400 ft
0%
grade = _________ = Through
156 235 134 = Right

855 831
600 ft 186
= Left
= Through + Right
96 600 ft = Left + Through

821
960 210 204
80
= Left + Right
0%
400 ft
400 ft
______________
Street = Left + Through + Right
I-99
_____________
Freeway 2%
grade = _________
D= 500 ft
The Question
What are the control delay and LOS for this interchange?

The Facts
There are no closely spaced intersections to this interchange. This
interchange has 3% heavy vehicles and a PHF of 0.97. During the analysis period,
there is no parking, and no buses, bicycles, or pedestrians utilize the interchange.
Extra distance traveled along each freeway ramp is 100 ft. The grade is 2% on the
NB and SB approaches.
Solution
O-Ds through this diamond interchange are calculated by using the worksheet
provided in Exhibit 34-169 in Section 4. The results of the O-D calculations and
the resulting PHF-adjusted values are presented in Exhibit 34-100.
Exhibit 34-100 Heavy Vehicle–

Example Problem 9: Adjusted PHF-Adjusted Adjusted Demand
O-D Table O-D Movement Demand (veh/h) Demand (veh/h) (pc/h)
A 179 185 191
B 169 174 179
C 122 126 130
D 228 235 242
E 93 96 99
F 78 80 82
G 94 97 100
H 119 123 127
I 509 525 541
J 529 545 561
K 0 0 0
L 0 0 0
M 0 0 0
N 0 0 0
Calculation of Approach Capacity and Control Delay

To estimate the delay of each approach to the roundabout, the procedures
outlined in Section 4 are used to estimate the entering and conflicting flow rates
and the resulting capacity of each approach. Exhibit 34-160 and Exhibit 34-161
are used to determine the entering and conflicting flow rates for each approach
of the interchange. For example, the northbound ramp movement (Number 13 in
Exhibit 34-160) consists of O-D Movements A, B, K, and M at a diamond
interchange (Exhibit 34-161). The conflicting flow (Number 12) consists of O-D
Movements D, E, I, and N. Exhibit 34-101 shows the entering and conflicting flow
for each approach, along with the corresponding capacity and delay.
Exhibit 34-101 Entering Flow Conflicting Flow Capacity Control Delay

Example Problem 9: Approach Approach (pc/h) (pc/h) (pc/h) (s/veh)
Capacity and Delay EB EXT 722 369 782 34.5
Calculations EB INT 882 0 1,130 13.4
WB EXT 788 289 846 33.8
WB INT 879 0 1,130 13.3
NB RAMP 370 882 468 30.9
SB RAMP 372 879 469 31.1
Notes: EXT = external, INT = internal, EB = eastbound, WB = westbound, NB = northbound, SB = southbound.

O-D Movement Control Delay and LOS

Delay for each O-D is estimated as the sum of approach delays for each
approach utilized by the O-D. For example, O-D Movement A will utilize the
northbound ramp approach and westbound internal through approach. Control
delays for these approaches are then summed to estimate control delay for O-D
Movement A. LOS for each O-D is assigned on the basis of Exhibit 23-14. The
resulting control delay and LOS for all movements are shown in Exhibit 34-102.
After extra distances are measured according to the Exhibit 23-8 discussion,
EDTT can be obtained from Equation 23-50 [i.e., EDTT = 100 / (1.47 × 35) + 0 =
1.9 s]. Interchangewide ETT is calculated by a weighted average of O-D
movement flow rates. This interchange operates at LOS D overall.
O-D Heavy Vehicle–Adjusted Control Delay EDTT ETT Demand × Exhibit 34-102
Movement Demand (veh/h) (s/veh) (s/veh) (s/veh) LOS ETT Example Problem 9: Control
A 174 44.2 1.9 46.1 D 8,805.1 Delay and LOS for Each O-D
B 168 30.9 −1.9 29.0 C 5,191.0 Movement
C 126 31.1 −1.9 29.2 C 3,796.0
D 547 44.5 1.9 46.4 D 11,228.8
E 177 47.9 1.9 49.8 D 4,930.2
F 84 34.5 −1.9 32.6 C 2,673.2
G 221 33.8 −1.9 31.9 C 3,190.0
H 194 47.1 1.9 49.0 D 6,223.0
I 911 47.9 0.0 47.9 D 25,913.9
J 881 47.1 0.0 47.1 D 26,423.1
Totals 2,252 98,374.3
EXAMPLE PROBLEM 10: OPERATIONAL ANALYSIS FOR TYPE SELECTION

The Interchange
An interchange is to be built at the junction of I-83 (NB/SB) and Archer Road
(EB/WB) in an urban area. The interchange type selection methodology described
in Section 3 is used.
The Question
Which interchange type is likely to operate better under the given demands?
The Facts
This interchange will have two-lane approaches with single left-turn lanes on
the arterial approaches. Freeway ramps will consist of two-lane approaches with
channelized right turns in addition to the main ramp lanes. Default saturation
flow rates for use in the type selection analysis are given in Exhibit 34-151. The
O-D movements of traffic through this interchange are shown in Exhibit 34-103.

Exhibit 34-103 O-D Movement Volume (veh/h)

Example Problem 10: O-D A 400
Demand Information for the B 350
Interchange C 400
D 550
E 150
F 200
G 225
H 185
I 600
J 800
K 2,500
L 3,200
M 0
N 10
Outline of Solution
Mapping O-D Flows into Interchange Movements
The primary objective of this example is to compare up to eight interchange
types against a given set of design volumes. The first step is to convert these O-D
flows into movement flows through the signalized interchange. The interchange
type methodology uses the standard NEMA numbering sequence for
interchange phasing, and Exhibit 34-152 in Section 3 demonstrates which O-Ds
make up each NEMA phase at the eight interchange types. Exhibit 34-104 shows
the corresponding volumes for this example on the basis of the O-Ds from
Exhibit 34-103. Since this interchange has channelized right turns, Exhibit 34-105
shows only the NEMA phasing volumes utilizing the signals.
Exhibit 34-104 Interchange NEMA Phase Movement Number

Example Problem 10: NEMA Type 1 2 3 4 5 6 7 8
Flows (veh/h) for the SPUI 185 800 400 400 150 1,025 560 350
Interchange TUDI /CUDI 185 950 -- 960 160 1,210 -- 750
CDI (I) 185 950 -- 960 -- 1,200 -- --
CDI (II) -- 1,150 -- -- 160 1,210 -- 750
Parclo A-4Q (I) -- 750 -- 960 -- 1,385 -- --
Parclo A-4Q (II) -- 1,310 -- -- -- 985 -- 750
Parclo A-2Q (I) -- 750 -- 960 200 1,385 -- --
Parclo A-2Q (II) 225 1,310 -- -- -- 985 -- 750
Parclo B-4Q (I) 185 950 -- -- -- 1,200 -- --
Parclo B-4Q (II) -- 1,150 -- -- 160 1,210 -- --
Parclo B-2Q (I) 185 950 -- -- -- 1,200 -- 400
Parclo B-2Q (II) -- 1,150 -- 350 160 1,210 -- --
Notes: SPUI = single-point urban interchange, TUDI = tight urban diamond interchange, CUDI = compressed
urban diamond interchange, CDI = conventional diamond interchange, Parclo = partial cloverleaf.
(I) and (II) indicate the intersections within the interchange type.
-- indicates that the movement does not exist in this interchange type.

Interchange NEMA Phase Movement Number Exhibit 34-105

Type 1 2 3 4 5 6 7 8 Example Problem 10: NEMA
SPUI 185 600 400 0 150 1,025 560 350 Flows for the Interchange
TUDI /CUDI 185 750 -- 560 160 1,210 -- 750 Without Channelized Right
CDI (I) 185 750 -- 560 -- 1,200 -- -- Turns
CDI (II) -- 1,150 -- -- 160 1,210 -- 750
Parclo A-4Q (I) -- 750 -- 560 -- 1,385 -- --
Parclo A-4Q (II) -- 1,150 -- -- -- 985 -- 750
Parclo A-2Q (I) -- 750 -- 560 200 1,385 -- --
Parclo A-2Q (II) 225 1,150 -- -- -- 985 -- 750
Parclo B-4Q (I) 185 750 -- -- -- 1,200 -- --
Parclo B-4Q (II) -- 1,150 -- -- 160 1,210 -- --
Parclo B-2Q (I) 185 750 -- -- -- 1,200 -- 400
Parclo B-2Q (II) -- 1,150 -- 350 160 1,210 -- --
Notes: (I) and (II) indicate the intersections within the interchange type.
-- indicates that the movement does not exist in this interchange type.
Computation of Critical Flow Ratios

Comparison between the eight intersection types begins with computation of
the critical flow ratio at each interchange type. The first intersection type to be
calculated is the SPUI by using Equation 34-1. On the basis of the default
saturation flow rate for a SPUI and the values for the NEMA phases, Exhibit 34-
106 shows the output from these calculations for a SPUI. The TUDI critical flow
ratios are calculated by using Equation 34-4. Exhibit 34-107 shows these
calculations for a 300-ft distance between the two TUDI intersections.
Value Signalized Right Turns Channelized Right Turns Exhibit 34-106

Critical flow ratio for the arterial Example Problem 10: SPUI
0.368 0.306 Critical Flow Ratio Calculations
movements, A
Critical flow ratio for the ramp
0.350 0.156
movements, R
Sum of critical flow ratios, Yc 0.718 0.462

Effective flow ratio for Example Problem 10: TUDI
concurrent phase when 0.070 0.070 Critical Flow Ratio Calculations
dictated by travel time, yt
Effective flow ratio for
0.070 0.070
concurrent Phase 3, y3
Effective flow ratio for
0.070 0.070
concurrent Phase 7, y7
Critical flow ratio for the arterial
0.461 0.294
movements, A
0.474 0.315
movements, R

Flow ratio for Phase 2 with Example Problem 10: CUDI
0.264 0.208
consideration of pre-positioning, y2 Critical Flow Ratio Calculations
Flow ratio for Phase 6 with
0.208 0.208
consideration of pre-positioning, y6
0.373 0.332
movements, A
0.267 0.156
movements, R

Exhibit 34-109
Value Signalized Right Turns Channelized Right Turns
Example Problem 10: CDI
Critical Flow Ratio Calculations Critical flow ratio for the arterial
0.373 0.333
movements at Intersection I, AI
0.282 0.165
movements at Intersection I, RI
Sum of critical flow ratios at
0.655 0.498
Intersection I, Yc,I
0.430 0.368
movements at Intersection II, AII
0.221 0.118
movements at Intersection II, RII
0.651 0.486
Intersection II, Yc,II
Maximum sum of critical flow ratios,
0.655 0.498
Yc
The CUDI critical flow ratios are calculated by using Equation 34-9. Exhibit
34-108 shows these calculations for a CUDI with the given O-D flows.
The CDI, Parclo A-4Q, Parclo A-2Q, Parclo B-4Q, and Parclo B-2Q all use
separate controllers. For these interchanges the critical flow ratios are calculated
for each intersection, and then the maximum is taken for the overall interchange
critical flow ratio. These numbers are all calculated by using Equation 34-14 and
the default saturation flows. Exhibit 34-109 through Exhibit 34-113 show the
calculations for these interchanges utilizing two controllers.
Exhibit 34-110 Value Signalized Right Turns Channelized Right Turns

Example Problem 10: Parclo Critical flow ratio for the arterial
A-4Q Critical Flow Ratio 0.385 0.333
Calculations Critical flow ratio for the ramp
0.282 0.282
0.667 0.615
0.364 0.333
0.208 0.111
0.572 0.444
0.667 0.615
Yc


Example Problem 10: Parclo
Critical flow ratio for the arterial A-2Q Critical Flow Ratio
0.502 0.451
movements at Intersection I, AI Calculations
0.282 0.165
0.784 0.616
0.430 0.452
0.221 0.111
0.651 0.563
Maximum sum of critical flow
0.784 0.616
ratios, Yc

Example Problem 10: Parclo
Critical flow ratio for the arterial B-4Q Critical Flow Ratio
0.373 0.333
movements at Intersection I, AI Calculations
0.000 0.000
0.373 0.333
0.430 0.368
0.000 0.000
0.430 0.368
0.430 0.368
Yc

Critical flow ratio for the arterial Example Problem 10: Parclo
0.373 0.333 B-2Q Critical Flow Ratio
Critical flow ratio for the ramp Calculations
0.111 0.111
0.484 0.444
0.430 0.368
0.103 0.103
0.533 0.471
Maximum sum of critical flow
0.533 0.471
ratios, Yc
Estimation of Interchange Delay

Estimation of interchange delay is the final step when interchange types are
compared. On the basis of the critical flow ratios calculated previously, Exhibit
34-159 in Section 3 can be used to calculate the delay at the eight interchange
types. Exhibit 34-114 shows the solutions to these calculations.

Exhibit 34-114 Intersection Interchange Delay dI (s/veh) Interchange Delay dI (s/veh)

Example Problem 10: Type Right Turns Signalized Right Turns Free or YIELD-Controlled
Interchange Delay for the SPUI 62.9 22.0
Eight Interchange Types TUDI 217.7 33.3
CUDI 35.9 27.4
CDI 26.6 21.7
Parclo A-4Q 26.2 21.6
Parclo A-2Q 47.4 29.0
Parclo B-4Q 11.9 11.3
Parclo B-2Q 30.7 29.0
Results
As demonstrated by Exhibit 34-114, a Parclo B-4Q would be the best
interchange type to select in terms of operational performance for the given O-D
flows at this interchange. For the final interchange type selection, however,
additional criteria should be considered, including those related to economic,
environmental, and land use concerns.
EXAMPLE PROBLEM 11: ALTERNATIVE ANALYSIS TOOL

This example presents a simulation analysis of the diamond interchange
configuration originally described in Example Problem 1. A few changes have
been made to introduce elements that are beyond the stated limitations of the
interchange ramp terminal procedures. The use of a typical simulation tool to
address the limitations is described in this section. The need to determine
performance measures from analysis of vehicle trajectories was emphasized in
Chapter 7, Interpreting HCM and Alternative Tool Results. Specific procedures
for defining measures in terms of vehicle trajectories were proposed to guide the
development of alternative tools. Pending further development, the example
presented in this chapter applied existing versions of alternative tools and thus
does not reflect the trajectory-based measures described in Chapter 7.
Operational Characteristics
A two-way STOP-controlled (TWSC) intersection was introduced 600 ft west
of the first signalized intersection of the interchange. Ramp metering signals
were installed on both of the freeway entrance ramps. Right-turn storage bays
were introduced on all approaches to the interchange that accommodated right
turns. The demand volumes were modified to introduce conditions that varied
from undersaturated to heavily oversaturated. The signal timing plan was
modified to accommodate the distribution of volumes. Exhibit 34-115 shows the
interchange configuration and demand volumes. The demand volumes are
referenced to the total directional arterial demand d, which varies from 600 to
1,800 veh/h. The turning movement volumes entering and leaving the arterial
have been balanced for continuity of traffic flow. The turning movements
entering and leaving the freeway were set at 25% of the total approach volumes
and were adjusted proportionally to match the arterial demand volumes. The
cross-street entry demand from the TWSC intersection was held constant at 100
veh/h in each direction, with 50% assigned to the left and right turns. No through
vehicles were assigned from the cross street at this intersection.

Exhibit 34-115
Example Problem 11:
Interchange Configuration and
Demand Volumes
Note: TWSC = two-way STOP control.
Exhibit 34-116 shows the signal timing plan for both intersections of the
diamond interchange. A simple three-phase operation at each intersection is
depicted in this table. No attempt has been made to optimize the phasing or
timing since the main purpose of this example is to demonstrate self-aggravating
phenomena that are not recognized by the Chapter 23 procedures. The ramp
metering signals installed on each of the entrance ramps were set to release a
single vehicle at 10-s intervals, giving a capacity of 360 veh/h for each ramp.
Movement Green (s) Yellow (s) All Red (s) Exhibit 34-116
Entry through/left 20 4 1 Example Problem 11: Signal
Entry and exit through/right 45 4 1 Timing Plan
Ramp 20 4 1
Cycle length (s) 100
Summary of Simulation Runs

Operation of this interchange was simulated by using demand volumes d
ranging from 600 veh/h (very undersaturated) to 1,800 veh/h (very
oversaturated). The volume increment was 200 veh/h. Thirty simulations were
run for each condition to capture stochastic variations inherent to simulation.
Two configurations were examined for each of the demand levels:
1. A single intersection at the west end of the diamond interchange and
2. The full diamond interchange with ramp metering.
Both of these configurations are illustrated in Exhibit 34-117. The west
intersection was examined separately to show the difference between a
signalized intersection operating independently and one operating as part of a
diamond interchange with mutual interactions between intersections at each end.

Exhibit 34-117
Example Problem 11: Physical
West intersection
Configurations Examined only
Full diamond
interchange with
ramp metering
Diamond Interchange Operation

Exhibit 34-118 illustrates the self-aggravating effects from interactions among
the two signals that make up the interchange and the ramp metering. Backup
and congestion are observed at high demands on all approaches. The left-turn
bays on the internal interchange segments spill over to block through traffic.
Backup from the ramp metering signals causes additional impediment to traffic
trying to leave the interchange.
Exhibit 34-118
Example Problem 11:
Congested Approaches to
Diamond Interchange
Excessive delays will be associated with the oversaturated operation.

However, for purposes of this example, the reduction in capacity is of more
interest because capacity reductions due to self-aggravating phenomena are not
fully recognized by the Chapter 23 methodology. Proper assessment of delay
with heavy oversaturation would require a more complex procedure involving
multiperiod analysis with possible consideration of route diversion due to the
excessive congestion. Therefore, this example will be limited to examining the

capacity reduction that results from interaction between the elements within this
system. The extent of the capacity reduction will be estimated by the relationship
between demand (input) and discharge (output) on the various segments.
Exhibit 34-119 shows the westbound arterial discharge from the diamond
interchange (through plus left turns) as a function of arterial demand d. Note that
the discharge tracks the demand at low volumes, which indicates that all arrivals
were accommodated. As the demand increases, the discharge levels off at a point
that indicates the capacity of the approach. When the approach is a part of an
isolated intersection, the capacity nears 1,600 veh/h. A much lower capacity
(about 850 veh/h) is attainable in the case of the diamond interchange with ramp
metering. A number of self-aggravating phenomena reduce the capacity. Some
westbound vehicles are unable to enter the east intersection because of backup
from internal westbound left-turn bay spillover. Other westbound vehicles are
unable to exit the interchange because of backup from the ramp metering signal
and because of blockage of the intersection by left-turning exit ramp vehicles.
The net result is a substantial reduction in capacity that would not be evident
from application of the Chapter 23 methodology.
Exhibit 34-120 shows the effect of the demand volume on the southbound
exit ramp discharge at the west signal of the diamond interchange. With an
isolated signal, the discharge levels off at the approach capacity. As shown, the
capacity is reduced slightly when the signal is part of a diamond interchange.
The reduction was not as apparent as it was for the arterial movements because
the blockage effects are not as significant. Some left turns were unable to enter
the intersection because of backup from the east signal. The right turns from the
ramp were not subject to any blockage effects.
2,000 Exhibit 34-119

Example Problem 11:
1,800 Discharge from the Diamond
Interchange Under the Full
1,600
Range of Arterial Demand
Westbound Discharge (veh/h)
1,400
1,200
1,000
800
600
400
200
0
600 800 1,000 1,200 1,400 1,600 1,800
Arterial Demand, d (veh/h)

Demand Diamond Intersection

Exhibit 34-120 1,000

Example Problem 11:
900
Discharge from the
Southbound Exit Ramp Under
800
the Full Range of Ramp
Demand
Ramp Discharge (veh/h)

700
600
500
400
300
200
100
0
300 400 500 600 700 800 900
Ramp Demand (veh/h)
Demand Intersection Diamond
TWSC Intersection Operation

The TWSC analysis procedures prescribed in Chapter 20 recognize the effects
of adjacent signalized intersections to some extent, but they do not address cases
in which an approach is blocked throughout part of a cycle by stationary queues
that prevent vehicles from entering on the minor street. This situation is depicted
in Exhibit 34-121, in which a stationary queue of eastbound vehicles backed up
from the west intersection of the diamond interchange has blocked the entry to
the intersection for three of the four minor-street movements.
Exhibit 34-121
Example Problem 11:
Congested Approaches to the
TWSC Intersection
Exhibit 34-122 shows the minor-street entry as a function of the arterial

demand. Unlike the other movements in this example, the minor-street demand
was kept constant throughout the entire range of arterial demand. According to a
well-established principle of TWSC analysis, the entry capacity for minor-street
movements diminishes with increasing major-street volumes. That phenomenon
is depicted clearly for northbound traffic in Exhibit 34-121. It is evident here that
capacity begins to drop below demand at about 800 veh/h in each arterial
direction. The southbound situation, on the other hand, presents some surprising

results. The southbound left turn is impeded by a queue of westbound vehicles

backed up from the interchange, as expected. The southbound right turn,
assisted by gaps created by the interchange signal, experiences an increase in
capacity, producing entry volumes that exceed the original demand. Animated
graphics indicate that some of the southbound left-turn vehicles were unable to
maneuver into the proper lane. The driver behavior model of the simulation tool
reassigned these vehicles to right turns because of excessive waiting times. This
effect provides a clear example of the difference between simulation modeling
and the analytical approach presented throughout the HCM.
Exhibit 34-122
Example Problem 11: Effect of
Arterial Demand on Minor-
Street Discharge at the TWSC
Intersection
(a) Northbound
(b) Southbound

EXAMPLE PROBLEM 12: FOUR-LEGGED RESTRICTED CROSSING

U-TURN INTERSECTION WITH MERGES
The Intersection
An RCUT with merges in a rural area has four approaches.
The Question
What is the LOS for each of the 12 movements at the intersection?
The Facts
The geometry is as pictured in Exhibit 23-42, with the main street running
east–west. The distance from the main intersection to each U-turn crossover is
2,000 ft. The storage bay length for each left-turn crossover is 300 ft. The PHF is
0.92. Free-flow speed on the major street is 60 mi/h. The truck percentages are
zero, and there are no significant grades on any approach. Exhibit 34-123 shows
the vehicular demands (veh/h).
Exhibit 34-123
Example Problem 12: Turning
Movement Demands
Solution
The solution follows the 10-step procedure outlined in Chapter 23. Once the
v/c ratio, 95% queue-to-storage ratio, and experienced travel time have been
determined for a movement, its LOS will be found by using Exhibit 23-13.
Determination of O-D Demands and Movement Demands

Exhibit 34-124 shows the demands (veh/h) redistributed to the different
junctions of the RCUT.
Exhibit 34-124
Example Problem 12:
Demands Converted to the
RCUT Geometry
Determination of Lane Groups

RCUTs with merges do not have signals, so there is no need to determine
lane or movement groups at each approach. Exhibit 34-125 shows the redistributed
demands converted to flow rates (veh/h) by using the PHF and Equation 23-57.

Exhibit 34-125
Example Problem 12: Flow
Rates in the RCUT Geometry
Determination of Lane Utilization

This step is not needed for an RCUT with merges.
Calculation of Signal Progression Adjustments

This step is not needed for an RCUT with merges.
Calculation of Additional Control-Based Adjustments

For an RCUT with merges, the analyst may use judgment to determine
whether significant weaving delay exists. When significant weaving delay exists,
the analyst must develop an estimate of this delay from field measurements or an
alternative tool and add it to the EDTT estimate calculated later.
Calculation of Junction-Specific Performance Measures

At an RCUT with merges that passes the weaving area tests in Step 5, control
delay is only experienced by the major-street left turns. Use of the methods of
Chapter 20 with the inputs listed above, and with default values for all other
factors provided, produces the following results:
• For the eastbound left turn (at the north main intersection), v/c = 0.18, 95%
queue length = 0.66 veh or 16.5 ft at 25 ft/veh, and control delay = 11.2
s/veh; and
• For the westbound left turn (at the south main intersection), v/c = 0.35, 95%
queue length = 1.58 veh or 39.5 ft at 25 ft/veh, and control delay = 15.0 s/veh.
Calculation of Extra Distance Travel Time

The bottom portion of Exhibit 23-48 shows that at a four-legged RCUT with
merges, extra travel distance is experienced by the left turns from the minor
street and by the through movements on the minor street. Both minor left turns
will experience the same extra distance travel time (EDTT) since the distances
from the main intersection to both U-turn crossovers are the same. Use of
Equation 23-58 results in the following EDTT:
𝐷𝑡 + 𝐷𝑓
𝐸𝐷𝑇𝑇 = +𝑎
1.47 × 𝐹𝐹𝑆
2,000 + 2,000
𝐸𝐷𝑇𝑇 = + 10 = 55.4 s/veh
1.47 × 60
Both minor-street through movements will experience the same EDTT, since
the distances from the main intersection to both U-turn crossovers are the same.
Use of Equation 23-58 results in the following EDTT:

2,000 + 2,000
𝐸𝐷𝑇𝑇 = + 15 = 60.4 s/veh
1.47 × 60
Calculation of Additional Weaving Delay

In this example problem, it is assumed that no significant weaving delay exists,
in the analyst’s judgment. Therefore, there are no adjustments to make in this step.
Calculation of Experienced Travel Time

Experienced travel time (ETT) is computed with Equation 23-60:
𝐸𝑇𝑇 = ∑ 𝑑𝑖 + ∑ 𝐸𝐷𝑇𝑇
The bottom portion of Exhibit 23-48 gives the following:

• For the EB left from the major street, ETT = 11.2 + 0 = 11.2 s/veh.
• For the WB left from the major street, ETT = 15.0 + 0 = 15.0 s/veh.
• For the major-street through movements, ETT = 0 + 0 = 0 s/veh.
• For the major-street right-turn movements, ETT = 0 + 0 = 0 s/veh.
• For the left turns from the minor street, ETT = (0 + 0) + 55.4 = 55.4 s/veh.
• For the through movements from the minor street, ETT = (0 + 0) + 60.4 =
60.4 s/veh.
• For the right turns from the minor street, ETT = 0 + 0 = 0 s/veh.
Determination of Level of Service

The LOS for each movement is obtained with Exhibit 23-13 (it has been
established that the v/c ratio was less than 1.0 at all junctions and that the queue-
to-storage ratios were well below 1.0 for the 300-ft bay lengths provided):
• For the eastbound left from the major street, LOS = B.
• For the westbound left from the major street, LOS = B.
• For the major-street through movements, LOS = A.
• For the major-street right-turn movements, LOS = A.
• For the minor-street left turns, LOS = E.
• For the minor-street through movements, LOS = E.
• For minor-street right turns, LOS = A.
Discussion
The minor-street left-turn and through movements experience LOS E
because of the distances from the main intersection to the U-turn crossovers and
the major-street free-flow speed. Chapter 23 explores the sensitivity of EDTT and
LOS to these factors. It shows that, over typical ranges, there is some change in
EDTT and LOS as a result of these factors but that achievement of a LOS better
than D or E for minor-street left-turn and through movements with this design
will be difficult.

EXAMPLE PROBLEM 13: THREE-LEGGED RESTRICTED CROSSING

U-TURN INTERSECTION WITH STOP SIGNS
The Intersection
An RCUT with STOP signs in a rural area has three approaches.
The Question
What is the LOS for each of the six movements at the intersection?
The Facts
The main street has two through lanes in each direction and runs north–
south. The distance from the main intersection to the U-turn crossover is 700 ft.
The storage bay lengths for the left-turn and U-turn crossovers are 400 ft. The
PHF is 0.90. The free-flow speed on the major street is 60 mi/h. The truck
percentage is 6% on all approaches. The grade on the EB approach is 2%, there
are no pedestrians, and there are no nearby traffic signals. Exhibit 34-126 shows
the vehicular demands (veh/h) and a diagram of the intersection.
Exhibit 34-126
Example Problem 13: O-D
Demands and Intersection
Diagram
Solution
v/c ratio, queue-to-storage ratio, and experienced travel time have been
determined for a movement, its LOS will be found with Exhibit 23-13.
Step 1: Determination of O-D Demands and Movement Demands

Exhibit 34-127(a) shows the demands (veh/h) redistributed to the various

Exhibit 34-127
Example Problem 13:
Movement Demands and Flow
Rates in the RCUT Geometry
(a) Demands (b) Flow Rates
Step 2: Determination of Lane Groups

RCUTs with STOP signs do not have traffic signals, so there is no need to
determine lane or movement groups at each approach. Exhibit 34-127(b) shows
the redistributed demands converted to flow rates (veh/h) on the basis of the
PHF and Equation 23-57. Exhibit 34-126 showed the RCUT lane configurations.
Step 3: Determination of Lane Utilization

This step is not needed for an RCUT with STOP signs.
Step 4: Calculation of Signal Progression Adjustments

This step is not needed for an RCUT with STOP signs.
Step 5: Calculation of Additional Control-Based Adjustments

For this RCUT with STOP signs, no field data on the base critical headway
and base follow-up time are available, so the solution will use the default values
suggested in Chapter 23.
Step 6: Calculation of Junction-Specific Performance Measures

The bottom of Exhibit 23-49 shows that, for a three-legged RCUT with STOP
signs, control delay is experienced by the major-street left-turn and minor-street
left-turn and right-turn vehicles at the main junction and by the minor-street left-
turn vehicles at the U-turn crossover. The methods of Chapter 20, with the inputs
listed above and default values for all other factors, provide the results shown in
Exhibit 34-128:

Main Intersection U-Turn Crossover Exhibit 34-128

Value EB R NB L NB T SB T SB R SB U NB T Example Problem 13: Control
Movement 12 1 2 5 6 4U 2 Delay Calculations from the
vi (veh/h) 344 189 1,167 889 156 167 1,189 Two-Way STOP-Controlled
vc,i (veh/h) 444 1,044 0 0 0 1,189 0 (TWSC) Intersections
tc,base (s) 6.9 4.1 * Methodology
tc,HV (s) 2.0 2.0 *
PHV 0.06 0.06 *
tc,G (s) 0.1 0.0 *
G (%) 2 0 *
tc,x (s) 7.22 4.22 4.4
tf,base (s) 3.3 2.2 *
tf,HV (s) 1.0 1.0 *
tf,x (s) 3.36 2.26 2.6
cp,x (veh/h) 537 638 483
v/c 0.64 0.30 0.35
T (h) 0.25 0.25 0.25
di (s/veh) 22.9 13.0 16.3
LOS C B C
Q95 (veh) 4.5 1.2 1.5
Lpc (ft) 25 25 25
LHV (ft) 45 45 45
Lh (ft) 26.2 26.2 26.2
Q95 (ft) 113 32 40
La (ft) NA 400 400
RQ% NA 0.08 0.10
Notes: EB = eastbound, NB = northbound, SB = southbound, R = right turn, L =left turn, T = through,
U = U-turn, and NA = not applicable.
*Not used in calculation; default critical headway and follow-up time for a U-turn crossover used instead.
See Chapter 20 for details of the two-way STOP-controlled intersections methodology.
See Chapter 31 for details of the queue storage ratio estimation procedure.
Step 7: Calculation of Extra Distance Travel Time

The bottom portion of Exhibit 23-49 shows that at a three-legged RCUT with
STOP signs, extra travel distance is experienced by the left turns from the minor
street. Use of Equation 23-59 gives the extra distance travel time (EDTT):
𝐷𝑡 + 𝐷𝑓
𝐸𝐷𝑇𝑇 =
1.47 × 𝐹𝐹𝑆
700 + 700
𝐸𝐷𝑇𝑇 = = 15.9 s/veh
1.47 × 60
Step 8: Calculation of Additional Weaving Delay

For an RCUT with STOP signs there are no adjustments to make in this step.
Steps 9–10: Calculation of Experienced Travel Time and Determination of Level

of Service
Use of the bottom portion of Exhibit 23-49 gives the calculations shown in
Exhibit 34-129. LOS for each movement is obtained with Exhibit 23-13. It was
shown in Exhibit 34-128 that the v/c ratio is less than 1.0 at all junctions and that
the queue-to-storage ratios are well below 1.0 for the 400-ft bay lengths
provided).

Exhibit 34-129 Value EB L EB R NB L NB T SB T SB R

Example Problem 13: vi (veh/h) 167 178 189 1,000 889 156
Experienced Travel Time di (s/veh)—first junction 22.9 22.9 13.0 0.0 0.0 0.0
Calculations and Level of di (s/veh)—second junction 16.3 None None None None None
Service EDTT (s/veh) 15.9 0.0 0.0 0.0 0.0 0.0
ETT (s/veh) 55.2 22.9 13.0 0.0 0.0 0.0
LOS E C B A A A
Note: EB = eastbound, NB = northbound, SB = southbound, R = right turn, L =left turn, and T = through.
Discussion
Interesting factors to examine in this problem are the base critical headway
and base follow-up time at the U-turn crossover and the minor-street left-turn
demand. Recalculation of the example by using the default values for base
critical headway and base follow-up time for minor-street left turns (6.5 s and 3.5
s, respectively) results in control delay at the U-turn crossover rising from 16.3 to
83.2 s/veh. In turn, this changes the ETT value for the minor-street left-turn
movement to 122.3 s/veh, which is LOS F. It is apparent that the base critical
headway and base follow-up time values used in the U-turn crossover analysis
could affect LOS by one level.
In general, the RCUT design requires extra travel time for the minor-street
left-turn and through movements while minimizing delays for the major-street
movements. Chapter 23 shows, for the conditions in this example, how far the
minor street can be pushed before it reaches LOS F. In this case, a demand of
more than 250 veh/h minor-street left turns in conjunction with 250 veh/h minor-
street right turns results in LOS F. If these are peak-period flows and typical K-
and D-factors apply, these demand levels translate to annual average daily traffic
values of 8,000 to 10,000 veh/day. Of course, better levels of service can be
achieved on the minor-street approach with an additional lane. Chapter 23 also
illustrates that minor-street left-turn LOS at an RCUT with STOP signs will rarely
achieve better than LOS D. It is apparent that the LOS constraint at an RCUT will
typically be the minor-street approach, which serves more movements than the
major-street left-turn crossover or the U-turn crossover.

EXAMPLE PROBLEM 14: FOUR-LEGGED RESTRICTED CROSSING

U-TURN INTERSECTION WITH SIGNALS
The Intersection
An RCUT with signals in a suburban area has four approaches.
The Question
What is the LOS for each of the 12 movements at the intersection and for the
facility as a whole?
The Facts
The main street runs north–south. The distance from the main intersections
to the U-turn crossovers is 800 ft. The storage bay lengths for the left-turn and U-
turn crossovers are 400 ft. The median is 40 ft wide. All crossovers have a single
lane. The major street has two through lanes and exclusive right-turn lanes at the
main junction in each direction. The minor street has two lanes on each of the
approaches to the main junctions. The PHF is 0.93. Free-flow speed on the major
street is 50 mi/h. The truck percentages are 3.7%. Grades are flat on all
approaches. There are no pedestrians, and there are no significant volumes
turning on a red signal. Exhibit 34-130 shows the vehicular demands (veh/h).
The signals are pretimed as part of a longer RCUT corridor. The arrival type
is 6 on the major street at the U-turn crossover signals in both directions and 3 for
the minor street. At both southbound signals, the cycle length is 90 s, with 60 s of
major-street green, 20 s of minor-street or crossover green, 4 s of yellow, and 1 s
of all-red. At both northbound signals, the cycle length is 60 s, with 25 s of major-
street green, 25 s of minor-street or crossover green, 4 s of yellow, and 1 s of all-red.
Exhibit 34-130
Movement Demands
Solution
v/c ratio, queue-to-storage ratio, and experienced travel time have been
determined for a movement, its LOS will be found with Exhibit 23-13.

Exhibit 34-131(a) shows the demands (veh/h) redistributed to the various

Exhibit 34-131
Example Problem 14:
Demands and Flow Rates in
the RCUT Geometry
(a) Demands (b) Flow Rates

Lane and movement groups at each approach are determined with the
methods of Chapter 19. Exhibit 34-131(b) shows the redistributed demands
converted to flow rates (veh/h) obtained by using the PHF and Equation 23-57.

With no field data on hand, the default lane distribution is applied to all
approaches to signals.

The top portion of Exhibit 23-51 is used to find arrival types for each
approach to each signal after the first signal encountered.

For this RCUT with signals, no field data are available on the saturation flow
rate for traffic in the U-turn crossover, so the solution will use the default value
of 0.85 suggested in Exhibit 23-52 for a 40-ft median width.

The top portion of Exhibit 23-48 shows that, for a four-legged RCUT with
signals, one to three increments of control delay are experienced by each
movement. The methods of Chapter 19 are applied to calculate these delays, on
the basis of the inputs listed above and defaults for all other values. The results
are shown in Exhibit 34-132.

95% Queue Control Delay Exhibit 34-132

Junction Movement v/c Length (veh) (s/veh) Example Problem 14: Control
North crossover SB through 0.92 4.4 7.6 Delay for Each Junction
WB crossover 0.40 5.0 33.3
West main SB through 0.89 3.2 5.4
intersection SB right turn 0.16 0.2 0.3
EB right turn 0.58 6.4 35.1
NB left turn 0.41 5.7 33.2
South crossover NB through 0.43 1.4 4.1
EB crossover 0.53 5.9 16.1
East main intersection NB through 0.32 1.7 6.4
NB right turn 0.51 3.1 9.1
WB right turn 0.31 2.4 12.4
SB left turn 0.09 0.8 10.8

The top portion of Exhibit 23-48 shows that at a four-legged RCUT with
signals, extra travel distance is experienced by the left turns and through
movements from the minor street. Use of Equation 23-59 gives the following
extra distance travel time (EDTT):
𝐷𝑡 + 𝐷𝑓
𝐸𝐷𝑇𝑇 =
1.47 × 𝑆𝑓
800 + 800
1.47 × 50

For an RCUT with signals, there are no adjustments to make in this step.

Use of the top portion of Exhibit 23-48 gives the results in Exhibit 34-133.
Control Delay (s/veh) by Exhibit 34-133

Traffic Control Device EDTT ETT Example Problem 14: ETT and
Movement First Second Third (s/veh) (s/veh) LOS LOS Results
NB left 4.1 33.2 None 0 37.3 D
SB left 7.6 10.8 None 0 18.4 B
NB through 4.1 6.4 None 0 10.5 B
SB through 7.6 5.4 None 0 13.0 B
NB right 4.1 9.1 None 0 13.2 B
SB right 7.6 0.3 None 0 7.9 A
EB left 35.1 16.1 6.4 21.8 79.4 E
WB left 12.4 33.3 5.4 21.8 72.9 E
EB through 35.1 16.1 9.1 21.8 82.1 F
WB through 12.4 33.3 0.3 21.8 67.8 E
EB right 35.1 None None 0 35.1 D
WB right 12.4 None None 0 12.4 B


Levels of service for each movement are shown above in Exhibit 34-133. The
results were obtained with Exhibit 23-13, after establishing that the v/c ratio was
less than 1.0 at all junctions and that the queue-to-storage ratios were well below
1.0 for the 400-ft bay lengths provided.
The ETT for the entire intersection is obtained from Equation 23-62:
∑(𝐸𝑇𝑇𝑗 × 𝑣𝑗 )
𝐸𝑇𝑇𝐼 =
∑ 𝑣𝑗
ETTI is 79,900 / 3,500 = 22.8 s/veh, which corresponds to LOS C.
Discussion
One of the concerns at an RCUT is the possibility of uneven lane distribution
on a multilane minor-street approach or a multilane U-turn crossover. The
results above were produced by assuming a relatively even lane distribution on
the two-lane minor-street approaches. On the westbound minor-street approach,
there was a demand of 200 veh/h to turn right and 130 veh/h to turn left or make
a through movement. Placing all of the right-turn vehicles in the right lane and
all of the other vehicles in the left lane would add just 0.3 s/veh of control delay
to those movements, which indicates that for situations like the one in this
example, lane distribution may not matter too much.
The effect of the saturation flow adjustment factor for U-turns can also be
examined. The default suggested in Exhibit 23-52 for this case, with a 40-ft-wide
median, is 0.85. If field data showed that the factor should be 0.8, control delay
for each movement using a crossover would increase by 0.7 to 0.9 s/veh from the
results in Exhibit 34-133. On the other hand, if field data showed that the factor
should be 0.9, the control delay for each movement using a crossover would
decrease by 0.6 to 0.7 s/veh, compared with the results in Exhibit 34-133. Overall,
the U-turn saturation flow adjustment factor only makes a small difference in
this problem.

EXAMPLE PROBLEM 15: FOUR-LEGGED MEDIAN U-TURN

INTERSECTION WITH STOP SIGNS
The Intersection
An MUT with STOP signs at the U-turn crossovers in a suburban area has
four approaches.
The Question
What is the LOS for each of the 12 movements at the intersection?
The Facts
The main street runs north–south. The distance from the main intersections
to the U-turn crossovers is 600 ft. The storage bay lengths for the left-turn and U-
turn crossovers are 500 ft. Both U-turn crossovers have a single lane. The major
street has two through lanes at the main junction, with shared right-turn lanes.
The minor street has one through lane and one exclusive right-turn lane on each
approach to the main junction. The PHF is 0.95. Free-flow speed on the major
street is 40 mi/h. The truck percentages are 2.6%. Grades are flat on all
approaches. There are 100 pedestrians per hour on each crosswalk at the main
junction, and there are no turns on red at the signal due to the pedestrians.
Exhibit 34-134 shows the vehicular demands (veh/h). The signal is actuated and
not coordinated. The yellow time is 4 s and the all-red is 1 s. Maximum green
times are 30 s for east–west phases and 50 s for north–south phases.
Exhibit 34-134
Movement Demands and
Average Interval Durations
Green (s) 39.8 18.9

Yellow (s) 4.0 4.0
Red (s) 1.0 1.0
Solution

Exhibit 34-135 shows the demands (veh/h) redistributed to the various
junctions of the MUT.

Exhibit 34-135
Example Problem 15:
Demands Converted to the
MUT Geometry

Lane and movement groups at each approach are determined with the
methods of Chapter 19. Exhibit 34-136 shows the redistributed demands
converted to flow rates (veh/h) obtained by using the PHF and Equation 23-57.
Exhibit 34-136
Rates in the MUT Geometry


With no field data on hand, the default lane distribution is applied to the
major-street approaches to the signal.

Because the signal is not coordinated, arrival types of 3 will be used on all
approaches to the signal.

For this MUT with STOP signs at the U-turn crossovers, no field data on the
base critical headway and no base follow-up time are available, so the solution
uses the default values suggested in Chapter 23.

The middle portion of Exhibit 23-50 shows that, for a four-legged MUT with
STOP signs at the U-turn crossovers, one to three increments of control delay are
experienced by each movement. The methods of Chapters 19 and 20 are applied,
by using the inputs listed above and defaults for all other values. The results are
95% Queue Control Delay Exhibit 34-137

Junction Movement v/c Length (veh) (s/veh) Example Problem 15: Control
North crossover WB crossover 0.78 7.1 34.6 Delay for Each Junction
Main intersection EB through 0.82 10.2 25.1
EB right turn 0.74 7.1 23.7
WB through 0.62 7.5 22.2
WB right turn 0.35 3.0 20.2
NB through 0.58 8.3 9.3
NB right turn 0.58 8.0 9.4
SB through 0.76 12.2 12.3
SB right turn 0.80 12.0 13.7
South crossover EB crossover 0.24 0.9 14.0

The middle portion of Exhibit 23-50 shows that at a four-legged MUT with
STOP signs at the U-turn crossovers, extra travel distance is experienced by the
left turns from the major and minor streets. Use of Equation 23-59 gives the extra
distance travel time (EDTT) as follows:
𝐷𝑡 + 𝐷𝑓
𝐸𝐷𝑇𝑇 =
1.47 × 𝑆𝑓
800 + 800
1.47 × 50

For an MUT, there are no adjustments to make in this step.


Use of the middle portion of Exhibit 23-50 gives the results in Exhibit 34-138.
Exhibit 34-138 Control Delay (s/veh) by

Example Problem 15: ETT and Traffic Control Device EDTT ETT
LOS Results Movement First Second Third (s/veh) (s/veh) LOS
NB left 9.3 34.6 13.7 20.4 78.0 E
SB left 12.3 14.0 9.4 20.4 56.1 E
NB through 9.3 None None 0 9.3 A
SB through 12.3 None None 0 12.3 B
NB right 9.4 None None 0 9.4 A
SB right 13.7 None None 0 13.7 B
EB left 23.7 14.0 9.3 20.4 67.4 E
WB left 20.2 34.6 12.3 20.4 87.5 F
EB through 25.1 None None 0 25.1 C
WB through 22.2 None None 0 22.2 C
EB right 23.7 None None 0 23.7 C
WB right 20.2 None None 0 20.2 C

LOS for each movement is shown above in Exhibit 34-138. The results were
obtained by using Exhibit 23-13, having established that the v/c ratio was less
than 1.0 at all junctions and that the queue-to-storage ratios were well below 1.0
for the 500-ft bay lengths provided.
Discussion
MUT and RCUT intersections are particularly aided by right turns and U-
turns on red because the demands for those movements are relatively higher
than at conventional intersections. If right turns on red were allowed from the
minor-street approaches in this case, where there are exclusive right-turn lanes,
the Chapter 23 example results in Part C show the effects on ETT. If 40% of the
right-turning volume (which includes the traffic that will eventually turn left) is
able to turn on red, with an estimated zero control delay, ETT will be reduced by
more than 11 s/veh for some of the minor-street movements, which will change
LOS by one level in some cases.

EXAMPLE PROBLEM 16: PARTIAL DISPLACED LEFT-TURN

INTERSECTION
The Intersection
The intersection of Speedway Boulevard (east–west) and Campbell Avenue
(north–south) has multiple failing movements and heavy left-turn demands.
Many of the nonfailing movements are close to failing, and future traffic growth
is a concern. Exhibit 34-139 provides the intersection volumes and channelization,
and Exhibit 34-140 provides the signalization information. Volumes (hourly flow
rates) listed in Exhibit 34-139 are only valid during the peak 15-min period.
Exhibit 34-139
Example Problem 16:
Intersection Volumes and
Channelization
(a) Peak 15-min Volumes (veh/h) (b) Lane Channelization
Exhibit 34-140
Example Problem 16:
Intersection Signalization
Green (s) 20.9 5.9 23.0 21.6 4.4 26.0

Yellow (s) 4.0 4.0 4.0 4.0 4.0 4.0
Red (s) 1.0 1.0 1.0 1.0 1.0 1.0
The Question
Will displacing the left turns on the major street significantly improve
performance of this intersection?
The Facts
No other signalized intersections exist within 1 mi. The intersection is
controlled by a fully actuated signal, with no right turns on red allowed. There
are no heavy vehicles, and the PHF is estimated to be 0.92. The start-up lost time
and the extension of effective green are both 2 s for all approaches. During the
analysis period, there is no parking, and no buses, bicycles, or pedestrians utilize
the intersection.

Solution
The analyst wishes to evaluate potential improvements when the east–west
left turns are displaced 350 ft upstream of the main intersection. These upstream
locations are now classified as the supplemental intersections. In the HCM
context, a DLT intersection analysis can be considered an extension of the urban
streets procedure. Thus, definitions of volume, geometric, and signalization data
for an urban street having three intersections are necessary at this stage.
Determination of Movement Demands

Exhibit 34-141 illustrates the demand volumes at each intersection in the
partial DLT configuration. The displaced eastbound and westbound left-turn
volumes are assumed to be zero at the main intersection, according to Step 1 of
the DLT computational procedure. At the western supplemental intersection,
eastbound through (709 veh/h) and right-turn (81 veh/h) demands at the main
intersection are combined into a single through (790 veh/h) demand. Similarly,
three feeding demands (northbound left, westbound through, and southbound
right) at the main intersection are combined into a westbound through (1,285
veh/h) demand. Similar flow aggregations are made at the eastern supplemental
intersection. Exhibit 34-142 illustrates lane geometries in the DLT configuration.
Exhibit 34-141
Rates at the Supplemental
and Main Intersections
Exhibit 34-142
Geometries at the
Supplemental and Main
Intersections
Determination of Lane Groups, Lane Utilization, and Signal Progression

Adjustments
Steps 2 through 4 of the DLT procedure involve lane group determination,
lane utilization, and arrival type adjustments, respectively. Lane group
determination and lane utilization are performed by the Chapter 19, Signalized
Intersections, procedures. Arrival type adjustments are handled by the flow
profile analysis from Chapter 18, Urban Street Segments.


In Step 5 of the DLT procedure, a right-turn saturation flow rate adjustment
factor is applied to the left-turn movements at the supplemental intersections. In
addition, signalization offsets must be set such that displaced left-turn vehicles
always arrive during the guaranteed green window at the main intersection. The
signalization information provided in Exhibit 34-140 should no longer be used in
a potential DLT configuration, because the major-street left-turn phases will no
longer exist at the main intersection. To ensure proper coordination, the
supplemental intersections must have the same cycle length as the main
intersection, and major-street through phases must now be treated as non-
actuated phases. Exhibit 34-143 provides the new timing plans at each
intersection. The new timing plans were generated by an alternative tool for
signal optimization.
Supplemental Intersection Timing Plans Exhibit 34-143

Example Problem 16:
Signalization at the DLT
Intersections
Green (s) 18.7 36.3 17.0 11.8 4.8 16.4 12.7 42.3
Yellow (s) 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0
Red (s) 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Main Intersection Timing Plan
After the overall new timing plans are determined, the signalization offsets
can be recalculated according to Step 5. The following steps represent the offset
computation process for DLT intersections in Chapter 23:
1. Determine the travel distance for (i.e., segment length of) the displaced
left-turn roadway TDDLT, in feet. The displaced left-turn roadway is the
roadway used by displaced left-turning vehicles as they travel from the
upstream crossover at the supplemental intersection to the stop bar at the
main intersection. In this case, the distance is 350 ft.
2. Compute the left-turn travel time TTDLT with Equation 23-63:
𝑇𝐷𝐷𝐿𝑇
𝑇𝑇𝐷𝐿𝑇 =
𝑆𝑓,𝐷𝐿𝑇 × 1.47
350
𝑇𝑇𝐷𝐿𝑇 = = 6.8 s
35 × 1.47
3. For the upstream supplemental intersection, obtain the duration between
the reference point and the start of the displaced left-turn phase LAGDLT,
in seconds. For the downstream main intersection, obtain the duration
between the reference point and the start of the major-street through
phase LAGTH, in seconds. These durations should be based on input phase
splits instead of output phase durations.
In this example, the reference point at all intersections is assumed to be
the end of the major-street through phase. From Exhibit 34-143, the
supplemental intersection’s displaced left-turn phases always begin

exactly when the major-street through phases end, so that LAGDLT is equal
to zero.
Exhibit 34-143 indicates that at the main intersection, after the major-street
through phase ends, the signal must cycle through all minor-street phases
before reaching a point where the major-street through phase begins.
However, Exhibit 34-143 illustrates average phase durations. To
determine the window of green time that is guaranteed to occur on the
major street, it is necessary to observe what the timing plan would be if
actuated phases were driven to their maximum durations. Exhibit 34-144
illustrates this timing plan.
Exhibit 34-144
Example Problem 16:
Maximum Phase Times at the
Main Intersection
Green (s) 8.0 21.0 1.0 15.0
Yellow (s) 4.0 4.0 4.0 4.0
Red (s) 1.0 1.0 1.0 1.0
Thus LAGTH is equal to 21 + 4 + 1 + 1 + 4 + 1 + 15 + 4 + 1 = 52 s. This means

that the major-street through phase begins 52 s after the reference point.
4. Obtain the offsets at the upstream supplemental intersection OSUPP and the
downstream main intersection OMAIN, both in seconds.
In this example, the initial offsets at all intersections are assumed equal to
0 s. When an existing DLT intersection having nonzero offsets is evaluated,
the existing offsets would be assigned here.
5. Compute the system start time of the displaced left-turn phase STDLT, in
seconds, for the upstream crossover at the supplemental intersection, by
𝑆𝑇𝐷𝐿𝑇 = 𝐿𝐴𝐺𝐷𝐿𝑇 + 𝑂𝑆𝑈𝑃𝑃
𝑆𝑇𝐷𝐿𝑇 = 0 + 0 = 0 s
6. Compute the system start time of the major-street through phase STTH at
the main intersection by using Equation 23-65:
𝑆𝑇𝑇𝐻 = 𝐿𝐴𝐺𝑇𝐻 + 𝑂𝑀𝐴𝐼𝑁
𝑆𝑇𝑇𝐻 = 52 + 0 = 52 s
7. Change OSUPP so that STTH is equal to STDLT + TTDLT by using Equation
23-66:
𝑂𝑆𝑈𝑃𝑃 = 𝑂𝑆𝑈𝑃𝑃 − 𝑆𝑇𝐷𝐿𝑇 + 𝑆𝑇𝑇𝐻 − 𝑇𝑇𝐷𝐿𝑇
𝑂𝑆𝑈𝑃𝑃 = 0 − 0 + 52 − 7 = 45 s
8. If the offset value is greater than the background cycle length value,
decrement the offset value by the cycle length C to obtain an equivalent
offset within the valid range.
In this example, the new offset value of 45 s is not greater than the cycle
length value of 65 s.

9. If any offset value is lower than zero, increment the offset value by the
cycle length to obtain an equivalent offset within the valid range.
In this example, the new offset value of 45 s is not lower than zero. Thus,
when the offset is set to 45 s at the supplemental intersections, displaced
left-turn vehicles are expected to pass through the main intersection
without stopping.

After the offset calculation in Step 5, Step 6 of the alternative intersection
procedure estimates the v/c ratio and control delay at each intersection. Steps 7
through 9 are not applicable to DLT intersections, and Step 10 is the LOS
determination.
For the conventional intersection design from Exhibit 34-139, intersection-
wide control delay is calculated as 64.1 s/veh by using Chapter 19 methods. For
the DLT intersection design from Exhibit 34-141, after Steps 1 through 5 of the
alternative intersection procedure are used to adjust the input data, v/c and
control delay for each isolated turn movement can be calculated by using
methods from Chapters 18 and 19. However at the overall DLT facility, turn
movement–specific control delays are encountered sequentially at each
intersection, as shown in Exhibit 34-145.
Move- Flows Delays Products Exhibit 34-145

Example Problem 16:
ment Orig. Int 1 Int 2 Int 3 Int 1 Int 2 Int 3 Int 1 Int 2 Int 3
Weighted Average Control
EB L 761 761 22.5 17,123 0 0 Delays
EB TH 437 859 437 1,352 0.4 41.9 2.5 344 18,310 3,380
EB R 422 422 42.5 0 17,935 0
WB L 486 486 25.7 0 0 12,490
WB TH 340 1,397 340 667 4.0 29.3 0.4 5,588 9,962 267
WB R 328 328 29.7 0 9,742 0
NB L 739 739 23.7 0 17,514 0
NB TH 439 439 19.8 0 8,692 0
NB R 425 425 19.8 0 8,415 0
SB L 500 500 26.2 0 13,100 0
SB TH 364 364 23.4 0 8,518 0
SB R 353 353 23.5 0 8,296 0
Total 5,594 159,675
Avg. 28.5
Notes: EB = eastbound, WB = westbound, NB = northbound, SB = southbound, TH = through, L = left, R = right,
Orig. = original (non-DLT) intersection, Int = intersection, Avg. = average.

Comparison of the conventional intersection delay of 64.1 s/veh with the
alternative intersection delay of 28.5 s/veh indicates that the alternative design is
expected to offer a 55% average delay reduction while processing the same
number (5,594) of vehicle trips. For DLT intersections, experienced travel time
(ETT) can be assumed as equal to control delay. According to the LOS thresholds
given in Chapter 19, Signalized Intersections, the overall DLT intersection would
operate at LOS C, in contrast to the conventional intersection operating at LOS E.
This raises the question of what might happen if left turns could be displaced on
all four intersection approaches. This is the subject of Example Problem 17.

Validity Checks
Chapter 23 cites a number of conditions that would invalidate the DLT
analysis method. If any of these conditions are met, the analysis results are
unreliable, and alternative tool analysis is recommended:
• Displaced left-turn vehicles are significantly delayed at the main
intersection,
• Displaced left-turn approach’s through and left-turning movements are
not served by exactly the same signal phasing and timing,
• Green times at the main intersection are not long enough to serve
displaced left-turning vehicle demands fully, or
• Side street green durations do not exceed the sum of (a) main street travel
time between supplemental and main intersections and (b) displaced left-
turn queue clearance time.
EXAMPLE PROBLEM 17: FULL DISPLACED LEFT-TURN INTERSECTION

The Intersection
The conventional intersection conditions in Example Problem 17 are identical
to those given in Example Problem 16, before DLT conversion.
The Question
Will displacement of left-turn movements on all four approaches
significantly improve performance of this intersection?
The Facts
The facts of the example problem are the same as in Example Problem 16.
Solution
The analyst wishes to evaluate potential improvements when left turns on all
four approaches are displaced 350 ft upstream of the main intersection. In this
case, two partial DLT analyses must be performed: one for the major street and
one for the minor street.
Determination of Movement Demands (East–West Partial DLT Analysis)

Exhibit 34-146 illustrates the major-street flow rates. Displaced left-turn
volumes are again assumed to be zero at the main intersection, according to Step
1 of the DLT computational procedure. Unlike partial DLT intersections, pseudo
right-turn modeling adjustments are needed at full DLT intersections. Minor-
street left-turn lanes have been converted to pseudo right-turn lanes on the
opposite side of the intersection. Similarly, minor-street left-turn volumes have
been combined with right-turn volumes on the opposite side of the intersection.
Exhibit 34-147 further illustrates the lane geometries at all three intersections in
the DLT configuration.

Exhibit 34-146
Rates at the Supplemental
and Main Intersections
Exhibit 34-147
Geometries at the
Supplemental and Main
Intersections
Determination of Lane Groups, Lane Utilization, and Signal Progression

Adjustments (East–West Partial DLT Analysis)
Steps 2 through 4 of the DLT procedure involve lane group determination,
lane utilization, and arrival type adjustments, respectively. Lane group
determination and lane utilization are performed by the Chapter 19, Signalized
Intersections, procedures. Arrival type adjustments should be handled by the
flow profile analysis from Chapter 18, Urban Street Segments.
Determination of Additional Control-Based Adjustments (East–West Partial DLT

Analysis)
In Step 5 of the DLT procedure, a right-turn saturation flow rate adjustment
factor is applied to the left-turn movements at the supplemental intersections. A
left-turn saturation flow rate adjustment factor is applied to both pseudo right-
turn movements at the main intersection. A start-up lost time of 0 s is assumed
for both pseudo right-turn movements at the main intersection.
Signalization offsets must then be set to allow displaced left-turn vehicles to
arrive during the guaranteed green window at the main intersection. The
signalization information provided in Exhibit 34-140 should no longer be used in
a potential DLT configuration, because the major-street left-turn phases will no
longer exist at the main intersection. To ensure proper coordination, the
supplemental intersections must have the same cycle length as the main
intersection. Because of the full DLT configuration, all phases at the main
intersection are nonactuated phases. Exhibit 34-148 illustrates new timing plans
(in units of seconds) at each intersection. The new timing plans were generated
by an alternative tool for signal optimization.

Exhibit 34-148 Supplemental Intersection Timing Plans

Example Problem 17: East–
West Signalization at the DLT
Intersections
Green (s) 13.4 21.6 17.0 18.0 9.2 25.8

Yellow (s) 4.0 4.0 4.0 4.0 4.0 4.0
Red (s) 1.0 1.0 1.0 1.0 1.0 1.0
After the overall new timing plans are determined, signalization offsets can
be recalculated according to Step 5. The following steps represent the offset
computation process for DLT intersections in Chapter 23:
1. Determine the travel distance for (i.e., segment length of) the displaced
left-turn roadway TDDLT, in feet. The displaced left-turn roadway is the
roadway used by displaced left-turning vehicles as they travel from the
upstream crossover at the supplemental intersection to the stop bar at the
main intersection. In this case, the distance is 350 ft.
2. Compute the left-turn travel time TTDLT by using Equation 23-63:
𝑇𝐷𝐷𝐿𝑇
𝑇𝑇𝐷𝐿𝑇 =
𝐹𝐹𝑆𝐷𝐿𝑇 × 1.47
350
𝑇𝑇𝐷𝐿𝑇 = = 6.8 s
35 × 1.47
3. For the upstream supplemental intersection, obtain the duration between
the reference point and the start of the displaced left-turn phase LAGDLT,
in seconds. For the downstream main intersection, obtain the duration
between the reference point and the start of the major-street through
phase LAGTH, in seconds. These durations should be based on input phase
splits instead of output phase durations.
In this example, the reference point at all intersections is assumed to be
the end of the major-street through phase. From Exhibit 34-148, the
supplemental intersection’s displaced left-turn phases always begin
exactly when the major-street through phases end, so that LAGDLT is equal
to zero.
From Exhibit 34-148 at the main intersection, after the major-street
through phase ends, the signal must cycle through the minor-street phase
before reaching a point where the major-street through phase begins. For
partial DLTs, it is necessary to observe what the timing plan would be if
actuated phases were driven to their maximum durations, but for full
DLTs, no phases are allowed to be actuated at the main intersection. Thus
LAGTH is equal to 18 + 4 + 1 = 23 s. This means that the major-street through
phase begins 23 s after the reference point.

4. Obtain the offsets at the upstream supplemental intersection OSUPP and the
downstream main intersection OMAIN, both in seconds.
For this example, the initial offsets at all intersections are assumed equal
to 0 s. When an existing DLT intersection having nonzero offsets is
evaluated, the existing offsets would be assigned here.
5. Compute the system start time of the displaced left-turn phase STDLT, in
seconds, for the upstream crossover at the supplemental intersection by
𝑆𝑇𝐷𝐿𝑇 = 𝐿𝐴𝐺𝐷𝐿𝑇 + 𝑂𝑆𝑈𝑃𝑃
𝑆𝑇𝐷𝐿𝑇 = 0 + 0 = 0 s
6. Compute the system start time of the major-street through phase STTH at
the main intersection by using Equation 23-65:
𝑆𝑇𝑇𝐻 = 𝐿𝐴𝐺𝑇𝐻 + 𝑂𝑀𝐴𝐼𝑁
𝑆𝑇𝑇𝐻 = 23 + 0 = 23 s
7. Change OSUPP so that STTH is equal to STDLT + TTDLT by using Equation
23-66:
𝑂𝑆𝑈𝑃𝑃 = 𝑂𝑆𝑈𝑃𝑃 − 𝑆𝑇𝐷𝐿𝑇 + 𝑆𝑇𝑇𝐻 − 𝑇𝑇𝐷𝐿𝑇
𝑂𝑆𝑈𝑃𝑃 = 0 − 0 + 23 − 7 = 16 s
8. If the offset value is greater than the background cycle length value,
decrement the offset value by the cycle length C to obtain an equivalent
offset within the valid range.
In this example, the new offset value of 16 s is not greater than the cycle
length value of 45 s.
9. If any offset value is lower than zero, increment the offset value by the
cycle length to obtain an equivalent offset within the valid range.
In this example, the new offset value of 16 is not lower than zero. Thus,
with offset values of 16 s at the east–west supplemental intersections,
displaced left-turn vehicles are expected to pass through the main
intersection without stopping. This completes the input data adjustments
for a partial DLT analysis in the east–west direction.
North–South Partial DLT Analysis

Input data adjustments must now be performed for a second partial DLT
analysis in the north–south direction. The cycle length of 45 s from the east–west
partial DLT analysis must now be applied to the north–south partial DLT
analysis. The main intersection timing plan from Exhibit 34-148 must not be
changed in the north–south partial DLT analysis.
Step 1 of the north–south partial DLT analysis is similar to what was
illustrated in Exhibit 34-146 and Exhibit 34-147. Steps 2 through 4 are again
handled by the Chapter 19, Signalized Intersections, and Chapter 18, Urban
Street Segments, procedures. In Step 5, a right-turn saturation flow rate
adjustment factor is again applied to the supplemental intersection left-turn
movements. A left-turn saturation flow rate adjustment factor is applied to both

pseudo right-turn movements at the main intersection. A start-up lost time of 0 s

is assumed for both pseudo right-turn movements at the main intersection.
Signalization offsets must now be set to allow displaced left-turn vehicles to
arrive during the guaranteed green window at the main intersection. Before the
offsets are calculated, green splits must be optimized in the north–south
direction, while constrained to the cycle length of 45 s. Exhibit 34-149 illustrates
new timing plans (in units of seconds) at each intersection. The new timing plans
were generated by an alternative tool for signal optimization.
After the overall new timing plans are determined, signalization offsets can be
recalculated according to Step 5. The north–south and east–west offset calculations
are mostly identical. However, LAGTH is now equal to 17 + 4 + 1 = 22 s, ultimately
leading to 15-s offsets at the north–south supplemental intersections. With offset
values of 15 s at the north–south supplemental intersections, displaced left-turn
vehicles are expected to pass through the main intersection without stopping.
Exhibit 34-149 Supplemental Intersection Timing Plans

Example Problem 17: North–
South Signalization at the DLT
Intersections
Green (s) 13.3 21.7 17.0 18.0 9.4 25.6

Yellow (s) 4.0 4.0 4.0 4.0 4.0 4.0
Red (s) 1.0 1.0 1.0 1.0 1.0 1.0

After the offset calculation in Step 5, Step 6 of the alternative intersection
procedure estimates the v/c ratio and control delay at each intersection. Steps 7
through 9 are not applicable to DLT intersections, and Step 10 is the LOS
determination.
For the conventional intersection design from Exhibit 34-139, intersectionwide
control delay is calculated as 64.1 s/veh by using Chapter 19’s methods.
For the DLT intersection design, after Steps 1 through 5 of the alternative
intersection procedure are used to adjust the input data, v/c ratio and control
delay for each isolated turn movement can be calculated with methods from
Chapter 19, Signalized Intersections, and Chapter 18, Urban Street Segments.
However, for the overall DLT facility, turn movement–specific control delays are
encountered sequentially at each intersection, as shown in Exhibit 34-150. To
avoid double counting, minor-street performance measures are not tabulated in
either of the two partial DLT analyses.
The full DLT delay computed here (29.0 s/veh) is similar to the partial DLT
delay (28.5 s/veh) from Example Problem 16. For DLT intersections, experienced
travel time can be assumed equal to control delay. According to Chapter 19’s
LOS thresholds, the overall DLT intersection would operate at LOS C, in contrast
to the conventional intersection operating at LOS E.

Since the major-street and minor-street demands were all relatively heavy in
Example Problems 16 and 17, the failure of the full DLT configuration to
outperform the partial DLT configuration was surprising. However, when the
same exercise was performed with 800-ft spacings between supplemental and
main intersections, the full DLT (25.3 s/veh) outperformed the partial DLT (28.4
s/veh) by more than 10%. This shows that the DLT results are sensitive to
intersection spacings and that intersection spacings should be taken into
consideration in designing a new DLT facility.
Exhibit 34-150
Move- Flows Delays
Example Problem 17:
ment Orig. Int 1 Int 2 Int 3 Int 4 Int 5 Int 1 Int 2 Int 3 Int 4 Int 5 Weighted Average Control
EB L 761 761 15.8 Delays
EB TH 437 859 437 1,352 0.6 14.5 10.4
EB R 422 422 14.6
WB L 486 486 17.5
WB TH 340 1,397 340 667 17.9 12.8 0.5
WB R 328 328 12.9
NB L 739 739 15.2
NB TH 439 439 864 1,618 13.1 0.6 14.2
NB R 425 425 13.2
SB L 500 500 17.4
SB TH 364 364 1,226 717 12.2 13.8 0.5
SB R 353 353 12.3
Total 5,594
Products
Movement Int 1 Int 2 Int 3 Int 4 Int 5
EB L 12,024 0 0 0 0
EB TH 515 6,337 14,061 0 0
EB R 0 6,161 0 0 0
WB L 0 0 8,505 0 0
WB TH 25,006 4,352 334 0 0
WB R 0 4,231 0 0 0
NB L 0 0 0 11,233 0
NB TH 0 5,751 0 518 22,976
NB R 0 5,610 0 0 0
SB L 0 0 0 0 8,700
SB TH 0 4,441 0 16,919 359
SB R 0 4,342 0 0 0
Total 162,373
Average 29.0
Notes: EB = eastbound, WB = westbound, NB = northbound, SB = southbound, TH = through, L = left, R = right,
Orig. = original (non-DLT) intersection, Int = intersection.
Validity Checks
Chapter 23 cites a number of conditions that would invalidate the DLT
analysis method. If any of these conditions are met, the analysis results are
unreliable, and alternative tool analysis is recommended:
• Displaced left-turn vehicles are significantly delayed at the main
intersection,
• The displaced left-turn approach’s through and left-turning movements
are not served by exactly the same signal phasing and timing,

• Green times at the main intersection are not large enough to serve
displaced left-turning vehicle demands fully, or
• Side street green durations do not exceed the sum of (a) main street travel
time between supplemental and main intersections and (b) displaced left-
turn queue clearance time.

3. OPERATIONAL ANALYSIS FOR

INTERCHANGE TYPE SELECTION
INTRODUCTION
The operational analysis for interchange type selection can be used to
evaluate the operational performance of various interchange types. It allows the
user to compare eight fundamental types of interchanges for a given set of
demand flows. The eight signalized interchange types covered by the
interchange type selection analysis methodology are as follows:
1. SPUI,
2. Tight urban diamond interchange (TUDI),
3. Compressed urban diamond interchange (CUDI),
4. Conventional diamond interchange (CDI),
5. Parclo A—four quadrants (Parclo A-4Q),
6. Parclo A—two quadrants (Parclo A-2Q),
7. Parclo B—four quadrants (Parclo B-4Q), and
8. Parclo B—two quadrants (Parclo B-2Q).
Other types of signalized interchanges cannot be investigated with this
interchange type selection analysis methodology. Also, the operational analysis
methodology does not distinguish between the TUDI, CUDI, and CDI types. In
general, the interchange type selection analysis methodology categorizes
diamond interchanges by the distance between the centerlines of the ramp
roadways that form the signalized intersections. This distance is generally
between 200 and 400 ft for the TUDI, between 600 and 800 ft for the CUDI, and
between 1,000 and 1,200 ft for the CDI.
The method is based on research (4). The research also provides a
methodology for selecting unsignalized interchanges. Since unsignalized
interchanges are not covered by Chapter 23, users should consult the original
source for this information.
The methodology is based on the estimation of the sums of critical flow
ratios through the interchange and their use to estimate interchange delay. A
combination of simulation and field data was used to develop critical
relationships for the methodology.
The sum of critical flow ratios is based on an identification of all flows served
during a particular signal phase and the determination of maximum flow ratios
among the movements served by that phase. The models are similar to those
used in Chapter 19 for signalized intersections; they are modified to take into
account the fact that each signal phase involves two signalized intersections.
Interchange delay is defined as the total of all control delays experienced by all
interchange movements involved in signalized ramp terminal movements
divided by the sum of all external movement flows. Additional information is
available in the source report (4).
Chapter 34/Interchange Ramp Terminals: Supplemental Operational Analysis for Interchange Type Selection
Because signalization is not specified for an interchange type selection

analysis, the following interchange types are assumed to be operated by a single
signal controller: SPUI, TUDI, and CUDI. All other types are assumed to be
operated by separate controllers at each signalized ramp terminal. In all cases,
optimal signal timing and phasing are assumed.
INPUTS AND APPLICATIONS

This interchange type selection analysis methodology can be used in several
ways:
1. For a given set of O-D interchange movements, eight basic types of
signalized interchanges may be compared on the basis of interchange
delay.
2. For a given type of interchange, the impact of intersection spacing on
interchange delay can be examined (within the range of applicability for
each interchange type).
3. For a given type of interchange, the impact of the number of lanes on
ramp and surface arterial approaches and the movements assigned to
these lanes can be examined, again by using interchange delay as the
measure of effectiveness.
For any of these applications, all interchange O-D movements must be
specified, generally by using full peak-hour volumes. The interchange type
selection methodology is not detailed enough to use flow rates or to consider
such factors as the presence of heavy vehicles.
In addition, for any given computation, the number of lanes assigned to each
phase movement and the distance between the centerlines of the two ramps,
measured along the surface arterial, must be specified.
SATURATION FLOW RATES

Implementation of the interchange type selection methodology requires the
adoption of default values for saturation flow rate. Research (3) suggests the use
of 1,900 veh/hg/ln for some basic cases. However, this is based on a suggested
base saturation flow rate of 2,000 pc/hg/ln, which is higher than the default
values suggested in Chapter 19, Signalized Intersections. For consistency with
the base saturation flow rate of 1,900 pc/hg/ln specified in Chapter 19 and to
recognize the impact of various movements on saturation flow rate, the default
values shown in Exhibit 34-151 are recommended for use in conjunction with the
interchange type selection methodology. Alternatively, if relevant information is
available, the default values provided in Chapter 19 (Exhibit 19-11 and Exhibit
19-12) may be used. Where turning movements are in shared lanes, the
“through” saturation flow rates should be used for analysis.
Operational Analysis for Interchange Type Selection Chapter 34/Interchange Ramp Terminals: Supplemental
Default Saturation Flow Rate (veh/hg/ln) Exhibit 34-151

Default Values of Saturation
Interchange Type Left Turns Through Right Turns
Flow Rate for Use with the
SPUI 1,800 1,800 1,800
Operational Analysis for
TUDI 1,700 1,800 1,800
Interchange Type Selection
CUDI 1,700 1,800 1,800
CDI 1,700 1,800 1,800
Parclo A-4Q 1,700 1,800 1,800
Parclo A-2Q 1,700 1,800 1,800
Parclo B-4Q 1,700 1,800 1,800
Parclo B-2Q 1,700 1,800 1,800
COMPUTATIONAL STEPS
Step 1: Mapping O-D Flows into Interchange Movements
Since the primary objective of an interchange type selection analysis is to
compare up to eight interchange types against a given set of design volumes,
conversion of a given set of design origin and destination volumes to movement
flows through the signalized interchange is necessary first. The methodology
identifies volumes by signal phase by using the standard NEMA numbering
sequence for interchange phasing. Thus, movements are numbered 1 through 8
on the basis of the signal phase that accommodates the movement. Not all
configurations and signalizations include all eight NEMA phases, and for some
interchange forms some movements are not signalized and do not, therefore,
contribute to interchange delay.
As for the operational analysis methodology, to simplify the mapping
process, the freeway is assumed to be oriented north–south and the surface
arterial east–west. If the freeway is oriented in the east–west direction, rotate the
interchange drawing or diagram clockwise until the freeway is in the north–
south direction. In rotating clockwise, the westbound freeway direction becomes
northbound and the eastbound freeway direction becomes southbound; the
northbound arterial direction becomes eastbound and the southbound arterial
direction becomes westbound. The methodology allows for separate
consideration of freeway U-turn movements through the interchange. Thus, 14
basic movements must be mapped for each interchange type.
For interchange types using two controllers, phase movements through the
left (Intersection I) and right (Intersection II) intersections of the interchange are
separately mapped and used in the procedure.
Exhibit 34-152 indicates the appropriate mapping of O-D demand volumes
into phase movement volumes for the eight covered interchange types. The
designation of the O-D demands is shown in Exhibit 34-162. The mapped phase
movement volumes are then used in Step 2 to compute critical flow ratios.
Exhibit 34-152 Interchange NEMA Phase Movement Number

Mapping of Interchange Type 1 2 3 4 5 6 7 8
Origins and Destinations into
SPUI H I+F A+M C E J+G D+N B
Phase Movements for
TUDI /CUDI H+M E+I+F -- D+C+N E+N H+J+G -- A+M+B
Operational Interchange Type
CDI (I) H+M E+I+F -- D+C+N -- J+A -- --
Selection Analysis
CDI (II) -- I+D -- -- E+N H+J+G -- A+M+B
Parclo A-4Q (I) -- E+I -- D+N+C -- J+A+M+H -- --
Parclo A-4Q (II) -- I+D+N+E -- -- -- J+H -- A+M+B
Parclo A-2Q (I) -- E+I -- D+N+C F J+A+H+M -- --
Parclo A-2Q (II) G I+D+E+N -- -- -- H+J -- A+M+B
Parclo B-4Q (I) H+M I+E+F -- -- -- J+A -- --
Parclo B-4Q (II) -- I+D -- -- E+N H+J+G -- --
Parclo B-2Q (I) H+M E+I+F -- -- -- J+A -- C
Parclo B-2Q (II) -- I+D -- B E+N H+J+G -- --
Notes: -- indicates that phase movement does not exist for this interchange configuration.
Bold indicates movements not included when they operate from a separate lane with YIELD or STOP control.
Step 2: Computation of Critical Flow Ratios

The subsections that follow detail the computation of the critical flow ratio Yc
for the interchange for the eight basic configurations covered by this
methodology.
Single-Point Urban Interchange

The phase movements in a SPUI are illustrated in Exhibit 34-153. The sum of
critical flow ratios is estimated as follows:
Equation 34-1 𝑌𝑐 = 𝐴 + 𝑅
with
𝑣1 𝑣2 𝑣5 𝑣6
Equation 34-2 𝐴 = max [( + ),( + )]
𝑠1 𝑛1 𝑠2 𝑛2 𝑠5 𝑛5 𝑠6 𝑛6
𝑣3 𝑣4 𝑣7 𝑣8
Equation 34-3 𝑅 = max [( + ),( + )]
𝑠3 𝑛3 𝑠4 𝑛4 𝑠7𝑛7 𝑠8 𝑛8
where
Yc = sum of the critical flow ratios,
vi = phase movement volume for phase i (veh/h),
ni = number of lanes serving phase movement i,
si = saturation flow rate for phase movement i (veh/hg/ln),
A = critical flow ratio for the arterial movements, and
R = critical flow ratio for the exit ramp movements.
Exhibit 34-153
Phase Movements in a SPUI
Source: Bonneson et al. (4 ).
Tight Urban Diamond Interchange

Phase movements in a TUDI are illustrated in Exhibit 34-154.
Exhibit 34-154
Phase Movements in a Tight
Urban or Compressed Urban
Diamond Interchange
The sum of critical flow ratios is computed as follows:

𝑌𝑐 = 𝐴 + 𝑅 Equation 34-4
with
𝑣2 𝑣4 𝑣5
𝐴 = max [( + ) − 𝑦3 , ( + 𝑦7 )] Equation 34-5
𝑠2 𝑛2 𝑠4 𝑛4 𝑠5 𝑛5
𝑣1 𝑣6 𝑣8 Equation 34-6
𝑅 = max [( + 𝑦3 ) , ( + − 𝑦7 )]
𝑠1 𝑛1 𝑠6 𝑛6 𝑠8 𝑛8
𝑣4
𝑦3 = min ( ,𝑦 ) Equation 34-7
𝑠4 𝑛4 𝑡
𝑣8
𝑦7 = min ( ,𝑦 ) Equation 34-8
𝑠8 𝑛8 𝑡
where y3 and y7 are the effective flow ratios for concurrent (or transition) Phases 3
and 7, respectively; and yt is the effective flow ratio for the concurrent phase
when dictated by travel time.
For preliminary design applications, the default values of Exhibit 34-155 are
recommended for yt. The distance between the two intersections is measured
from the centerline of the left ramp roadway to the centerline of the right ramp
roadway.
Distance Between Intersections Dʹ(ft) Default Value for yt Exhibit 34-155

200 0.050 Default Values for yt
300 0.070
400 0.085
For Phase Movements 2 and 6, the number of assigned lanes (n2 and n6) is
related to the arterial left-turn bay design. If the left-turn bay extends back to the
external approach to the interchange, the number of lanes on these external
approaches is the total number of approaching lanes, including the left-turn bay.
If the left-turn bay is provided only on the internal arterial link, n2 or n6, or both,
would not include this lane.
Compressed Urban Diamond Interchange

Exhibit 34-154 illustrates the phase movement volumes for a CUDI. They are
the same as for a TUDI. The sum of critical flow ratios is computed as follows:
Equation 34-9 𝑌𝑐 = 𝐴 + 𝑅
with
𝑣1 𝑣5
Equation 34-10 𝐴 = max [( + 𝑦2 ) , ( + 𝑦6 )]
𝑠1 𝑛1 𝑠5 𝑛5
𝑣4 𝑣8
Equation 34-11 𝑅 = max ( , )
𝑠4 𝑛4 𝑠8 𝑛8
𝑣2 𝑣5
Equation 34-12 𝑦2 = max ( , )
𝑠2𝑛2 𝑠2
𝑣8 𝑣1
Equation 34-13 𝑦6 = max ( , )
𝑠8𝑛8 𝑠6
where y2 and y6 are the flow ratios for Phases 2 and 6, respectively, with
consideration of pre-positioning.
All Interchanges with Two Signalized Intersections and Separate Controllers

These interchange types include CDI, Parclo A-4Q, Parclo A-2Q, Parclo B-4Q,
and Parclo B-2Q. The computation of the maximum sum of critical volumes is
the same for each. Each has two signalized intersections, and each is generally
operated with two controllers.
While the equations for estimating the maximum sum of critical volumes are
the same, the phase movement volumes differ for each type of interchange, as
was indicated in Exhibit 34-152. Exhibit 34-156 through Exhibit 34-158 illustrate
the phase movements for each of these interchange types.
Exhibit 34-156
Phase Movements in a CDI
Exhibit 34-157
Phase Movements in Parclo A-
2Q and A-4Q Interchanges
Source: Messer and Bonneson (3 ).
Exhibit 34-158
Phase Movements in Parclo B-
2Q and B-4Q Interchanges
Source: Messer and Bonneson (3 ).
For all conventional diamond, Parclo A, and Parclo B interchanges, the sum
of critical flow ratios is computed as follows:
𝑌𝑐,𝑚𝑎𝑥 = max(𝑌𝑐,I , 𝑌𝑐,II ) Equation 34-14
with
𝑌𝑐,I = 𝐴I + 𝑅I Equation 34-15
𝑌𝑐,II = 𝐴II + 𝑅II Equation 34-16
𝑣1 𝑣2 𝑣5 𝑣6 Equation 34-17
𝐴I,II = max [( + ),( + )]
𝑠1𝑛1 𝑠2 𝑛2 𝑠5𝑛5 𝑠6 𝑛6
𝑣4 𝑣8 Equation 34-18
𝑅I,II = max ( , )
𝑠4 𝑛4 𝑠8 𝑛8
where
Yc,I = sum of the critical flow ratios for Intersection I,
Yc,II = sum of the critical flow ratios for Intersection II,
Yc,max = sum of the critical flow ratios for the interchange,
AI = critical flow ratio for the arterial movements for Intersection I,
AII = critical flow ratio for the arterial movements for Intersection II,
AI,II = critical flow ratio for the arterial movements for the interchange,
RI = critical flow ratio for the exit-ramp movements for Intersection I,
RII = critical flow ratio for the exit-ramp movements for Intersection II, and
RI,II = critical flow ratio for the exit-ramp movements for the interchange.
Note that when values of AI, AII, RI, and RII are computed, the movement
volumes vary for Intersections I and II, even though the phase movement
designations are the same (Exhibit 34-152).
Some of the phase movement volumes do not exist in either Intersection I or
II. A value of 0 is used for the volume in each case where this occurs.
Step 3: Estimation of Interchange Delay

Interchange delay for each interchange type or design is estimated by using
regression models that were developed primarily from simulation output but
validated with a limited amount of field data (4). In each case, two delay
estimators are provided on the basis of the control of the off-ramp right-turn
movements:
• Case A, used where the right-turn movements from freeway off-ramps
are controlled by the signal.
• Case B, used where the right-turn movements from freeway off-ramps
have a separate lane or lanes that are either free (uncontrolled) or
controlled by a YIELD sign.
For SPUIs, a third condition is added. Where the right turns from the
freeway ramps are controlled by a signal and right turn on red is allowed, both
cases are used, and the results are weighted by the proportions of right turns
made during the red and green indications. Since the signal timing is unknown
for an interchange type selection application, the assumption of a 50%/50% split
is recommended.
This modification, applied only to SPUIs, is necessary due to difficulties
experienced in simulating right turn on red at these interchanges.
Exhibit 34-159 gives the delay equations used to estimate interchange delay
for the eight interchange types covered by the interchange type selection
procedure. In each case, the variables used are defined as follows:
d = interchange delay (s/veh);
Yc = critical or controlling flow ratio from Step 1; and
Dʹ = distance between the two intersections, measured between the
centerlines of the two ramp roadways along the surface arterial (ft).
Exhibit 34-159 also shows the ranges of Dʹ over which these equations are
valid. They generally represent the normal design range for these interchange
types. These equations should be used with great caution beyond these ranges.
Inter- Valid Case B: Exhibit 34-159

change Range Case A: Right Turns Free or Estimation of Interchange
Type of Dʹ (ft) Right Turns Signalized YIELD-Controlled Delay dI for Eight Basic
𝑌𝑐 𝑌𝑐 Interchange Types
SPUI 150–400 15.1 + (16.0 + 0.01𝐷) ( ) 15.1 + (5.9 + 0.008𝐷) ( )
1 − 𝑌𝑐 1 − 𝑌𝑐
𝑌𝑐 𝑌𝑐
TUDI 200–400 13.4 + 14.2 ( ) 13.4 + 12.8 ( )
1 − 𝑌𝑐 1 − 𝑌𝑐
𝑌𝑐 𝑌𝑐
CUDI 600–800 19.2 + [9.4 − 0.011(𝐷 − 700)] ( ) 19.2 + [8.6 − 0.009(𝐷 − 700)] ( )
1 − 𝑌𝑐 1 − 𝑌𝑐
𝑌𝑐 𝑌𝑐
CDI 900–1,300 17.1 + [5.0 − 0.011(𝐷 − 1,100)] ( ) 17.1 + [4.6 − 0.009(𝐷 − 1,100)] ( )
1 − 𝑌𝑐 1 − 𝑌𝑐
𝑌𝑐 𝑌𝑐
Parclo A-4Q 700–1,000 11.7 + [7.8 − 0.011(𝐷 − 800)] (
1 − 𝑌𝑐
) 11.7 + [6.6 − 0.009(𝐷 − 800)] (
1 − 𝑌𝑐
)
𝑌𝑐 𝑌𝑐
Parclo A-2Q 700–1,000 19.1 + [8.3 − 0.011(𝐷 − 800)] (
1 − 𝑌𝑐
) 19.1 + [8.3 − 0.009(𝐷 − 800)] (
1 − 𝑌𝑐
)
𝑌𝑐 𝑌𝑐
Parclo B-4Q 1,000–1,400 9.3 + [3.5 − 0.011(𝐷 − 1,200)] ( ) 9.3 + [3.4 − 0.009(𝐷 − 1,200)] ( )
1 − 𝑌𝑐 1 − 𝑌𝑐
𝑌𝑐 𝑌𝑐
Parclo B-2Q 1,000–1,400 26.2 + [3.9 − 0.011(𝐷 − 1,200)] ( ) 26.2 + [3.2 − 0.009(𝐷 − 1,200)] ( )
1 − 𝑌𝑐 1 − 𝑌𝑐
Delay estimates can be related to LOS. For consistency, the same criteria as
used for the operational analysis methodology (4) are applied. Because LOS F is
based on a v/c ratio greater than 1.00 or a queue storage ratio greater than 1.00,
this interchange type selection methodology will never predict LOS F, because it
does not predict these ratios. Users should be exceedingly cautious of results
when interchange delay exceeds 85 to 90 s/veh.
In evaluating alternative interchange types, the exact distance, Dʹ, may not be
known for each of the alternatives. It is recommended that all lengths be selected
at the midpoint of the range shown in Exhibit 34-159 for this level of analysis.
Interpretation of Results
The output of the interchange type selection procedure for signalized
interchanges is a set of delay predictions for (a) various interchange types, (b)
various distances Dʹ between the two intersections, or (c) various numbers and
assignments of lanes on ramps and the surface arterials.
Although a lower interchange delay is generally better, a final choice must
consider a number of other criteria that are not part of this methodology,
including the following:
• Availability of right-of-way,
• Environmental impacts,
• Social impacts,
• Construction cost, and
• Benefit–cost analysis.
This methodology provides valuable information that can be used, in
conjunction with other analyses, in making an appropriate choice of an interchange
type and some of the primary design parameters. However, the final design will
be based on many other criteria in addition to the output of this methodology.
Users are also cautioned that while the definition of interchange delay is
similar for the interchange type selection methodology and the operational analysis
methodology, different modeling approaches to delay prediction were taken, and
there is no guarantee that the results of the two methodologies will be consistent.
4. O-D AND TURNING MOVEMENTS
O-D AND TURNING MOVEMENTS FOR INTERCHANGES WITH

ROUNDABOUTS
Roundabouts are generally analyzed with the procedures of Chapter 22 of the
HCM. This chapter provides guidance for translating O-D demands into
movement demands at a roundabout to apply the procedures of Chapter 22.
Exhibit 34-160 defines the movements traveling through an interchange with
two roundabouts, while Exhibit 34-161 lists the O-D demands contributing to
each of these movements. For example, for diamond interchanges, O-D
Movements G, H, and J constitute Movement 15 in Exhibit 34-160.
In analyzing interchanges with roundabouts, Exhibit 34-160 and Exhibit 34-
161 should be used to establish the roundabout movements. The procedures of
Chapter 22 should then be applied to estimate the capacity and delay for each
roundabout approach. Finally, Exhibit 23-14 should be used to determine the
LOS for each O-D demand through the interchange.
Exhibit 34-160
Illustration and Notation of
O-D Demands at an
Interchange with
Roundabouts
O-D and Turning Movements Chapter 34/Interchange Ramp Terminals: Supplemental

Movement Diamond Parclo A-2Q Parclo B-2Q Parclo B-4Q Exhibit 34-161
1 C, D, L, N C, D, N -- C Notation of O-D Demands at
Interchanges with
2 D, H, L, M, N D, N H, M, N H, M
Roundabouts
3 E, F, I E, F E, F, I E, F, I
4 D, E, F, H, I, L, M, N D, E, F, I, N E, F, H, I, M E, F, H, I, M
5 -- -- C D, N
6 -- F C --
7 A, H, J, M A, H, J, M A, H, J, M A, H, J, M
8 J, M A, F, H, J, M A, C, H, J, M A, H, J, M
9 -- -- A, B, M A, M
10 -- G B -
11 D, E, I, N D, E, I, N D, E, I, N D, E, I, N
12 D, E, I, N D, E, G, I, N B, D, E D, E, I, N
13 A, B, K, M A, B, M -- B
14 A, E, K, M, N A, M E, N E, N
15 G, H, J G, H, J G, H, J G, H, J
16 A, E, G, H, J, K, M, N A, G, H, J, M E, G, H, J, N E, G, H, J, N
Movement SPUI Parclo AB-4Q Parclo A-4Q Parclo AB-2Q
1 C, D, L, N C C, D, N --
2 D, H, L, M, N H, M D, N H, M
3 E, F, I E, F, I E, F, I E, F, I
4 D, E, I, N E, F, H, I, M D, E, F, I, N E, F, H, I, M
5 A, B, K, M D, N -- C, D, N
6 A, E, K, M, N -- -- C
7 G, H, J A, H, J, M A, H, J, M A, H, J, M
8 A, H, J, M A, H, J, M A, H, J, M A, C, H, J, M
9 -- -- -- --
10 -- -- -- G
11 -- D, E, I, N D, E, I, N D, E, I, N
12 -- D, E, I, N D, E, I, N D, E, G, I, N
13 -- A, B, M A, B, M A, B, M
14 -- A, M A, M A, M
15 -- G, H, J G, H, J G, H, J
16 -- A, G, H, J, M A, G, H, J, M A, G, H, J, M
Note: -- indicates movements that do not exist for a given interchange form.
Chapter 34/Interchange Ramp Terminals: Supplemental O-D and Turning Movements

O-D AND TURNING MOVEMENTS FOR CONVENTIONAL INTERCHANGES

Exhibit 34-162 illustrates how O-D movements can be obtained from turning
movements for each type of interchange considered in this methodology. Exhibit
34-163 through Exhibit 34-177 provide the corresponding calculations for
obtaining turning movements from O-D movements.
Exhibit 34-162
O-D Flows for Each
Interchange Configuration

Input Output Exhibit 34-163

Intersection I Intersection II Worksheet for Obtaining O-D
Turning Turning Movements from Turning
Move- Volume Move- Volume Volume Movements for Parclo A-2Q
Approach ment (veh/h) ment (veh/h) O-D Movement Calculation (veh/h) Interchanges
EXT-LT LT A = (NB LT) – (NB UT)
Eastbound
RT INT-RT B = NB RT
(EB)
EXT-TH INT-TH C = SB RT
LT EXT-LT D = (SB LT) – (SB UT)
Westbound
INT-RT RT E = (EB INT-RT) – (SB UT)
(WB)
INT-TH EXT-TH F = EB EXT-LT
LT LT G = WB EXT-LT
Northbound RT RT H = (WB INT-RT) – (NB UT)
(NB) TH TH I = (EB INT-TH) – (SB LT) + (SB UT)
UT UT J = (WB INT-TH) – (NB LT) + (NB UT)
LT LT K
Southbound RT RT L
(SB) TH TH M = NB UT
UT UT N = SB UT
Notes: LT = left turn, RT = right turn, UT = U-turn, TH = through, INT = internal, EXT = external.
The flows of the two U-turn movements from the freeway (SB UT and NB UT) are user-specified.
Shading indicates movements that do not occur in this interchange form.
Exhibit 34-164
Worksheet for Obtaining O-D
Input Output Movements from Turning
Intersection I Intersection II Movements for Parclo A-4Q
Turning Turning Interchanges
Move- Volume Move- Volume Volume
Approach ment (veh/h) ment (veh/h) O-D Movement Calculation (veh/h)
LT LT A = (NB LT) – (NB UT)
Eastbound
EXT-RT INT-RT B = NB RT
(EB)
LT LT D = (SB LT) – (SB UT)
Westbound
INT-RT EXT-RT E = (EB INT-RT) – (SB UT)
(WB)
INT-TH EXT-TH F = EB EXT-RT
LT LT G = WB EXT-RT
Northbound RT RT H = (WB INT-RT) – (NB UT)
LT LT K
Southbound RT RT L
UT UT N = SB UT
The flows of the two U-turn movements from the freeway (SB UT and NB UT) are user-specified.

Exhibit 34-165
Input Output
Movements from Turning Intersection I Intersection II
Movements for Parclo AB-2Q Turning Turning
Interchanges Move- Volume Move- Volume Volume
LT LT A = (NB LT(II)) – (NB UT(II))
Eastbound
EXT-RT INT-RT B = NB RT(II)
(EB)
EXT-TH INT-TH C = NB LT(I)
INT-LT EXT-LT D = (NB RT(I)) – (NB UT(I))
Westbound
RT RT E = (EB INT-RT) – (NB UT(I))
(WB)
LT(I) LT(II) G = WB EXT-LT
Northbound RT(I) RT(II) H = (WB INT-LT) – (NB UT(II))
(NB) TH TH I = (EB INT-TH) – (NB RT(I)) + (NB UT(I))
UT(I) UT(II) J = (WB INT-TH) – (NB LT(II)) + (NB UT(II))
LT LT K
Southbound RT RT L
(SB) TH TH M = NB UT(II)
UT UT N = NB UT(I)
The flows of the two U-turn movements from the freeway [NB UT(I) and NB UT(II)] are user-specified.
Input Output
Exhibit 34-166
Intersection I Intersection II
Movements from Turning Turning Turning
Movements for Parclo AB-4Q Move- Volume Move- Volume Volume
Interchanges Approach ment (veh/h) ment (veh/h) O-D Movement Calculation (veh/h)
LT LT A = (NB LT(II)) – (NB UT(II))
Eastbound
EXT-RT INT-RT B = NB RT(II)
(EB)
EXT-TH INT-TH C = SB RT(I)
INT-LT LT D = (NB RT(I)) – (NB UT(I))
Westbound
RT EXT-RT E = (EB INT-RT) – (NB UT(I))
(WB)
LT LT(II) G = WB EXT-LT
Northbound RT(I) RT(II) H = (WB INT-LT) – (NB UT(II))
(NB) TH TH I = (EB INT-TH) – (NB RT(I)) + (NB UT(I))
UT(I) UT(II) J = (WB INT-TH) – (NB LT(II)) + (NB UT(II))
LT LT K
Southbound RT(I) RT L
(SB) TH TH M = NB UT(II)
UT UT N = NB UT(I)
The flows of the two U-turn movements from the freeway [NB UT(I) and NB UT(II)] are user-specified.


Movements from Turning
Turning Turning
Movements for Parclo B-2Q
Interchanges
LT INT-LT A = (SB RT) – (SB UT)
Eastbound
EXT-RT RT B = SB LT
(EB)
EXT-TH INT-TH C = NB LT
INT-LT LT D = (NB RT) – (NB UT)
Westbound
RT EXT-RT E = (EB INT-LT) – (NB UT)
(WB)
INT-TH EXT-TH F = (EB EXT-RT)
LT LT G = (WB EXT-RT)
Northbound RT RT H = (WB INT-LT) – (SB UT)
(NB) TH TH I = (EB INT-TH) – (NB RT) + (NB UT)
UT UT J = (WB INT-TH) – (SB RT) + (SB UT)
LT LT K
Southbound RT RT L
(SB) TH TH M = SB UT
UT UT N = NB UT
The flows of the two U-turn movements from the freeway (NB UT and SB UT) are user-specified.

Turning Turning
Interchanges
LT INT-LT A = (SB RT(II)) – (SB UT)
Eastbound
EXT-RT RT B = NB RT(II)
(EB)
EXT-TH INT-TH C = SB RT(I)
INT-LT LT D = (NB RT(I)) – (NB UT)
Westbound
RT EXT-RT E = (EB INT-LT) – (NB UT)
(WB)
LT LT G = WB EXT-RT
Northbound RT(I) RT(II) H = (WB INT-LT) – (SB UT)
(NB) TH TH I = (EB INT-TH) – (NB RT(I)) + (NB UT)
UT UT J = (WB INT-TH) – (SB RT(II)) + (SB UT)
LT LT K
Southbound RT(I) RT(II) L
(SB) TH TH M = SB UT
UT UT N = NB UT

Exhibit 34-169 Input Output

Worksheet for Obtaining O-D Intersection I Intersection II
Turning Turning
Movements for Diamond
Interchanges
LT INT-LT A = (NB LT) – (NB UT)
Eastbound
EXT-RT RT B = NB RT
(EB)
INT-LT LT D = (SB LT) – (SB UT)
Westbound
RT EXT-RT E = (EB INT-LT) – (SB UT)
(WB)
LT LT G = WB EXT-RT
Northbound RT RT H = (WB INT-LT) – (NB UT)
LT LT K = NB TH
Southbound RT RT L = SB TH
UT UT N = SB UT

Worksheet for Obtaining O-D Turning Volume Volume
Movements from Turning Approach Movement (veh/h) O-D Movement Calculation (veh/h)
Movements for SPUIs LT A= NB LT
Eastbound
RT B= NB RT
(EB)
TH C= SB RT
LT D= SB LT
Westbound
RT E= EB LT
(WB)
TH F= EB RT
LT G= WB RT
Northbound RT H= WB LT
(NB) TH I= EB TH
UT J= WB TH
LT K= NB TH
Southbound RT L= SB TH
(SB) TH M
UT N
Notes: LT = left turn, RT = right turn, UT = U-turn, TH = through.
The flow of the two U-turn movements from the freeway (NB UT and SB UT) are user-specified.

Worksheet for Obtaining O-D Intersection I Intersection II
Turning Movements from O-D Move- Volume Turning Movement Volume Turning Movement Volume
Movements for Parclo A-2Q ment (veh/h) Approach Calculation (veh/h) Calculation (veh/h)
and Parclo A-4Q Interchanges A EXT-LT = F LT
Eastbound
B RT INT-RT = E+N
(EB)
C EXT-TH = I+E INT-TH = I+D
D LT EXT-LT = G
Westbound
E INT-RT = H+M RT
(WB)
F INT-TH = J+A EXT-TH = J+H
G LT LT = A+M
H Northbound RT RT = B
I (NB) TH TH
J UT UT = M
K LT = D+N LT
L Southbound RT = C RT
M (SB) TH TH
N UT = N UT


Intersection I Intersection II Worksheet for Obtaining
O-D
Turning Movements from O-D
Move- Volume Turning Movement Volume Turning Movement Volume
Movements for Parclo AB-2Q
ment (veh/h) Approach Calculation (veh/h) Calculation (veh/h)
Interchanges
A LT LT
Eastbound
B EXT RT = F INT-RT = E+N
(EB)
D INT-LT = H+M EXT-LT = G
Westbound
E RT RT
(WB)
G LT(I) = C LT(II) = A+M
H Northbound RT(I) = D+N RT(II) = B
I (NB) TH TH
J UT(I) = N UT(II) = M
K LT LT
L Southbound RT RT
M (SB) TH TH
N UT UT

O-D
Movements for Parclo AB-4Q
Interchanges
A LT LT
Eastbound
B EXT RT = F INT-RT = E+N
(EB)
D INT-LT = H+M LT
Westbound
E RT EXT-RT = G
(WB)
G LT LT(II) = A+M
I (NB) TH TH
J UT(I) = N UT(II) = M
K LT LT
L Southbound RT(I) = C RT
M (SB) TH TH
N UT UT

O-D
Interchanges
A LT INT-LT = E+N
Eastbound
B EXT RT = F RT
(EB)
D INT-LT = H+M LT
Westbound
E RT EXT-RT = G
(WB)
G LT = C LT
H Northbound RT = D+N RT
I (NB) TH TH
J UT = N UT
K LT LT = B
L Southbound RT RT = A+M
M (SB) TH TH
N UT UT = M


Worksheet for Obtaining Intersection I Intersection II
O-D
Interchanges
A LT INT-LT = E+N
Eastbound
B EXT RT = F RT
(EB)
D INT-LT = H+M LT
Westbound
E RT EXT-RT = G
(WB)
G LT LT
I (NB) TH TH
J UT = N UT
K LT LT
L Southbound RT(I) = C RT(II) = A+M
M (SB) TH TH
N UT UT = M

Worksheet for Obtaining Intersection I Intersection II
O-D
Movements for Diamond
Interchanges
A LT INT-LT = E+N
Eastbound
B EXT RT = F RT
(EB)
D INT-LT = H+M LT
Westbound
E RT EXT-RT = G
(WB)
G LT LT = A+M
H Northbound RT RT = B
I (NB) TH TH = K
J UT UT = M
K LT = D+N LT
L Southbound RT = C RT
M (SB) TH = L TH
N UT = N UT

Worksheet for Obtaining O-D Volume Volume
Turning Movements from O-D Movement (veh/h) Approach Turning Movement Calculation (veh/h)
Movements for SPUIs
A LT = E
Eastbound
B RT = F
(EB)
C TH = I
D LT = H
Westbound
E RT = G
(WB)
F TH = J
G LT = A
H Northbound RT = B
I (NB) TH = K
J UT
K LT = D
L Southbound RT = C
M (SB) TH = L
N UT
Notes: LT = left turn, RT = right turn, UT = U-turn, TH = through.

5. REFERENCES
1. Elefteriadou, L., C. Fang, R. P. Roess, E. Prassas, J. Yeon, X. Cui, A. Kondyli,

H. Wang, and J. M. Mason. Capacity and Quality of Service of Interchange Ramp
Terminals. Final Report, National Cooperative Highway Research Program
Project 3-60. Pennsylvania State University, University Park, March 2005.
2. Elefteriadou, L., A. Elias, C. Fang, C. Lu, L. Xie, and B. Martin. Validation and
Enhancement of the Highway Capacity Manual’s Interchange Ramp Terminal
Methodology. Final Report, National Cooperative Highway Research Program
Project 3-60A. University of Florida, Gainesville, 2009.
3. Messer, C. J., and J. A. Bonneson. Capacity of Interchange Ramp Terminals. Final
Report, National Cooperative Highway Research Program Project 3-47.
Texas A&M Research Foundation, College Station, April 1997.
4. Bonneson, J., K. Zimmerman, and M. Jacobson. Review and Evaluation of
Interchange Ramp Design Considerations for Facilities Without Frontage Roads.
Research Report 0-4538-1. Cooperative Research Program, Texas
Transportation Institute, Texas A&M University System, College Station,
2004.
5. Federal Highway Administration. EDC2 Intersection and Interchange
Geometrics website. https://www.fhwa.dot.gov/everydaycounts/edctwo/
2012/. Accessed Dec. 30, 2014.
6. National Transportation Communications for ITS Protocol: Object Definitions for
Actuated Traffic Signal Controller (ASC) Units–1202. National Electrical
Manufacturers Association, Rosslyn, Va., Jan. 2005.
Chapter 34/Interchange Ramp Terminals: Supplemental References

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 35
PEDESTRIANS AND BICYCLES: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 35-1

Example Problem 1: Pedestrian LOS on Shared-Use and Exclusive
Paths ............................................................................................................... 35-2
Example Problem 2: Bicycle LOS on a Shared-Use Path ................................ 35-4
Chapter 35/Pedestrians and Bicycles: Supplemental Contents

LIST OF EXHIBITS
Exhibit 35-1 List of Example Problems ....................................................................35-2
Contents Chapter 35/Pedestrians and Bicycles: Supplemental

1. INTRODUCTION
Chapter 35 is the supplemental chapter for Chapter 24, Off-Street Pedestrian VOLUME 4: APPLICATIONS
GUIDE
and Bicycle Facilities, which is found in Volume 3 of the Highway Capacity 25. Freeway Facilities:
Manual. It provides two example problems demonstrating the calculation of Supplemental
pedestrian and bicycle level of service (LOS) for off-street paths. Segments: Supplemental
Supplemental
28. Freeway Merges and
Supplemental
Supplemental
Supplemental
32. STOP-Controlled
Intersections:
Supplemental
33. Roundabouts:
Supplemental
35. Pedestrians and
Bicycles: Supplemental
38: Network Analysis
Chapter 35/Pedestrians and Bicycles: Supplemental Introduction

2. EXAMPLE PROBLEMS

List of Example Problems Problem Description Application
1 Pedestrian LOS on shared-use and exclusive paths Operational analysis
2 Bicycle LOS on a shared-use path Planning analysis
EXAMPLE PROBLEM 1: PEDESTRIAN LOS ON SHARED-USE AND

EXCLUSIVE PATHS
The Facts
The parks and recreation department responsible for an off-street shared-use
path has received several complaints from pedestrians that the volume of
bicyclists using the path makes walking on the path an uncomfortable
experience. The department wishes to quantify path operations and, if necessary,
evaluate potential solutions.
The following information was collected in the field for this path:
• Qsb = bicycle volume in same direction = 100 bicycles/h;
• Qob = bicycle volume in opposing direction = 100 bicycles/h;
• v15 = peak 15-min pedestrian volume = 100 pedestrians;
• PHF = peak hour factor = 0.83;
• Sp = average pedestrian speed = 4.0 ft/s (2.7 mi/h);
• Sb = average bicycle speed = 16.0 ft/s (10.9 mi/h); and
• No pedestrian platooning was observed.

The shared-use path pedestrian LOS methodology requires pedestrian and
bicycle speeds and bicycle demand, all of which are available from the field
measurements just given.
Step 2: Calculate Number of Bicycle Passing and Meeting Events

The number of passing events Fp is determined from Equation 24-5:
𝑄𝑠𝑏 𝑆𝑝
𝐹𝑝 = (1 − )
𝑃𝐻𝐹 𝑆𝑏
100 bicycles/h 4.0 ft/s
𝐹𝑝 = (1 − )
0.83 16.0 ft/s
𝐹𝑝 = 90 events/h
The number of meeting events Fm is determined from Equation 24-6:
𝑄𝑜𝑏 𝑆𝑝
𝐹𝑚 = (1 + )
𝑃𝐻𝐹 𝑆𝑏
100 bicycles/h 4.0 ft/s
𝐹𝑚 = (1 + )
0.83 16.0 ft/s
Example Problems Chapter 35/Pedestrians and Bicycles: Supplemental

𝐹𝑚 = 151 events/h
The total number of events is calculated from Equation 24-7:
𝐹 = (𝐹𝑝 + 0.5𝐹𝑚 )
𝐹 = (90 + 0.5(151))
𝐹 = 166 events/h
Step 3: Determine Shared-Use Path Pedestrian LOS

The shared-use path LOS is determined from Exhibit 24-4. The value of F,
166 events/h, falls into the LOS E range. Because this LOS is rather low, what
would happen if a parallel, 5-ft-wide, pedestrian-only path were provided?
Step 4: Compare Exclusive-Path Pedestrian LOS

Step 4.1: Determine Effective Walkway Width
Assuming no obstacles exist on or immediately adjacent to the path, the
effective width would be the same as the actual width, or 5 ft. If common
amenities like trash cans and benches will be located along the path, they should
be placed at least 3 ft and 5 ft, respectively, off the path to avoid affecting the
effective width. These distances are based on data from Exhibit 24-9.
Step 4.2: Calculate Pedestrian Flow Rate

Because a peak 15-min pedestrian volume was measured in the field, it is not
necessary to use Equation 24-2 to determine v15. The unit flow rate for the
walkway vp is determined from Equation 24-3 as follows:
𝑣15
𝑣𝑝 =
15 × 𝑊𝐸
100
𝑣𝑝 =
15 × 5
𝑣𝑝 = 1.33 p/ft/min
Step 4.3: Calculate Average Pedestrian Space

Average pedestrian space is determined from Equation 24-4, including
applying a conversion from seconds to minutes:
𝑆𝑝
𝐴𝑝 =
𝑣𝑝
𝐴𝑝 = (4.0 ft/s)(60 s/min)/(1.33 p/ft/min)
𝐴𝑝 = 180 ft 2 /p
Step 4.4: Determine LOS

Because no pedestrian platooning was observed, Exhibit 24-1 should be used
to determine LOS. A value of 180 ft2/min corresponds to LOS A.
Chapter 35/Pedestrians and Bicycles: Supplemental Example Problems

Discussion
The existing shared-use path operates at LOS E for pedestrians. Pedestrian
LOS would increase to LOS A if a parallel, 5-ft-wide pedestrian path were
provided.
EXAMPLE PROBLEM 2: BICYCLE LOS ON A SHARED-USE PATH

The Facts
A new shared-use path is being planned. On the basis of data from a similar
facility in the region, planners estimate the path will have a peak hour volume of
340 users, a peak hour factor of 0.90, and a 50/50 directional split. The path will
be 10 ft wide, without obstacles or a centerline. The segment analyzed here is 3
mi long.

Facility and overall demand data are available but not the mode split of users
or the average mode group speed. Those values will need to be defaulted by
using Exhibit 24-6. On the basis of the default mode split and the estimated
directional split, the directional flow rate by mode is as follows:
• Directional bicycle flow rate = (340 users/h  0.5  0.55)/0.90 = 104
bicycles/h;
• Directional pedestrian flow rate = (340  0.5  0.20)/0.90 = 38 p/h;
• Directional runner flow rate = (340  0.5  0.10)/0.90 = 19 runners/h;
• Directional inline skater flow rate = (340  0.5  0.10)/0.90 = 19 skaters/h;
and
• Directional child bicyclist volume = (340  0.5  0.05)/0.90 = 9 child
bicyclists/h.
From Exhibit 24-6, average mode group speeds μ and standard deviations σ
are as follows:
• Bicycle: μ = 12.8 mi/h, σ = 3.4 mi/h;
• Pedestrian: μ = 3.4 mi/h, σ = 0.6 mi/h;
• Runner: μ = 6.5 mi/h, σ = 1.2 mi/h;
• Inline skater: μ = 10.1 mi/h, σ = 2.7 mi/h; and
• Child bicyclist: μ = 7.9 mi/h, σ = 1.9 mi/h.
Step 2: Calculate Active Passings per Minute

Active passings per minute must be calculated separately for each mode by
using Equation 24-9 through Equation 24-11. The path segment length L is 3 mi,
and the path is considered as broken into 300 pieces, each of which has a length
dx of 0.01 mi.
For a given modal user in the path when the average bicyclist enters, the
probability of being passed is expressed by Equation 24-9. The average
probability of passing within each piece j can be estimated as the average of the
probabilities at the start and end of each piece, as expressed by Equation 24-10.

The probability of passing a bicycle at the end of the first 0.01-mi piece of
path (i.e., at x = 0.01 mi) is derived from a normal distribution of bicycle speeds
with a mean speed μ and a standard deviation σ.
𝑥 0.01
𝐹(𝑥) = 𝑃 [𝑣𝑏𝑖𝑘𝑒 < 𝑈 (1 − )] = 𝑃 [𝑣𝑏𝑖𝑘𝑒 < 12.8 (1 − )]
𝐿 3
𝐹(𝑥) = 𝑃[𝑣𝑏𝑖𝑘𝑒 < 12.76] = 0.4950
The probability of passing a bicycle at the start of the first 0.01-mi piece of
path is
𝑥 − 𝑑𝑥 0.01 − 0.01
𝐹(𝑥 − 𝑑𝑥) = 𝑃 [𝑣𝑏𝑖𝑘𝑒 < 𝑈 (1 − )] = 𝑃 [𝑣𝑏𝑖𝑘𝑒 < 12.8 (1 − )]
𝐿 3
𝐹(𝑥 − 𝑑𝑥) = 𝑃[𝑣𝑏𝑖𝑘𝑒 < 12.8] = 0.5000
Next, the average probability of passing in the first piece is
𝑃(𝑣𝑏𝑖𝑘𝑒 ) = 0.5[𝐹(𝑥 − 𝑑𝑥) + 𝐹(𝑥)]
𝑃(𝑣𝑏𝑖𝑘𝑒 ) = 0.5[0.5000 + 0.4950] = 0.4975
The expected number of times the average bicyclist passes users of mode i
over the entire path segment is determined by multiplying P(vi) by the density of
users of mode i and summing over all pieces of the segment. The number of
active passings per minute is then obtained by dividing the result by the number
of minutes required for the bicyclist to traverse the path segment, as given by
Equation 24-11:
𝑛
𝑞𝑖 1
𝐴𝑖 = ∑ 𝑃(𝑣𝑖 ) × × 𝑑𝑥
𝜇𝑖 𝑡 𝑗
𝑗=1
For the first mode, adult bicyclists, for the first piece, the expected active
passings per minute is
104 1
𝐴𝑏𝑖𝑘𝑒,1 = 0.4975 × × (0.01) = 0.0029
12.8 14
Repeating this procedure for all pieces from n = 1 to n = 300 and summing the
results yields
Active bicycle passings per minute = 0.0029 + 𝐴𝑏𝑖𝑘𝑒,2 + ⋯ + 𝐴𝑏𝑖𝑘𝑒,𝑛 = 0.18
When the same methodology is applied for each mode, the following active
passings per minute are found for the other modes:
• Pedestrians, 1.74;
• Runners, 0.31;
• Inline skaters, 0.09; and
• Child bicyclists, 0.10.
Total active passings are then determined by using Equation 24-12:
𝐴 𝑇 = ∑ 𝐴𝑖
𝑖
Total passings per minute = 0.18 + 1.74 + 0.31 + 0.09 + 0.10 = 2.42

Step 3: Calculate Meetings per Minute

Meetings per minute of users already on the path segment M1 are calculated
for each mode i with Equation 24-13:
𝑈 𝑞𝑖
𝑀1 = ∑
60 𝜇𝑖
𝑖
𝑀1 = (12.8/60) × [(104/12.8) + (38/3.4) + (19/6.6) + (19/10.1) + (9/7.9)]
𝑀1 = 5.36
Meetings per minute of users in the opposing direction not yet on the path
segment at the time the average bicyclist enters must be calculated separately for
each mode. For the number of bicycles passed per minute, the section of path
beyond the study segment is considered as broken into n pieces, each of which
has length dx = 0.01 mi, and a total segment length equivalent to L (3 mi). For the
first piece ending at x = 0.01 mi, Equation 24-14 gives
𝑈 12.8
𝐹(𝑋) = 𝑃 (𝑣𝑏𝑖𝑘𝑒 > 𝑋 ) = 𝑃 (𝑣𝑏𝑖𝑘𝑒 > 0.01 × )
𝐿 3
𝐹(𝑋) = 𝑃(𝑣𝑏𝑖𝑘𝑒 > 0.4267) = 0.99992
𝑈 12.8
𝐹(𝑋 − 𝑑𝑥) = 𝑃 (𝑣𝑏𝑖𝑘𝑒 > (𝑋 − 𝑑𝑥) ) = 𝑃 (𝑣𝑏𝑖𝑘𝑒 > 0 × )
𝐿 3
𝐹(𝑋 − 𝑑𝑥) = 𝑃(𝑣𝑏𝑖𝑘𝑒 > 0) = 1.00000
Applying Equation 24-10 and Equation 24-15 then gives the probability of
passing in the first piece:
𝑃(𝑣𝑏𝑖𝑘𝑒 ) = 0.5[𝐹(𝑋 − 𝑑𝑥) + 𝐹(𝑥)]
𝑃(𝑣𝑏𝑖𝑘𝑒 ) = 0.5[0.99992 + 1.00000] = 0.99996
𝑛
𝑞𝑏𝑖𝑘𝑒 1
𝑀2,𝑏𝑖𝑘𝑒,j = ∑ 𝑃(𝑣𝑂,𝑏𝑖𝑘𝑒 ) × × 𝑑𝑥
𝜇𝑏𝑖𝑘𝑒 𝑡 𝑗
𝑗=1
M2,𝑏𝑖𝑘𝑒,1 = 0.99996 × (104/12.8) × (1/14) × 0.01 = 0.0058

Repeating this procedure for all pieces from n = 1 to n = 300 and summing the
results yields
𝑀2,𝑏𝑖𝑘𝑒 = meetings of bicycles per minute = 0.0058 + 𝑀2,𝑏𝑖𝑘𝑒,2 + ⋯ + 𝑀2,𝑏𝑖𝑘𝑒,𝑛
𝑀2,𝑏𝑖𝑘𝑒 = 1.55
When the foregoing procedure is repeated for the other modes, the following
meetings per minute are found for each mode:
• Pedestrians, 0.63;
• Runners, 0.32;
• Inline skaters, 0.31; and
• Child bicyclists, 0.16.

Total meetings are then determined by using Equation 24-16:
𝑀𝑇 = (𝑀1 + ∑ 𝑀2,𝑖 )
𝑖
Total meetings per minute = 5.36 + 1.55 + 0.63 + 0.32 + 0.31 + 0.16 = 8.33
Step 4: Determine the Number of Effective Lanes

From Exhibit 24-14, a 10-ft-wide path has two effective lanes.
Step 5: Calculate the Probability of Delayed Passing

From Step 4, it is clear that a path with a width of 10 ft will operate as two
lanes. Therefore, delayed passings per minute must be calculated separately for
each of the 25 modal pairs by using Equation 24-17 and Equation 24-20. For
instance, considering the probability of a delayed passing of a bicyclist as a result
of an opposing bicyclist overtaking a pedestrian gives the following:
𝑃𝑛,𝑖 = 1 − 𝑒 −𝑝𝑖 𝑘𝑖
100 104
−( )×( )
𝑃𝑛,𝑏𝑖𝑘𝑒 = 1 − 𝑒 5,280 12.8 = 1 − 0.8574 = 0.1426
100 38
−( )×( )
𝑃𝑛,𝑝𝑒𝑑 = 1 − 𝑒 5,280 3.4 = 1 − 0.8092 = 0.1908
Substituting into Equation 24-20 yields Pbike-ped,ds:
2
𝑃𝑛,𝑝𝑒𝑑 𝑃𝑛,𝑏𝑖𝑘𝑒 + 𝑃𝑛,𝑝𝑒𝑑 (1 − 𝑃𝑛,𝑏𝑖𝑘𝑒 )
𝑃𝑏𝑖𝑘𝑒−𝑝𝑒𝑑,𝑑𝑠 =
1 − 𝑃𝑛,𝑝𝑒𝑑 𝑃𝑛,𝑏𝑖𝑘𝑒 (1 − 𝑃𝑛,𝑝𝑒𝑑 )(1 − 𝑃𝑛,𝑏𝑖𝑘𝑒 )
0.1908 × 0.1426 + 0.1908(1 − 0.1426)2
𝑃𝑏𝑖𝑘𝑒−𝑝𝑒𝑑,𝑑𝑠 = = 0.1707
1 − (0.1908 × 0.1426)(1 − 0.1908)(1 − 0.1426)
Step 6: Determine Delayed Passings per Minute

Step 5 is performed for each of the 25 modal pairs. Equation 24-33 is used to
determine the total probability of delayed passing:
𝑃𝑇𝑑𝑠 = 1 − ∏(1 − 𝑃𝑚,𝑑𝑠 )

𝑚
𝑃𝑇𝑑𝑠 = 1 − (1 − 0.1707) × (1 − 𝑃𝑏𝑖𝑘𝑒−𝑟𝑢𝑛𝑛𝑒𝑟,𝑑𝑠 ) × ⋯ × (1 − 𝑃𝑚,𝑑𝑠 ) = 0.8334
Thus, the probability of delayed passing is 83.34%.
Equation 24-34 is used to determine the total number of delayed passings per
minute:
𝐷𝑃𝑚 = 𝐴 𝑇 × 𝑃𝑇𝑑𝑠 × 𝑃𝐻𝐹
𝐷𝑃𝑚 = 2.42 × 0.8334 × 0.90 = 1.82

Step 7: Calculate LOS

Equation 24-35 is used to determine the bicycle LOS (BLOS) score for the
path:
𝐵𝐿𝑂𝑆 = 5.446 − 0.00809𝐸 − 15.86𝑅𝑊 − 0.287𝐶𝐿 − 𝐷𝑃
1
𝐵𝐿𝑂𝑆 = 5.446 − 0.00809[8.33 + (10 × 2.42)] − 15.86 ( ) − 0.287(0)
10
− (min [𝐷𝑃𝑚 × 0.5, 1.5]) = 2.69
Because the bicyclist perception index is between 2.5 and 3.0, the path
operates at LOS D according to Exhibit 24-5.
Results
The results indicate that the path would operate close to its functional
capacity. A slightly wider path would provide three effective lanes and a better
LOS.

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 36
CONCEPTS: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 36-1
2. PRESENTATION OF RESULTS ......................................................................... 36-2

Guidance on the Display of HCM Results ...................................................... 36-2
Presenting Results to Facilitate Interpretation ................................................ 36-3
Graphic Representation of Results ................................................................... 36-4
3. MEASURING TRAVEL TIME RELIABILITY IN THE FIELD...................... 36-7

Measurement of Travel Time Reliability ......................................................... 36-7
Data Sources for Travel Time Reliability ......................................................... 36-7
Recommended Method for Computing Reliability by Using
Roadway-Based Spot Measurement Detectors ......................................36-11
Recommended Method for Computing Reliability by Using
Probe Vehicles .............................................................................................36-13
4. RELIABILITY VALUES FOR SELECTED U.S. FACILITIES ....................... 36-15

Data Sources .......................................................................................................36-15
Reliability Statistics for a Cross Section of U.S. Facilities .............................36-15
Reliability Statistics for Florida Freeways ......................................................36-20
5. VEHICLE TRAJECTORY ANALYSIS.............................................................. 36-22

Introduction ........................................................................................................36-22
Trajectory Analysis Examples ..........................................................................36-24
Estimating Performance Measures from Vehicle Trajectory Data ..............36-37
6. REFERENCES ....................................................................................................... 36-52
Chapter 36/Concepts: Supplemental Contents

LIST OF EXHIBITS
Exhibit 36-1 Example of a Graphic Display of LOS ...............................................36-4

Exhibit 36-2 Example of a Thematic Graphic Display of LOS ..............................36-5
Exhibit 36-3 Example Presentation of Planning Analysis Results ........................36-5
Exhibit 36-4 Three-Dimensional Reliability Box .....................................................36-7
Exhibit 36-5 Spot Speed (Vertical) Sampling of Loop Detectors ..........................36-9
Exhibit 36-6 Time–Space (Diagonal) Sampling of Probe Vehicle Detectors .....36-10
Exhibit 36-7 Comparison of Loop Detector and Probe Cumulative Travel
Time Distributions .............................................................................................36-10
Exhibit 36-8 Rankings of U.S. Facilities by Mean TTI and PTI (A.M. Peak,
Midday, and P.M. Peak Combined) ................................................................36-16
Exhibit 36-9 Rankings of U.S. Facilities by Mean TTI and PTI (A.M. Peak) .....36-16
Exhibit 36-10 Rankings of U.S. Facilities by Mean TTI and PTI (Midday) ........36-17
Exhibit 36-11 Rankings of U.S. Facilities by Mean TTI and PTI (P.M.
Peak) ....................................................................................................................36-17
Exhibit 36-12 Freeway Reliability Values: Weekday A.M. Peak Period ............36-18
Exhibit 36-13 Freeway Reliability Values: Weekday Midday Periods ..............36-18
Exhibit 36-14 Freeway Reliability Values: Weekday P.M. Peak Period ............36-19
Exhibit 36-15 Urban Street Reliability Values: Weekday A.M. Peak Period .....36-19
Exhibit 36-16 Urban Street Reliability Values: Weekday Midday Periods .......36-20
Exhibit 36-17 Urban Street Reliability Values: Weekday P.M. Peak Period .....36-20
Exhibit 36-18 Florida Freeway Reliability Statistics .............................................36-21
Exhibit 36-19 Vehicle Data Stored for Each Time Step ........................................36-23
Exhibit 36-20 Basic Signalized Intersection Example ...........................................36-25
Exhibit 36-21 Trajectory Plots for Uniform Arrivals and Departures ................36-25
Exhibit 36-22 Introducing Randomness into the Simulation ..............................36-26
Exhibit 36-23 Cycle Failure Example ......................................................................36-27
Exhibit 36-24 Oversaturated Signal Approach .....................................................36-28
Exhibit 36-25 Queue Backup from a Downstream Signal ...................................36-29
Exhibit 36-26 Trajectory Plot for More Complex Signal Phasing .......................36-30
Exhibit 36-27 Weaving Segment Description and Animated
Graphics View ....................................................................................................36-31
Exhibit 36-28 Trajectory Plot for Freeway Links ...................................................36-32
Exhibit 36-29 Trajectory Plot for Entrance and Exit Ramp Links .......................36-33
Exhibit 36-30 Entrance Ramp Merging Segment Graphics View .......................36-34
Exhibit 36-31 Trajectory Plot for All Freeway Lanes in the Merge Area ...........36-34
Contents Chapter 36/Concepts: Supplemental

Exhibit 36-32 Trajectory Plot for Freeway Lane 1 (Rightmost) in the

Merge Area ......................................................................................................... 36-35
Exhibit 36-33 Trajectory Plot for Freeway Lane 2 (Center) in the
Merge Area ......................................................................................................... 36-35
Exhibit 36-34 Trajectory Plot for Freeway Lane 3 (Leftmost) in the
Merge Area ......................................................................................................... 36-35
Exhibit 36-35 Trajectory Plot for Acceleration and Deceleration Lanes ............ 36-36
Exhibit 36-36 Addition of Intermediate Nodes for Continuous
Trajectory Plots .................................................................................................. 36-37
Exhibit 36-37 Trajectory Plot for Acceleration Lane and Freeway Lane 1 ........ 36-37
Exhibit 36-38 Trajectories for Several Cycles on a Signalized Approach .......... 36-45
Exhibit 36-39 Example Trajectory Analysis Plots ................................................. 36-45
Exhibit 36-40 Analysis of a Full and a Partial Stop .............................................. 36-46
Exhibit 36-41 BOQ Analysis by Time Step ............................................................ 36-47
Exhibit 36-42 BOQ Histogram ................................................................................. 36-48
Exhibit 36-43 Accumulated Delay by Various Definitions ................................. 36-49
Exhibit 36-44 Delay Analysis for All Vehicles on a Segment .............................. 36-50
Exhibit 36-45 Longitudinal Analysis of Delay for a Selected Vehicle in a
Weaving Area..................................................................................................... 36-50
Exhibit 36-46 Example Spatial Analysis by Lane ................................................. 36-51
Chapter 36/Concepts: Supplemental Contents

1. INTRODUCTION
Chapter 36 is the supplemental chapter for Volume 1, Concepts, of the VOLUME 4: APPLICATIONS
GUIDE
Highway Capacity Manual (HCM). 25. Freeway Facilities:
Supplemental
Section 2 supplements material in Chapter 7, Interpreting HCM and 26. Freeway and Highway
Alternative Tool Results. It provides information on the recommended number Segments: Supplemental
of significant digits to use in presenting results and guidance on presenting Supplemental
analysis results to decision makers, the public, and practitioners. 28. Freeway Merges and
Sections 3 and 4 supplement material in Chapter 4, Traffic Operations and 29. Urban Street Facilities:
Supplemental
Capacity Concepts. Section 3 provides guidance on measuring travel time 30. Urban Street Segments:
reliability in the field, and Section 4 presents travel time reliability values for Supplemental
selected freeway and arterial facilities as an aid to analysts in interpreting travel Supplemental
time reliability performance measures. 32. STOP-Controlled
Intersections:
Section 5 supplements Chapters 4 and 7. It provides expanded guidance on Supplemental
33. Roundabouts:
the use of vehicle trajectory analysis as a means by which performance measures Supplemental
can be consistently estimated by various alternative analysis tools. 34. Interchange Ramp
Supplemental
36. Concepts:
Supplemental
Chapter 36/Concepts: Supplemental Introduction

2. PRESENTATION OF RESULTS
GUIDANCE ON THE DISPLAY OF HCM RESULTS

Tabular values and calculated results are displayed in a consistent manner
throughout the HCM. Analyst adherence to these conventions is suggested. A
key objective is to use the number of significant digits that is reasonable, to
indicate to users, decision makers, and other viewers that the results are not
extremely precise but take on the precision and accuracy associated with the
input variables used. This guidance applies primarily to inputs and final outputs;
intermediate results in a series of calculations should not be rounded unless
specifically indicated by a particular methodology.
Input Values
Following is a list of representative (not exhaustive) input variables and the
suggested number of digits for each.
• Volume (whole number);
• Grade (whole number);
• Lane width (one decimal place);
• Percentage of heavy vehicles (whole number);
• Peak hour factor (two decimal places);
• Pedestrian volume (whole number);
• Bicycle volume (whole number);
• Parking maneuvers (whole number);
• Bus stopping (whole number);
• Green, yellow, all-red, and cycle times (one decimal place);
• Lost time/phase (whole number); and
• Minimum pedestrian time (one decimal place).
Adjustment Factors
Factors interpolated from tabular material can use one more decimal place
than is presented in the table. Factors generated from equations can be taken to
three decimal places.
Service Volume Tables

When volumes for service volume tables are rounded, the precision used
should be no greater than the nearest 10 vehicles or passenger cars for hourly
tables and no greater than the nearest 100 vehicles or passenger cars for daily
tables.
Free-Flow Speed
For a base free-flow speed (FFS), show the value to the nearest 1 mi/h. If the
FFS has been adjusted for various conditions and is considered an intermediate
calculation, show speed to the nearest 0.1 mi/h.
Presentation of Results Chapter 36/Concepts: Supplemental

Speeds
For threshold values that define level of service (LOS), show speed to the
nearest 1 mi/h. For intermediate calculations of speed, use one decimal place.
Volume-to-Capacity and Demand-to-Capacity Ratios

Show volume-to-capacity and demand-to-capacity ratios with two decimal
places.
Delay
In computing delay, show results with one decimal place. In presenting
delay as a threshold value in LOS tables, show a whole number.
Density
Show density results with one decimal place.
Pedestrian Space
Show pedestrian space values with one decimal place.
Occurrences and Events

For all event-based items, use values to a whole number. These items include
parking maneuvers, buses stopping, and passing and meeting events along a
pedestrian or bicycle path.
General Factors
In performing all calculations on a computer, the full precision available
should be used. Intermediate calculation outputs should be displayed to three
significant digits throughout. For the measure that defines LOS, the number of
significant digits presented should exceed by one the number of significant digits
shown in the LOS table.
PRESENTING RESULTS TO FACILITATE INTERPRETATION

Several performance measures can result from HCM analyses. Performance measures
selected should be related to
Determination of the appropriate measures will depend on the transportation the problem being addressed.
need being studied. However, decision-making situations generally can be
divided into those involving the public (e.g., city councils and community
groups) and those involving technicians (e.g., state and local engineering and
planning staff).
The HCM is highly technical and complex. The results of the analyses can be
difficult for people to interpret for decision making unless the data are carefully
organized and presented. In general, the results should be presented as simply as
possible. The presentation might use a small set of performance measures and
provide the data in an aggregate form without losing the ability to relate to the
underlying variations and factors that generated the results.
The LOS concept was created, in part, to make presentation of results easier
than if numerical values of service measures were reported directly. In many
cases, analysts and decision makers prefer to see one service measure rather than
multiple performance measures. At the same time, relying solely on LOS results
Chapter 36/Concepts: Supplemental Presentation of Results

in making recommendations or decisions can lead to important information

available from other performance measures being overlooked. Despite the
limitations to its usefulness, the LOS concept remains a part of the HCM because
of its acceptance by the public and decision makers.
Decision makers who represent the public usually prefer measures that their
constituents can understand. The public can relate to LOS results, which describe
relative differences in highway operations. Unit delay (e.g., seconds per vehicle)
and travel speed are also readily understood. However, volume-to-capacity ratio,
density, percent time-spent-following, and vehicle hours of travel are not
measures to which the public easily relates. In the selection of measures to
present, recognition by the analyst of the orientation of the decision maker and
the context in which the decision will be made is important. In general, these
measures can be differentiated as system user or system manager oriented.
GRAPHIC REPRESENTATION OF RESULTS

Historically, data and analysis results have been presented primarily in
tables. However, results may be best presented as pictures and supplemented
only as necessary with the underlying numbers in some situations. Graphs and
charts should be conceived and fashioned to aid in interpretation of the meaning
behind the numbers (1).
Present results to make them Most performance measures in the HCM are quantitative, continuous
very plain (obvious) to the
audience. variables. However, LOS values result from step functions and do not lend
themselves to graphing. When they are placed on a scale, LOS results must be
given an equivalent numeric value, as shown in Exhibit 36-1, which presents the
LOS for a group of intersections. The LOS letter is indicated, and shaded (or
colored) areas indicate intersections that are below, at, or above the analysis
objective of LOS D. The size of the indicator at each intersection shows the
relative control delay value for the indicated LOS.
Exhibit 36-1
2nd Ave
3rd Ave
1st Ave
Example of a Graphic Display

of LOS
B C C
Alder St
C E E
Beech St
D F E
Chestnut St
The issue is whether the change in value between successive LOS values (i.e.,
the interval) should be equal. For example, is conversion of LOS A to F to a scale
of 0 through 5 appropriate? Should the numerical equivalent assigned to the

difference of the thresholds between LOS A and B be the same as the difference
between LOS E and F? These questions have not been addressed in research,
except in the area of traveler perception models. Furthermore, LOS F is not given
an upper bound. Therefore, a graph of LOS should be considered ordinal, not
interval, because the numeric differences between the levels would not appear
significant.
However, it is difficult to refrain from comparing the differences. A scale
representing the relative values of the LOS letters would have to incorporate the
judgment of the analyst and the opinions of the public or decision makers—a
difficult task. A thematic graphic presentation avoids this issue. In Exhibit 36-2,
for example, shading is used to highlight analysis periods and basic freeway
segments that do not meet the objective LOS (in this case, D).
Start Time Segment 1 Segment 2 Segment 3 Segment 4 Exhibit 36-2

5:00 p.m. A B B A Example of a Thematic
5:15 p.m. B B D A Graphic Display of LOS
5:30 p.m. B B F A
5:45 p.m. B D F A
6:00 p.m. B F F A
6:15 p.m. D F E A
6:30 p.m. D E C A
6:45 p.m. B B B A
Further simplification of the presentation can be achieved by converting LOS

letters into general descriptors of conditions. For example, Exhibit 36-3 shows a
map of a portion of a downtown area, where street segments have been labeled
by the analyst as “not congested” (e.g., LOS A, B, or C), “becoming congested”
(e.g., LOS D or E), or “congested” (e.g., LOS F). (Note that these represent the
analyst’s choice of how to interpret and present the results; the HCM does not
define specific levels of congestion.) This type of presentation is particularly
useful for planning applications where many inputs into the HCM method have
been defaulted and therefore the results may be less precise.
Exhibit 36-3
Example Presentation of
Planning Analysis Results
Source: City of Milwaukee.
Chapter 36/Concepts: Supplemental Presentation of Results

The HCM provides valuable assistance in making transportation

management decisions in a wide range of situations. It offers the user a selection
of performance measures to meet a variety of needs. The analyst should
recognize that using the HCM involves mixing art with science. Sound judgment
is needed not only for interpreting the values produced but also for summarizing
and presenting the results.

3. MEASURING TRAVEL TIME RELIABILITY IN THE FIELD
This section provides a recommended method for measuring travel time

reliability in the field. The intent is to provide a standardized method for
gathering and reporting travel time reliability for freeways and arterials directly
from field sensors, which can be used for validating estimates of reliability
produced by the HCM method and for consistently comparing reliability across
facilities.
MEASUREMENT OF TRAVEL TIME RELIABILITY

Measuring travel time reliability in the field involves the development of the
three-dimensional reliability box. The three dimensions of reliability are the
study section of the facility, the daily study period, and the reliability reporting
period (Exhibit 36-4). For example, travel time reliability can be computed for a
1-mi length of freeway during the afternoon peak hour for all nonholiday
weekdays in a year.
Exhibit 36-4
Three-Dimensional Reliability
Box
DATA SOURCES FOR TRAVEL TIME RELIABILITY

Travel time reliability (and travel times generally) may be measured by
recording a sample of the vehicle travel times over a fixed length of facility
(probe vehicle method) or by recording the spot speeds of all vehicles as they
pass over a set of stationary detectors. The latter method will be called for
convenience the “spot measurement detector method”; many technologies are
available (loops, radar, video, etc.) for measuring spot speeds.
Chapter 36/Concepts: Supplemental Measuring Travel Time Reliability in the Field

Measuring reliability is all about measuring variability, so the larger the

sample (in terms of number of vehicles and hours of the year), the more
confidence one can have in the result.
Travel time, like demand, exhibits strong daily and weekly cyclic patterns.
There may also be strong seasonal patterns to both demand and travel time. To
obtain a useful estimate of the travel time distribution for any given hour of the
day or day of the week, a sufficient sample of that hour and that day (and that
season, if seasonality is significant) must be obtained to estimate the mean and
the standard deviation of the travel time for that hour (and day of the week)
within an acceptable range of accuracy. A reference provides details and
examples of computing the required sample size to estimate the mean of the
travel time distribution for the hour (3).
Estimating the standard deviation of the travel time distribution generally
requires a much larger sample than estimating the mean to the same precision.
To estimate the standard deviation of a normal distribution to within 10% of its
true value at the 95% confidence level will require on the order of 200 samples of
travel time for the hour (close to a year’s worth of nonholiday, weekday data).
Only 50 samples are needed to estimate the standard deviation to within 20% of
its true value at a 95% confidence level (4).
Note that travel time is not normally distributed, so the minimum sample
sizes described here should be considered as providing lower confidence levels
than the 95% confidence level cited from the literature for the normal
distribution.
Roadway-Based Spot Measurement Detectors

Spot measurement detectors can be as close as ⅓ to ½ mi apart, but they can
be much farther apart. However, as detector spacing increases, the assumption
that speeds are constant over the entire distance becomes more problematic.
While an upper limit on spacing has not been established by research, detector
spacing of ½ mi or less is greatly preferred.
Single detectors will measure the time a vehicle spends within the detector’s
detection zone and will divide this time by the estimated average vehicle length
(supplied by the operator) to arrive at the estimated speed of the vehicle.
Pairs of detectors will measure the lag between the time the leading edge of
the vehicle arrives at the first detector and the time the leading edge arrives at
the second detector. The distance between the two detectors is divided by the
time difference between the arrival of the leading edge of the vehicle at the
upstream detector and its arrival at the downstream detector to obtain the
vehicle speed for the short distance between the two detectors.
Probe Vehicles
Electronic toll tag or Bluetooth readers can be deployed at certain segments
of freeway so that time stamps of vehicles crossing at these locations can be
tracked. When a vehicle with a toll tag or a discoverable Bluetooth device crosses
locations with readers, identification of the same vehicle can be matched with
Measuring Travel Time Reliability in the Field Chapter 36/Concepts: Supplemental

different time stamps and corresponding locations. Then the travel time between
a pair of toll tag reader locations can be obtained.
In addition, “crowd-sourced” data may be available. To obtain such data, the
movements of vehicles and people carrying various GPS-equipped
telecommunication devices are monitored anonymously. The observed point
speed data or the point-to-point travel times are filtered, converted into average
travel times, and archived for later retrieval. The Federal Highway
Administration’s (FHWA’s) National Performance Management Research Data
Set is one example of a crowd-sourced database of travel times (5).
For point-to-point measurements of travel time, the analyst will need to
develop and apply a filtering algorithm that removes vehicles from the sample
that take an excessive amount of time to appear at the downstream detector
because they have left the facility to stop for errands between the two detectors.
The closer together the two readers, the tighter the filtering criterion can be.
Comparison of Sampling Methods

Spot detectors (e.g., loops) take a vertical sample of the facility time–space
diagram, while probe vehicle (e.g., electronic toll collection) detectors take a
diagonal sample of the facility time–space diagram (compare Exhibit 36-5 and
Exhibit 36-6).
At the time of writing, the probe data available from vendors resemble
detector data more closely than true probe data. The data may have started out as
recorded positions of selected vehicles traveling on a facility, but the processed
data that analysts receive are speeds on a link. Consequently, vendor-supplied
data at present do not look at all like the Bluetooth or toll tag data collected by
agencies.
Exhibit 36-5
Spot Speed (Vertical)
Sampling of Loop Detectors

Exhibit 36-6
Time–Space (Diagonal)
Sampling of Probe Vehicle
Detectors
Since the two measurement methods sample the three-dimensional

reliability space differently, they will produce slightly different estimates of the
travel time reliability distribution, as illustrated for one freeway in Exhibit 36-7.
However, the differences between the methods will generally be less than the
differences in reliability between different peak periods.
Exhibit 36-7
Comparison of Loop Detector
and Probe Cumulative Travel
Time Distributions
Source: Kittelson & Associates, Inc.

Note: I-80 westbound, Contra Costa County, California.
Each method has its strengths and weaknesses, and neither method is always
the best. A dense network of loop detectors may produce better estimates than a
sparse network of toll tag readers. The reverse may also be true. Thus the choice
of method is contingent on the density of the detection available for each method.
Similarly, crowd-sourced data may be superior or inferior to field detector–
based measuring methods, depending on the sample size and the gaps in the
crowd-sourced data and the density and reliability of the field detectors.

RECOMMENDED METHOD FOR COMPUTING RELIABILITY BY USING

ROADWAY-BASED SPOT MEASUREMENT DETECTORS
The recommended method for computing travel time reliability statistics for
freeways by using stationary sensors of spot speeds and volumes is described
below. Because of the highly varying nature of speeds by distance from signal on
urban streets, this method is not recommended for urban streets.
1. Define reliability study bounds. Select facility direction, length, study period,
and reliability reporting period. The analyst should select the reliability
reporting period appropriate for the purposes of the analysis. This may be
all the nonholiday weekdays of a year (approximately 250 days out of the
year) if the analyst is evaluating the reliability of a facility that has regular
recurring weekday congestion. It may be the summer or winter weekends
of a year if the analyst is evaluating a facility with regular recreational
travel congestion.
2. Download data. Download lane-by-lane vehicle speeds and volumes
aggregated or averaged to 5-min periods for all mainline speed detectors
for the selected study direction, within the selected facility length and
study period, and for all days included in the reliability reporting period.
3. Quality check data.
a. If the system fills gaps in detector data (e.g., detectors down) with
estimates, remove data with less than 70% observed rating.
b. Remove unrealistic speeds from the data set. Analysts will need to
review the data and use local knowledge to determine what is
unreasonable. In addition, FHWA provides guidance on quality
control for detector data (6).
c. Gaps in data are treated as nonobservations.
4. Compute 5-min vehicle miles traveled (VMT).
a. For each detector station, identify the length of facility represented by
the detector. This is usually half the distance to the upstream detector
station plus half the distance to the downstream detector, but it can be
a different value based on local knowledge of the facility.
b. Sum volumes across all lanes at the detector station for 5-min time
periods.
c. Neglect periods when the detector is not functioning.
d. VMT(t, d) = V(t, d) × L(d), where VMT(t, d) = vehicle miles traveled
during time period t measured at detector station d; L(d) = length
represented by detector station d (mi), and V(t, d) = sum of lane
volumes (veh) measured at detector station d during time period t.
5. Compute 5-min vehicle hours traveled (VHT).
a. VHT(t, d) = VMT(t, d) / S(t, d), where VHT(t, d) = vehicle hours traveled
during time period t measured at lane detector station d and S(t, d) =
arithmetic average speed of vehicles (mi/h) measured during time
period t at lane detector station d.
b. Neglect periods when the detector is not functioning.

6. Compute the FFS for the facility. For a facility analysis, the use of data from
continuously operating devices (roadway detectors or probe vehicles) is
the preferred method, as described below. However, the analyst should
be satisfied with the quality of the data from the suggested time periods
before proceeding. For performance monitoring of multiple facilities or
complete roadway systems, the analyst may wish to establish FFS in other
ways, mainly to establish a consistent base from which to track trends.
For example, if monitoring is performed on an annual basis, calculation of
FFS every year on a facility may lead to different values for each year.
One way to address this problem is to use the empirical method given
below in the first year of the monitoring program to set the FFS for all
years. Other methods include picking a constant FFS on the basis of
agency policy for that facility type or speed limit. The “agency policy”
FFS reflects in some way the agency’s performance objectives for the
facility. Whatever method is used, the analyst should clearly specify it.
a. Select a nonholiday weekend (or other period known to the analyst to
be a light-flow period without congestion).
b. For each detector, obtain 5-min speeds for 7 to 9 a.m. on a typical
weekend morning (or other uncongested, light-flow period).
c. Neglect periods when the detector is not functioning.
d. Quality control for excessively high speeds or excessively low
volumes as discussed earlier.
e. Identify the average (mean) speed during the observed light-flow
period. That is the FFS for the detector.
f. Convert speed to segment travel times.
g. Sum segment times to obtain facility free-flow travel times.
7. Compute the VMT and VHT for each time period.
𝑉𝑀𝑇𝑡 = ∑ 𝑉𝑀𝑇𝑡,𝑑
𝑑
𝑉𝐻𝑇𝑡 = ∑ 𝑉𝐻𝑇𝑡,𝑑
𝑑
8. Compute the travel time index (TTI) for the facility for each time period.
𝑉𝐻𝑇𝑡
𝑇𝑇𝐼𝑡 =
𝑉𝐻𝑇𝐹𝐹𝑡
where VHTFFt is the VHT that would occur during time period t if all
vehicles traveled at the FFS:
𝑉𝑀𝑇𝑡
𝑉𝐻𝑇𝐹𝐹𝑡 =
𝐹𝐹𝑆
9. Develop a distribution of the TTIt values for the facility for the entire analysis
period. Each TTIt value becomes an observation in the distribution. All
performance measures are derived from this distribution. The statistics
and percentiles are calculated by using VMTt as a weight; this is done to
account for the fact that the TTIs in each time period are based on a
different number of vehicles.

RECOMMENDED METHOD FOR COMPUTING RELIABILITY BY USING

PROBE VEHICLES
The recommended method for computing travel time reliability statistics for TMC segments are industry-
standard roadway sections
freeways and arterials by using probe vehicles and Bluetooth, toll tag, or license used in communicating traffic
plate readers is described below. The instructions assume that the data are information to drivers (for
example, via a vehicle’s
obtained from a commercial vendor of historical traffic message channel (TMC) navigation system).
segment speed data.
1. Define reliability study bounds. Select the facility direction, length, study
period, and reliability reporting period. The analyst should select the
reliability reporting period appropriate for the purposes of the analysis.
This may be all the nonholiday weekdays of a year (approximately 250
days out of the year) if the analyst is evaluating the reliability of a facility
that has regular recurring weekday congestion. It may be the summer or
winter weekends of a year if the analyst is evaluating a facility with
regular recreational travel congestion.
2. Download data. Download TMC segment speeds (or travel times if
Bluetooth or toll tag reader data are being used) aggregated or averaged
to 5-min (or similar) periods for all mainline segments for the selected
study direction and selected facility length, for all study periods and days
included in the reliability reporting period.
3. Quality check data.
a. If travel time data (e.g., Bluetooth or toll tag reader data) are being
used, convert data to speeds for error-checking purposes.
b. Remove unrealistic speeds from the data set. Analysts will need to
review the data and use local knowledge to determine what is
unreasonable.
4. Compute facility travel times for each analysis period.
a. For each TMC (or Bluetooth or toll tag reader) segment, identify its
length in miles (to the nearest 0.01 mi).
b. Divide the segment length by speed to obtain the segment travel time
for each analysis period (skip this step if Bluetooth or toll tag travel
time data are being used).
c. Sum the segment travel times to obtain the facility travel time for each
time period.
5. Compute FFS for the facility. Steps 5a to 5g below are only applicable to
freeway facilities, as urban street segment reference speeds or probe
vehicle speeds under low-volume conditions may include traffic signal
delays not included in the HCM definition of FFS. For urban street
facilities, FFS can be established by use of an alternate method, including
(a) picking a constant FFS on the basis of agency policy for a given facility
type or speed limit; (b) establishing FFS on the basis of the actual speed
limit (e.g., speed limit plus a constant); and (c) measuring speeds at
locations not influenced by traffic control or junctions (e.g., midsegment
on urban streets).

Vendor-supplied urban street a. If the segment reference speed provided by the commercial vendor is
reference speeds may include
traffic signal delays not reliable, that can be used for the FFS. If it is not reliable, perform the
included in the HCM definition following steps.
of FFS.
b. Select a nonholiday weekend (or other period known to the analyst to
be a light-flow period without congestion).
c. For each segment, obtain speeds for 5-min time periods for 7 to 9 a.m.
on a typical weekend morning (or other uncongested, light-flow
period).
d. Quality control for excessively high speeds or travel times as
explained earlier.
e. Identify the average (mean) speed. That is the FFS for the segment.
f. Convert the segment speed to segment travel times (segment length
divided by segment speed).
g. Sum the segment times to obtain facility free-flow travel times.
6. Compute the VMT and VHT for each time period.
𝑉𝑀𝑇𝑡 = ∑ 𝑉𝑀𝑇𝑡,𝑑
𝑑
𝑉𝐻𝑇𝑡 = ∑ 𝑉𝐻𝑇𝑡,𝑑
𝑑
7. Compute the TTI for the facility for each time period.
𝑉𝐻𝑇𝑡
𝑇𝑇𝐼𝑡 =
𝑉𝐻𝑇𝐹𝐹𝑡
where VHTFFt is the VHT that would occur during time period t if all
vehicles traveled at the FFS:
𝑉𝑀𝑇𝑡
𝑉𝐻𝑇𝐹𝐹𝑡 =
𝐹𝐹𝑆
8. Develop a distribution of the TTIt values for the facility for the entire analysis
period. Each TTIt value becomes an observation in the distribution. All
performance measures are derived from this distribution. The statistics
and percentiles are calculated by using VMTt as a weight; this is done to
account for the fact that the TTIs in each time period are based on a
different number of vehicles.

4. RELIABILITY VALUES FOR SELECTED U.S. FACILITIES
DATA SOURCES
Reliability data for 1 year of nonholiday weekday travel time were obtained
from the following sources:
• 2-min traffic speed data in the I-95 corridor for 2010 (7), and
• 5-min traffic speed data in California for 2010 (8).
The base travel time for
The first data set includes freeway and urban street reliability data for states freeways was an empirically
and metropolitan areas in the I-95 corridor (i.e., U.S. East Coast). The average measured free-flow travel time.
For urban streets, the base
speed of traffic was measured every 2 min for each TMC road segment (9). Road travel time corresponded to
segments vary but generally terminate at a decision point for the driver (e.g., the 85th percentile highest
speed observed during off-
intersection, start of left-turn pocket, ramp merge or diverge). Traffic speeds are peak hours. Therefore, the
obtained by monitoring the positions of GPS units in participating vehicles. A free-flow reference speeds
used in these data sets do not
“free-flow reference speed” is established for each TMC segment on the basis of correspond exactly to the FFS
empirical observations. It may not correspond exactly to the FFS that would be that an HCM method would
produce.
estimated by the HCM’s analytical or field-measurement methods.
The California data include freeway reliability data for the state’s major
metropolitan areas, plus reliability data for one urban street in Chula Vista. The
data come from two sources: toll tag readers and loop detectors. California’s
system provides a function for stringing together a series of loop detector station
speeds into an estimate of the overall average speed for the facility. The loop
detector data used to compute an average speed for each segment of the facility
are offset by the time taken by the average vehicle to traverse the upstream
segment. Thus for a selected direction of travel, the average speed of vehicles in
Segment 1 is used to compute the average travel time t for the selected time
period (e.g., 5 min) for that segment starting at time T = 0. The mean speed is
computed for the next downstream segment for the 5-min period starting at T = 0
+ t. The resulting mean travel times are then added together to get the average
travel time of vehicles for the 5-min period starting their trip at 0 < T < 5 min.
RELIABILITY STATISTICS FOR A CROSS SECTION OF U.S. FACILITIES

Exhibit 36-8 through Exhibit 36-11 show the distribution of 50th percentile
travel time index (TTI50), mean travel time index (TTImean), and planning time
index (PTI or TTI95) observed in the data set of U.S. freeways and urban streets
described above, for all analysis periods combined, the 2-h a.m. peak period, the
2-h midday period, and the 2-h p.m. peak period, respectively. Exhibit 36-11 is an
expanded version of Exhibit 11-3 in Chapter 11, Freeway Reliability and Strategy
Assessment. The exhibits provide values in 5 percentile increments and include a
combined set of values.
Because the free-flow reference speeds used in these data sets do not exactly
correspond to the FFS estimates that an HCM analytical method or field-
measurement technique would produce, the TTI values presented in these
exhibits should be interpreted as being relative to the stated reference speed.
TTIs calculated by using the HCM definition of FFS could be different, but the
general patterns observed would be similar.
Chapter 36/Concepts: Supplemental Reliability Values for Selected U.S. Facilities

Exhibit 36-8 Freeways Urban Streets

Rankings of U.S. Facilities by Percentile Rank TTI50 TTImean PTI TTI50 TTImean PTI
Mean TTI and PTI (A.M. Peak, Minimum 1.01 1.02 1.07 1.03 1.06 1.23
Midday, and P.M. Peak Worst 95% 1.02 1.05 1.09 1.09 1.12 1.27
Combined) Worst 90% 1.02 1.06 1.13 1.13 1.15 1.29
Worst 85% 1.04 1.06 1.14 1.15 1.16 1.32
Worst 80% 1.05 1.08 1.17 1.17 1.20 1.33
Worst 75% 1.05 1.08 1.22 1.19 1.20 1.35
Worst 70% 1.05 1.09 1.25 1.19 1.22 1.36
Worst 65% 1.06 1.10 1.30 1.20 1.22 1.39
Worst 60% 1.07 1.12 1.34 1.20 1.23 1.41
Worst 55% 1.08 1.15 1.39 1.21 1.23 1.42
Worst 50% 1.10 1.16 1.47 1.23 1.26 1.44
Worst 45% 1.11 1.19 1.57 1.24 1.27 1.47
Worst 40% 1.13 1.23 1.73 1.25 1.28 1.49
Worst 35% 1.14 1.30 1.84 1.25 1.29 1.52
Worst 30% 1.17 1.33 1.97 1.26 1.30 1.54
Worst 25% 1.20 1.39 2.24 1.30 1.34 1.60
Worst 20% 1.26 1.43 2.71 1.33 1.36 1.63
Worst 15% 1.31 1.51 2.90 1.35 1.38 1.70
Worst 10% 1.59 1.78 3.34 1.39 1.47 1.84
Worst 5% 1.75 1.97 3.60 1.45 1.54 1.98
Maximum 2.55 2.73 4.73 1.60 1.66 2.55
Source: Derived from directional values in Exhibit 36-12 through Exhibit 36-17. Entries are the lowest value for a
category.
Note: TTI50 = 50th percentile travel time index (50th percentile travel time divided by base travel time).
TTImean = mean travel time index (mean travel time divided by base travel time).
PTI = planning time index (95th percentile travel time divided by base travel time).
For freeways, the base travel time is the free-flow travel time. For urban streets, the base travel time
corresponds to the 85th percentile highest speed observed during off-peak hours.
Exhibit 36-9 Freeways Urban Streets

Rankings of U.S. Facilities by Percentile Rank TTI50 TTImean PTI TTI50 TTImean PTI
Mean TTI and PTI (A.M. Peak) Minimum 1.01 1.02 1.07 1.03 1.06 1.24
Worst 95% 1.01 1.03 1.08 1.08 1.12 1.24
Worst 90% 1.03 1.05 1.12 1.12 1.13 1.27
Worst 85% 1.04 1.06 1.14 1.13 1.15 1.29
Worst 80% 1.04 1.08 1.14 1.14 1.16 1.29
Worst 75% 1.05 1.08 1.17 1.15 1.16 1.31
Worst 70% 1.06 1.09 1.24 1.16 1.17 1.33
Worst 65% 1.07 1.10 1.36 1.18 1.20 1.35
Worst 60% 1.08 1.11 1.40 1.19 1.20 1.37
Worst 55% 1.08 1.16 1.47 1.19 1.21 1.39
Worst 50% 1.09 1.17 1.53 1.20 1.23 1.41
Worst 45% 1.11 1.19 1.58 1.20 1.24 1.42
Worst 40% 1.12 1.21 1.70 1.22 1.26 1.44
Worst 35% 1.13 1.21 1.78 1.24 1.27 1.50
Worst 30% 1.15 1.25 1.89 1.24 1.28 1.52
Worst 25% 1.20 1.42 2.13 1.25 1.29 1.54
Worst 20% 1.28 1.48 2.61 1.26 1.29 1.57
Worst 15% 1.54 1.83 3.17 1.26 1.29 1.66
Worst 10% 1.72 1.93 3.55 1.28 1.31 1.71
Worst 5% 1.95 2.08 3.92 1.35 1.36 1.84
Maximum 2.17 2.73 4.66 1.38 1.49 2.13
category.
Reliability Values for Selected U.S. Facilities Chapter 36/Concepts: Supplemental

Freeways Urban Streets Exhibit 36-10

Percentile Rank TTI50 TTImean PTI TTI50 TTImean PTI Rankings of U.S. Facilities by
Minimum 1.02 1.03 1.07 1.05 1.07 1.23 Mean TTI and PTI (Midday)
Worst 95% 1.02 1.04 1.08 1.08 1.10 1.27
Worst 90% 1.02 1.05 1.11 1.15 1.18 1.28
Worst 85% 1.02 1.06 1.14 1.16 1.18 1.30
Worst 80% 1.03 1.06 1.15 1.18 1.20 1.33
Worst 75% 1.04 1.08 1.17 1.19 1.21 1.34
Worst 70% 1.05 1.08 1.20 1.19 1.22 1.37
Worst 65% 1.05 1.09 1.21 1.20 1.22 1.39
Worst 60% 1.05 1.09 1.24 1.20 1.23 1.41
Worst 55% 1.06 1.11 1.26 1.21 1.23 1.42
Worst 50% 1.06 1.12 1.32 1.22 1.24 1.45
Worst 45% 1.07 1.13 1.34 1.24 1.27 1.47
Worst 40% 1.09 1.15 1.37 1.25 1.29 1.48
Worst 35% 1.09 1.15 1.43 1.25 1.30 1.51
Worst 30% 1.10 1.17 1.51 1.27 1.32 1.53
Worst 25% 1.12 1.26 1.65 1.30 1.34 1.57
Worst 20% 1.14 1.30 1.92 1.31 1.34 1.60
Worst 15% 1.16 1.32 2.41 1.32 1.35 1.63
Worst 10% 1.17 1.42 2.85 1.33 1.38 1.63
Worst 5% 1.21 1.46 3.16 1.35 1.42 1.86
Maximum 1.31 1.76 3.96 1.47 1.55 2.01
category.
Freeways Urban Streets Exhibit 36-11

Percentile Rank TTI50 TTImean PTI TTI50 TTImean PTI Rankings of U.S. Facilities by
Minimum 1.01 1.05 1.10 1.13 1.14 1.32 Mean TTI and PTI (P.M. Peak)
Worst 95% 1.03 1.06 1.14 1.13 1.15 1.35
Worst 90% 1.04 1.06 1.22 1.18 1.21 1.35
Worst 85% 1.05 1.08 1.24 1.20 1.22 1.36
Worst 80% 1.05 1.09 1.28 1.20 1.22 1.37
Worst 75% 1.06 1.10 1.31 1.21 1.23 1.40
Worst 70% 1.07 1.14 1.32 1.22 1.23 1.41
Worst 65% 1.11 1.16 1.38 1.23 1.25 1.42
Worst 60% 1.14 1.23 1.59 1.24 1.26 1.44
Worst 55% 1.14 1.30 1.72 1.24 1.27 1.47
Worst 50% 1.17 1.31 1.85 1.25 1.28 1.49
Worst 45% 1.20 1.34 1.94 1.25 1.29 1.50
Worst 40% 1.21 1.36 2.06 1.31 1.33 1.52
Worst 35% 1.23 1.38 2.25 1.34 1.36 1.59
Worst 30% 1.26 1.41 2.46 1.35 1.38 1.64
Worst 25% 1.29 1.48 2.62 1.39 1.44 1.68
Worst 20% 1.35 1.57 2.77 1.41 1.49 1.78
Worst 15% 1.61 1.71 2.93 1.41 1.52 1.83
Worst 10% 1.70 1.86 3.26 1.49 1.56 1.88
Worst 5% 1.76 1.99 3.54 1.56 1.60 2.10
Maximum 2.55 2.73 4.73 1.60 1.66 2.55
category.
Exhibit 36-12 through Exhibit 36-14 present the source freeway data for the
a.m. peak, midday, and p.m. peak periods, respectively. Exhibit 36-15 through

Exhibit 36-17 present the source urban street data for the a.m. peak, midday, and
p.m. peak periods, respectively.
Exhibit 36-12 Length FFRS Avg. Travel

Freeway Reliability Values: Location Freeway (mi) (mi/h) Direction Time (min) TTImean PTI
Weekday A.M. Peak Period Delaware I-495 11.5 65 NB 11.0 1.03 1.08
Delaware I-495 11.6 65 SB 11.1 1.03 1.07
Delaware I-95 13.4 60 NB 14.6 1.10 1.37
Delaware I-95 13.1 61 SB 13.5 1.05 1.13
Los Angeles I-10 4.6 64 EB 4.5 1.06 1.12
Los Angeles I-10 4.6 65 WB 4.5 1.08 1.14
Los Angeles I-210 4.6 66 EB 4.9 1.17 1.57
Los Angeles I-210 4.6 69 WB 4.6 1.16 1.57
Maryland I-495 ES 26.5 63 SB 28.0 1.10 1.42
Maryland I-495 ES 26.7 62 NB 31.1 1.20 1.71
Maryland I-495 WS 15.4 60 NB 18.3 1.19 1.68
Maryland I-495 WS 15.3 61 SB 26.9 1.78 2.71
Pennsylvania I-76 3.7 51 EB 4.7 1.08 1.22
Pennsylvania I-76 3.6 49 WB 6.5 1.49 3.06
Philadelphia I-76 3.7 51 EB 4.7 1.08 1.22
Philadelphia I-76 3.6 49 WB 6.5 1.79 3.06
Sacramento US-50 6.0 69 EB 5.7 1.10 1.27
Sacramento US-50 6.0 71 WB 6.2 1.21 1.78
Sacramento I-80 12.4 68 EB 11.5 1.06 1.14
Sacramento I-80 12.4 67 WB 12.0 1.09 1.17
San Diego I-5 10.6 71 NB 11.1 1.23 1.81
San Diego I-5 10.6 72 SB 9.1 1.02 1.07
San Diego I-15 3.9 70 NB 4.7 1.41 2.10
San Diego I-15 3.9 69 SB 7.3 1.58 3.38
San Francisco I-880 4.6 71 NB 4.6 1.17 1.47
San Francisco I-880 4.8 67 SB 8.2 1.92 3.57
Notes: FFRS = free-flow reference speed, calculated empirically; may not exactly match the HCM-defined FFS.
TTImean = mean travel time index (mean travel time divided by free-flow travel time).
PTI = planning time index (95th percentile travel time divided by free-flow travel time).
NB = northbound, SB = southbound, EB = eastbound, WB = westbound, ES = east side, WS = west side.

Freeway Reliability Values: Location Roadway (mi) (mi/h) Direction Time (min) TTImean PTI
Weekday Midday Periods Delaware I-495 11.5 65 NB 11.0 1.03 1.07
Delaware I-495 11.6 65 SB 11.3 1.05 1.11
Delaware I-95 13.4 60 NB 13.9 1.05 1.20
Delaware I-95 13.1 61 SB 13.8 1.08 1.34
Los Angeles I-10 4.6 64 EB 4.5 1.06 1.15
Los Angeles I-10 4.6 65 WB 4.5 1.08 1.14
Los Angeles I-210 4.6 66 EB 4.8 1.16 1.32
Los Angeles I-210 4.6 69 WB 4.4 1.10 1.18
Maryland I-495 ES 26.5 63 SB 27.2 1.07 1.31
Maryland I-495 ES 26.7 62 NB 28.2 1.09 1.42
Maryland I-495 WS 15.4 60 NB 20.5 1.34 2.69
Maryland I-495 WS 15.3 61 SB 19.8 1.30 2.26
Sacramento US-50 6.0 69 EB 5.8 1.11 1.20
Sacramento US-50 6.0 71 WB 5.9 1.15 1.47
Sacramento I-80 12.4 68 EB 11.8 1.09 1.25
Sacramento I-80 12.4 67 WB 11.9 1.08 1.14
San Diego I-5 10.6 71 NB 9.3 1.03 1.07
San Diego I-5 10.6 72 SB 9.5 1.06 1.21
San Diego I-15 3.9 70 NB 3.8 1.13 1.23
San Diego I-15 3.9 69 SB 4.1 1.24 1.61

Length FFRS Avg. Travel Exhibit 36-14

Location Roadway (mi) (mi/h) Direction Time (min) TTImean PTI Freeway Reliability Values:
Delaware I-495 11.5 65 NB 11.4 1.06 1.23 Weekday P.M. Peak Period
Delaware I-495 11.6 65 SB 12.0 1.10 1.39
Delaware I-95 13.4 60 NB 14.6 1.10 1.29
Delaware I-95 13.1 61 SB 16.8 1.30 1.83
Los Angeles I-10 4.6 64 EB 5.1 1.20 1.31
Los Angeles I-10 4.6 65 WB 4.9 1.16 1.28
Los Angeles I-210 4.6 66 EB 4.5 1.08 1.35
Los Angeles I-210 4.6 69 WB 4.2 1.06 1.15
Maryland I-495 ES 26.5 63 SB 33.3 1.31 1.85
Maryland I-495 ES 26.7 62 NB 33.7 1.31 1.98
Maryland I-495 WS 15.4 60 NB 41.8 2.73 4.73
Maryland I-495 WS 15.3 61 SB 30.6 2.02 3.67
Sacramento US-50 6.0 69 EB 7.0 1.35 2.12
Sacramento US-50 6.0 71 WB 7.7 1.51 2.74
Sacramento I-80 12.4 68 EB 13.9 1.28 1.84
Sacramento I-80 12.4 67 WB 12.1 1.09 1.31
San Diego I-5 10.6 71 NB 9.4 1.05 1.22
San Diego I-5 10.6 72 SB 13.1 1.47 2.45
San Diego I-15 3.9 70 NB 4.7 1.18 2.97
San Diego I-15 3.9 69 SB 3.8 1.14 1.50
Length FFRS Avg. Travel Exhibit 36-15

Location Roadway (mi) (mi/h) Direction Time (min) TTImean PTI Urban Street Reliability
California Telegraph Canyon Rd. 4.4 45 EB 6.19 1.06 1.24 Values: Weekday A.M. Peak
California Telegraph Canyon Rd. 4.4 45 WB 6.57 1.12 1.42 Period
Delaware US-202 3.8 42 NB 6.97 1.28 1.55
Delaware US-202 3.9 44 SB 6.52 1.20 1.41
Maryland Hwy 175 7.4 38 NB 13.92 1.20 1.32
Maryland Hwy 175 7.4 38 SB 14.00 1.21 1.35
Maryland Hwy 193 5.9 33 EB 13.75 1.26 1.45
Maryland Hwy 193 5.9 33 WB 13.72 1.27 1.52
Maryland Hwy 198 10.1 42 EB 16.51 1.13 1.24
Maryland Hwy 198 10.2 41 WB 16.95 1.15 1.27
Maryland Hwy 355 4.2 30 NB 10.37 1.23 1.38
Maryland Hwy 355 4.2 30 SB 12.57 1.49 2.13
Maryland Randolph Rd. 6.7 35 EB 14.13 1.22 1.36
Maryland Randolph Rd. 6.7 35 WB 15.28 1.31 1.71
Maryland US-40 4.1 41 EB 7.00 1.16 1.29
Maryland US-40 4.2 39 WB 8.50 1.29 1.85
Pennsylvania US-1 8.0 33 NB 19.68 1.36 1.67
Pennsylvania US-1 7.6 32 SB 18.18 1.29 1.52
Philadelphia Hwy 611 3.4 20 NB 13.26 1.29 1.58
Philadelphia Hwy 611 3.3 19 SB 12.89 1.25 1.41
South Carolina US-378 5.5 44 EB 8.61 1.16 1.29
South Carolina US-378 5.4 45 WB 8.37 1.16 1.31
NB = northbound, SB = southbound, EB = eastbound, WB = westbound.
The base travel time corresponds to the 85th percentile highest speed observed during off-peak hours.


Urban Street Reliability Location Roadway (mi) (mi/h) Direction Time (min) TTImean PTI
Values: Weekday Midday California Telegraph Canyon Rd. 4.4 45 EB 6.27 1.07 1.23
Periods California Telegraph Canyon Rd. 4.4 45 WB 6.46 1.10 1.28
Delaware US-202 3.8 42 NB 7.28 1.34 1.63
Delaware US-202 3.9 44 SB 6.93 1.28 1.47
Maryland Hwy 175 7.4 38 NB 13.93 1.20 1.33
Maryland Hwy 175 7.4 38 SB 14.17 1.23 1.38
Maryland Hwy 193 5.9 33 EB 14.29 1.31 1.52
Maryland Hwy 193 5.9 33 WB 13.99 1.29 1.49
Maryland Hwy 198 10.1 42 EB 17.13 1.18 1.29
Maryland Hwy 198 10.2 41 WB 17.47 1.18 1.27
Maryland Hwy 355 4.2 30 NB 12.02 1.42 1.87
Maryland Hwy 355 4.2 30 SB 13.07 1.55 2.01
Maryland US-40 4.1 41 EB 7.44 1.23 1.47
Maryland US-40 4.2 39 WB 8.01 1.22 1.42

Urban Street Reliability Location Roadway (mi) (mi/h) Direction Time (min) TTImean PTI
Values: Weekday P.M. Peak California Telegraph Canyon Rd. 4.4 45 EB 6.71 1.14 1.35
Period California Telegraph Canyon Rd. 4.4 45 WB 6.73 1.15 1.35
Delaware US-202 3.8 42 NB 7.42 1.36 1.62
Delaware US-202 3.9 44 SB 6.84 1.26 1.43
Maryland Hwy 175 7.4 38 NB 14.20 1.23 1.36
Maryland Hwy 175 7.4 38 SB 14.81 1.28 1.49
Maryland Hwy 193 5.9 33 EB 16.39 1.50 1.83
Maryland Hwy 193 5.9 33 WB 15.67 1.45 1.69
Maryland Hwy 198 10.1 42 EB 18.53 1.27 1.50
Maryland Hwy 198 10.2 41 WB 17.81 1.21 1.32
Maryland Hwy 355 4.2 30 NB 14.03 1.66 2.11
Maryland Hwy 355 4.2 30 SB 13.47 1.60 1.89
Maryland US-40 4.1 41 EB 9.40 1.56 2.55
Maryland US-40 4.2 39 WB 8.04 1.22 1.41
RELIABILITY STATISTICS FOR FLORIDA FREEWAYS

Exhibit 36-18 presents reliability statistics for a cross section of Florida
freeways (10). The data were gathered and reported for the p.m. peak period
(4:30 to 6:00 p.m.) and are not aggregated over the length of the facility. The data
consist of spot speeds that have been inverted into travel time rates (min/mi).
The reliability statistics for Florida are reported separately from the rest of
the United States because Florida was testing a variety of definitions of FFS in the

research from which these data were obtained (10). Florida usually sets the FFS
for its freeways as the posted speed limit plus 5 mi/h. However, a speed of 5 mi/h
less than the posted speed limit and a policy speed of 40 mi/h were also being
tested for reliability computation purposes. The following statistics are
presented:
• Four different TTIs (50th, 80th, 90th, and 95th percentile TTIs) based on a
definition of FFS of the posted speed plus 5 mi/h;
• Two policy indices, one based on the 50th percentile speed and a target
speed of the posted speed minus 5 mi/h, the other based on the 50th
percentile speed and a speed of 40 mi/h;
• A buffer time index based on the 95th percentile speed and the mean
speed; and
• A misery index based on the average of the highest 5% of travel times and
a free-flow travel time derived from the posted speed plus 5 mi/h.
Policy Policy Buffer Exhibit 36-18

TTI95 Index Index Time Misery Florida Freeway Reliability
Location TTI50 TTI80 TTI90 (PTI) Alt. 1 Alt. 2 Index Index Statistics
I-95 NB at NW 19th St. 1.00 1.36 1.69 2.01 1.27 1.75 2.02 2.22
I-95 SB at NW 19th St. 1.08 1.19 1.58 2.01 1.27 1.75 1.86 2.48
I-95 NB, S of Atlantic Blvd. 1.03 1.28 1.73 2.23 1.27 1.75 2.16 2.74
I-95 SB, S of Atlantic Blvd. 1.10 1.36 1.89 2.37 1.27 1.75 2.15 2.93
SR 826 NB at NW 66th St. 2.40 2.82 3.07 3.35 1.33 1.50 1.39 3.69
SR 826 SB at NW 66th St. 1.01 1.28 2.63 4.06 1.33 1.50 4.02 4.62
SR 826 WB, W of NW 67th Ave. 1.04 1.08 1.21 1.77 1.33 1.50 1.70 2.10
SR 826 EB, W of NW 67th Ave. 0.98 1.00 1.02 1.04 1.33 1.50 1.07 1.10
I-4 EB, W of World Dr. 0.97 1.04 1.06 1.08 1.27 1.75 1.12 1.12
I-4 WB, W of World Dr. 1.02 1.09 1.49 1.90 1.27 1.75 1.86 2.22
I-4 EB, W of Central Florida Pkwy. 1.06 1.13 1.18 1.31 1.27 1.75 1.24 1.56
I-4 WB, W of Central Florida
1.05 1.36 1.63 1.81 1.27 1.75 1.72 2.03
Pkwy.
I-275 NB, N of MLK Jr Blvd. 1.45 1.71 1.91 2.16 1.33 1.50 1.49 2.58
I-275 SB, N of MLK Jr Blvd. 0.97 1.01 1.04 1.12 1.33 1.50 1.15 1.28
I-275 NB, N of Fletcher Blvd. 1.05 1.07 1.11 1.21 1.33 1.50 1.16 1.35
I-275 SB, N of Fletcher Blvd. 0.96 0.98 0.99 1.00 1.33 1.50 1.04 1.01
I-10 EB, E of Lane Ave. 0.93 0.96 0.98 0.99 1.33 1.50 1.07 1.01
I-10 WB, E of Lane Ave. 0.97 1.10 1.24 1.46 1.33 1.50 1.51 1.87
I-95 NB, S of Spring Glen Rd. 1.04 1.09 1.26 1.77 1.27 1.75 1.70 2.00
I-95 SB, S of Spring Glen Rd. 1.16 1.30 1.42 1.60 1.27 1.75 1.38 1.88
Minimum 0.93 0.96 0.98 0.99 1.27 1.50 1.04 1.01
Average 1.11 1.26 1.51 1.81 1.30 1.63 1.64 2.09
Maximum 2.40 2.82 3.07 4.06 1.33 1.75 4.02 4.62
Source: Adapted from Kittelson & Associates, Inc. (10).
Notes: TTIxx = travel time index based on the percentile speed indicated in the subscript and a free-flow speed
defined as the posted speed plus 5 mi/h.
PTI = planning time index.
Policy Index Alternative 1 = index based on the 50th percentile speed and a target speed of the posted
speed minus 10 mi/h.
Policy Index Alternative 2 = index based on the 50th percentile speed and a target speed of 40 mi/h.
Buffer time index = index based on the ratio of the 95th percentile and mean travel speeds.
Misery index = index based on the ratio of (a) the average of the highest 5% of travel times and
(b) a free-flow travel time defined as the posted speed plus 5 mi/h.
N = north, S = south, E = east, W = west, NB = northbound, SB = southbound, EB = eastbound, WB =
westbound.

5. VEHICLE TRAJECTORY ANALYSIS
INTRODUCTION
Overview
This section contains expanded guidance for the use of alternative traffic
analysis tools (mostly microsimulation tools) in assessing the performance of
highway facilities. An important part of the guidance deals with the use of
vehicle trajectory analysis as the “lowest common denominator” for comparing
performance measures from different tools. Material on vehicle trajectory
analysis is also included in the following chapters:
• Chapter 4, Traffic Operations and Capacity Concepts, introduces the concept
of individual vehicle trajectory analysis. A growing school of thought
suggests that comparing results between traffic analysis tools and
methods is possible only through analyzing vehicle trajectories as the
“lowest common denominator.” Vehicle trajectories can be used to
develop performance measures that are consistent with HCM definitions,
with field measurement techniques, and with each other. Examples of
vehicle trajectory plots were shown that illustrate the visual properties of
vehicle trajectories.
• Chapter 7, Interpreting HCM and Alternative Tool Results, explores the use of
vehicle trajectory analysis in defining and estimating consistent
performance measures. First, it introduces the mathematical properties of
trajectories as an extension of the visual properties. Next, it identifies the
performance measures that can be computed from trajectories and
explores their compatibility with the performance measures estimated by
the computational procedures presented throughout the HCM.
Chapter 7 presents general guidelines for defining and comparing measures
from different traffic analysis tools. Those guidelines are expanded in this section
through presentation of more specific trajectory analysis procedures by which
consistent performance measures can be estimated. The trajectory analysis
procedures described in this section were developed and tested by
postprocessing the external trajectory files produced by a typical simulation tool.
The postprocessor features and the process by which the procedures were
developed are described elsewhere (11).
Several examples of the analysis of vehicle trajectories on both interrupted-
and uninterrupted-flow facilities are presented here. These examples
demonstrate the complexities that can arise, for example, in multilane situations,
multiphase operations, situations in which the demand exceeds the capacity, and
situations in which vehicles are unable to access a desired lane because of
congestion. Specific procedures are then proposed and demonstrated with
additional examples.
Vehicle Trajectory Analysis Chapter 36/Concepts: Supplemental

Mathematical Properties of Vehicle Trajectories

As was pointed out in Chapter 7, an analysis of vehicle trajectories requires a
mathematical representation that includes a set of properties associated with
each vehicle at specific points in time and space. Some of the material on
mathematical properties of vehicle trajectories presented in this section is also
included in Chapter 7. It is repeated here to provide a convenient introduction to
the topic of vehicle trajectory analysis. A graphic representation of the path of an
individual vehicle in space and time is also repeated here as Exhibit 36-19.
Exhibit 36-19
Vehicle Data Stored for Each
Time Step
Many properties can be associated with a specific vehicle at a point in time.

Some properties are required for the accurate determination of performance
measures from trajectories. Others are used for different purposes, such as safety
analysis.
Basic Trajectory Properties

The basic trajectory properties from which all the required performance
measures can be estimated include the following information for each vehicle
within the facility boundaries and for each time step within the analysis period:
• Vehicle identification: Vehicle identification is required to distinguish a
specific vehicle from all other vehicles within the facility boundaries.
• Position: This property is the most basic of all, and many other properties
may be derived from it. A one-dimensional position is sufficient to
produce performance measures. Some question remains about a universal
representation of position, because different tools specify the position in
different ways. A common reference point for position needs to be
established. A reference point that indicates the relative position of the
vehicle in the link would be desirable to enable developers to produce
uniform measures.
• Link or segment: A link or segment is required to associate performance
measures with a specific link or analysis segment for reporting purposes.
• Lane: In multilane facilities, knowledge of the lane in which the vehicle is
traveling is important because headways, densities, and other measures
must be estimated by lane. It is also necessary for identifying lane
changes.
Chapter 36/Concepts: Supplemental Vehicle Trajectory Analysis

Static Vehicle and Facility Parameters

Some required properties can be derived from the basic properties with
knowledge of certain parameters that are constant with respect to time:
• Vehicle length: Required to convert headways to gaps, and
• Link end positions: Required to determine the position of the vehicle with
respect to the upstream or downstream end of the link.
Some simulation tools repeat this static information in each record to avoid
the need for an external parameter file.
Derived Trajectory Properties

The remainder of the required trajectory properties can be derived from the
basic properties as follows:
• Instantaneous speed: This property can be determined from the relative
positions of the vehicle at time t and time t – t on the assumption of a
constant acceleration during t. However, since most tools update vehicle
positions from the speeds, speed is commonly included as a basic
trajectory property.
• Instantaneous acceleration: This property can be determined from the
relative speeds of the vehicle at time t and time t – t on the assumption
of a constant acceleration during t. However, since most tools update
vehicle speeds from the acceleration, acceleration is commonly included
as a basic trajectory property.
TRAJECTORY ANALYSIS EXAMPLES

This section demonstrates the ability of alternative analysis tools to quantify
trajectory properties. Several examples are presented for both uninterrupted- and
interrupted-flow facilities.
Basic Signalized Intersection

The first example is very basic. The intersection configuration involves two
single-lane, one-way streets as shown in Exhibit 36-20. To simplify the situation
even more, the simulation parameters are adjusted to enforce a uniform
operation. Essentially, all the randomness inherent in simulation is removed. A
simulation of uniform conditions would not normally produce useful results, but
this example provides a good starting point for illustrating the nature of vehicle
trajectory plots.
A trajectory plot showing two cycles of simulated operation for this example
is presented in Exhibit 36-21(a). This form is the classic one that appears often in
the literature to support discussion related to queue accumulation and discharge.
A copy of the exhibit used in Chapter 31, Signalized Intersections: Supplemental,
to illustrate the basic traffic signal principles is also included as Exhibit 36-21(b).
The two figures are different in that the first was produced directly from the
vehicle trajectory data while the second was drawn by hand. The ability to
reproduce the classic representation from controlled conditions will provide a
measure of confidence in the validity of future examples involving much more
complicated situations.

Demand volume Signal timing Exhibit 36-20

Cycle length: 60 s Basic Signalized Intersection
750 veh/h on each approach
Green: 26 s Example
Intergreen: 4 s
Uniform parameters
Vehicles generated from a uniform
distribution
No speed, headway, deceleration, or
start-up lost time variation
All vehicles are 16-ft passenger cars
Maximum deceleration = 8 ft/s2
1,000 Exhibit 36-21

Trajectory Plots for Uniform
900 Arrivals and Departures
800
Area represented by the HCM
uniform delay equation
700
CYCLE 1 CYCLE 2
Stop line
600
Position (ft)
500
400
300
200
100
Note the similarity between the
0 trajectories obtained from the
0 10 20 30 40 50 60 70 80 90 100 file (above) and those
Time (s) developed manually in Chapter
31 (below) to illustrate the
basic principles of signalized
(a) Plot Produced from Simulation
intersection operation.
(b) Plot Produced by Hand

Restoring Randomness to the Simulation

To simplify the discussion, the first example was presented with all
randomness removed from the operation. Subsequent examples are more
realistic in their treatment of traffic flow. Vehicles are generated at entry points
from a Poisson distribution, and the simulation tool’s default parameters for
randomizing driver behavior are applied.
Exhibit 36-22 shows a sample trajectory plot for the same operation depicted
in Exhibit 36-21. As expected, the individual trajectories follow the same pattern
as the uniform case, except that some spacings and speeds are not as consistent.
The trajectory lines do not cross each other in this example because the example
uses a single-lane approach and overtaking is not possible.
Exhibit 36-22 1,000

Introducing Randomness into
the Simulation 900
800
700
600
Distance (ft)
500
400
300
200
100
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160
Time (s)
Vehicle Trajectories for Oversaturated Operation

Up to this point, the examples have involved volume-to-capacity ratios less
than 1.0, in which all vehicles arriving on a given cycle were able to clear on the
same cycle. Saturation levels close to and above 1.0 present a different picture.
Three cases are presented here:
1. Cycle failure, occurring when saturation approaches 1.0 and residual
queues build on one cycle but are resolved on the next cycle;
2. Oversaturated operation, a situation in which the link has a demand volume
exceeding the link’s capacity and queues extend throughout the approach
link; and
3. Undersaturated operation, in which queues extend to an upstream link for a
part of a cycle because of closely spaced intersections.

Cycle Failure
A cycle failure example is presented in Exhibit 36-23. This trajectory plot
shows a situation in which some vehicles arriving in Cycle 1 were unable to clear
until Cycle 2. This condition is identified from the trajectory plot for four stopped
vehicles (i.e., horizontal trajectory lines) that were forced to stop again before
reaching the stop line. These vehicles became the first four vehicles in the queue
for Cycle 2. Fortunately, the arrivals during Cycle 2 were few enough that all
stopped vehicles were able to clear the intersection before the beginning of the
red phase. A closer inspection of Exhibit 36-23 shows that one more vehicle,
which was not stopped, was also able to clear.
1,000 Exhibit 36-23

Cycle Failure Example
Vehicles stopped
in Cycle 1 that
800 did not clear until
Cycle 2
CYCLE 1 CYCLE 2 CYCLE 3

600
Distance (ft)
400
Last vehicle stopped in

200
Cycle 2 cleared before
the red
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190
Time (s)
Severely Oversaturated Operation

Oversaturated operation was produced by increasing the demand volume to
the point where it exceeded the capacity of the approach. The increased demand
produced a queue that extended the length of the link. Inspection of the
animated graphics showed that the queue did, in fact, back up beyond the link
entry point.
The vehicle trajectory plot for this operation is presented in Exhibit 36-24.
The move-up process is represented in the trajectories. Vehicles entering the link
require up to three cycles to clear the intersection. The implications for control
delay computations when the queue occupies a substantial proportion of the link
are discussed in Chapter 7.

Exhibit 36-24 1,000

Oversaturated Signal
Approach
800
Stop line
600
Distance (ft) 400
200
Upstream end of link

0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
Time (s)
A larger question is what to do with the vehicles denied entry during the
analysis period. The answer is that, as indicated in Chapter 19, Signalized
Intersections, the analysis period must be long enough to include a period of
uncongested operation at each end. The delay to vehicles denied entry to this
link will be accounted for in upstream links during the period. The upstream
links must include a holding area outside the system. Some tools include the
delay to vehicles denied entry and some do not. If a tool is used that does not
include denied-entry delay, fictitious links must be built into the network
structure for that purpose.
Queue Backup from a Downstream Signal

Even when an approach is not fully saturated, queues might back up from a
downstream signal for a portion of the cycle. This happens when intersections
are closely spaced. An example of queue backup within a cycle is shown in
Exhibit 36-25.
The two-intersection configuration for this example is shown in Exhibit 36-
25(a). The graphics screen capture shows that vehicles that would normally pass
through the upstream link are prevented from doing so by queues that extend
beyond the end of the downstream link for a portion of the cycle. The question is
how to treat the resulting delay.
By the definitions given to this point, the delay in the upstream link would
be assigned to the upstream link, even though the signal on the downstream link
was the primary cause. The important thing is not to overlook any delay and to
assign all delay somewhere and in a consistent manner. With simulation
modeling, the only practical place to assign delay consistently is the link on
which the delay occurred. Subtle complexities make it impractical to do
otherwise. For example, the root cause of a specific backup might not be the
immediate downstream link. The backup might be secondary to a problem at
some distant location in the network at some other point in time.

Queue backs up from Exhibit 36-25

downstream signal into Queue Backup from a
upstream link Downstream Signal
(a) Simulation Graphics Representation
(b) Vehicle Trajectory Representation
More Complex Signal Phasing

Up to this point only simple signal phasing has been considered. Many
applications involve simulating more complex phasing on urban streets. As an
example of a more complex situation, a left turn moving on both a protected and
a permitted phase is examined.
Exhibit 36-26 shows the trajectory plot for an eastbound left-turn movement
from an exclusive lane controlled by a signal with both protected and permitted
phases. In this case, the upstream link is the eastbound approach to the
intersection and the downstream link is the northbound approach to the next
intersection. Because the distance on a trajectory plot is one-dimensional, the
distance scale is linear, even though the actual route takes a right-angle bend.

Exhibit 36-26
Trajectory Plot for More
Complex Signal Phasing
Even with an undersaturated operation, this trajectory plot is substantially

more involved than the previous ones. Several phenomena are identified in the
exhibit, including the following:
1. Cross-street traffic entering the downstream link on the northbound
phase: These vehicles do not appear on the upstream link because they
are on a different link. They enter the downstream link at the stop line on
the red phase for the left-turn movement of interest.
2. Left turns on the protected phase, shown as solid lines on the trajectory
plot: The protected left-turn phase takes place immediately after the red
phase. The left-turning vehicles begin to cross the stop line at that point.
3. Left turns on the permitted phase, shown as broken lines on the trajectory
plot: The permitted left-turn phase takes place immediately after the
protected phase. There is a gap in the trajectory plot because the left-
turning vehicles must wait for oncoming traffic to clear.
4. Left-turn “sneakers”: Explicit identification of a sneaker on the trajectory
plot is not possible; however, the last left turn to clear the intersection on
the permitted phase is probably a sneaker if it enters at the end of the
permitted phase.
5. Left-turn vehicles that enter the link in the through lane and change into
the left lane somewhere along the link: These vehicles are identified by
trajectories that begin in the middle of the link.
6. Through vehicles that enter the link in the left-turn lane and change into
the through lane somewhere along the link: These vehicles are identified
by trajectories that end abruptly in the middle of the link.
The trajectory plot shown for this example is more complex than the
previous plots; however, performance can be analyzed in the same way.

Freeway Examples
Freeway trajectories follow the same definitions as surface street trajectories,
but the queuing patterns differ because they are created by car-following
phenomena and not by traffic signals. The performance measures of interest also
differ. There is no notion of control delay on freeways because there is no control.
The level of service on uninterrupted-flow facilities is based on traffic density
expressed in units of vehicles per mile per lane. In some cases, such as merging
segments, the density in specific lanes is of interest.
Two cases are examined. The first deals with a weaving segment, and the
second deals with merging at an entrance ramp.
Weaving Segment Example

Simulation Network Structure
The problem description, link–node structure, and animated graphics view
for the weaving segment example are shown in Exhibit 36-27. The scenario is the
same as that used in Example Problem 1 in Chapter 27, Freeway Weaving:
Supplemental. There are two lanes on the freeway and on each ramp. The two
ramp lanes are connected by full auxiliary lanes.
Exhibit 36-27
Weaving Segment Description
and Animated Graphics View
LS = 1,500 ft
v FF = 1,815 veh/h
v RF = 1,037 veh/h
v FR = 692 veh/h
v RR = 1,297 veh/h
v = 4,841 veh/h
1 2 3 4
1,000 ft 1,500 ft 1,000 ft
63
9f
t
0f t
71
6
5
Note: LS = length of segment, VFF = vehicles entering from freeway and leaving to freeway, VRF = vehicles
entering from ramp and leaving to freeway, VFR = vehicles entering from freeway and leaving to ramp, VRR
= vehicles entering from ramp and leaving to ramp, veh/h = vehicles per hour.

Vehicle Trajectories for the Freeway Lanes

The vertical (i.e., distance) axis of the trajectory plot provides a linear one-
dimensional representation of a series of connected links. The links can follow
any pattern as long as some of the vehicles leaving one link flow into the next
link. The analysis tool accommodates a maximum of eight connected links. When
multiple links are connected to a node (as is usually the case), different
combinations of links may be used to construct a multilink trajectory analysis.
The route configuration must be designed with the end product in mind.
Sometimes multiple routes must be examined to obtain a complete picture of the
operation.
There are two entry links and two exit links to the weaving segment, giving
four possible routes for analysis. Two routes are examined in this example. The
first route, which is represented in Exhibit 36-28, shows the traffic entering the
weaving segment from the freeway and leaving to the freeway (VFF in Exhibit 36-
27), represented by Links 1–2–3–4. The second route will be examined in the next
subsection.
Exhibit 36-28 3,500

Trajectory Plot for Freeway
Links Vehicle exiting
3,000 to ramp
Exit ramp
2,500
Slower vehicle overtaken

by 2 other vehicles
Distance (ft)
2,000
Vehicles entering
from ramp
1,500
1,000
Entrance ramp
Link density at time t =

500 number of vehicles on the link at time
t divided by the link length
0
0 5 10 15 20 25 30 35 40 45 50 55
Time (s)
In this multilane plot, in contrast to previous plots, some of the trajectory

lines might cross each other because of different speeds in different lanes. One
such instance is highlighted in Exhibit 36-28. This figure also shows vehicles that
enter and leave the weaving segment on the ramps. Because the ramps are not
part of the selected route, the ramp vehicles appear on the trajectory plot only on
the link that represents the weaving segment. Examples of ramp vehicles are
identified in the figure.
The definition of link density (vehicles per mile) is also indicated in Exhibit
36-28. Density as a function of time t is expressed in vehicles per mile and is
determined by counting the number of vehicles within the link and dividing by
the link length in miles. Average lane density (vehicles per mile per lane) on the

link may then be determined by dividing the link density by the number of lanes.
To obtain individual lane densities, the trajectory analysis must be performed on
each lane. The analysis must also be performed on a per lane basis to examine
individual vehicle headways.
Vehicle Trajectories for the Entrance and Exit Ramps

By specifying the links on the route as 5–2–3–6 instead of 1–2–3–4, the
trajectories for vehicles entering and leaving the weaving segment on the ramps
(VRR in Exhibit 36-27) can be examined. This trajectory plot is shown in Exhibit 36-
29. This figure is similar to Exhibit 36-28, except that the vehicles that do not
appear outside the weaving segment are those on the freeway links instead of the
ramp links.
Two other routes can also be constructed, one for vehicles entering from the
freeway and leaving to the exit ramp, VFR, as 1–2–3–6, and one for those entering
from the ramp and leaving to the freeway, VRF, as 5–2–3–4. These plots are not
included here.
3,000 Exhibit 36-29

Trajectory Plot for Entrance
Continuation on and Exit Ramp Links
freeway
2,500
Exit ramp
2,000
Distance (ft)
1,500 Entry from

freeway
1,000
Entrance ramp
500
0
0 5 10 15 20 25 30 35 40 45 50
Time (s)
Entrance Ramp Merging Example

Merging segments provide another good example of vehicle trajectory
analysis on a freeway. The merging vehicles affect freeway operation differently
in each lane, so each lane must be examined independently.
Simulation Network Structure

The same node structure used in the weaving segment example is used here.
The lane configuration has been changed to be more representative of a merge
operation. Three lanes have been assigned to the freeway and one lane to the
entrance ramp. The demand volumes have been specified to provide a near-
saturated operation to observe the effects of merging under these conditions. A
graphic view of the operation is presented in Exhibit 36-30.

Exhibit 36-30
Entrance Ramp Merging
Segment Graphics View
Trajectory Plots for All Lanes

Exhibit 36-31 shows a trajectory plot for all freeway lanes combined within
the merge area. The operation is clearly heterogeneous, with a mixture of fast
and slow speeds. Many trajectory lines cross each other, and not much can be
done in the way of analysis with these data.
Exhibit 36-31
Trajectory Plot for All Freeway
Lanes in the Merge Area
Trajectory Plots for Individual Lanes

Clearly, each lane must be examined individually. Exhibit 36-32, Exhibit 36-
33, and Exhibit 36-34 show selected trajectories for Lanes 1, 2, and 3, respectively,
from a later point in time in the simulation. Because these plots represent
individual lanes, the trajectory lines do not cross each other. The effect of the
merging operation is observable (and predictable) in these three figures.
In Lane 1, freeway speeds are low upstream of the merge point. Merging
vehicles enter the freeway slowly but pick up speed rapidly downstream of the
merge point bottleneck. The merging vehicles enter the freeway from the
acceleration lane, which begins at 1,000 ft on the distance scale. The merging
vehicle trajectories before entry onto the freeway are not shown in Exhibit 36-32
because those vehicles are either on a different link or in a different lane.

Exhibit 36-32
Lane 1 (Rightmost) in the
Merge Area
Exhibit 36-33
Lane 2 (Center) in the Merge
Area
Exhibit 36-34
Lane 3 (Leftmost) in the
Merge Area

In Lane 2, the freeway speeds are higher but still well below the FFS,
indicating that the merge operation affects the second lane as well. Some vehicles
enter Lane 2 in the vicinity of the acceleration lane, but they are generally
vehicles that have left Lane 1 to avoid the friction. Both Lane 1 and Lane 2 show
several discontinuous trajectories that indicate lane changes. The Lane 3
operation is much more homogeneous and speeds are higher, indicating a much
smaller effect of the merging operation.
Trajectory Plots for Ramp Vehicles

To configure a trajectory route covering the entrance ramp vehicles, the ramp
and acceleration lane, which were not represented in Exhibit 36-32 through
Exhibit 36-34, must be selected in place of the upstream freeway link. The
acceleration lane number must first be identified from the simulation tool’s
output. Because of the selected tool’s unique and somewhat creative lane
numbering scheme, the acceleration lane will be Lane 9. To cover both the ramp
and the acceleration lane, Lane 9 must be selected on the freeway link (2–3).
The trajectory plot for this route is shown in Exhibit 36-35. The results are not
what might be anticipated. Vehicles are observed on the ramp and in the
acceleration lane, but they disappear as soon as they enter the freeway. More
vehicles eventually appear toward the end of the freeway link. The vehicles
disappear because Lane 9 was selected for the freeway link, so vehicles in Lane 1
do not show up on the plot. The vehicles that reappear at the end of the link are
those leaving the freeway at the downstream exit. They reappear at that point
because the deceleration lane at the end of the link is also assigned as Lane 9.
This plot is not particularly useful, except that it illustrates the complexities of
trajectory analysis.
Exhibit 36-35
Trajectory Plot for
Acceleration and Deceleration
Lanes

To obtain a continuous plot of ramp vehicles, nodes must be added to the

network at the points where the acceleration and deceleration lanes join the
freeway. These nodes are shown as Nodes 7 and 8 in Exhibit 36-36. A continuous
route may then be configured as 5–2–7–8–3–4. Selected trajectories from the
trajectory plot for this route are shown in Exhibit 36-37. This plot shows the
entering vehicles on the ramp as they pass through the acceleration lane onto the
freeway. There are some discontinuities in the trajectories because of the
different point at which vehicles leave the acceleration lane.
1 2 7 8 3 4 Exhibit 36-36
Addition of Intermediate
Nodes for Continuous
Trajectory Plots
6
5
Exhibit 36-37
Trajectory Plot for
Acceleration Lane and
Freeway Lane 1
ESTIMATING PERFORMANCE MEASURES FROM VEHICLE

TRAJECTORY DATA
The preceding subsections demonstrated that the production of vehicle
trajectory plots that can be interpreted and analyzed is possible. This subsection
focuses on computation of the performance measures from a mathematical
analysis of the data represented in these plots.
Trajectory Analysis Procedures Overview

One development goal for the HCM 2010 was the creation of a set of
computational procedures by which developers of simulation tools could
produce performance measures that are consistent among different tools and, to
the extent possible, compatible with the HCM’s deterministic procedures. The
procedures presented here were designed to be implemented easily by using the

common trajectory properties described previously and illustrated by examples.

Developers of simulation tools are encouraged to implement these procedures,
and users of simulation tools are encouraged to consider the extent to which the
procedures have been implemented in the traffic analysis tool selection process
described in Chapter 6, HCM and Alternative Analysis Tools.
Requirements for Trajectory Analysis Algorithm Development

A basic set of guidelines for computing uniform performance measures from
vehicle trajectory analysis was introduced in Chapter 7, Interpreting HCM and
Alternative Tool Results. Since these requirements are also incorporated into the
specific computational procedures proposed in this chapter, they are repeated
here to promote a better understanding of the procedures. The general guidelines
suggested in Chapter 7 include the following:
1. The trajectory analysis procedures are limited to analysis of trajectories
produced by the traffic flow model of each simulation tool. The nature of
the procedures must not suggest the need for developers to change their
driver behavior or traffic flow modeling logic.
2. If the procedures for estimating a particular measure cannot be
satisfactorily defined to permit a valid comparison between the HCM and
other modeling approaches, such comparisons should not be made.
3. All performance measures that accrue over time and space should be
assigned to the link and time interval in which they occur. Subtle
complexities make it impractical to do otherwise. For example, the root
cause of a specific delay might not be within the link or the immediate
downstream link. The delay might be secondary to a problem at some
distant location in the network and in a different time interval.
4. The spatial and temporal boundaries of the analysis domain must include
a period that is free of congestion on all sides. This principle is also stated
in Chapter 10, Freeway Facilities Core Methodology, and in Chapter 19,
Signalized Intersections, for multiperiod signalized intersection analysis.
To ensure that delays to vehicles denied entry to the system during a
given period are properly recognized, creation of fictitious links outside
the physical network to hold such vehicles might be necessary. A more
detailed discussion of spatial and temporal boundaries is provided in
Chapter 7.
5. It is important to ensure that the network has been properly initialized or
“seeded” before trajectory analysis is performed. When the warm-up
periods are set and applied, simulation tools typically start with an empty
network and introduce vehicles until the vehicular content of the network
stabilizes. Trajectory analysis should not begin until stability has been
achieved. If the simulation period begins with oversaturated conditions,
stability may never be achieved. See the discussion in Chapter 7 on
temporal and spatial boundaries.
In addition to the general guidelines, some requirements must be addressed
here to promote the development of trajectory analysis procedures that can be

applied in a practical manner by the developers of simulation tools. The

following requirements are suggested:
1. The algorithms must be suitable for computation “on the fly.” They must
not require information from a future time step that would complicate the
data handling within the simulation process.
2. Arbitrary thresholds for determining parameters should be kept to a
minimum because of the difficulty of obtaining acceptance throughout
the user community for specific thresholds. When arbitrary thresholds
cannot be avoided, they should be justified to the extent possible by
definitions in the literature, and above all, they should be applied
consistently for different types of analysis.
3. Computationally complex and time-consuming methods should be
avoided to minimize the additional load on the model. Methods should
be developed to simplify situations with many special cases because of
the difficulty of enumerating all special cases.
4. The same definitions, thresholds, and logic should be used for
determination of similar parameters in different computational
algorithms for longitudinal and spatial analysis.
Summary of Computational Procedures

Several performance measures were examined in Chapter 7, and general
guidelines for comparing measures produced by different tools were presented.
Previous material in this section has demonstrated the potential for development
of uniform measures by individual vehicle trajectory analysis and has proposed
some requirements for development of the analysis procedures. Specific
procedures for analyzing vehicle trajectories are now presented and
demonstrated with additional examples.
Thresholds for Computation of Performance Measures

Elimination of arbitrary and user-specified values is an important element of
standardization. Avoidance of arbitrary thresholds was identified earlier as a
requirement for the development of trajectory analysis procedures. Avoidance of
all arbitrary thresholds is desirable. If thresholds cannot be avoided, they should
be justified in terms of the literature. When no such justification exists, they
should at least be established on the basis of consensus and applied consistently.
The following thresholds cannot be avoided in vehicle trajectory analysis.
Car Length
The following is stated in Chapter 31, Signalized Intersections: Supplemental:
A vehicle is considered as having joined the queue when it
approaches within one car length of a stopped vehicle and is
itself about to stop. This definition is used because of the
difficulty of keeping track of the moment when a vehicle comes
to a stop.
So, for estimation of queue-related measures, a value that represents one car
length must be chosen. For the purposes of this section, a value of 20 ft is used.

Stopped-Vehicle State
One example of an arbitrary threshold is the speed at which a vehicle is
considered to have come to a stop. Several arbitrary thresholds have been
applied for this purpose. To maintain consistency with the definition of the
stopped state applied in other chapters of the HCM, a speed less than 5 mi/h is
used here for determining when a vehicle has stopped.
Moving-Vehicle States
Other states in addition to the stopped state that must be defined
consistently for vehicle trajectory analysis include the following:
• The uncongested state, in which a vehicle is moving in a traffic stream
that is operating below its capacity;
• The congested state, in which the traffic stream has reached a point that is
at or slightly above its capacity, but no queuing from downstream
bottlenecks is present; and
• The severely constrained state, in which downstream bottlenecks have
affected the operation.
These states apply primarily to uninterrupted flow. A precise definition
would require complex modeling algorithms involving capacity computations or
“look ahead” features, both of which would create a computational burden.
Therefore, an easily applied approximation must be sought. Threshold speeds
are a good candidate for such an approximation.
These states can be thought of conveniently in terms of speed ranges. To
avoid specifying arbitrary speeds as absolute values, use of the target speed of
each vehicle as a reference is preferable. The target speed is the speed at which
the driver prefers to travel. It differs from the FFS in the sense that most
simulation tools apply a “driver aggressiveness” factor to the FFS to determine
the target speed. In the absence of accepted criteria, three equal speed ranges are
applied for the purposes of this section. Thus, the operation is defined as
uncongested if the speed is above two-thirds of the target speed. It is defined as
severely constrained when the speed is below one-third of the target speed, and
it is considered congested in the middle speed range. This stratification is used to
produce performance measures directly (e.g., percent of time severely
constrained). It is also used in computing other performance measures (e.g.,
release from a queue).
Computational Procedures for Stop-Related Measures

The two main stop-related measures are number of stops and stopped delay.
The beginning of a stop is defined in the same way for both measures. The end of
a stop is treated differently for stopped delay and number of stops. For stopped
delay, the end of a stop is established as soon as the vehicle starts to move (i.e.,
its speed reaches 5 mi/h or greater). For determining the number of stops, some
hysteresis is required. For purposes of this section, after a vehicle is stopped a
subsequent stop is not recognized until it leaves the severely constrained state
(i.e., its speed reaches one-third of the target speed).

Because subsequent stops are generally made from a lower speed, they can
be expected to have a smaller impact on driver perception, operating costs, and
safety. Recognizing this fact, the National Cooperative Highway Research
Program (NCHRP) 03-85 project proposed a “proportional stop” concept (11), in
which the proportion of a subsequent stop is based on the relative kinetic energy
loss and is therefore proportional to the square of the speed from which the stop
was made. Thus, each time a vehicle speed drops below 5 mi/h, the number of
stops is incremented by (Smax/Starget)2 , where Smax is the maximum speed attained
since the last stop and Starget is the target speed.
This procedure has not been applied in practice. It is mentioned here because
it offers an interesting possibility for the use of simulation to produce measures
that could be obtained in the field but could not be estimated by the macroscopic
deterministic models described in the HCM. The procedure is illustrated by an
example later in this section.
Computational Procedures for Delay-Related Measures

The procedures for computing delay from vehicle trajectories involve
aggregating all delay measures over each time step. Therefore, the results take
the form of aggregated delay and not unit delay, as defined in Chapter 7. To
determine unit delays, the aggregated delays must be divided by the number of
vehicles involved in the aggregation. Partial trips made over a segment during
the time period add some complexity to the unit delay computations.
The following procedures should be used to compute the various delay-
related measures from vehicle trajectories:
• Time step delay: The delay on any time step is, by definition, the length of
the time step minus the time the vehicle would have taken to cover the
distance traveled in the step at the target speed. This value is easily
determined and is the basis for the remainder of the delay computations.
• Segment delay: Segment delay is the time actually taken to traverse a
segment minus the time that would have been taken to traverse the
segment at the target speed. The segment delay on any step is equal to the
time step delay. Segment delays accumulated over all time steps in which
a vehicle is present on the segment represent the segment delay for that
vehicle.
• Queue delay: Queue delay is equal to the time step delay on any step in
which the vehicle is in a queued state; otherwise, it is zero. Queue delays
are accumulated over all time steps while the vehicle is in a queue.
• Stopped delay: Stopped delay is equal to the time step delay on any step in
which the vehicle is in a stopped state; otherwise, it is zero. Because a
vehicle is considered to be “stopped” if it is traveling at less than a
threshold speed, a consistent definition of stopped delay requires that the
travel time at the target speed be subtracted. Time step delays
accumulated over all time steps in which the vehicle was in the stopped
state represent the stopped delay.

• Control delay: Control delay is the additional travel time caused by

operation of a traffic control device. It cannot be computed directly from
the vehicle trajectories in a manner consistent with the procedures given
in Chapters 19 and 31 for signalized intersection analysis. However, it is
an important measure because it is the basis for determining the level of
service on a signalized approach.
Queue delay computed from The queue delay computed from vehicle trajectories provides a reasonable
trajectory analysis provides the
most appropriate approximation of control delay when the following conditions are met:
representation of control delay.
1. The queue delay is caused by a traffic control device, and
2. The identification of the queued state is consistent with the definitions
provided in this section.
Computational Procedures for Queue-Related Measures

Procedures for computing queue-related measures begin with determining
whether each vehicle in a segment is in a queued state. A vehicle is in a queued
state if it has entered a queue and has not yet left it. The beginning of a queued
state occurs when
• The gap between a vehicle and its leader is less than or equal to 20 ft,
• The vehicle speed is greater than or equal to the leader speed, and
• The vehicle speed is less than or equal to one-third of the target speed
(i.e., the speed is severely constrained).
A separate case must be created to accommodate the first vehicle to arrive at
the stop line. If the link is controlled (interrupted-flow case), the beginning of the
queued state also occurs when
• No leader is present on the link,
• The vehicle is within 50 ft of the stop line, and
• The vehicle is decelerating or has stopped.
These rules have been found to cover all the conditions encountered.
The ending of the queued state also requires some rules. For most purposes,
the vehicle should be considered to remain in the queue until it leaves the link.
The analysis is done on a link-by-link basis. In the case of queues that extend
over multiple links, a vehicle leaving a link immediately enters the queue on the
next link. Experience with trajectory analysis has shown that other conditions
need to be applied to supplement this rule. Thus, the end of the queued state also
occurs when
• The vehicle has reached two-thirds of the target speed (i.e., uncongested
operation), and
• The leader speed is greater than or equal to the vehicle speed or the
vehicle has no leader in the same link.
The additional conditions cover situations in which, for example, a vehicle
escapes a queue by changing lanes into an uncongested lane (e.g., through
vehicle caught temporarily in a turn bay overflow).

Chapters 19 and 31 offer the following guidance on estimating queue length:

1. The maximum queue reach (i.e., back of queue, or BOQ) is a more useful
measure than the number of vehicles in the queue, because the BOQ
causes blockage of lanes. The maximum BOQ is reached when the queue
has almost dissipated (i.e., has zero vehicles remaining).
2. A procedure is prescribed to estimate average maximum BOQ on a
signalized approach.
Because of its macroscopic nature, the HCM queue estimation procedure
cannot be applied directly to simulation. On the other hand, simulation can The BOQ at any time step will
produce additional useful measures because of its higher level of detail. The first be determined by the position
of the last queued vehicle on
step in queue length determination has already been dealt with by setting up the the link plus the length of that
rules for determining the conditions that indicate when a vehicle is in a queue. vehicle.
The next step is to determine the position of the last vehicle in the queue.
The BOQ on any step is a relatively simple thing to determine. The trick is to
figure out how to accumulate the individual BOQ measures over the entire
period. Several measures can be produced.
1. The maximum BOQ at some percentile value—for example, 95%;
2. The maximum BOQ on any cycle at some percentile value—for example,
95%;
3. The historical maximum BOQ (i.e., the longest queue recorded during the
period);
4. The probability that a queue will back up beyond a specified point; and
5. The proportion of time that the queue will be backed up beyond a
specified point.
Some of these measures are illustrated later in an example.
Computational Procedures for Density-Related Measures

The uninterrupted-flow procedures described in the HCM base their LOS
estimates on the density of traffic in terms of passenger cars per mile per lane
(pc/mi/ln). In one case (freeway merges and diverges), the density is estimated
only for the two lanes adjacent to the ramp.
Density computations do not require a detailed analysis of the trajectory of
each vehicle. They are best made by simply counting the number of vehicles in
each lane on a given segment, recognizing that the results represent actual
vehicles and not passenger cars.
For comparable results, the simulated densities must be converted to
pc/mi/ln, especially if simulation tools are used to evaluate the LOS on a
segment. Because the effect of heavy vehicles on the flow of traffic is treated
microscopically, there is no notion of passenger car equivalence in simulation
modeling. In addition, traffic flow models may differ among the various
simulation tools in their detailed treatment of heavy vehicles. Therefore, a simple
conversion process that will ensure full compatibility with the HCM’s LOS
estimation procedures cannot be prescribed. One possible method for developing
passenger car equivalence conversion factors involves multiple simulation runs:

1. Use the known demand flow rates, v, and truck proportions to obtain the
resulting segment density in vehicles per mile per lane (veh/mi/ln), d1.
2. Use the known demand flow rates, v, with passenger cars only to obtain
the resulting segment density in veh/mi/ln, d2.
3. Determine the heavy vehicle equivalence factor as fHV = d2/d1.
4. Set the demand flow rates to v/fHV with passenger cars only to obtain the
resulting segment density in pc/mi/ln.
This process is more precise because it adheres to the definition of passenger
car equivalence. Unfortunately, it is too complicated to be of much practical
value. However, two methods could produce a more practical approximation.
Both require determining the heavy vehicle adjustment factor, fHV, by the method
prescribed in Chapter 12 for basic freeway segments. This method is also
referenced and used in the procedural chapters covering other types of freeway
segments. The simplest approximation may be obtained by running the
simulation with known demand flow rates and truck proportions and then
dividing the simulated density by fHV. Another approximation involves dividing
the demand flow rates by fHV before running the simulation with passenger cars
only. The resulting densities are then expressed in pc/mi/ln. The second method
conforms better to the procedures prescribed in Chapters 11 to 13, but the first
method is probably easier to apply.
Follower density is an emerging density-based measure for two-lane highways
(12, 13). It is defined as the number of followers per mile per lane. A vehicle can
be classified as following when
• The gap between the rear and the front ends of the leading and following
vehicles, respectively, are shorter than or equal to 3 s; and
• The speed of the following vehicle is not more than 12 mi/h lower than
that of the preceding vehicle.
The follower density can be derived from point measurements by means of
the following formula:
Equation 36-1 Follower density = % followers × flow rate / time mean speed
This performance measure can be computed by the procedures in Chapter 15
and can easily be computed by vehicle trajectory analysis.

Analysis of a Signalized Approach

The simple approach to a signalized intersection (Exhibit 36-20) is now
converted to a two-lane approach with a length of 2,000 ft. A 10-min (600-s)
analysis period is used. The cycle length is 60 s, giving 10 cycles for inspection.
The analysis period would normally be longer, but 10 min is adequate for
demonstration purposes.
Trajectory Plots
The trajectory plot for the first few cycles is shown in Exhibit 36-38. The
vehicle track selected for later analysis is also shown in this exhibit.
Exhibit 36-38
Trajectories for Several Cycles
on a Signalized Approach
Two individual trajectory analysis plots are shown in Exhibit 36-39. The first
plot shows the trajectories of two vehicles where the progress of the subject
vehicle is constrained by its leader. The second plot shows the speed and
acceleration profiles for the subject vehicle.
50 50
2,000
45 45
Exhibit 36-39
40 40 Example Trajectory Analysis
1,500
35 35 Plots
Acceleration (ft/s )
30 30
2
Speed (ft/s)
Distance (ft)
25 25
20 20
1,000
Leader 15 15
10 10
Subject vehicle 5 5
500 0 0
-5 -5
-10 -10
-15 -15
0
510 520 530 540 550 560 570 580 590 600 610
500 510 520 530 540 550 560 570 580 590 600 610 620
Time (s)
Time (s) Speed Acceleration
(a) Subject Vehicle and Leader Vehicle Trajectories (b) Speed and Acceleration Profile
of Subject Vehicle

Analysis of Stops
An example of the analysis of a single vehicle selected from the entire
trajectory plot is shown in Exhibit 36-40. With the definition of a partial stop
based on the NCHRP 03-85 kinetic energy loss concept, the total stop value was
1.81 because the second stop was made from a lower speed.
Exhibit 36-40 1,200

Analysis of a Full and a Partial Selected vehicle
Stop
1,000
800
Distance (ft)
Stop line
600
400
200
0
500 520 540 560 580 600 620 640 660 680 700 720
Time (s)
(a) Vehicle Trajectories

50
45
40
35
30
Speed (ft/s)
25
20
15
10
0
530 535 540 545 550 555 560 565 570 575 580
Time (s)
Vehicle speed Maximum speed since stop
(b) Selected Vehicle Speed
Segment delay Queue delay Stop delay Number of stops

34.64 s 33.23 s 20 s 1.81
(c) Performance Measures for Selected Vehicle

Queuing Analysis
Exhibit 36-41(a) illustrates the queue length (BOQ) per step for one lane of
the signalized approach over all the time steps in the period. The 10 cycles are
discernible in this figure. Also, a considerable variation in the cyclical maximum
BOQ is evident.
The percentile instantaneous BOQ and the percentile maximum BOQ per
cycle should be distinguished. For the instantaneous BOQ, the individual
observation is the BOQ on any step, so the sample size is the number of steps
covered (600 in this case). For cyclical maximum BOQ, the individual
observation is the maximum BOQ in any cycle, so the sample size is the number
of cycles (10 in this case). The maximum BOQ in any cycle can be determined
only by inspecting the plotted instantaneous values. No procedure is proposed
here for automatic extraction of the maximum cyclical BOQ from the
instantaneous BOQ data.
A statistical analysis showing the average BOQ, the 95th percentile BOQ
(based on 2 standard deviations past the average value), and the historical
maximum BOQ is presented in Exhibit 36-41(b). One important question is
whether the 95% BOQ can be represented statistically on the basis of the
standard deviation, assuming a normal distribution. The BOQ histogram
showing the distribution of instantaneous BOQ for the 600 observations is shown
in Exhibit 36-42. The appearance of this histogram does not suggest any
analytical distribution; however, the relationship between the 95% BOQ and the
historical maximum appears to be reasonable for this example.
500 Exhibit 36-41

BOQ Analysis by Time Step
450
400
350
Back of Queue (ft)
300
250
200
150
100
50
0
0 60 120 180 240 300 360 420 480 540 600
Time (s)
(a) BOQ Plot
Average queue Standard deviation 95th percentile queue Maximum queue

174 ft 110 ft 395 ft 440 ft
(b) Queue-Related Performance Measures

Exhibit 36-42
BOQ Histogram
The queue length on an isolated approach that is close to saturation will have
a near uniform distribution (i.e., equal probability of all lengths between zero
and the maximum). The standard deviation of a uniform distribution is greater
than one-half of the mean, so the 95th percentile estimator (mean value plus 2
standard deviations) will be greater than the maximum value. This situation
raises some doubt about the validity of basing the 95th percentile BOQ on the
standard deviation, especially with cyclical queuing.
Delay Analysis for a Single Trajectory

A comparison of the accumulated delay by all definitions for the selected
vehicle track indicated in Exhibit 36-38 is presented in Exhibit 36-43(a). The
relationships between segment delay, queue delay, and stopped delay are
evident in this figure. The segment delay begins to accumulate before the vehicle
approaches the intersection because of midsegment interactions that reduce the
speed below the target speed. The queue delay begins to accumulate as the
vehicle enters the queue, and the stopped delay begins to accumulate a few
seconds later. The stopped delay ceases to accumulate as soon as the vehicle
starts to move, but the queue delay continues to accumulate until the vehicle
leaves the link.
The time step delay analysis plots shown in Exhibit 36-43(b), based on 1-s
time steps, provide additional insight into the operation. The time step delay is
close to zero as the vehicle enters the segment, indicating that the speed is close
to the target speed. Small delays begin to accumulate in advance of the
intersection. The accumulation becomes more rapid when the vehicle enters the
queue. The periods when the vehicle is in the stopped and the queued state are
also shown in this figure.

25 Exhibit 36-43
Accumulated Delay by Various
Definitions
20
Accumulated Delay (s)
15
10
0
140 145 150 155 160 165 170 175 180 185 190 195 200
Elapsed Simulation Time (s)
Segment delay Queue delay Stopped delay
(a) Accumulated Delay
(b) Time Step Delay
As was indicated previously, the value of control delay cannot be

determined by simulation in a manner that is comparable with the procedures
prescribed in Chapters 19 and 31. Because this segment terminates at a signal, it
is suggested that the queue delay would provide a reasonable estimate of control
delay because the queue delay offers a close approximation to the delay that
would be measured in the field.

Delay Analysis for All Vehicles on the Segment

The preceding example dealt with accumulated delay of a single vehicle
traversing the segment. A useful delay measure requires the accumulation of
delay to all vehicles traversing the segment during the period. An example is
shown in Exhibit 36-44. In keeping with the recommendations offered elsewhere
(14), only vehicles that traversed the entire link during the period are included in
this analysis. Therefore, the number of vehicles analyzed (210) is lower than the
number of vehicles that were actually on the link during the period (286).
Exhibit 36-44 Segment Delay (s) Queue Delay (s) Stop Delay (s) No. of Stops
Delay Analysis for All Vehicles Lane 1 3,128 2,562 1,957 95.4
on a Segment Lane 2 3,400 2,793 2,047 96.2
Total 6,529 5,355 4,004 191.6
Average per vehicle 31.09 25.50 19.07 0.91
Analysis of a Freeway Segment

A performance analysis of the freeway weaving area originally shown in
Exhibit 36-27 is presented here. A single vehicle is selected from the trajectory
plot and its trajectory is analyzed. The results are shown in Exhibit 36-45. The
analysis produced segment delay and queue delay. This segment was very
congested, as indicated by the trajectory plot. No stopped delay was produced
because the vehicle never actually came to a stop (i.e., its speed stayed above 5
mi/h).
Exhibit 36-45 3,500

Longitudinal Analysis of Delay
for a Selected Vehicle in a
3,000
Weaving Area
Subject
vehicle
2,500
Distance (ft)
2,000
1,500
1,000
500
0
500 510 520 530 540 550 560 570 580 590 600
Time (s)
(a) Vehicle Trajectories
Segment delay Queue delay Stopped delay

39.58 s 37.01 s 0s
(b) Delay-Related Performance Measures for Subject Vehicle

A spatial analysis of the entire segment can also be performed to produce the
following measures by lane:
• Average density over the segment,
• Percent slow vehicles (i.e., traveling at less than two-thirds the target
speed),
• Percent queued vehicles,
• Average queue length (measured from front of queue to BOQ),
• Average BOQ position,
• Maximum BOQ position, and
• Percent of time steps when the queue overflowed the segment.
The results are presented in tabular form in Exhibit 36-46. The values are
presented by lane, and the exhibit note presents combined density values for
Lanes 1 and 2 for compatibility with the HCM definition of merge area density.
Acceleration Exhibit 36-46

Lane 1 Lane 2 Lane 3 Lane Example Spatial Analysis by
Average density (veh/mi/ln) 73.4 51.0 43.6 9.9 Lane
Percent slow vehicles (%) 88.4 68.5 41.5 65.7
Percent queued vehicles (%) 63.4 22.0 2.4 26.7
Average queue length (ft) 600 215 15 40
Average back of queue (ft) 1,471 1,119 135 562
Maximum back of queue (ft) 1,497 1,497 1,492 1,474
Percent overflow 66.1 29.6 0.5 0.17
Note: Average Lane 1 and Lane 2 density is 62.2 veh/mi/ln.

6. REFERENCES
Some of these references can 1. Tufte, E. R. The Visual Display of Quantitative Information. Graphics Press,
Reference Library in Volume 4. Cheshire, Conn., 1983.
3. National Institute of Statistics and Sematech. E-Handbook of Statistical
Methods. http://www.itl.nist.gov/div898/handbook/index.htm. Accessed
March 5, 2021.
4. Greenwood, J., and M. Sandomire. Sample Size Required for Estimating the
Standard Deviation as a Percent of Its True Value. Journal of the American
Statistical Association, Vol. 45, No. 250, June 1950, pp. 257–260.
5. Federal Highway Administration. National Performance Management
Research Data Set (NPMRDS) Technical Frequently Asked Questions.
http://www.ops.fhwa.dot.gov/freight/freight_analysis/perform_meas/vpds/n
pmrdsfaqs.htm. Accessed April 24, 2015.
6. Turner, S. Quality Control Procedures for Archived Operations Traffic Data:
Synthesis of Practice and Recommendations. Final Report, Contract DTFH61-97-
C-00010. Federal Highway Administration, Washington, D.C., March 2007.
7. INRIX and I-95 Corridor Coalition. I-95 Vehicle Probe Data website.
http://www.i95coalition.org/i95/VehicleProbe/tabid/219/Default.aspx.
Accessed Aug. 10, 2012.
8. California Department of Transportation. California Performance
Measurement System (PeMS) website. http://pems.dot.ca.gov/. Accessed
Aug. 10, 2012.
9. INRIX. Traffic Scorecard Methodology website.
http://www.inrix.com/scorecard/methodology.asp. Accessed Aug. 10, 2012.
10. Kittelson & Associates, Inc. Comparison of Freeway Travel Time Index and Other
Travel Time Reliability Measures. Florida Department of Transportation,
Tallahassee, May 2012.
11. Courage, K. G., S. Washburn, L. Elefteriadou, and D. Nam. Guidance for the
Use of Alternative Traffic Analysis Tools in Highway Capacity Analyses. National
Cooperative Highway Research Program Project 03-85 Final Report.
University of Florida, Gainesville, 2010.
12. Van As, S. C., and A. Van Niekerk. The Operational Analysis of Two-Lane
Rural Highways. Presented at 23rd Annual Southern African Transport
Conference, Pretoria, South Africa, July 2004.
13. Catbagan, J. L., and H. Nakamura. Probability-Based Follower Identification
in Two-Lane Highways. Presented at 88th Annual Meeting of the
Transportation Research Board, Washington, D.C., 2009.
References Chapter 36/Concepts: Supplemental

14. Dowling, R. Traffic Analysis Toolbox Volume VI: Definition, Interpretation, and
Calculation of Traffic Analysis Tools Measures of Effectiveness. Report FHWA-
HOP-08-054. Federal Highway Administration, Washington, D.C., Jan. 2007.
Chapter 36/Concepts: Supplemental References

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 37
ATDM: SUPPLEMENTAL
CONTENTS
1. INTRODUCTION.................................................................................................. 37-1
2. TYPES OF ATDM STRATEGIES ....................................................................... 37-2

Overview .............................................................................................................. 37-2
Roadway Metering .............................................................................................. 37-2
Congestion Pricing .............................................................................................. 37-3
Traveler Information Systems ............................................................................ 37-4
Managed Lanes .................................................................................................... 37-5
Speed Harmonization ......................................................................................... 37-6
Traffic Signal Control .......................................................................................... 37-7
Dynamic Lane Grouping .................................................................................... 37-7
Reversible Center Lanes ..................................................................................... 37-9
Specialized Applications of ATDM Strategies................................................. 37-9
3. EFFECTS OF SHOULDER AND MEDIAN LANE STRATEGIES ............. 37-11

Open Shoulders as Auxiliary Lanes Between Adjacent On- and
Off-Ramps ................................................................................................... 37-11
Open Shoulders to Buses Only ........................................................................ 37-11
Open Shoulders to HOVs Only ....................................................................... 37-12
Open Right Shoulders to All Traffic ............................................................... 37-12
Open Median Shoulder to Buses Only ........................................................... 37-12
Open Median Shoulder to HOVs Only .......................................................... 37-12
Open Median Shoulder to All Traffic ............................................................. 37-12
4. EFFECTS OF RAMP-METERING STRATEGIES .......................................... 37-13

Capacity of Ramp-Metered Merge Sections................................................... 37-13
Locally Dynamic Ramp Metering ................................................................... 37-13
5. EFFECTS OF ADAPTIVE SIGNALS................................................................ 37-14
6. EFFECTS OF DYNAMIC LANE GROUPING ................................................ 37-16
7. EFFECTS OF REVERSIBLE CENTER LANES................................................ 37-18
Chapter 37/ATDM: Supplemental Contents

8. PLANNING AN ATDM PROGRAM ...............................................................37-20

Travel Demand Management Plans ................................................................37-20
Weather-Responsive Traffic Management Plans ..........................................37-21
Traffic Incident Management Plans ................................................................37-22
Work Zone Transportation Management Plans ............................................37-24
Special Event Management Plans ....................................................................37-27
9. REFERENCES .......................................................................................................37-28
Contents Chapter 37/ATDM: Supplemental

LIST OF EXHIBITS
Exhibit 37-1 Freeway Ramp Metering, SR-94, Lemon Grove, California ............ 37-2
Exhibit 37-2 Minnesota Dynamic Pricing for HOT Lanes ..................................... 37-3
Exhibit 37-3 San Francisco Bay Area Traffic Map .................................................. 37-4
Exhibit 37-4 HOV Lane .............................................................................................. 37-5
Exhibit 37-5 Variable Speed Limit Signs, Rotterdam, The Netherlands ............. 37-6
Exhibit 37-6 Illustrative Use of Dynamic Lane Grouping to Reduce
Left-Turn Congestion .......................................................................................... 37-8
Exhibit 37-7 Upstream DLG Signage in Maryland................................................. 37-8
Exhibit 37-8 Reversible Center Lanes in Utah ......................................................... 37-9
Exhibit 37-9 Illustrative Adaptive Signal Effects on Daily
Traffic Operations .............................................................................................. 37-15
Exhibit 37-10 Illustrative RCL Delay Reductions by Scenario ............................ 37-19
Exhibit 37-11 Possible Incident Management Strategies ..................................... 37-24
Chapter 37/ATDM: Supplemental Contents

1. INTRODUCTION
Chapter 37 presents additional information about the following aspects of VOLUME 4: APPLICATIONS
GUIDE
active traffic and demand management (ATDM): 25. Freeway Facilities:
Supplemental
• An overview of typical ATDM strategies for managing demand, capacity, 26. Freeway and Highway
and the performance of the highway and street system; Segments: Supplemental
• Guidance on analyzing shoulder lane, median lane, ramp metering, Supplemental
28. Freeway Merges and
adaptive signal control, dynamic lane grouping, and reversible center lane Diverges: Supplemental
strategies using the Highway Capacity Manual (HCM); and 29. Urban Street Facilities:
Supplemental
• Guidance on designing an ATDM program. 30. Urban Street Segments:
Supplemental
Chapter 11, Freeway Reliability Analysis, and Chapter 17, Urban Street 31. Signalized Intersections:
Supplemental
Reliability and ATDM, provide methods for analyzing the effects of ATDM 32. STOP-Controlled
strategies on freeway and urban street operations, respectively. Intersections:
Supplemental
33. Roundabouts:
Supplemental
Supplemental
Chapter 37/ATDM: Supplemental Introduction

2. TYPES OF ATDM STRATEGIES
OVERVIEW
More in-depth and up-to-date This section provides brief overviews of typical ATDM strategies for
information on ATDM
strategies is available at the managing demand, capacity, and the performance of the highway and street
Federal Highway system. The strategies described here are intended to be illustrative rather than
Administration’s website:
http://www.ops.fhwa.dot.gov/ definitive. ATDM strategies constantly evolve as technology advances.
atdm.
ROADWAY METERING
Roadway metering treatments store surges in demand at various points in
the transportation network. Typical examples of roadway metering include
freeway on-ramp metering, freeway-to-freeway ramp metering, freeway
mainline metering, peak period freeway ramp closures, and arterial signal
metering. Exhibit 37-1 illustrates a freeway ramp-metering application.
Exhibit 37-1
Freeway Ramp Metering,
SR-94, Lemon Grove,
California
Source: FHWA (1 ).
X
Roadway metering may be highly dynamic or comparatively static. A

comparatively static roadway metering system would establish some preset
metering rates on the basis of historical demand data, periodically monitor
system performance, and adjust the rates to obtain satisfactory facility
performance. A static metering system, unlike a dynamic system, would not
generally be considered an ATDM strategy. A highly dynamic system may
monitor system performance on a real-time basis and automatically adjust
metering rates by using a predetermined algorithm in response to changes in
observed facility conditions. Preferential treatment of high-occupancy vehicles
(HOVs) may be part of a roadway metering strategy.
Roadway metering may be applied on freeways or arterials. On arterials,
metering might be accomplished through “gating,” in which an upstream signal
is used to control the number of vehicles reaching downstream signals. Surges in
Types of ATDM Strategies Chapter 37/ATDM: Supplemental

demand are temporarily stored at the upstream signal and released later when
the downstream signals can better serve the vehicles.
CONGESTION PRICING
Congestion or value pricing is the practice of charging tolls for the use of all The objective of congestion
pricing is to preserve reliable
or part of a facility or a central area according to the severity of congestion. The operating speeds on the tolled
tolls may vary by distance traveled, vehicle class, and estimated time savings. facility.
The objective of congestion pricing is to preserve reliable operating speeds on the

tolled facility with a tolling system that encourages drivers to switch to other
times of the day, other modes, or other facilities when demand starts to approach
facility capacity. Exhibit 37-2 shows an example of congestion pricing in
Minnesota.
Exhibit 37-2
Minnesota Dynamic Pricing for
HOT Lanes
Source: FHWA (2 ) (courtesy of Minnesota Department of Transportation).
Congestion pricing may use different degrees of responsiveness and

automation. Some implementations may use a preset schedule under which the
toll varies by the same amount for preset times during the day and week. The
implementation may be monitored on a regular schedule and the pricing
adjusted to achieve or maintain desired facility performance. An ATDM-based
implementation of congestion pricing may monitor facility performance much
more frequently and use automatic or semiautomatic dynamic pricing to vary
the toll on the basis of a predetermined algorithm according to the observed
performance of the facility.
Chapter 37/ATDM: Supplemental Types of ATDM Strategies

High-occupancy toll (HOT) lanes (also called express lanes) are tolled lanes
adjacent to general purpose lanes. HOT lanes allow motorists to pay tolls to enter
the lanes to avoid congested nontoll lanes. HOVs may be allowed to enter the
lanes for free or at a reduced toll rate.
Central area pricing is an Central area pricing and dynamic parking pricing are examples of an
areawide implementation of
congestion pricing. areawide implementation of congestion pricing. Central area pricing imposes
tolls on vehicles entering or traveling within a central area street network during
certain hours of certain days. The fee varies by time of day and day of week or
according to real-time measurements of congestion within the central area. The
toll may be reduced or waived for certain vehicle types, such as HOVs, or for
residents of the zone.
TRAVELER INFORMATION SYSTEMS

Traveler information is an integration of technologies allowing the general
public to access real-time or near-real-time data on incident conditions, travel
time, speed, and possibly other information. Traveler information enhances
awareness of current and anticipated traffic conditions on the transportation
system. Traveler information may be tailored to one or more specific modes of
travel, such as auto, truck, bus, bicycle, or pedestrian.
Traveler information can be grouped into three types (pretrip, in vehicle, and
roadside) according to when the information is made available and how it is
delivered to the driver.
Pretrip information is obtained from various sources and transmitted to
motorists before the start of their trip through various means. Exhibit 37-3
illustrates Internet-based dissemination of travel information.
Exhibit 37-3
San Francisco Bay Area
Traffic Map
Source: © 2009 Metropolitan Transportation Commission (http://traffic.511.org).

In-vehicle information may involve route guidance or dissemination of

incident and travel time conditions to the en route vehicle. Route guidance
involves Global Positioning System–based real-time data acquisition to calculate
the most efficient routes for drivers. This technology allows individual vehicles
and their occupants to receive optimal route guidance via in-dash displays or
portable devices. A number of third-party vendors offer smartphone apps that
provide real-time travel time and route guidance information to drivers,
including re-routing information to avoid areas of high congestion. Some state
departments of transportation and metropolitan planning organizations have
agreements with such vendors to better provide real-time information to the
traveling public.
Roadside messages consist of dynamic message signs (also called changeable
or variable message signs) and highway advisory radio (also called traveler
advisory radio) that display or transmit information on road conditions for
travelers while they are en route.
MANAGED LANES
Managed lanes include reversible lanes, HOV lanes, HOT lanes, truck lanes,
bus lanes, speed harmonization, temporary closures for incidents or maintenance,
and temporary use of shoulders during peak periods (see Exhibit 37-4). HOT
lanes are described above under congestion pricing, and speed harmonization is
described in the next section.
HOV lanes assign a portion of the roadway capacity to vehicles that carry the
most people on the facility or that in some other way meet societal objectives for
reducing the environmental impacts of vehicular travel. HOV lanes may operate
24 hours a day, 7 days a week, or they may be limited to the peak periods when
demand is greatest. The minimum vehicle-occupancy requirement for the HOV
lanes may be adjusted in response to operating conditions to preserve
uncongested HOV lane operation.
Exhibit 37-4
HOV Lane
Source: FHWA (3 ).

Reversible lanes provide additional capacity for directional peak flows

depending on the time of the day. Reversible lanes on freeways may be located
in the center of a freeway with gate control on both ends. On interrupted-flow
facilities, reversible lanes may be implemented through lane-use control signals
and signs that open and close lanes by direction.
The temporary use of shoulders during peak periods by all or a subset of
vehicle types can provide additional capacity in a bottleneck section and improve
overall facility performance. Part-time shoulder use by buses in queuing
locations can substantially reduce bus delays by enabling them to proceed along
the roadway without having to wait in the mainline queue.
SPEED HARMONIZATION
The objective of speed harmonization is to improve safety and facility
operations by reducing the shock waves that typically occur when traffic
abruptly slows upstream of a bottleneck or for an incident. The reduction of
shock waves decreases the probability of secondary incidents and reduces the
loss of capacity associated with incident-related and recurring traffic congestion.
Changeable speed limit or speed advisory signs are typically used to
implement speed harmonization. Exhibit 37-5 shows an example of variable
speed limit signs used for speed harmonization in the Netherlands. The speed
restrictions may apply uniformly across all lanes or may vary by lane. The same
lane signs may be used to close individual lanes upstream of an incident until the
incident is cleared (this practice is not strictly speed harmonization).
The variable speed limit may be advisory or regulatory. Advisory speeds
indicate a recommended speed, which drivers may exceed if they believe doing
so is safe under prevailing conditions. Regulatory speed limits may not be
exceeded under any conditions.
Exhibit 37-5
Variable Speed Limit Signs,
Rotterdam, The Netherlands
Source: FHWA Active Traffic Management Scan, Jessie Yung.

TRAFFIC SIGNAL CONTROL

Signal timing optimization is the single most cost-effective action that can be
taken to improve a roadway corridor’s capacity and performance (4). Signal
timing is as important as the number of lanes in determining the capacity and
performance of an urban street.
Traffic signal timing optimization and coordination minimize the stops,
delay, and queues for vehicles at individual and multiple signalized
intersections.
Traffic signal preemption and priority provide special timing for certain
classes of vehicles (e.g., buses, light rail vehicles, emergency response vehicles,
and railroad trains) using the intersection. Preemption interrupts the regular
signal operation. Priority either extends or advances the time when a priority
vehicle obtains the green phase, but generally the priority is within the constraints
of the regular signal-operating scheme.
Traffic-responsive operation and adaptive control provide different levels of
automation when adjusting signal timing in response to variations in demand.
Traffic-responsive operation selects from a prepared set of timing plans on the
basis of the observed level of traffic in the system. Adaptive traffic signal control
involves detecting traffic in advance, predicting its arrival at the downstream
signal, and adjusting the downstream signal operation based on that prediction.
DYNAMIC LANE GROUPING

Dynamic lane grouping (DLG) involves the dynamic changing of allowable
turning movements in each lane; also known as lane assignments, or lane
channelization (5). For example, consider an intersection approach with three
through lanes and one left-turn lane. When the intersection experiences heavy
left-turn demands with relatively low through demand, dynamically converting
a through lane into a left-turn lane would better align capacity supply and traffic
demand across all lanes, leading to an overall reduction in delay. Volume demands
at intersections can vary greatly between weekdays (6), resulting in significant
operational effects (7) similar to those caused by site-to-site variation (8). DLG
allows a better alignment of roadway capacities with time-varying traffic demands.
Exhibit 37-6 illustrates an example of left-turn congestion before DLG, and
improved flow after the eastbound lane grouping is changed as a result of DLG.
DLG can be implemented in the field by installing dynamic message signs
(DMSs) that can change their lane assignment displays in real time (9). Black-
and-white regulatory lane-use signs are essential to convey lane use, both at and
in advance of the intersection, when the lane configuration is not defined by
default rules of the road. The advance notification is especially critical in a
dynamic situation where drivers may not expect a change. The use of traffic
signal indications can supplement but not replace the regulatory signs. Exhibit
37-7 illustrates an example of upstream signage for DLG. In this case, an
overhead DMS is helpful because it sits directly above the affected lane.
Regardless of which lane is dynamically controlled, similar DLG signage is
needed at the intersection stop line to reinforce the message.

Exhibit 37-6
Illustrative Use of Dynamic
Lane Grouping to Reduce
Left-Turn Congestion
(a) Eastbound left-turn congestion prior to dynamic lane grouping
(b) Eastbound left-turn operation after dynamic lane grouping
Source: Derived from Hale et al. (9 ).
Exhibit 37-7
Upstream DLG Signage in
Maryland
Source: Courtesy of David Hale.

REVERSIBLE CENTER LANES

Reversible center lanes (RCLs) dynamically switch the direction of one or
more travel lanes on roadways with highly directional peaking characteristics, to
better match roadway capacity with demand. Exhibit 37-8 shows an implementation
with overhead DMSs. In Washington, DC, despite issues with lane violations and
higher crash rates than on other corridors (thought to be related to a prohibition on
the use of overhead signs and mast arms in key view corridors in the city), RCLs
have shown potential for improving traffic conditions (10). A study in Utah
described benefits that included increased peak-direction peak-period
throughput, and similar or increased throughput in the off-peak direction (11).
Simulation studies have been used to integrate the design and operation of RCLs
on arterial roadways (12).
Exhibit 37-8
Reversible Center Lanes in
Utah
Source: Avenue Consultants (11).
SPECIALIZED APPLICATIONS OF ATDM STRATEGIES

ATDM strategies are often applied to the day-to-day operation of a facility.
Incident management and work zone management are example applications of
one or more ATDM strategies to address specific facility conditions. Employer-
based demand management is an example of private-sector applications in
which traveler information systems may be an important component.
Incident Management
Traffic incident management (TIM) is “the coordinated, preplanned use of
technology, processes, and procedures to reduce the duration and impact of
incidents, and to improve the safety of motorists, crash victims and incident
responders” (13 ). An incident is “any non-recurring event that causes a reduction
X
in capacity or an abnormal increase in traffic demand that disrupts the normal

operation of the transportation system” (13 ). Such events include traffic crashes,
X
disabled vehicles, spilled cargo, severe weather, and special events such as
sporting events and concerts. ATDM strategies may be included as part of an
overall incident management plan to improve facility operations during and
after incidents.

Work Zone Management

Work zone management has the objective of moving traffic through the
working area with as little delay as possible consistent with the safety of the
workers, the safety of the traveling public, and the requirements of the work
being performed. Transportation management plans are a collection of
administrative, procedural, and operational strategies used to manage and
mitigate the impacts of a work zone project. The plan may have three
components: a temporary traffic control plan, a transportation operations plan,
and a public information plan. The temporary traffic control plan describes the
control strategies, traffic control devices, and project coordination. The
transportation operations plan identifies the demand management, corridor
management, work zone safety management, and the traffic or incident
management and enforcement strategies. The public information plan describes
the public awareness and motorist information strategies (14). ATDM strategies
X14
can be important components of a transportation management plan.
Employer-Based Demand Management

Employer-based demand management consists of cooperative actions taken
by employers to reduce the impacts of recurring or nonrecurring traffic
congestion on employee productivity. For example, a large employer may
implement work-at-home or stay-at-home days in response to announced snow
days; “spare the air” days; or traffic alerts concerning major construction
projects, incidents, and highway facility closures. Another company may
contract for or directly provide regular shuttle van service to and from transit
stations. Flexible or staggered work hours may be implemented to enable
employees to avoid peak commute hours. Rideshare-matching services and
incentives may be implemented by the employer to facilitate employee
ridesharing.
Employers may use components of a traveler information system to
determine appropriate responses to changing traffic conditions. Employees can
use traveler information systems in their daily commuting choices.

3. EFFECTS OF SHOULDER AND MEDIAN LANE STRATEGIES
This section provides details on the free-flow speed and capacity

adjustments associated with temporary shoulder and median lane strategies.
OPEN SHOULDERS AS AUXILIARY LANES BETWEEN ADJACENT

ON- AND OFF-RAMPS
This strategy involves opening a shoulder lane for use by all vehicles
entering at the upstream on-ramp or exiting at the downstream off-ramp. Some
through vehicles may temporarily use the auxiliary lane to try to jump ahead of
the queue.
The capacity of an auxiliary lane is assumed by the Chapter 10 freeway
facilities method to be the same as that of a regular lane; however, utilization of
the auxiliary lane may be lower than that of a through lane. In addition, the
freeway method does not provide a capacity for shoulder lanes. Until the HCM
has specific information on the capacities of auxiliary shoulder lanes, this
procedure assumes that the capacity of an auxiliary shoulder lane is one-half that
of a normal freeway through lane.
Because the freeway facilities method does not recognize individual lane
capacities, computation of an average capacity for freeway sections with
auxiliary shoulder lanes across all lanes is necessary.
𝐶𝑎𝑝𝑆ℎ𝑙𝑑𝑟(𝑠) + 𝐶𝑎𝑝𝑀𝐹𝑙𝑎𝑛𝑒𝑠(𝑠) × 𝑀𝐹𝑙𝑎𝑛𝑒𝑠(𝑠)
𝐴𝑣𝑒𝐶𝑎𝑝(𝑠) = Equation 37-1
1 + 𝑀𝐹𝑙𝑎𝑛𝑒𝑠(𝑠)
where
AveCap(s) = average capacity per lane for section s (veh/h/ln),
CapShldr(s) = capacity per shoulder lane for section s (veh/h/ln),
CapMFlanes(s) = capacity per mixed-flow lane in section s (veh/h/ln), and
MFlanes(s) = number of mixed-flow lanes in section s (integer).
The number of lanes on the freeway segments between adjacent on- and off-
ramps is increased by one for the shoulder lane.
Until the HCM has more specific information for shoulder lanes, free-flow
speeds on auxiliary shoulder lanes are assumed in this procedure to be the same
as for regular through lanes.
OPEN SHOULDERS TO BUSES ONLY

This strategy involves opening a shoulder lane to buses only. The same
procedure and assumptions as described above for auxiliary shoulder lanes are
used to compute freeway section capacities, lanes, and free-flow speeds where
buses are allowed on shoulders, with the following exception: the capacity of the
shoulder lane is the number of buses per hour using the shoulder lane or the
user-specified capacity, whichever is less (the user can override the default
capacity).
Chapter 37/ATDM: Supplemental Effects of Shoulder and Median Lane Strategies

OPEN SHOULDERS TO HOVs ONLY

This strategy involves opening a shoulder lane to buses, vanpools, and
carpools (HOVs) only. The same procedure and assumptions as described above
for auxiliary shoulder lanes are used to compute freeway section capacities,
lanes, and free-flow speeds where HOVs are allowed on shoulders, with the
following exception: the capacity of the shoulder lane is the number of HOVs per
hour using the shoulder lane or the user-specified capacity, whichever is less.
OPEN RIGHT SHOULDERS TO ALL TRAFFIC

This strategy involves opening a shoulder lane to all vehicles. The same
used to compute freeway section capacities, lanes, and free-flow speeds where all
vehicles are allowed on shoulders, with the following exception: the capacity of
the shoulder lane is as specified by the user.
OPEN MEDIAN SHOULDER TO BUSES ONLY

This strategy involves opening a median lane to buses only. The same
used to compute freeway section capacities, lanes, and free-flow speeds, with the
following exception: the capacity of the median lane is the number of buses per
hour using the shoulder lane or the user-designated capacity, whichever is less.
OPEN MEDIAN SHOULDER TO HOVs ONLY

This strategy involves opening a median lane to HOVs (buses, vanpools,
carpools) only. The same procedure and assumptions as described above for
auxiliary shoulder lanes are used to compute freeway section capacities, lanes,
and free-flow speeds, with the following exception: the capacity of the median
lane is the number of HOVs per hour using the shoulder lane or the user-
designated capacity, whichever is less.
OPEN MEDIAN SHOULDER TO ALL TRAFFIC

This strategy involves opening a median lane to all traffic. The same
used to compute freeway section capacities, lanes, and free-flow speeds, with the
following exception: the capacity of the median lane is as designated by the user.
Effects of Shoulder and Median Lane Strategies Chapter 37/ATDM: Supplemental

4. EFFECTS OF RAMP-METERING STRATEGIES
This section provides details on the capacity adjustments associated with

ramp-metering strategies.
CAPACITY OF RAMP-METERED MERGE SECTIONS

A capacity adjustment factor of 1.03 is recommended to be applied to
freeway merge segments in the Chapter 10 freeway facilities method for those
times when ramp metering is in operation (15).
LOCALLY DYNAMIC RAMP METERING

For locally dynamic ramp metering, an adaptation of the ALINEA algorithm
(16) is used to estimate the ramp-metering rate for each analysis period for each
scenario:
(𝐶𝑀 − 𝑉𝑀(𝑡))
𝑅(𝑡) = Equation 37-2
𝑁𝑅
subject to
𝑀𝑖𝑛𝑅𝑎𝑡𝑒 < 𝑅(𝑡) < 𝑀𝑎𝑥𝑅𝑎𝑡𝑒
𝑉𝑅(𝑡) + 𝑄𝑅(𝑡 − 1) − 𝑄𝑅𝑆
𝑅(𝑡) >
𝑁𝑅
where
R(t) = ramp-metering rate for analysis period t (veh/h/ln),
NR = number of metered lanes on ramp (integer),
CM = capacity of downstream section (veh/h),
VM(t) = volume on upstream section for analysis period t (veh/h),
VR(t) = volume on ramp during analysis period t (veh/h),
QR(t – 1) = queue on ramp at end of previous analysis period t – 1 (veh),
QRS = queue storage capacity of ramp (veh),
MinRate = user-defined minimum ramp-metering rate (veh/h/ln) (default value
is 240 veh/h/ln), and
MaxRate = user-defined maximum ramp-metering rate (veh/h/ln) (default value
is 900 veh/h/ln).
Chapter 37/ATDM: Supplemental Effects of Ramp-Metering Strategies

5. EFFECTS OF ADAPTIVE SIGNALS
This section provides guidance on modeling adaptive signal control using

HCM methods and summarizes the results of a simulation-based evaluation of
adaptive signals.
MODELING ADAPTIVE SIGNAL CONTROL USING HCM METHODS

Because each vendor’s adaptive signal control system uses its own
proprietary algorithms, it has not been possible to develop a generalized method
adaptive signal control method for the HCM (17). The recommended analysis
approach is to use a simulation tool that can work with a proprietary application
programming interface (API) that replicates the algorithms used by the adaptive
control system being analyzed.
ILLUSTRATIVE CASE STUDY RESULTS

This section summarizes the results of an evaluation of the potential benefits
of adaptive signal control on urban street operations. Simulation models
calibrated to three real-world urban street facilities were used to evaluate the
benefit of adaptive signal control in terms of average delay, average number of
stops, travel speed, and travel time reliability. The simulation emulated what an
adaptive control logic might accomplish, but without using any specific
commercial algorithms. Days with rain and snow were included in some of the
simulated days, based on weather records for the metropolitan area in which the
urban street facilities are located.
Exhibit 37-9 summarizes the 24-hour effects of adaptive signal control on the
three facilities. “Moving speed” excludes stopped delays, while “average speed”
includes stopped delays. Although there were consistently positive benefits
according to a wide range of performance measures, the magnitudes of these
benefits were less consistent (e.g., delay reductions between 3% and 24%, travel
time index [TTI] reductions between 3% and 13%).
Effects of Adaptive Signals Chapter 37/ATDM: Supplemental

Moving Average Stopped Exhibit 37-9

Direction TTI PTI Speed Speed Delay Stop Time Vehicles Illustrative Adaptive Signal
Effects on Daily Traffic
Corridor #1
Operations
East -1.5% 1.0% 0.7% 5.4% -5.9% -5.7% -0.6%
West -33.2% -31.9% -0.4% 4.9% -42.9% -34.9% -1.3%
Major -8.6% -4.2% 0.2% 5.3% -13.2% -12.5% -0.9%
Minor -3.0% -2.8% 1.6% 5.1% -6.8% -8.0% -0.6%
All -3.4% -2.3% 0.7% 5.4% -5.9% -5.7% -0.6%
Corridor #2
East -17.0% -18.0% 11.9% 22.4% -32.7% -28.1% -1.2%
West -2.7% -5.8% 0.9% 1.3% -6.0% -3.4% -1.1%
Major -16.2% -17.9% 8.0% 13.5% -28.8% -24.5% -1.0%
Minor -11.9% -8.4% 1.8% 5.4% -21.7% -25.0% -0.6%
All -12.7% -9.3% 2.9% 7.0% -23.6% -24.8% -0.7%
Corridor #3
North -2.4% -4.7% 0.9% 1.0% -10.4% -7.7% 3.4%
South -43.5% -47.7% 26.3% 20.7% -30.6% -20.4% -2.7%
Major -37.8% -54.1% 16.0% 12.0% -29.5% -19.3% -0.9%
Minor -2.4% -3.5% 0.1% 1.2% -1.5% -1.9% -0.5%
All -4.7% -6.7% 1.1% 2.1% -3.0% -2.7% -0.5%
Note: TTI = travel time index, PTI = planning time index.
Results from the simulation indicated that adaptive signal benefits were
often more significant on days when rain occurred, a finding consistent with
research (18). Rain was found to exhibit the most variability in facility
performance across the model repetitions. In contrast, snow produced a much
tighter distribution of performance indicators across repetitions, because speeds
were generally very low for all vehicles. Therefore, although adaptive control did
not improve average performance under severe weather conditions, it did
improve performance reliability.
Chapter 37/ATDM: Supplemental Effects of Adaptive Signals

6. EFFECTS OF DYNAMIC LANE GROUPING
This section provides guidance on modeling DLG using HCM methods and
conditions under which DLG may provide operational benefits.
MODELING DLG USING HCM METHODS

The following inputs to the motorized vehicle signalized intersection
procedure described in Chapter 19 may need to be modified when comparing
intersection operations with DLG to operations without DLG:
• Number of lanes by movement
• Lane utilization adjustment factor
• Turn bay length
• Left-turn operational mode
• Right-turn operational mode
• Phase sequence
• Right-turn-on-red flow rate
• Signal timing parameters

This section summarizes the results of an evaluation of DLG on intersection
and urban street operations. HCM analysis methods were applied to a
hypothetical urban street facility. Tested parameters included the number of
through lanes (2 or 3) and a range of traffic demands that included both over-
and undersaturated conditions. Intersection signal timings were optimized for
each tested scenario (17). The following results were observed:
1. Significant DLG benefits (in terms of reduction in average intersection
delay) only occurred when (a) the turn movement volume-to-capacity
(v/c) ratio exceeded 0.95, and (b) when the parallel through movement v/c
ratio was less than [(N−1)/N]−0.05, where N is the number of exclusive
through lanes prior to DLG treatment. For example, if there are three
exclusive through lanes prior to treatment, the through movement v/c
ratio should be less than [(3−1)/3]−0.05 = 0.62.
2. Benefits were much higher when the effective green–to–cycle length
(g/C) ratio was high on the DLG approach prior to DLG treatment (i.e.,
when DLG movements received green during most of the cycle).
3. Benefits increased when adding a new turn lane allowed the turn
movement to go from significantly oversaturated to undersaturated.
4. Benefits increased when cycle lengths were re-optimized to
accommodate the new lane grouping; however, this tactic may not be
effective when other congested intersections exist along the urban street
and which govern the background cycle length.
5. Similar benefits were observed for left turns and right turns, and for two
exclusive through lanes versus three exclusive through lanes.
Effects of Dynamic Lane Grouping Chapter 37/ATDM: Supplemental

6. Benefits increased when there was good progression quality on the DLG
approach, prior to DLG treatment.
7. Shared-lane DLG produced no significant benefits for right turns unless
right-turn-on-red was allowed, and produced no significant benefits for
left turns.
Reference (17) also describes an application of the HCM’s Chapter 17 urban
street reliability method to evaluate a corridor using DLG.
Chapter 37/ATDM: Supplemental Effects of Dynamic Lane Grouping

7. EFFECTS OF REVERSIBLE CENTER LANES
This section provides guidance on modeling RCLs using HCM methods and
summarizes the results of a simulation-based evaluation of RCLs.
MODELING RCLS USING HCM METHODS

The following inputs to the motorized vehicle urban street segment
procedure described in Chapter 18 may need to be modified when comparing
roadway operations with RCLs to operations without RCLs:
• Number of lanes by movement group at boundary intersections
• Number of midsegment through lanes
• Left-turn volumes
• Median type
• Number of access point approaches on the opposite side
• Signal timing parameters

This section summarizes the results of an evaluation of RCLs on the
operation of an urban street. A simulation model calibrated to a real-world urban
street was used to evaluate the benefit of a hypothetical RCL operation in terms
of average delay, average speed, and average number of stops. Four different
directional demand splits (60/40, 70/30, 80/20, 90/10) and two different demand
levels (existing total volume and twice the existing total volume) were simulated.
The roadway was simulated with two travel lanes in each direction without RCL
operation, and with three peak-direction and one off-peak-direction travel lanes
with RCL operation. Two signal timing scenarios were also tested with RCL
operation: (a) the existing signal timing, and (b) optimized signal timing for the
new lane configuration (17).
The simulation results are illustrated in Exhibit 37-10 for two demand levels
(existing and twice the existing) and three operational scenarios (base case, RCL
with the existing signal timing, and RCL with optimized signal timings). Note
that the y-axis scale is different in the two graphs.
Effects of Reversible Center Lanes Chapter 37/ATDM: Supplemental

Exhibit 37-10
Illustrative RCL Delay Percent
Reductions by Scenario
(a) Existing Demand Level (b) Twice the Existing Demand Level
Similar benefits were observed through other performance measures,

including vehicle stops, speeds, and travel times. RCL benefits for the simulated
corridor were maximized at higher demand levels and higher directional splits.
However the tipping point at which RCL benefits are viewed as significant may
depend on a large number of additional factors that affect urban street
operations. These factors could include signal spacing, progression quality,
distribution of major and minor street demands, and corridor reliability.
Chapter 37/ATDM: Supplemental Effects of Reversible Center Lanes

8. PLANNING AN ATDM PROGRAM
ATDM strategies are combined into an overall ATDM program to address

challenges to the efficient operation of the highway system. The ATDM program
will have different plan elements to address specific challenges to the system:
• The travel demand management (TDM) plan element will address how
demand management will be used to address recurring congestion on the
facility.
• The weather traffic management plan element will identify the ATDM
strategies to be used during weather events. The weather traffic
management plan will have a TDM component targeted to special
weather events.
• The TIM plan element will identify the ATDM strategies to be used for
incidents. The TIM will have a TDM component for managing demand on
the facility during incidents.
• The work zone traffic management plan element will identify the ATDM
strategies to be used for work zones. The work zone traffic management
plan will have a TDM component for managing demand while work
zones are present.
• Facilities located next to major sporting and entertainment venues may
also have a special event management plan with ATDM strategies
identified to support management of traffic before and after major events.
TRAVEL DEMAND MANAGEMENT PLANS

FHWA’s Travel Demand The Federal Highway Administration’s (FHWA) Travel Demand Management
Management Toolbox is
available at Toolbox website provides resources to help manage traffic congestion by better
http://ops.fhwa.dot.gov/tdm/ managing demand. These resources include publications, web links, and training
toolbox.htm.
offerings. Demand management strategies include the following (19):
• Technology accelerators:
o Real-time traveler information,
o National 511 phone number, and
o Electronic payment systems;
• Financial incentives:
o Tax incentives,
o Parking cash-out,
o Parking pricing,
o Variable pricing,
o Distance-based pricing, and
o Incentive reward programs;
• Travel time incentives:
o HOT lanes,
Planning an ATDM Program Chapter 37/ATDM: Supplemental

o Signal priority systems, and

o Preferential parking;
• Marketing and education:
o Social marketing and
o Individualized marketing;
• Mode-targeted strategies:
o Guaranteed ride home,
o Transit pass programs, and
o Shared vehicles;
• Departure time–targeted strategies:
o Worksite flextime and
o Coordinated event or shift scheduling;
• Route-targeted strategies:
o Real-time route information,
o In-vehicle navigation, and
o Web-based route-planning tools;
• Trip reduction–targeted strategies:
o Employer telework programs and policies and
o Compressed workweek programs; and
• Location- and design-targeted strategies:
o Transit-oriented development,
o Live near your work, and
o Proximate commute.
FHWA’s guide on this topic (19) should be consulted for more information
on designing the TDM element of an ATDM program.
WEATHER-RESPONSIVE TRAFFIC MANAGEMENT PLANS

Weather-responsive traffic management involves the implementation of
traffic advisory, control, and treatment strategies in direct response to or in
anticipation of developing roadway and visibility issues that result from
deteriorating or forecast weather conditions (20).
Weather-responsive traffic management strategies include the following:
• Motorist advisory, alert, and warning systems;
• Speed management strategies;
• Vehicle restriction strategies;
• Road restriction strategies;
• Traffic-signal control strategies;
• Traffic incident management;
Chapter 37/ATDM: Supplemental Planning an ATDM Program

• Personnel and asset management; and

• Agency coordination and integration.
FHWA’s report on this topic (20) should be consulted for additional
information on the design and selection of weather-responsive traffic
management strategies.
TRAFFIC INCIDENT MANAGEMENT PLANS

An FHWA handbook (21) provides information on the design of TIM plans.
TIM is “the coordinated, preplanned use of technology, processes, and
procedures to reduce the duration and impact of incidents, and to improve the
safety of motorists, crash victims and incident responders.” An incident is “any
non-recurring event that causes a reduction in capacity or an abnormal increase
in traffic demand that disrupts the normal operation of the transportation
system” (13). Such events include traffic crashes, disabled vehicles, spilled cargo,
severe weather, and special events such as sporting events and concerts. ATDM
strategies may be included as part of an overall TIM plan to improve facility
operations during and after incidents.
An agency’s incident management plan documents the agency’s strategy for
dealing with incidents. It is, in essence, a maintenance of traffic plan (MOTP) for
incidents and unplanned work zones. The responses available to the agency are
more limited for incident management and by definition must be real-time,
dynamic responses to each incident as it presents itself. The agency’s incident
MOTP ensures that adequate resources are prepositioned and interagency
communications are established to respond rapidly and effectively to an
incident. The TIM plan may include measures in effect 24 hours a day and 7 days
a week, weekdays only, weekday peak periods, or any other periods of time or
days of the week that are the focus of the TIM plan.
Incidents Defined and Classified

An incident is an unplanned disruption to the capacity of the facility.
Incidents do not need to block a travel lane to disrupt the capacity of the facility.
They can be a simple distraction within the vehicle (e.g., spilling coffee), on the
side of the road, or in the opposite direction of the facility.
Incidents can be classified according to the response resources and
procedures required to clear the incident. This classification helps in identifying
strategic options for improving incident management.
Section 6I.01 of the 2009 Manual on Uniform Traffic Control Devices (MUTCD)
(22) classifies incidents according to their expected duration:
• Extended-duration incidents are those expected to persist for more than 24
h and should be treated like work zones.
• Major incidents have expected durations of more than 2 h.
• Intermediate incidents have expected durations of 0.5 h up to and
including 2 h.
• Minor incidents are expected to persist for less than 30 min.

Stages of Incident Management

Incident management is the systematic, planned, and coordinated use of
human, institutional, mechanical, and technical resources to reduce the duration
and impact of incidents. Incident management has several stages:
• Detection;
• Verification;
• Response;
• Motorist information; and
• Site management, consisting of
o Traffic management,
o Investigation, and
o Clearance.
Detection is the first notice the agency receives that there may be an incident
on the facility. Detection may occur via 911 calls, closed-circuit TV cameras, or
detector feeds to a transportation management center or to maintenance or
enforcement personnel monitoring the facility.
Verification confirms an incident has occurred; collects additional information
on the nature of the incident; and refines the operating agency’s understanding
of the nature, extent, and location of the incident for an effective response.
A response is selected after an incident is verified, and the appropriate
resources are dispatched to the incident. A decision is also made as to the
dissemination of information about the incident to the motoring public.
Motorist information informs drivers not at the site about the location and
severity of the incident to enable them to anticipate conditions at the site and
give them the opportunity to divert and avoid the site.
Site management refers to the management of resources to remove the
incident and reduce the impact on traffic flow and safety. This stage involves the
following three major tasks:
• Traffic management, which is the control and safe movement of traffic
through the incident zone;
• Investigation, which documents the causes of traffic incidents for safety
evaluation and legal and insurance purposes; and
• Clearance, which refers to the safe and timely removal of any wreckage or
spilled material from the roadway.
An incident management plan has the following strategic and tactical
program elements (21):
• Management objectives and performance measurement;
• Designated interagency teams’ membership, roles, and responsibilities;
• Response and clearance policies and procedures; and
• Responder and motorist safety laws and equipment.

Incident Response and Clearance Strategies

The incident management plan will designate the responder roles and
responsibilities, establish an incident command system with a unified command
across agencies, identify who is responsible for bringing which equipment and
resources to the incident site, establish response and clearance procedures by
responding agency and by incident type, and identify state and local laws that
apply to incident clearance procedures.
Exhibit 37-11 presents a menu of possible incident management strategy
improvements that an agency may wish to evaluate by using the ATDM analysis
procedure (23). The expected effect of each class of strategies on highway
capacities and speeds is included in this exhibit.
Exhibit 37-11 Strategy Description

Possible Incident Management
Improved detection Closed-circuit TV, routine service patrol, or other continuously monitored
Strategies
and verification incident detection system to spot incidents more quickly and verify the
required resources to clear the incident. Enhanced 911, automated
positioning systems, motorist aid call boxes, and automated collision
notification systems are included.
Traveler information 511 systems, traveler information websites, media partnerships, dynamic
system message signs, standardized dynamic message sign message sets, and
usage protocols to improve the information available to travelers.
Response Personnel and equipment resource lists, towing and recovery vehicle
identification guide, instant tow dispatch procedures, towing and recovery
zone–based contracts, enhanced computer-aided dispatch, dual or
optimized dispatch procedures, motorcycle patrols, equipment staging
areas or prepositioned equipment.
Scene management Incident command system, response vehicle parking plans, high-visibility
and traffic control safety apparel and vehicle markings, on-scene emergency lighting
procedures, safe and quick clearance laws, effective traffic control
through on-site traffic management teams, overhead lane-closure signs,
variable speed limits, end-of-queue advance warning systems, alternate
route plans.
Quick clearance and Abandoned-vehicle laws, safe and quick clearance laws, service patrols,
recovery vehicle-mounted push bumpers, incident investigation sites, noncargo
vehicle fluid-discharge policy, fatality certification and removal policy,
expedited crash investigation, quick clearance using fire apparatus,
towing and recovery quick clearance incentives, major incident response
teams.
Source: Adapted from Carson (23).
WORK ZONE TRANSPORTATION MANAGEMENT PLANS

Work zone management has the objective of moving traffic through the
working area with as little delay as possible, consistent with the safety of the
workers, the safety of the traveling public, and the requirements of the work
being performed. Transportation management plans are a collection of
administrative, procedural, and operational strategies used to manage and
mitigate the impacts of a work zone project.
The work zone MOTP may have three components: a temporary traffic
control plan, a transportation operations plan, and a public information plan.
The temporary traffic control plan describes the control strategies, traffic control
devices, and project coordination. The transportation operations plan identifies
the demand management, corridor management, work zone safety management,

and the traffic and incident management and enforcement strategies. The public
information plan describes the public awareness and motorist information
strategies (13). ATDM strategies can be important components of a
transportation management plan (14).
The work zone MOTP codifies the agency’s management strategy. It has the
following elements:
• Construction approach: staging, sequencing, lane and ramp closure
alternatives, alternative work schedules (e.g., night, weekend).
• Traffic control operations: a mix of dynamic (ATDM) and static measures
consisting of speed limit reductions, truck restrictions, signal timing
(coordination and phasing), reversible lanes, and physical barriers.
• Public information: a mix of dynamic (ATDM) and static pretrip and en
route information (e.g., 511, newspapers, meetings, websites, closed-
circuit television over the Internet), plus on-site information signing such
as static signs, changeable or variable message signs, and highway
advisory radio.
• TDM: employer-based and other incentives (in addition to public
information) for use of alternative modes of travel, including park-and-
ride.
• Incident management and enforcement: generally, ATDM measures specified
in an incident management plan (i.e., an incident MOTP), such as traffic
management centers, intelligent transportation systems, emergency
service patrols, hazardous materials teams, and enhanced police
enforcement. A particularly aggressive incident MOTP may be put in
place for work zones.
Construction Approach
The work zone MOTP must consider several alternative construction
approaches (including traffic maintenance) and recommend the construction
approach that best meets the agency’s objectives for the construction project.
Traffic maintenance approaches to be considered in the work zone MOTP
include the following:
1. Complete closure of the work area for a short time versus partial closure
for a longer time,
2. Nighttime versus daytime lane closures, and
3. Off-peak versus peak hour lane closures.
Traffic Control Operations

The traffic control element of the MOTP specifies work zone speed-limit
reductions, signal timing changes (if needed), reversible lanes (e.g., flagging),
and the locations of physical barriers and cones. The traffic control elements may
be dynamic, responding in real time to changing conditions, or they may be more
static, operating at prespecified times of the day.

MUTCD Section 6G.02 defines work zone types according to duration and
time of day (22):
• Duration Type A: long-term stationary work that occupies a location more
than 3 days;
• Duration Type B: intermediate-term stationary work that occupies a
location more than one daylight period up to 3 days, or nighttime work
lasting more than 1 h;
• Duration Type C: short-term stationary daytime work that occupies a
location for more than 1 h within a single daylight period;
• Duration Type D: short-duration work that occupies a location up to 1 h;
and
• Duration Type E: mobile work that moves intermittently or continuously.
Work zones are further categorized by MUTCD Section 6G.03 according to
their location on the facility. Work zones within the traveled way (Location Type
E) are further subdivided by facility type (22):
• Location Type A: outside the shoulder (Section G6.06);
• Location Type B: on the shoulder with no encroachment (Section G6.07);
• Location Type C: on the shoulder with minor encroachment, leaving at
least a 10-ft lane (Section G6.08);
• Location Type D: within the median (Section G6.09); and
• Location Type E: within the traveled way of
o A two-lane highway (Section 6G.10),
o An urban street (Section 6G.11),
o A multilane non-access-controlled highway (Section 6G.12),
o An intersection (Section 6G.13), or
o A freeway or an expressway (Section 6G.14).
Each work zone type has an associated typical application of temporary
traffic controls. They are described in MUTCD Section 6H-1 (22).
Public Information Element

The public information element is intended to provide the public with
pretrip and en route information and with preconstruction and during-
construction information on the work zone so the public can plan accordingly.
The intent is to encourage travelers who can to reschedule or reroute their trip to
avoid the work zone during periods of peak closures. Public information
includes 511 alerts; press interviews; public information meetings; project update
websites; and on-site web-accessible closed-circuit cameras, variable message
signs, and highway advisory radio.

Travel Demand Management Element

In coordination with the public information element, the TDM element
identifies incentives, such as park-and-ride lots, that will be provided for
travelers using alternative modes. The public information element and the TDM
element differ in that the public information is neutral, leaving it to the traveler
to choose how to respond. The TDM element provides monetary and service
incentives to encourage a particular subset of choices.
Incident Management and Enforcement Element

Incident management includes the development of incident management
plans for the work zone. The plans describe coordination with traffic
management centers, the use of intelligent transportation systems devices,
deployment of emergency service patrols in the work zone, and enhanced police
enforcement. Enforcement may be strengthened with speed limit feedback signs
and other devices.
SPECIAL EVENT MANAGEMENT PLANS

Special event management deals with moving people and traffic to and from
special event locations, such as a sports stadium, concert hall, or arena. The
objective is to get people and traffic onto and off of the site with minimal
backups onto the public transportation system and in a reasonable time. Traffic
control officers, temporary cones and signs, reversible lanes, and special signal
control plans are often part of a special event management plan (24).
A special event management plan typically has the following components:
• Preevent ingress control,
• During-event access control, and
• Postevent egress control.
The special event management plan will deploy a combination of temporary
signing, lane controls, signal timing plans, and personnel to move traffic into and
out of the event venue, much like a short-term work zone.

9. REFERENCES
Many of these references are 1. Ramp Management and Control: A Primer. Report FHWA-HOP-06-080. Federal
available in the Technical
Reference Library in Volume 4. Highway Administration, Washington, D.C., Jan. 2006.
2. Technologies That Complement Congestion Pricing: A Primer. Report FHWA-
HOP-08-043. Federal Highway Administration, Washington, D.C., Oct. 2008.
3. Managed Lanes: A Primer. Report FHWA-HOP-05-031. Federal Highway
Administration, Washington, D.C., 2005.
4. National Signal Timing Optimization Project: Summary Evaluation Report.
Federal Highway Administration, Washington, D.C., and University of
Florida, Gainesville, May 1982.
5. Su, P., X. Jiang, R. Jagannathan, and D. Hale. Dynamic Lane Grouping at
Signalized Intersections: Selecting the Candidates and Evaluating
Performance. ITE Journal, Vol. 85, No. 11, 2015, pp. 43–47.
6. Levinson, H., D. Sullivan, and R. Bryson. Effects of Urban Traffic Volume
Variations on Service Levels. Presented at 85th Annual Meeting of the
7. Hellinga, B., and Z. Abdy. Impact of Day-to-Day Variability of Peak-Hour
Volumes on Signalized Intersection Performance. Presented at 86th Annual
Meeting of the Transportation Research Board, Washington, D.C., 2007.
8. Tarko, A.P., and R.I. Perez-Cartagena. Variability of Peak Hour Factor at
Intersections. In Transportation Research Record: Journal of the Transportation
Research Board, No. 1920, Transportation Research Board of the National
Academies, Washington, D.C., pp. 125–130.
9. Hale, D., R. Jagannathan, M. Xyntarakis, P. Su, X. Jiang, J. Ma, J. Hu, and C.
Krause. Traffic Bottlenecks: Identification and Solutions. Report FHWA-HRT-16-
064. Federal Highway Administration, Washington, D.C., 2016.
10. Dey, S., J. Ma, and Y. Aden. Reversible Lane Operation for Arterial Roadways:
The Washington, DC, USA Experience. ITE Journal, Vol. 81, No. 5, 2011, pp.
26–35.
11. Avenue Consultants. 5400 South Flex Lanes Before/After Evaluation. Utah
Department of Transportation, Salt Lake City, 2013.
12. Zhao, J., W. Ma, Y. Liu, and X. Yang. Integrated design and operation of
urban arterials with reversible lanes. Transportmetrica B: Transport Dynamics,
2014, pp. 130–150.
13. Balke, K. N. Traffic Incident Management in Construction and Maintenance Work
Zones. Report FHWA-HOP-08-056. Federal Highway Administration,
Washington, D.C., Jan. 2009.
14. Jeannotte, K., and A. Chandra. Developing and Implementing Transportation
Management Plans for Work Zones. Report FHWA-HOP-05-066. Federal
Highway Administration, Washington, D.C., Dec. 2005.
References Chapter 37/ATDM: Supplemental

15. Zhang, L., and D. Levinson. Ramp Metering and Freeway Bottleneck
Capacity. In Transportation Research Part A, Vol. 44, 2010, pp. 218–235.
16. Papageorgiou, M., H. Hadj-Salem, and J.-M. Blosseville. ALINEA: A Local
Feedback Control Law for On-Ramp Metering. In Transportation Research
Record 1320, Transportation Research Board, National Research Council,
Washington, D.C., 1991, pp. 58–64.
17. Hale, D., H. Mahmassani, and A. Mittal. Active Transportation and Demand
Management (ATDM) Analytical Methods for Urban Streets Final Report. Report
FHWA-HOP-16-088. Federal Highway Administration, Washington, D.C.,
2016.
18. Stevanovic, A., M. Zlatkovic, and I. Dakic. Comparison of adaptive traffic
control benefits for recurring and nonrecurring traffic conditions. Presented
at the 22nd ITS World Congress, Bordeaux, France, 2015.
19. Association for Commuter Transportation, UrbanTrans Consultants, Parsons
Brinckerhoff, and ESTC. Mitigating Traffic Congestion: The Role of Demand-Side
Strategies. Report FHWA-HOP-05-001. Federal Highway Administration,
Washington, D.C., Oct. 2004.
20. Gopalakrishna, D., F. Kitchener, and K. Blake. Developments in Weather
Responsive Traffic Management Strategies. Report FHWA-JPO-11-086. Federal
Highway Administration, Washington, D.C., June 2011.
21. Owens, N., A. Armstrong, P. Sullivan, C. Mitchell, D. Newton, R. Brewster,
and T. Trego. Traffic Incident Management Handbook. Report FHWA-HOP-10-
013. Federal Highway Administration, Washington, D.C., Jan. 2010.
22. Manual on Uniform Traffic Control Devices for Streets and Highways. Federal
Highway Administration, Washington, D.C., 2009.
http://mutcd.fhwa.dot.gov. Accessed Mar. 5, 2021.
23. Carson, J. L. Best Practices in Traffic Incident Management. Report FHWA-
HOP-10-050. Federal Highway Administration, Washington, D.C., Sept.
2010.
24. Carson, J. L., and R. G. Bylsma. NCHRP Synthesis of Highway Practice 309:
Transportation Planning and Management for Special Events. Transportation
Chapter 37/ATDM: Supplemental References

HIGHWAY CAPACITY
MANUAL 7th Edition
TRBsales@nas.edu).



CHAPTER 38
NETWORK ANALYSIS
CONTENTS
1. INTRODUCTION.................................................................................................. 38-1
Overview ............................................................................................................. 38-1
Chapter Organization ........................................................................................ 38-1
Related HCM Content ........................................................................................ 38-1
2. CONCEPTS ............................................................................................................. 38-3

Overview ............................................................................................................. 38-3
Spillback Impact on Freeways .......................................................................... 38-3
Spillback Impact on Urban Streets ................................................................... 38-6
Lane-by-Lane Analysis ...................................................................................... 38-7
Performance Measurement for Networks and O-D Pairs ............................. 38-8
3. METHODOLOGY ................................................................................................. 38-9

Scope of the Methodology ................................................................................. 38-9
Computational Steps .........................................................................................38-12
4. EXAMPLE PROBLEMS....................................................................................... 38-28

Example Problem 1: O-D–Based Travel Time Estimation ...........................38-28
Example Problem 2: On-Ramp Spillback Analysis .......................................38-38
Example Problem 3: Off-Ramp Queue Spillback Analysis ..........................38-65
Example Problem 4: On-Ramp Queue Spillback Analysis into a
Single-Lane Roundabout ...........................................................................38-74
5. REFERENCE.......................................................................................................... 38-80
APPENDIX A: OFF-RAMP QUEUE SPILLBACK ANALYSIS ....................... 38-81

Capacity Checks.................................................................................................38-82
Queue Length Estimation .................................................................................38-83
Queue Storage Ratios and Spillback Chcecks ................................................38-86
Off-Ramp Queue Spillback Evaluation ..........................................................38-88
References .........................................................................................................38-129
Chapter 38/Network Analysis Contents

APPENDIX B: ON-RAMP QUEUE SPILLBACK ANALYSIS .......................38-130

Demand Estimation ........................................................................................ 38-130
Capacity Estimation ....................................................................................... 38-137
Evaluation of On-Ramp Queue Spillback Impacts .................................... 38-137
References ........................................................................................................ 38-154
APPENDIX C: LANE-BY-LANE ANALYSIS FOR FREEWAY

FACILITIES ............................................................................................................38-155
Lane-by-Lane Flow Models by Segment Type ........................................... 38-155
Lane Flow Ratio Distribution as a Function of the Demand-to-
Capacity Ratio .......................................................................................... 38-159
Checking for Negative Flows and Lane Capacities ................................... 38-160
Speed–Flow Curves by Lane and Segment Type ....................................... 38-162
Application Examples .................................................................................... 38-165
References ........................................................................................................ 38-173
Contents Chapter 38/Network Analysis

LIST OF EXHIBITS
Exhibit 38-1 Off-Ramp Components ........................................................................ 38-4

Exhibit 38-2 Definition of Spillback Regimes .......................................................... 38-4
Exhibit 38-3 Capacity Adjustment Factors (CAFBL) for Through Lanes
Adjacent to Blocked Lanes during Queue Spillback....................................... 38-6
Exhibit 38-4 Queue Influence Area with Increased Turbulence........................... 38-6
Exhibit 38-5 Length of Queue Influence Area as a Function of the
Segment Free-Flow Speed (FFS) ........................................................................ 38-6
Exhibit 38-6 Queue Spillback from an On-Ramp into Urban Street
Intersections ......................................................................................................... 38-7
Exhibit 38-7 Required Input Data, Potential Data Sources, and Default
Values for the Network Analysis Methodology............................................ 38-11
Exhibit 38-8 Default Spillback Regimes as a Function of Ramp Geometry
and Driver Aggressiveness .............................................................................. 38-12
Exhibit 38-9 Network Analysis Methodology Flowchart .................................... 38-12
Exhibit 38-10 Sample Study Network, with Multiple Origins and
Destinations ........................................................................................................ 38-13
Exhibit 38-11 Potential Bottlenecks Constraining the Ramp Terminal
Demand ............................................................................................................... 38-14
Exhibit 38-12 Potential Bottlenecks Constraining the On-Ramp Demand ....... 38-15
Exhibit 38-13 Spillback Check Procedure for Off-Ramps .................................... 38-17
Exhibit 38-14 Spillback Check Procedure for On-Ramps .................................... 38-18
Exhibit 38-15 Probability of Lane Choice for Entry/Exit Segments on
Freeway Facilities .............................................................................................. 38-20
Exhibit 38-16 Illustration of Lane Choice Probabilities Along a Freeway
Facility ................................................................................................................. 38-20
Exhibit 38-17 Speed–Flow Curves for Freeway Ramps ....................................... 38-22
Exhibit 38-18 Sample Calculation of Total Travel Time Using Multiperiod
Analysis............................................................................................................... 38-24
Exhibit 38-19 Reference Input Values for O-D Analysis under Free-Flow
Conditions .......................................................................................................... 38-24
Exhibit 38-20 List of Example Problems ................................................................ 38-28
Exhibit 38-21 Example Problem 1: Network Interchanges, Intersections,
and O-D Points................................................................................................... 38-28
Exhibit 38-22 Example Problem 1: Freeway Facility Segmentation and
O-D Entry and Exit Points ................................................................................ 38-29
Exhibit 38-23 Example Problem 1: O-D Matrix..................................................... 38-29
Exhibit 38-24 Example Problem 1: Urban Street Facilities .................................. 38-30

Exhibit 38-25 Example Problem 1: List of Intersections, Ramps, and

Segments Traversed for O-D Pair D-H ...........................................................38-30
Exhibit 38-26 Example Problem 1: Input Data for Freeway Facility
Analysis ...............................................................................................................38-31
Exhibit 38-27 Example Problem 1: Input Data for Intersection Analysis –
Archer Rd. WB ...................................................................................................38-31
Exhibit 38-28 Example Problem 1: Input Data for Segment Analysis –
Archer Rd. WB ...................................................................................................38-32
Exhibit 38-29 Example Problem 1: Input Data for Intersection Analysis –
NW 39th Ave. EB ...............................................................................................38-32
Exhibit 38-30 Example Problem 1: Input Data for Segment Analysis –
NW 39th Ave. EB ...............................................................................................38-32
Exhibit 38-31 Example Problem 1: On-Ramp Demands Along the
Freeway Facility .................................................................................................38-33
Exhibit 38-32 Example Problem 1: Freeway Segment LOS .................................38-33
Exhibit 38-33 Example Problem 1: Off-Ramp Demands Along the
Freeway Facility .................................................................................................38-33
Exhibit 38-34 Example Problem 1: Off-Ramp Queue Length Estimation
and Queue Storage Checks ...............................................................................38-34
Exhibit 38-35 Example Problem 1: Flow Distribution and Speeds for
Freeway Segments .............................................................................................38-35
Exhibit 38-36 Example Problem 1: Estimated Speeds by Segment Based
on Lane Choice Probability and Speeds .........................................................38-35
Exhibit 38-37 Example Problem 1: Urban Street Segment Speeds .....................38-36
Exhibit 38-38 Example Problem 1: Urban Streets Segment Travel Times .........38-36
Exhibit 38-39 Example Problem 1: Freeway Segment Travel Times ..................38-36
Exhibit 38-40 Example Problem 1: Ramp Roadway Travel Times .....................38-36
Exhibit 38-41 Example Problem 1: Cumulative Travel Time Computation......38-37
Exhibit 38-42 Example Problem 2: Network Interchanges, Intersections,
and O-D Points ...................................................................................................38-38
Exhibit 38-43 Example Problem 2: Freeway Facility Segmentation and
O-D Entry and Exit Points ................................................................................38-39
Exhibit 38-44 Example Problem 2: Urban Street Facility .....................................38-39
Exhibit 38-45 Example Problem 2A: Signalized Intersection Geometry:
I-10 EB Ramps ....................................................................................................38-40
Exhibit 38-46 Example Problem 2A: Phasing Sequence: I-10 EB Ramps ...........38-40
Exhibit 38-47 Example Problem 2A: Demand Flow Rates (veh/h): I-10 EB
Ramps ..................................................................................................................38-41
Exhibit 38-48 Example Problem 2A: Other Input Data: I-10 EB Ramps ............38-41
Exhibit 38-49 Example Problem 2A: Freeway Facility Segments .......................38-41

Exhibit 38-50 Example Problem 2A: Freeway Facility Geometric Features ...... 38-42
Exhibit 38-51 Example Problem 2A: Calculation of NB Right Turn
Capacity for a Single Cycle: Analysis Period 2 .............................................. 38-44
Exhibit 38-52 Example Problem 2A: NBR Capacity by Analysis Period ........... 38-44
Exhibit 38-53 Example Problem 2A: Calculation of the On-Ramp
Demand vR Based on the Intersection Operation .......................................... 38-45
Exhibit 38-54 Example Problem 2A: Freeway Facility Demand Inputs ............ 38-45
Exhibit 38-55 Example Problem 2A: Freeway Facility LOS ................................ 38-45
Exhibit 38-56 Example Problem 2A: Spillback Check: I-10 EB On-Ramp ......... 38-46
Exhibit 38-57 Example Problem 2A: Freeway Segment 5 Merge Capacity
and Queue Lengths ........................................................................................... 38-47
Exhibit 38-58 Example Problem 2A: Freeway Performance During
Analysis Period 4 with and without the Queue Storage Constraint .......... 38-47
Exhibit 38-59 Example Problem 2A: Estimated Queue Lengths and
Merge Capacities During Analysis Period 2 .................................................. 38-48
Exhibit 38-60 Example Problem 2A: Discharge Flow Rates into the On-
Ramp for Each Phase Throughout the Cycle During Analysis
Period 2 ............................................................................................................... 38-51
Exhibit 38-62 Example Problem 2A: Discharge Flow Rates into the On-
Ramp for Each Phase Throughout the Cycle During Analysis
Period 3 ............................................................................................................... 38-54
Exhibit 38-63 Example Problem 2A: Calculation of the Spillback Capacity
Reduction Factor for the SBL Movement for Analysis Period 3 ................. 38-55
Exhibit 38-65 Example Problem 2A: Calculation of the Spillback Capacity
Reduction Factor for the SBL Movement for Analysis Period 4 ................. 38-56
Exhibit 38-66 Example Problem 2A: Performance Measure Comparison
with and without Consideration of Spillback Effects ................................... 38-57
Exhibit 38-67 Example Problem 2B: TWSC Intersection Geometry: I-10 EB
Ramps .................................................................................................................. 38-58
Exhibit 38-68 Example Problem 2B: Calculation of the On-Ramp Demand
vR Based on the Intersection Operation .......................................................... 38-59
Exhibit 38-69 Example Problem 2B: Queue Accumulation Plot
Calculations for the On-Ramp ......................................................................... 38-60
Exhibit 38-70 Example Problem 2B: Queue Accumulation Polygon for the
On-Ramp............................................................................................................. 38-60

Exhibit 38-71 Example Problem 2B: Performance Measure Comparison

with and without Consideration of Spillback Effects—Analysis
Period 3 ...............................................................................................................38-61
Exhibit 38-72 Example Problem 2C: AWSC Intersection Geometry: I-10
EB Ramps ............................................................................................................38-62
Exhibit 38-73 Example Problem 2C: Calculation of the On-Ramp Demand
vR Based on the Intersection Operation ..........................................................38-63
Exhibit 38-74 Example Problem 2C: Spillback Occurrence Check .....................38-63
Exhibit 38-75 Example Problem 2C: Queue Accumulation Plot
Calculations for the On-Ramp .........................................................................38-64
Exhibit 38-76 Example Problem 2C: Queue Accumulation Polygon for
the On-Ramp ......................................................................................................38-64
Exhibit 38-77 Example Problem 2C: Equivalent Capacities and Headways
for the On-Ramp: Analysis Period 3 ...............................................................38-64
Exhibit 38-78 Example Problem 2C: Performance Measure Comparison
with and without Consideration of Spillback Effects—Analysis
Period 3 ...............................................................................................................38-65
Exhibit 38-79 Example Problem 3: Study Site .......................................................38-65
Exhibit 38-80 Example Problem 3: Freeway Facility Geometry .........................38-66
Exhibit 38-81 Example Problem 3: Traffic Demands ............................................38-66
Exhibit 38-82 Example Problem 3: Freeway Facility 1 (I-75) LOS ......................38-67
Exhibit 38-83 Example Problem 3: Freeway Facility 2 (SR-826) LOS .................38-67
Exhibit 38-84 Example Problem 3: Queue Length and Storage Ratio at the
SR-826 On-Ramp ................................................................................................38-68
Exhibit 38-85 Example Problem 3: Link–Node Structure for Spillback
Analysis: I-75 SB.................................................................................................38-69
Exhibit 38-86 Example Problem 3: Queued Vehicles and Total Number of
Vehicles RNV in the Ramp: Analysis Period 2 ...............................................38-70
Exhibit 38-87 Example Problem 3: Ramp Capacity RSTG and Ramp Input
RI: Analysis Period 2 .........................................................................................38-70
Exhibit 38-88 Example Problem 3: Ramp Capacity RSTG and Ramp Input
RI: Analysis Period 3 .........................................................................................38-71
Exhibit 38-89 Example Problem 3: Spillback Queue Length on I-75 SB:
Analysis Period 3 ...............................................................................................38-72
Exhibit 38-90 Example Problem 3: Available Queue Storage on I-75 SB ...........38-72
Exhibit 38-91 Example Problem 3: Back of Queue Length, Including
Queue Influence Area, at the End of Analysis Period 3 ...............................38-73
Exhibit 38-92 Example Problem 4: Study Interchange Schematic ......................38-74
Exhibit 38-93 Example Problem 4: Roundabout Flows and Queues .................38-76
Exhibit 38-94 Example Problem 4: Roundabout Approach Priority Order ......38-76

Exhibit 38-A1 Off-Ramp Queue Spillback Check Flowchart .............................. 38-81

Exhibit 38-A2 Capacity of Ramp Roadways (pc/h) .............................................. 38-82
Exhibit 38-A3 Examples of Unbalanced Ramp Lane Usage ............................... 38-84
Exhibit 38-A4 Illustrative Assignment of Intersection Lane Groups to
Ramp Lanes ........................................................................................................ 38-85
Exhibit 38-A5 Expanded Link–Node Structure to Evaluate Off-Ramp
Segments ............................................................................................................. 38-89
Exhibit 38-A6 Example Off-Ramp Geometry with Heavy Left-Turn
Demand at a Signalized Intersection .............................................................. 38-90
Exhibit 38-A7 Off-Ramp Queue Spillback Regimes ............................................. 38-91
Exhibit 38-A8 Freeway Facility Oversaturated Analysis Procedure,
Adapted for Off-Ramp Queue Spillback Evaluation .................................... 38-96
Exhibit 38-A9 Capacity Adjustment Factors for Lane Blockage CAFBL ........... 38-100
Exhibit 38-A10 Equivalent Segment Capacity for Unblocked Lanes When
Lane Blockage Occurs ..................................................................................... 38-100
Exhibit 38-A11 Maximum Off-Ramp Queue Storage Length at Diverge
Segments with Regime 3 or 4 Queue Spillback and No Shoulder
Available ........................................................................................................... 38-102
Exhibit 38-A12 Maximum Off-Ramp Queue Storage Length at Diverge
Segments with Regime 3 or 4 Queue Spillback and Shoulder
Available ........................................................................................................... 38-102
Exhibit 38-A13 Node Structure for Example 1 .................................................... 38-103
Exhibit 38-A16 Default Spillback Regime as a Function of Ramp
Geometry and Driver Aggressiveness .......................................................... 38-105
Exhibit 38-A17 Queue Influence Area with Increased Turbulence ................. 38-105
Exhibit 38-A18 Queue Influence Area as Function of the Segment Free-
Flow Speed ....................................................................................................... 38-106
Exhibit 38-A19 Capacity of Ramp Roadways (pc/h) .......................................... 38-106
Exhibit 38-A20 Freeway Ramp Speed–Flow Curves ......................................... 38-107
Exhibit 38-A21 Ramp Density at Capacity as a Function of Ramp FFS .......... 38-107
Exhibit 38-A22 Reference Equations for Back-of-Queue Length
Estimation ......................................................................................................... 38-108
Exhibit 38-A23 Selection of a Cycle Reference Point to Determine the
Initial Number of Vehicles Within the Approach ....................................... 38-109
Exhibit 38-A24 Example Signalized Intersection Approach from an Off-
Ramp ................................................................................................................. 38-109
Exhibit 38-A25 Assignment of Green Times to Time Steps .............................. 38-110

Exhibit 38-A26 Illustration of Mainline Flow Rate Split into Blocked and
Unblocked Lanes .............................................................................................38-111
Exhibit 38-A27 Procedure for Evaluating the Impact of Queue Spillback
on Upstream Nodes and Determining the Queue Length within
Upstream Segments .........................................................................................38-118
Exhibit 38-A28 Potential Effects of an Off-Ramp Queue on Node i ................38-119
Exhibit 38-A29 Distribution of pi as Function of Distance from the
Diverge Point, for a 3-Lane Segment ............................................................38-120
Exhibit 38-A30 Illustration of Lane-Change Maneuvers within the Queue
Influence Area in a 4-Lane Segment under Regime 3 .................................38-121
Exhibit 38-A31 Illustration of Lane-Change Maneuvers within the Queue
Influence Area in a 4-Lane Segment under Regime 4 .................................38-121
Exhibit 38-A32 Effect of Queue Spillback on the Discharge Capacity of an
Upstream On-Ramp ........................................................................................38-123
Exhibit 38-A33 Illustration of Different Density Values within One
Diverge Segment ..............................................................................................38-124
Exhibit 38-B1 Procedure for Detecting Spillback Occurrence at an On-
Ramp .................................................................................................................38-131
Exhibit 38-B2 Schematic of Movements Turning to an On-Ramp from a
TWSC Intersection ...........................................................................................38-133
Exhibit 38-B3 Schematic of Movements Turning to an On-Ramp from an
AWSC Intersection ..........................................................................................38-135
Exhibit 38-B4 Schematic of Movements Turning to an On-Ramp from a
Roundabout ......................................................................................................38-135
Exhibit 38-B5 Signalized Intersection Methodology With Adjustments to
Address On-Ramp Queue Spillback .............................................................38-138
Exhibit 38-B6 Typical Signalized Intersection Ramp Terminal in a
Diamond Interchange .....................................................................................38-139
Exhibit 38-B7 Estimation of Freeway On-Ramp Merging Capacity ................38-140
Exhibit 38-B8 Sample Intersection for Calculation of a QAP for the On-
Ramp .................................................................................................................38-141
Exhibit 38-B9 On-Ramp Queue Accumulation Polygon During Queue
Spillback ............................................................................................................38-141
Exhibit 38-B10 Illustration of Cooperative Behavior in Unsignalized
Intersections with Queue Spillback ...............................................................38-143
Exhibit 38-B11 TWSC intersections Core Methodology with Adjustments
to Address On-Ramp Queue Spillback .........................................................38-144
Exhibit 38-B12 On-Ramp Queue Accumulation Polygon: TWSC
Intersection .......................................................................................................38-145
Exhibit 38-B13 AWSC Intersection Core Methodology with Adjustments
to Address On-Ramp Queue Spillback .........................................................38-148

Page 38-viii Version 7.0
Exhibit 38-B14 Roundabouts Methodology With Adjustments to Address

On-Ramp Queue Spillback ............................................................................. 38-149
Exhibit 38-B15 Required Data and Potential Data Sources for
Roundabout Queue Spillback Evaluation .................................................... 38-149
Exhibit 38-B16 Example Priority Order for a Roundabout Upstream of an
On-Ramp........................................................................................................... 38-150
Exhibit 38-C1 Lane Flow Distribution Model Coefficients for Basic,
Merge, and Diverge Segments ....................................................................... 38-157
Exhibit 38-C2 Lane Flow Distribution Model Coefficients for Weaving
Segments ........................................................................................................... 38-159
Exhibit 38-C3 LFR Distribution for a Sample 2-Lane Basic Freeway
Segment ............................................................................................................. 38-159
Segment ............................................................................................................. 38-160
Segment ............................................................................................................. 38-160
Exhibit 38-C6 Check for Negative Lane Flows ................................................... 38-161
Exhibit 38-C7 Check for Lane Capacity ............................................................... 38-162
Exhibit 38-C8 Multipliers to Estimate Lane FFS from Segment FFS ................ 38-163
Exhibit 38-C9 Multipliers to Estimate Lane Capacity from Segment
Capacity for Basic, Merge, and Diverge Segments ..................................... 38-163
Exhibit 38-C10 Comparison of Speed–Flow Curves by Lane and for the
Segment ............................................................................................................. 38-168
Exhibit 38-C11 Example of LFR Calculation for a Weaving Segment ............. 38-168
Exhibit 38-C12 Comparison of Predicted and Field-Measured Lane-by-
Lane Speeds ...................................................................................................... 38-173

Version 7.0 Page 38-ix
1. INTRODUCTION
OVERVIEW VOLUME 4: APPLICATIONS

GUIDE
This chapter provides methodologies for evaluating the interactions between 25. Freeway Facilities:
Supplemental
freeways and urban streets and the effects of spillback from one facility to 26. Freeway and Highway
another. This chapter’s methodology can be applied to a network of Segments: Supplemental
interconnected freeways and to freeway-to-arterial connections. It can also be Supplemental
applied when the freeway–arterial interchange consists of signalized 28. Freeway Merges and
intersections, STOP-controlled intersections, or roundabouts. This chapter’s 29. Urban Street Facilities:
analysis tools provide travel times and speeds for networks and for origin– Supplemental
destination pairs (O-D) within these networks. Supplemental
The methodology builds on the analysis methods of individual points and Supplemental
segments and extends them in several ways to consider spillback effects from the 32. STOP-Controlled
Intersections:
downstream facility. First, because spillback affects each lane differently, the
Supplemental
analysis is conducted on a lane-by-lane basis. Second, supplemental performance 33. Roundabouts:
Supplemental
measures are provided at the network level and at the O-D level for
undersaturated and oversaturated conditions. Travel time measures are also Terminals: Supplemental
provided for segments and facilities, and their values are consistent with the 35. Pedestrians and Bicycles:
Supplemental
analysis methods described in other parts of the Highway Capacity Manual 36. Concepts: Supplemental
(HCM). 37. ATDM: Supplemental
CHAPTER ORGANIZATION
Section 2 provides the performance measures used at the network level and An O-D pair represents the
includes example calculations of O-D travel time and network travel time. route between two specific
points in the analysis network.
Section 3 describes procedures to evaluate spillback impact on a freeway due Points are defined in Chapter
2, Applications.
to congestion on a ramp or urban street.
Section 4 describes procedures to evaluate spillback impact on an urban
street due to congestion on the freeway or on-ramp.
Section 5 provides case studies to illustrate the application of this chapter’s
methods.
A series of appendices provide detailed information about specific models
and analysis steps.
RELATED HCM CONTENT

Other HCM content related to this chapter includes:
• Chapters 10 and 25, which present the freeway facilities analysis
methodology;
• Chapters 12, 13, and 14, which present the freeway segment
methodologies for basic freeway segments, freeway weaving segments,
and freeway merge and diverge segments, respectively;
• Chapter 26, which provides additional details for basic freeway segments
capacity measurement and driver population factors;
• Chapters 16 and 18, which provide methodologies for evaluating urban
street facilities and urban street segments, respectively;
Chapter 38/Network Analysis Introduction

• Chapters 19, 20, 21, and 22, which provide analysis tools for signalized
intersections, two-way STOP-controlled intersections, all-way STOP-
controlled intersections, and roundabouts, respectively; and
• Chapter 23, which provides methods for evaluating ramp terminals and
alternative intersections.
Introduction Chapter 38/Network Analysis

2. CONCEPTS
OVERVIEW
This section discusses concepts related to spillback on the freeway, spillback
on the urban street, lane-by-lane analysis, and performance measurement for
networks and O-Ds. Concepts related to freeway analysis and urban street
analysis are described in their respective chapters elsewhere in the HCM.
SPILLBACK IMPACT ON FREEWAYS

Spillback on the freeway may occur either due to inadequate capacity of the
ramp roadway or due to inadequate capacity at the ramp terminal (typically the
intersection at the downstream interchange). The capacity of the ramp roadway
is defined as the off-ramp’s maximum allowable hourly flow rate based on its
geometric characteristics (mainly the number of lanes and the free-flow speed).
The capacity of the ramp terminal is defined as the capacity of the signalized or
unsignalized approach to the surface street.
The methodology compares demand and capacity at the off-ramp and at the
ramp terminal to determine whether oversaturation conditions will occur. If
demand exceeds capacity at either of those two locations, then the queue length
is estimated and compared to the available storage on the ramp and along the
deceleration lane. When the queue extends beyond the ramp roadway, blockage
may occur on one or more mainline freeway lanes. In that case, the methodology
estimates the impact of this queue spillback along the freeway by reducing the
segment capacity dependent on the number of blocked lanes and the effects of
that blockage on adjacent lanes.
Off-Ramp Elements
A freeway off-ramp typically consists of three components, as illustrated in
Exhibit 38-1.
• Deceleration lane(s), measured from the beginning of the taper of the
auxiliary lane to the gore.
• Ramp roadway, connecting the deceleration lane and the downstream
ramp terminal and measured from the gore to the taper of the ramp
terminal.
• Ramp terminal, connecting the ramp roadway to the urban street facility
and measured from the point where additional lanes are added to the
intersection approaches to the stop bar of the approach. This component’s
length should be at least as long as the approach’s turn bay lengths. The
ramp terminal can be uncontrolled, STOP- or YIELD-controlled, or
signalized.
When the ramp connects two freeway facilities, the downstream ramp
terminal is replaced by the merge section of the on-ramp, with no storage length.
Chapter 38/Network Analysis Concepts

Exhibit 38-1
Off-Ramp Components
Queue Spillback Regimes

The impact of queue spillback on the freeway mainline varies as a function of
the queue length and the lanes blocked. Five spillback regimes are defined, as
illustrated in Exhibit 38-2.
Exhibit 38-2
Definition of Spillback
Regimes
(a) Regine 0: No queue or queue contained within the ramp roadway
(b) Regine 1: Queue within the deceleration lane (c) Regime 2: Queue along the shoulder
(d) Regime 3: Queue in the rightmost lane (e) Regime 4: Queue blockage of the adjacent lane
Regime 0
Under this regime, shown in Exhibit 38-2(a), there are no queues in the ramp
roadway or the queue, if it exists, is contained within the ramp roadway
boundaries. There are no operational effects in the ramp influence area.
Regime 1
Under this regime, shown in Exhibit 38-2(b), the queue ends within the
deceleration lane and does not spill back into the mainline freeway. Deceleration
lanes typically serve as a transition zone between speeds on the mainline
(typically 55–75 mi/h) and advisory speeds posted along the off-ramp roadway
(typically 20–50 mi/h). When queues begin to form on the deceleration lane, the
Concepts Chapter 38/Network Analysis

available deceleration distance is reduced, and speeds begin to be affected in the

rightmost lane.
Regime 2
Under this regime, shown in Exhibit 38-2(c), the queue of vehicles extends
upstream beyond the deceleration lane, but sufficient lateral clearance on the
right-hand shoulder allows for additional queue storage. In this case, the
deceleration lane does not serve as a transition zone and drivers decelerate and
join the back of the queue more abruptly, resulting in turbulence and reduced
speeds in the rightmost lane. If no lateral clearance exists immediately upstream
of the deceleration lane, Regime 2 conditions are not possible. In some cases, this
regime does not occur even if storage is available; this occurrence is site-specific
and depends on local driver behavior.
Regime 3
Under this regime, shown in Exhibit 38-2(d), the queue extends to the
rightmost lane of the freeway mainline. This regime may occur either when no
shoulder is available for additional queue storage, or when drivers choose to
queue in the rightmost lane once the deceleration lane is entirely occupied. Non-
exiting vehicles on the rightmost lane are delayed or change lanes, which causes
increased turbulence and reduced speeds in the two rightmost lanes.
Regime 4
Under this regime, shown in Exhibit 38-2(e), the queue blocks the rightmost
lane, and drivers occasionally or often use the adjacent freeway mainline lane
next to the rightmost freeway mainline lane to force their way into the queue,
thus blocking an additional lane. During this regime, mainline speed and
capacity are significantly reduced.
Capacity Adjustment Factors

The effects of spillback vary by site and analysis period due to driver
behavior and site geometry. Data collection has shown drivers block the adjacent
lane at some sites, but not at others, regardless of the queue spillback length at a
given site.
For unblocked lanes adjacent to those completely or temporarily blocked, the
methodology uses a “friction factor” in the form of a capacity adjustment factor
CAFBL. This adjustment factor is applied only to segments where Regime 3 or
Regime 4 occur.
The values for CAFBL are equal to the CAFs given for incidents in Chapter 11,
Freeway Reliability Analysis (Exhibit 11-23), as there are no data currently
available to accurately assess friction impacts on capacity for this case. These
values may be conservative, because capacities during incidents may be also be
reduced due to rubbernecking and the presence of police vehicles. Exhibit 38-3
presents adjustment factors to be applied to determine the capacity of through
lanes adjacent to blocked lanes during queue spillback. This adjustment factor is
not applicable to 2-lane segments under Regime 4, as there are no unblocked lanes.

Exhibit 38-3 Directional Lanes 1 Queued Lane 2 Queued Lanes

Capacity Adjustment Factors 2 0.70 N/A
(CAFBL) for Through Lanes 3 0.74 0.51
Adjacent to Blocked Lanes 4 0.77 0.50
during Queue Spillback 5 0.81 0.67
6 0.85 0.75
7 0.88 0.80
8 0.89 0.84
A capacity adjustment factor CAFUP is applied to the queue influence area

(QIA) upstream of the back of the queue (Exhibit 38-4). Within this area,
additional turbulence exists due to increased lane changing, which results in a
reduction of capacity.
Exhibit 38-4
Queue Influence Area with
Increased Turbulence
The length of the QIA is estimated as function of the segment free-flow speed
Additional discussion on the
determination of the Queue (FFS), as shown in Exhibit 38-5. During undersaturated operations, drivers have
Influence Area (QIA) is adequate warnings about the presence of a ramp through signage and navigation
presented in Appendix A.
aids and can position themselves according to their destination. However, when
queue spillback occurs ,drivers can only detect a downstream queue visually and
therefore have less time to react, resulting in more aggressive lane changes and
additional turbulence.
Exhibit 38-5 Segment Free-Flow Speed (mi/h) Queue Influence Area (ft)
Length of Queue Influence 50 810
Area as a Function of the 55 900
Segment Free-Flow Speed 60 980
(FFS) 65 1,060
70 1,140
75 1,220
SPILLBACK IMPACT ON URBAN STREETS

Spillback onto urban streets may occur due to oversaturated conditions on
freeways. Exhibit 38-6(a) illustrates spillback at a signalized intersection, while
Exhibit 38-6(b) illustrates spillback at a roundabout. Using the procedures of
Chapter 13, Freeway Merge and Diverge Segments, or Chapter 12, Freeway
Weaving Segments, the analyst can determine whether oversaturated conditions
will occur for a given freeway segment during a given analysis period. This
chapter’s methodology provides an estimate of the discharge rate from the
intersection to the on-ramp during congested conditions, while also considering
any effects from ramp metering. Estimating this discharge rate is necessary to
estimate the resulting queue length along the on-ramp. If the ramp is metered,
the metering rate should be used instead.

Exhibit 38-6
Queue Spillback from an On-
Ramp into Urban Street
Intersections
(a) Signalized intersection spillback (b) Roundabout spillback
The on-ramp queue length also depends on the upstream demands. In the
example shown in Exhibit 38-6(a), three movements contribute to this demand:
northbound (NB) right, southbound (SB) left, and eastbound (EB) through. If the
NB right movement is very heavy or has the right-of-way for a significant
amount of time, the SB left movement may not have much opportunity to
contribute to the demand and may spill back upstream, affecting the adjacent SB
through movement as well as the upstream intersection. Thus, in the case of
signalized intersections, the relative contribution of demands to the queue length
will depend on the relative demands of these movements and the respective
signal timings and right-of-way allocation. The discharge rate of these upstream
intersection movements will depend on the on-ramp’s storage availability during
the respective signal phase. The analysis estimates the additional lost time due to
the presence of the downstream queue and adjusts the effective green of the
affected movements.
In the roundabout example shown in Exhibit 38-6(b), the same three
movements contribute to the on-ramp demand. However, in this case, the
movements have priority in the following order: (a) SB left, (b) EB through, and
(c) NB right. A high-priority movement with a heavy demand may constrain the
entry capacity of lower-priority movements, resulting in total throughput that is
lower than the sum of the three contributing movement demands.
LANE-BY-LANE ANALYSIS
Spillback affects each lane of a facility differently. For example, when
spillback occurs at a freeway off-ramp, the rightmost lanes of the freeway may be
blocked, while the leftmost lanes operate in free-flow conditions. Therefore, the
methodology estimates operating conditions by lane as well as by segment. The
lane-by-lane performance metrics are also used to obtain O-D–based travel times.
The lane-by-lane analysis provides lane flow ratios (LFRs) representing the The demand flow rates by lane
are estimated as a percentage
percentage of the entering demand by lane. The LFR is a function of the segment- of the segment demand.
wide volume-to-capacity (v/c) ratio and values are provided for each segment
type (basic, merge, diverge, and weaving). In addition, FFS, speed, and capacity
are estimated by lane. When the facility becomes oversaturated, speeds are
estimated using the method of Chapter 10, Freeway Facilities, which is based on
interactions between successive segments.

PERFORMANCE MEASUREMENT FOR NETWORKS AND O-D PAIRS

When evaluating network and O-D performance, it is necessary to have a
common performance measure across the different types of facilities forming the
network. Therefore, the methodology estimates travel time by segment and lane,
and aggregates these times for O-D pairs and for the network.
Chapter 16, Urban Street Facilities, provides tools for obtaining speeds for all
urban street segments, and these speeds are used in the network analysis
methodology. Chapter 10, Freeway Facilities Core Methodology, determines
operational performance based on the density and speed of each freeway
segment in the network. Each segment’s average travel time can be derived from
its average speed. The average travel time for the entire facility is then the sum of
the segments’ average travel times. However, the travel time for some O-D pairs
cannot be accurately obtained this way, as the travel path may predominantly or
exclusively use specific lanes. The speeds in these lanes could differ substantially
from the average segment speed, especially during congested conditions caused
by off-ramp bottlenecks. For example, motorists exiting at a congested off-ramp
will experience a much different segment travel time than motorists continuing
on the freeway using the leftmost lanes of the same segment.
Therefore, the O-D–based analysis along a freeway network incorporates:
• Prevailing speeds by individual lanes—A set of models has been developed
for estimating the speeds and capacities of each lane for each type of
freeway segment.
• Selected travel lanes for each O-D pair—The set of lanes used by an O-D pair
in every segment of the freeway facility is also necessary to calculate the
corresponding travel times within each segment. For every feasible O-D
pair, the set of lanes that may be selected are obtained and considered in
the travel time estimation.

3. METHODOLOGY
This chapter’s methodology provides tools for evaluating the performance of

networks consisting of freeway and urban street facilities. It also provides
methods to evaluate the interactions between freeway and urban street facilities
and to assess the impact of queue spillback if it occurs. The methodology is based
on lane-by-lane analysis for freeway facilities. For signalized and unsignalized
intersections, the methodology relies on lane group analysis, while for urban
street segments there is no differentiation between travel lanes. The methodology
provides travel times and speeds for the network, each segment, and by O-D pair.
HCM chapters that address segments and
SCOPE OF THE METHODOLOGY facilities are:
10. Freeway Facilities Core
The methodology builds on the freeway facility and urban street facility Methodology
methods, and therefore incorporates the scope and all aspects of those chapters’ 12. Basic Freeway and Multilane
Highway Segments
methodologies. This chapter’s methodology can evaluate interconnected freeway
13. Freeway Weaving Segments
facilities as well as interconnected freeway and urban street facilities. It can 14. Freeway Merge and Diverge
consider signalized intersections, two-way STOP-controlled intersections, all-way Segments
STOP-controlled intersections, and roundabouts, as well as a wide range of 16. Urban Street Facilities
18. Urban Street Segments
interchange ramp terminal configurations.
19. Signalized Intersections
20. Two-Way STOP-Controlled
Spatial and Temporal Limits Intersections
The spatial scope of the analysis is a function of the network to be studied, 21. All-Way STOP-Controlled
Intersections
the extent of congestion, and the specific O-D pairs of interest. The external links 22. Roundabouts
to the network should remain uncongested throughout the study period. 23. Ramp Terminals and Alternative
Intersections
The definition of analysis boundaries, in practical terms, follows the
guidance of Chapter 10, Freeway Facilities Core Methodology. The analysis’
spatial extent should be sufficiently long to fully contain the formation and
dissipation of all queues within the network. Similarly, the analysis’ temporal
scope must be compatible with the selected O-D pairs, the study period, and the
duration of congestion. The first and last analysis periods should be free of
congestion. The methodology can perform multi-period analysis when an O-D
travel time is longer than 15 minutes.
The methodology of this chapter generates the following performance
measures:
• Freeway facilities:
o Flow, free-flow speed (FFS), operating speed, and capacity for
individual lanes
o Expected travel speed along each segment
• Urban street facilities:
o Travel time along each segment
o Expected travel speed along each segment
Chapter 38/Network Analysis Methodology

• Networks:
o Total and free-flow travel times
o Travel time index (TTI)
o Average speed
Strengths of the Methodology

The strengths of the methodology include:
1. The methodology evaluates the effects of spillback from one facility to
another and considers the interactions between urban streets and
freeways.
2. The methodology evaluates oversaturated and undersaturated conditions
by lane, by segment, and for the entire network.
3. The methodology produces travel times and other performance measures
by O-D pair within the network.
4. The methodology tracks the formation and dissipation of queues across
lanes, segments, and facilities.
5. The methodology can be used to evaluate the impacts of modifications to
one facility on an adjacent facility.
Limitations
The methodology has the following limitations:
1. Multiple overlapping breakdowns or bottlenecks cannot be fully
evaluated by this methodology. Consult Chapter 6, HCM and Alternative
Analysis Tools, for a discussion of simulation and other models.
2. Demand is an input into the process, and the methodology does not
address any changes in demand that are due to traffic operation conditions.
3. Managed lanes can be analyzed as part of the freeway system. However,
the interaction of managed lane operations with spillback conditions are
not addressed.
4. The methodology does not explicitly consider alternative intersection and
interchange designs, such as diverging diamond interchanges (DDIs) and
single-point urban interchanges (SPUIs). However, it can be extended to
consider these designs, assuming turning movements, demands, and
queues can be accurately estimated for the movements of interest.
5. The methodology does not consider two-lane roundabouts and their
interaction with freeway on-ramps.
6. The HCM’s reliability methods cannot be applied to network analysis
because the process for developing reliability scenarios is different for
freeways and arterials.
Methodology Chapter 38/Network Analysis


General Input Data
A network analysis requires details concerning each freeway and urban
street segment’s geometric characteristics, as well as each segment’s demand
characteristics during each analysis time period. Exhibit 38-7 shows the data
inputs required for an operational analysis of a network, potential sources for
these data, and suggested default values.
Required Input Potential Data Source(s) Suggested Default Value Exhibit 38-7
Trajectory Parameters by O-D Pair Required Input Data, Potential
Data Sources, and Default
Origin and destination points Set by analyst Must be provided
Values for the Network
Route between origin and destination Set by analyst Must be provided
Analysis Methodology
Freeway Facilities
Input data for facility and segment As shown in Chapters 10, As shown in Chapters 10, 12,
methods 12, 13, and 14 13, and 14
Ramp access density
Road geometry Must be provided
(number of ramps within 1 mi)
Grade (%) Road geometry Must be provided
Urban Street Facilities
Input data for facility and segment As shown in Chapters 16 As shown in Chapters 16 and
methods and 18–23 18–23
Urban street segments: corresponding Set by analyst, according to
Must be provided
movement at downstream intersection the selected route
Off-Ramp Queue Spillback
Off-ramp queue spillback: Road geometry, Function of diverge geometry
expected number of queued lanes field observations and driver aggressiveness
Length of available shoulder (ft) Road geometry Must be provided
Off-ramp detailed geometry Road geometry Must be provided
On-Ramp Queue Spillback
On-ramp metering rate (veh/h)
Field data Must be provided
(if applicable)
On-ramp detailed geometry Road geometry Must be provided
Roundabouts: exit capacity (pc/h) Field data, past counts 1,300 pc/h
Off-Ramp Queue Spillback: Expected Number of Queued Lanes

If queue spillback from the off-ramp is expected to extend beyond the length
of the deceleration lane, the expected prevailing spillback regime (3 or 4) must be
provided by the analyst.
Field observations (1) have shown that locations experiencing recurring
queue spillback always have the same type of spillback regime when the queue
extends beyond the deceleration lane (Regime 3 or 4). Regime 4 occurs often at
ramp junctions with a lane drop. At these locations, the exiting traffic can access
the off-ramp with a single lane change. Therefore, drivers are more likely to wait
until they are closer to the exit to change lanes, blocking the adjacent through
lane. However, not all lane drop exits experience a Regime 4 queue spillback.
Generally, Regime 4 occurs more frequently in locations with more aggressive
driver behavior. Local information and driver behavior should be taken into
consideration in determining the prevailing regime at a given site.
For operational analyses of existing locations, it is recommended that the
analyst provide the expected spillback regime based on observed field
conditions. For planning-level purposes where no field data is available, Exhibit
38-8 provides the expected queue spillback regime as a function of the number of
exiting lanes and driver aggressiveness.

Exhibit 38-8 Driver Aggressiveness

Default Spillback Regimes as a Ramp Geometry Low Medium High
Function of Ramp Geometry
Diverge with deceleration lane Regime 3 Regime 3 Regime 3
and Driver Aggressiveness
Diverge with lane drop Regime 3 Regime 4 Regime 4
COMPUTATIONAL STEPS
This section describes the methodology’s computational steps. Exhibit 38-9
illustrates the process used to evaluate network operations.
Exhibit 38-9
Network Analysis Methodology
Flowchart

Step 1: Define Spatial and Temporal Analysis Scope

The first step in the analysis requires identifying the spatial and temporal
extent of the network to be evaluated. To accurately evaluate traffic operations, it
is essential that the spatial and temporal extent of congestion is contained within
the network. If an initial analysis determines that queues extend beyond the
limits of the network, the analysis area should be modified accordingly to
contain all congestion effects.
The analyst should also select the O-D pairs to be evaluated, if any, along
with the respective set of links to be traveled for each selected O-D pair. Exhibit
38-10 illustrates a sample network with six possible O-D nodes.
Exhibit 38-10
Sample Study Network, with
Multiple Origins and
Destinations
Step 2: Provide Input Parameters for Freeway and Urban Street

Analysis
The urban street and freeway facilities are first modeled separately using the
methodologies from their respective chapters. If multiple facilities of the same
type are to be analyzed (for example, two distinct urban street facilities), each
facility must first be modeled separately. The performance measures for each of
these facilities must also be computed at this step, as they are used next to
analyze the freeway–arterial interactions.
Step 3: Balance Demands at Freeway–Urban Street Interface

When urban street and freeway facilities are modeled independently, the
analyst is required to provide demand flow rate values for both facilities. In the
case of an interface between a freeway and an urban street, the demand flows
traveling through a freeway ramp and the demands at the ramp terminal should
be the same when conducting a network analysis.
The presence of any bottlenecks upstream of the freeway exit may reduce the
demand to the ramp junction. If the total off-ramp demand is greater than the
ramp roadway capacity, the intersection demand will be reduced accordingly.
Similarly, any movements operating above capacity at the ramp–urban street
junction would constrain the demand to the downstream freeway on-ramp.
This process must be performed for every time period in the analysis,
starting from the upstream end of the facility. When demand exceeds capacity at
any given location, the downstream demands must be recalculated considering
the throughput from the bottleneck.

Off-Ramp Demand vs. Downstream Ramp Terminal Demand

The turning movement demand at the ramp terminal downstream of an off-
ramp can be metered by insufficient capacity of the upstream freeway or of the
ramp roadway, as shown by bottlenecks 1 and 2, respectively, in Exhibit 38-11. If
these bottlenecks are not active, the sum of intersection demands vlt and vrt is
equal to the off-ramp demand vR.
Exhibit 38-11
Potential Bottlenecks
Constraining the Ramp
Terminal Demand
However, if the demand at the upstream freeway segment or at the off-ramp

exceeds its capacity, the flow that will reach the ramp terminal will be lower than
the off-ramp demand vR. In this case, one or both of the following adjustments are
performed.
Insufficient Capacity at a Bottleneck Freeway Segment

To balance demands, the off-ramp flow OFRF is first aggregated for a 15-
minute analysis period as follows:
𝑆
𝑇
Equation 38-1 𝑣𝑅,𝑎𝑑𝑗 = ∑𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝)
𝑆
𝑖=1
where
vR,adj = adjusted demand at the subject off-ramp (pc/h),
T = number of time steps in 1 h,
S = number of computational time steps in an analysis period
(typically 240 for time steps of 15 s), and
The parameter OFRF(i, t, p) is OFRF (i, t, p) = actual flow that can exit at off-ramp i during time step t in time
defined as the “actual flow that
can exit at off-ramp i during interval p.
time step t in time interval p”
(Chapter 25). It can account If the freeway facility operates under undersaturated conditions, the value of
for the effects of bottlenecks vR,adj is equal to the off-ramp demand vR.
upstream of the off-ramp that
can meter the traffic that If the subject freeway facility operates under oversaturated conditions, the
arrives to the ramp.
off-ramp demand may be metered at an upstream bottleneck segment. The
Oversaturated Segment Evaluation methodology in Chapter 25, Freeway and
Highway Segments: Supplemental, provides equations to estimate the off-ramp
flow parameter OFRF at every 15-s time step (Equations 25-23 through 25-25).
Insufficient Capacity at the Ramp Roadway

If the total demand at a freeway exit is greater than the capacity of the ramp
roadway cR, the flow that will reach the downstream ramp terminal will be
constrained by the ramp roadway capacity. For each movement 𝑖 at the
intersection, the adjusted demand is calculated as follows:

𝑣𝑅 𝑐𝑅
𝑣𝑖, 𝑎𝑑𝑗 = 𝑣𝑖 × × min ( , 1) Equation 38-2
∑𝑖 𝑣𝑖 𝑣𝑅,𝑎𝑑𝑗
where
vi,adj = adjusted demand for movement 𝑖 at the downstream intersection
(pc/h);
vi = demand for movement 𝑖 at the downstream intersection (pc/h);
vR = off-ramp demand (pc/h);
vR,adj = adjusted off-ramp demand, from Equation 38-1) (pc/h); and
cR = capacity of ramp roadway, from Exhibit 14-12 (pc/h).
On-Ramp Demand vs. Upstream Ramp Terminal Demand

At a freeway merge segment the on-ramp demand flow rate vR can be
constrained by the following bottlenecks:
1. Insufficient capacity of one or more movements in the ramp terminal, and
2. Insufficient capacity at the ramp roadway.
These potential bottlenecks are illustrated in Exhibit 38-12. If capacity is not
exceeded at any of those locations, the on-ramp demand vR is equal to the sum of
intersection demands that contribute to the on-ramp (in Exhibit 38-12, vNBR, vEBT,
and vSBL).
Exhibit 38-12
Potential Bottlenecks
Constraining the On-Ramp
Demand
However, if capacity is exceeded at any of those locations, the flow that will
reach the freeway merge will be lower than the on-ramp demand vR and
adjustments should be made to the respective volumes.
If any of the ramp terminal movements that discharge into the on-ramp
operate over capacity, the total throughput to the on-ramp is given by
𝑁
1
𝑣𝑅,𝑎𝑑𝑗 = × ∑ min(𝑣𝑖 , 𝑐𝑖 ) Equation 38-3
𝑓𝐻𝑉
1

where
To keep the units of vR,adj = adjusted on-ramp demand (pc/h);
intersection throughput
(veh/h) and freeway on-ramp
demand (pc/h) consistent,
vHV = heavy vehicle adjustment factor (decimal), from Equation 12-10;
Equation 38-3 applies the
heavy vehicle adjustment vi = demand for movement i at the intersection (veh/h);
factor given by Equation 12-10.
ci = capacity of movement i at the intersection (veh/h); and
N = number of intersection movements that discharge into the on-ramp.
If the total on-ramp demand vR is greater than the ramp roadway capacity cR,
the adjusted on-ramp demand is:
Equation 38-4 𝑣𝑅,𝑎𝑑𝑗 = min(𝑣𝑅 , 𝑐𝑅 )
where
vR,adj = adjusted on-ramp demand (pc/h);
vR = on-ramp demand (pc/h); and
cR = ramp roadway capacity (pc/h), from Exhibit 14-12.
Step 4A: Check for Queue Spillback (Off-Ramp)

During this step, the methodology evaluates the network to determine
whether queue spillback exists from a freeway off-ramp. The analysis is first
conducted using 15-min analysis periods (single-period or multi-period) to
determine whether queue spillback is expected to occur. If spillback is expected,
Step 5 will perform an analysis using 15-s time steps; if not, Step 5 will use 15-
min intervals.
Exhibit 38-13 summarizes the process for conducting a spillback check at off-
ramps. The process evaluates whether the spillback originates from the demand
to the ramp roadway, from the demand to the ramp junction at the surface street,
or from the downstream freeway on-ramp. Based on this determination, the
procedure uses the demand and the capacity for the analysis interval, along with
the previous queue length, to calculate the anticipated queue length for the
current interval. The detailed calculations for an off-ramp spillback check are
presented in Appendix A.

Exhibit 38-13
Spillback Check Procedure for
Off-Ramps
Step 4B: Check for Queue Spillback (On-Ramp)

Queue spillback into a surface street intersection (or upstream freeway facility)
can occur when the freeway merge segment has insufficient capacity to process
the ramp demand. During this step, the methodology evaluates the network to
determine whether queue spillback occurs from a freeway on-ramp onto
upstream facilities. Exhibit 38-14 illustrates the process for conducting a queue
spillback analysis at on-ramps.
When the freeway facility operates in oversaturated conditions (i.e., at least
one segment with LOS F), on-ramp queues are computed using the Freeway
Facility Oversaturated Segment evaluation procedure (Chapter 25), using the
parameter ONRQ (Equation 25-21). The parameter ONRQ(i, t, p) is defined as the
unmet demand, in vehicles, that is stored as a queue on the on-ramp roadway at
node i during time step t in time interval p.
Appendix B details the calculations used to estimate the on-ramp demand
based on the intersection operation, as well as the procedures for conducting the
on-ramp spillback check.

Exhibit 38-14
Spillback Check Procedure for
On-Ramps

Step 5A: Compute Operating Speeds for Individual Lanes Along the
Freeway Facility
The operational performance of a freeway facility is determined from the
density and speed of each segment along the facility. The average travel time for
a given segment can be derived from its average speed.
For a network analysis, the segment speed is function of:
• Estimated speeds for individual lanes, and
• Probability that a lane will be selected for the subject O-D pair.
To estimate the speeds and capacities for individual lanes, a set of models
have been developed for each type of freeway segment considering the total
number of mainline freeway lanes. These models are valid only for
undersaturated conditions, and they predict the lane flow ratio (LFR) for each
lane. These models are of the form:
𝐿𝐹𝑅𝑖 = 𝑎 × ln(𝑣/𝑐) + 𝑏 Equation 38-5
𝑁−1
𝐿𝐹𝑅𝑁 = 1 − ∑ 𝐿𝐹𝑅𝑖 Equation 38-6

1
where
a = multiplicative calibration parameter, from Equation 38-C3, Equation The full methodology to predict lane-by-
lane speeds on freeway facilities is given
38-C5, or Equation 38-C7;
in Appendix C.
b = additive calibration parameter, from Equation 38-C4, Equation 38-C6,
or Equation 38-C8;
LFRi = share of the total flow on lane 𝑖, where 𝑖 ranges from 1 to N – 1;
N = total number of lanes in the segment;
LFRN = share of the total flow on the leftmost lane (lane N); and
v/c = volume-to-capacity ratio (0 < v/c ≤ 1).
Using the calculated LFRs, the methodology next estimates the lane-by-lane
free-flow speeds and capacities. These values are used to obtain each lane’s
speed using the speed–flow model given by Equation 12-1.
The LFR models, their coefficients, and the procedures for estimating lane-
by-lane free-flow speeds, capacities, and speeds are provided in Appendix C.
These models can be used to analyze basic, merge, diverge, and weaving
segments with two to four lanes.
Freeway segments with 5 or more lanes were not modeled due to insufficient
data. Limited field observations for these facilities indicate that flow distributions
become more homogeneous in wider segments. Therefore, the flow distribution
for these segments can be estimated as:
𝑣
𝐿𝐹𝑅𝑖 = Equation 38-7
𝑁
where
LFRi = share of the total flow on lane 𝑖, where 𝑖 ranges from 1 to N – 1;
v = segment entering demand (pc/h); and
N = total number of lanes in the segment.

For all segment types, the share of flow is estimated on the mainline
upstream of the segment. The oversaturated portion of the speed–flow curve
(when density is greater than the density at capacity) cannot be addressed by the
speed–flow models, as this is a limitation of the existing methods. The lane-by-
lane flows for oversaturated conditions are estimated using the procedures of
Chapter 25, adjusted to determine the incoming and outgoing flow on a lane-by-
lane basis. However, if off-ramp queue spillback occurs in the freeway facility,
then the methodology in Appendix A is used to determine the lane-by-lane flow
distribution.
The probability that a given lane is selected when traveling between a given
O-D pair depends on the segment’s location. For segments where a motorist
driving between the O-D pair would enter (merge segment) or leave the freeway
facility (diverge segment), the probability of lane selection is shown in Exhibit 38-
15 (assuming right-side ramps).
Exhibit 38-15 Number of Lanes in the Segment

Probability of Lane Choice for Lane Choice Probability for Lane i 2 3 4+
Entry/Exit Segments on p1 0.90 0.90 0.90
Freeway Facilities p2 0.10 0.05 0.05
p3 — 0.05 0.05
p4+ — — 0.00
For other segments within the freeway facility, the probability pi,j of choosing
a given lane i on segment j is equal to the lane’s LFR, defined as the percentage of
the total flow assigned to lane i:
Equation 38-8 𝑝𝑖,𝑗 = 𝐿𝐹𝑅𝑖,𝑗
This concept is illustrated in Exhibit 38-16 for a 3-lane freeway facility with
nine segments. The exhibit shows the lane choice probabilities for the O-D pair
where the traveler enters the freeway facility on segment 2 (merge) and leaves
the freeway on segment 8 (diverge). For segments 2 and 8, the choice
probabilities for lanes 1, 2 and 3 are 0.90, 0.05 and 0,05 respectively, from Exhibit
38-15. For segments 3 through 8, the lane choice probabilities are equal to the
LFR (Equation 38-3), calculated for each lane of each segment.
Exhibit 38-16
Illustration of Lane Choice
Probabilities Along a Freeway
Facility
Each segment’s speed is then computed as the sum of products of speeds for
each lane and the corresponding probability of lane choice:
𝑁
Equation 38-9 𝑆𝑒,𝑗 = ∑(𝑝𝑖,𝑗 × 𝑠𝑖,𝑗 )

𝑖=1

where
Se,j = expected speed for segment j (mi/h),
N = total number of lanes in the segment,
pi,j = probability that lane i is selected (decimal) on segment j, and
Si = speed of lane i (mi/h), from Equation 38-C12.
A special case occurs when a weaving segment is defined as an entry or exit
segment. The methodology for estimating lane-by-lane speeds cannot evaluate
auxiliary lanes in weaving segments; therefore, the expected speed equation
should be replaced by the average speed of weaving vehicles in the segment.
Additional details are provided in Appendix C: Lane-by-Lane Analysis for
Freeway Facilities.
Step 5B: Compute Speeds for Urban Street Segments

For urban street facilities, the speed along each segment is calculated using
the procedures of Chapter 18, Urban Streets Segments. However, for the
intersection at the ramp junction, the control delay value for the corresponding
movement (typically a right- or left-turn movement into the on-ramp) must be
used in the analysis. If there is queue spillback from the on-ramp into the urban
street intersection, the increased control delay of the movement towards the on-
ramp is obtained using the methodology given in Appendix B.
Step 6: Compute Travel Times for Each Segment

This step calculates the travel times for each segment using the speeds
obtained in Steps 5A and 5B by dividing each segment’s length by its respective
speed:
𝐿𝑖
𝑇𝑇𝑖 = Equation 38-10
1.47 × 𝑆𝑖
where
TTi = travel time for segment i (s);
Li = length of segment i (mi); and
Si = speed for segment i (mi/h), depending on the facility type:
if a freeway, Si = Se (expected speed), from Equation 38-9; or
if an urban street, Si = St,seg (travel speed), from Equation 18-15.
Step 7: Compute Travel Time for Freeway Ramps

Ramp speeds can be obtained using the following equation:
𝑣𝑅
𝑆ramp = (1 − 0.109 × ) × 𝑆𝐹𝑅 Equation 38-11
1000
where
Sramp = ramp speed (mi/h),
vR = ramp demand flow rate (pc/h), and
SFR = ramp free-flow speed (mi/h).

The speed–flow relationship for ramps is linear and speed decreases with
higher ramp flows, as shown in Exhibit 38-17. The maximum allowed values of
vR are bounded by ramp capacity, consistent with guidance provided by Chapter
14, Freeway Merge and Diverge Segments (Exhibit 14-12).
Exhibit 38-17
Speed–Flow Curves for
Freeway Ramps
For ramps with lower free-flow

speeds, the threshold value of
45 pc/h/ln for density at
capacity is not feasible as it
would result in ramp capacity
values inconsistent with Exhibit
14-12. The value of density at
capacity increases as free-flow
speed decreases. These values
are used to estimate ramp
queue lengths when off-ramp
queue spillback occurs, as
discussed in Appendix A.
The travel time along freeway ramps is calculated by dividing the ramp
length by its respective speed. When an O-D pair includes an off-ramp, the
control delay for the corresponding movement at the at-grade intersection must
also be added to the off-ramp travel time. This calculation is consistent with the
urban street facility methods, where each segment’s travel time includes the
control delay of the corresponding movement at the downstream intersection.
For off-ramps:
𝐿𝑅
Equation 38-12 𝑇𝑇ramp = + 𝑑𝑖 + 5
1.47 × 𝑆ramp
For on-ramps:
𝐿𝑅
Equation 38-13 𝑇𝑇ramp = +5
1.47 × 𝑆ramp
The constant value of 5 in where
Equation 38-12 and Equation
38-13 accounts for delay due TTramp = ramp travel time (s),
to vehicle deceleration and
acceleration along off-ramps Sramp = ramp speed (mi/h),
and on-ramps, respectively.
LR = ramp length (ft), and
di = control delay at the downstream ramp terminal for the O-D pair’s
corresponding movement (s) (applicable for off-ramps only).
In the case of queue spillback, Appendix A describes the procedure for
estimating the ramp speed, which is similar to the Oversaturated Segment
Evaluation method described in Chapter 25.

If the ramp roadway is the bottleneck, the off-ramp flow will be constant
(equal to the ramp roadway capacity), with the prevailing density equal to the
ramp density at capacity. The ramp speed is then computed as equal to the ramp
free-flow speed.

If the downstream intersection is the bottleneck, queues will build at the

intersection and limit the number of vehicles that can exit the ramp roadway and
enter the intersection. As a result, the number of vehicles NV stored inside the
ramp roadway will increase until the limit value of jam density is reached. The
NV parameter for the ramp roadway is computed every time step (15 s) and then
aggregated to the 15-min analysis period to compute the ramp roadway’s
average density. Similarly, the flow through the ramp roadway is aggregated to
15-min analysis periods and the speed at the off-ramp is obtained through
Equation 38-A14 (similar to Equation 12-1):
𝑆 = 𝐹𝐹𝑆𝑎𝑑𝑗 𝑣𝑝 ≤ 𝐵𝑃
𝑐𝑎𝑑𝑗 Equation 38-14
𝑎
𝑆 = 𝐹𝐹𝑆𝑎𝑑𝑗 − 𝑐 𝐵𝑃 < 𝑣𝑝 ≤ 𝑐
𝑎
where
S = mean speed of a basic segment (mi/h),
FFSadj = adjusted free-flow speed (mi/h),
vp = adjusted 15-min demand flow rate (pc/h/ln),
BP = breakpoint (pc/h/ln),
cadj = adjusted segment capacity (pc/h/ln),
Dc = density at capacity, typically 45 pc/mi/ln, and
c = base segment capacity (pc/h/ln).

If the on-ramp demand is greater than the merge capacity or the throughput
of any active ramp metering, the number of vehicles stored in the on-ramp will
increase at every time step by the difference between the vehicles that are
discharged from the upstream intersection and the number of vehicles that are
discharged into the freeway. Similar to off-ramp bottlenecks that form due
constraints at a downstream intersection, on-ramp flow and density are
computed at every time step (15 s) and then aggregated to a 15-min analysis
period. The speed is then computed through Equation 38-A14.
Step 8: Compute Travel Times for the Network and Each O-D Pair
This step computes the total travel time TTO-D for the network as the sum of
the travel times over all segments along the route. For multiperiod analysis, it is
important to select the travel time for the correct analysis period at each segment,
as a long O-D route may encompass several analysis periods. Exhibit 38-18
presents a sample calculation for a facility with two analysis periods (15 min
each). The first segment in the O-D route is traversed during analysis period 1,
and the cumulative travel time column is updated with the respective value.
Subsequent segments follow the same procedure until the cumulative travel time
exceeds the length of the first analysis period (900 s). For the next segment in the
network, travel times from analysis period 2 are added to the cumulative travel
time column. This procedure is then repeated until the final segment is reached.

The total travel time is obtained as the last value in the cumulative travel time
column.
Exhibit 38-18 Segment Travel Time (s) Selected Active Cumulative

Sample Calculation of Total Analysis Analysis Travel Time Analysis Travel Time
Travel Time Using Multiperiod Segment ID Period 1 Period 2 (s) Period (s)
Analysis
1 34 28 34 1 34
2 26 29 26 1 60
3 73 86 73 1 133
4 345 390 345 1 478
5 185 195 185 1 663
6 310 359 310 1 973
7 240 240 240 2 1,213
8 120 122 122 2 1,335
9 20 18 18 2 1,353
10 45 53 53 2 1,406
Total travel time (s): 1,406
Note: Cells shaded in gray highlight the analysis period applicable to each segment along the O-D route.
Step 9: Compute Performance Measures for Segments

The last step in the methodology computes performance measures for each
segment in the network, using the methods of the respective segment chapters.
In addition, the mean travel time index TTImean,O-D for a specific O-D pair can be
calculated for each segment and for the network by dividing the total travel time
for the O-D pair TTO-D by the respective free-flow total time (Equation 38-14). The
free-flow travel time for the O-D pair TTFF,O-D can be obtained by repeating Steps
1 through 8 for free-flow conditions. Exhibit 38-19 provides guidance on key
input parameters to be considered for such analysis.
𝑇𝑇𝑂-𝐷
Equation 38-15 𝑇𝑇𝐼mean,𝑂-𝐷 =
𝑇𝑇𝐹𝐹,𝑂-𝐷
where
TTO-D = total travel time for a specific O-D pair (s), and
TTFF,O-D = free-flow travel time for a specific O-D pair (s).
Exhibit 38-19 Performance Measure Reference Parameter Input Value

Reference Input Values for Freeway Facilities
O-D Analysis under Free-Flow
Lane flow ratio (LFR) v/c 0.1
Conditions
Speed by lane Free-flow speed by lane FFSi Equation 38-C9
Urban Street Segments
Travel speed Running time Equation 18-7
Urban Street Intersections
Control delay: signalized intersections Demand-to-capacity ratio X 0
Control delay: TWSC intersections Movement demand vi 0
Control delay: AWSC intersections Demand-to-capacity ratio X 0
Control delay: roundabouts Demand-to-capacity ratio X 0
Freeway Ramps
Ramp speed Ramp free-flow speed Analyst input
Note: TWSC = two-way STOP-controlled, AWSC = all-way STOP-controlled.
The computation of free-flow performance measurements for different

facility types is discussed next.

Freeway Facilities
At free-flow, the speed of freeway segments is computed as being equal to
their free-flow speed. When a lane-by-lane analysis is applied, the methodology
computes the free-flow speed for each lane (Equation 38-C9).
Next, the probabilities of lane choice on each segment are calculated for each
segment. If the subject segment is a entry or exit segment (segments where the
driver on a particular O-D route enters or leaves the freeway facility, as
illustrated in Exhibit 38-16), the lane choice probabilities are obtained from
Exhibit 38-15. For other segments, the lane choice probability is equal to the LFR
(Equation 38-5). When calculating the LFR under free-flow conditions, a v/c value
of 0.1 is recommended to provide results consistent with field data. Due to the
logarithmic form of the LFR equation, using v/c = 0 is mathematically infeasible,
and very low v/c values yield unrealistic results.
Urban Street Segments

The travel speed along urban street segments (Equation 38-16, based on
Equation 18-15) is calculated as a function of the segment running time (Equation
38-17, based on Equation 18-7), as shown:
3,600𝐿
𝑆𝑇,𝑠𝑒𝑔 = Equation 38-16
5,280(𝑡𝑅 + 𝑑𝑡 )
𝑁𝑎𝑝
6.0 − 𝑙1 3,600𝐿 Equation 38-17
𝑡𝑅 = 𝑓𝑥 + 𝑓 + ∑ 𝑑𝑎𝑝,𝑖 + 𝑑other
0.0025𝐿 5,280 𝑆𝑓 𝑣
𝑖=1
where
ST,seg = travel speed of through vehicles for the segment (mi/h);
L = segment length (ft);
tR = segment running time (s);
dt = control delay at the downstream intersection (s);
l1 = start-up lost time (2 s if signalized, 2.5 s if STOP- or YIELD-controlled);
fx = control-type adjustment factor, from Equation 18-8;
Sf = segment free-flow speed (mi/h);
fv = proximity adjustment factor, from Equation 18-6;
Nap = number of influential access points along the segment (points), from
Equation 18-8;
dap,i = delay due to left and right turns into access point intersection i (s/veh);
and
dother = delay due to other sources along the segment (e.g., curb parking or
pedestrians) (s/veh).
As shown by Equation 38-17, the running time along an urban street segment
is not directly affected by variations in demand. Therefore, free-flow running
time is calculated according to Equation 38-17. The only parameter in the segment

travel speed that accounts for congestion is the control delay for the specific O-D
related movement at the downstream intersection, which is discussed next.
Urban Street Intersections

When intersections are analyzed as part of an urban street facility, the
computed control delay is taken into account when estimating the travel speed of
the upstream segment. Even at free-flow, intersections still experience a small
amount of delay intrinsic to their operation.
The control delay for a given lane at a signalized intersection is provided by
Equation 19-18, repeated here as Equation 38-18:
Equation 38-18 𝑑 = 𝑑1 + 𝑑2 + 𝑑3
where
d = control delay (s/veh);
d1 = uniform delay (s/veh), from Equation 19-19;
d2 = incremental delay (s/veh), from Equation 19-26; and
d3 = initial queue delay (s/veh), from Equation 19-44.
Under free-flow conditions, the values of d2 and d3 are equal to zero.
Therefore, the free-flow control delay is equal to the value of uniform delay (d1)
computed for a demand-to-capacity ratio X of 0.
Two-Way STOP-Controlled Intersections

The control delay d for two-way STOP-controlled (TWSC) intersections (Rank
2 through Rank 4 movements) is computed by Equation 20-64, repeated here as
Equation 38-19:
3,600 𝑣
3,600 𝑣𝑥 𝑣
2 ( )( 𝑥 )
𝑐 𝑐
− 1 + √(
𝑥 𝑚,𝑥 𝑚,𝑥
Equation 38-19 𝑑= + 900𝑇 − 1) + +5
𝑐𝑚,𝑥 𝑐𝑚,𝑥 𝑐𝑚,𝑥 450𝑇
[ ]
where
d = control delay (s/veh);
vx = flow rate for movement x (veh/h);
cm,x = capacity of movement x (veh/h), from Chapter 20; and
T = analysis time period (0.25 h for a 15-min period) (h).
Under free-flow conditions, the demand vx is set to zero, which allows
Equation 38-19 to be reduced to the following:
3,600
Equation 38-20 𝑑= +5
𝑐𝑚,𝑥

All-Way STOP-Controlled Intersections

The control delay for all-way STOP-controlled intersections is computed by
Equation 21-30, repeated here as Equation 38-21:
ℎ𝑑 𝑥
𝑑 = 𝑡𝑠 + 900𝑇 [𝑥 − 1 + √(𝑥 − 1)2 + ]+5 Equation 38-21
450𝑇
where
d = average control delay (s/veh);
x = vhd/3,600 = degree of utilization (unitless);
ts = service time (s), from Equation 21-29;
hd = departure headway (s), from Equation 21-28; and
T = length of analysis period (0.25 h for a 15-min period) (h).
Under free-flow conditions, the degree of utilization x is set to zero, which
allows Equation 38-21 to be reduced to the following:
𝑑 = 𝑡𝑠 + 5 Equation 38-22

The estimation of the service time ts requires an iterative and
computationally intensive procedure described in Chapter 21, All-Way STOP-
Controlled Intersections. It should be performed setting x to 0.
Roundabouts
The control delay for roundabouts is computed by Equation 22-17, repeated
here as Equation 38-23:
3,600
3,600 √(𝑥 ( 𝑐 )𝑥
𝑑= + 900𝑇 [𝑥 − 1 + 2
− 1) + ] + 5 × min[𝑥, 1] Equation 38-23
𝑐 450𝑇
where
d = average control delay (s/veh);
x = volume-to-capacity ratio of the subject lane;
c = capacity of the subject lane (veh/h), from Equation 22-14; and
T = time period (h) (T = 0.25 h for a 15-min analysis).
Similar to TWSC intersections, setting the volume-to-capacity ratio x to 0
reduces Equation 38-23 to a simpler form:
3,600
𝑑= Equation 38-24
𝑐𝑚,𝑥
Freeway Ramps
Freeway ramp speeds at free-flow are equal to the ramp free-flow speed SFR
provided by the analyst and do not require additional adjustments.

4. EXAMPLE PROBLEMS
This section presents four example problems (Exhibit 38-20) illustrating the
evaluation of networks and addressing several cases of spillback onto freeways
and onto urban street facilities.

List of Example Problems Application
Problem Description
1 O-D–based travel time estimation Operational analysis
2 On-ramp spillback check for different ramp terminal types Operational analysis
2a Signalized intersection Operational analysis
2b Two-way STOP-controlled intersection Operational analysis
2c All-way STOP-controlled intersection Operational analysis
3 Queue spillback analysis for a freeway-to-freeway ramp Operational analysis
4 On-ramp queue spillback into a single-lane roundabout Operational analysis
EXAMPLE PROBLEM 1: O-D–BASED TRAVEL TIME ESTIMATION

An 8.72-mi section of northbound (NB) I-75 in Gainesville, Florida is
evaluated to obtain selected O-D travel times. Four consecutive interchanges are
evaluated: (a) Williston Rd., (b) Archer Rd., (c) Newberry Rd., and (d) NW 39th
Ave., as shown in Exhibit 38-21.
Exhibit 38-21
Example Problem 1: Network
Interchanges, Intersections,
and O-D Points
Example Problems Chapter 38/Network Analysis

When a freeway facility is analyzed in isolation, each on-ramp is a unique

origin and each off-ramp is a unique destination. However, for a network
analysis approach, O-D routes must be expanded to include turning movements
at the urban street intersections. Each turning movement is part of a different
O-D route with a distinct travel time.
The subject freeway facility has three freeway lanes throughout its entire
length. Four on-ramps and four off-ramps are connected to surface streets.
Exhibit 38-22 provides a schematic representation of the freeway facility and its
19 component segments, with five possible origins (A and merge segments 4, 8,
13, and 18) and five possible destinations (J and diverge segments 2, 6, 11, and 16).
Exhibit 38-22
Example Problem 1:
Freeway Facility Segmentation
and O-D Entry and Exit Points
The analysis steps for evaluating this network are discussed below.
Step 1: Define Spatial and Temporal Analysis Scope

The first step is to select the network’s origin and destination nodes. For each
selected O-D pair, the methodology identifies the segments traversed and
estimates their travel times.
As shown in Exhibit 38-21, the network has 9 nodes, which produce the 72
O-D pairs shown in Exhibit 38-23. This example problem estimates the travel time
from Archer Rd. East (D) to NW 39th Ave. East (H), highlighted in Exhibit 38-21.
Destinations Exhibit 38-23

Origins A B C D E F G H J Example Problem 1:
A — A-B A-C A-D A-E A-F A-G A-H A-J O-D Matrix
B B-A — B-C B-D B-E B-F B-G B-H B-J
C C-A C-B — C-D C-E C-F C-G C-H C-J
D D-A D-B D-C — D-E D-F D-G D-H D-J
E E-A E-B E-C E-D — E-F E-G E-H E-J
F F-A F-B F-C F-D F-E — F-G F-H F-J
G G-A G-B G-C G-D G-E G-F — G-H G-J
H H-A H-B H-C H-D H-E H-F H-G — H-J
J J-A J-B J-C J-D J-E J-F J-G J-H —
The average total travel time between each O-D pair can be obtained by
adding the average travel times on each segment and ramp roadway traversed,
plus the average delay experienced at all intersections along the route.
The O-D route from node D to node H will traverse two urban street
facilities, as shown in Exhibit 38-24:
• Archer Rd. westbound (WB), consisting of two urban street segments and
two signalized intersections (SW 40th Blvd. and I-75 NB on-ramp); and
• NW 39th Ave. eastbound (EB), consisting of one urban street segment and
two signalized intersections (I-75 NB off-ramp and NW 95th Blvd).
Chapter 38/Network Analysis Example Problems

Exhibit 38-24
Example Problem 1:
Urban Street Facilities
(a) Urban Street Facility 1: Archer Rd. WB (b) Urban Street Facility 2: NW 39th Ave. EB
Notes: The movements whose control delay will be used in the analysis are indicated in parentheses.
WBR = westbound right turn, WBT = westbound through, NBR = northbound right turn, EBT = eastbound
through.
The O-D route also includes the freeway facility (I-75 NB), starting at
segment 8 and ending at segment 16, as shown in Exhibit 38-22. The on-ramp
and off-ramp at the boundary ends of the facility are also included in the travel
time evaluation.
Exhibit 38-25 lists the segments, ramps, and intersections traversed for the
route connecting O-D pair D-H.
Exhibit 38-25 Facility 1: Archer Rd WB Facility 2: I-75 NB Facility 3: NW 39th Ave.

Example Problem 1: Intersections Ramp Intersections
List of Intersections, Ramps, (Movements) Segments Junctions Segments (Movements) Segments
and Segments Traversed for 8, 9, 10, 11, I-75 NB to
O-D Pair D-H SW 40th Blvd. SW 37th Blvd. to Archer Rd. I-75 NB
12, 13, 14, NW 95th
(WB through) SW 40th Blvd. On-Ramp (NB right turn)
15, 16 Blvd.
I-75 NB SW 40th Blvd. to NW 39th Ave. NW 95th Blvd.
— —
(WB right turn) I-75 NB Off-Ramp (EB through)
The analysis’ temporal scope must also be defined. Given the short length of
the subject network, a single-period analysis will be performed initially. The
average travel time obtained will be compared to the 15-min analysis period
length; if the travel time is longer than 15 min, the study’s temporal scope will be
reevaluated.
Step 2: Provide Input Parameters for Freeway and Urban Street

Analysis
For this step, the facilities connecting the subject O-D pair must be modeled
individually using their respective HCM methods.
Freeway Facility: I-75 NB

The freeway facility was divided into 19 segments for capacity analysis and
modeled as described in Chapter 10, Freeway Facilities Core Methodology. The
detailed input data for each segment are presented in Exhibit 38-26.

Ramp Accel/ Ramp Exhibit 38-26

Segment Length Mainline Flow Grade Flow Rate Decel Lane Length Example Problem 1:
ID Type (ft) Rate (veh/h) (%) (veh/h) Length (ft) (ft) Input Data for Freeway
1 Basic 2,220 4,800 0 — — — Facility Analysis
2 Diverge 1,500 4,800 −2 480 800 900
3 Basic 990 4,320 0 — — —
4 Merge 1,500 4,240 0.5 580 1,124 1,000
5 Basic 1,600 4,900 3 — — —
6 Diverge 1,500 4,900 0 364 541 1,650
7 Basic 1,800 4,536 0 — — —
8 Merge 1,500 4,536 1.7 868 438 2,250
9 Basic 6,300 5,404 0 — — —
10 Basic 5,385 5,404 0 — — —
11 Diverge 1,500 5,404 −1 936 490 660
12 Basic 2,014 4,468 0 — — —
13 Merge 1,500 4,468 1.8 380 1,443 1,850
14 Basic 6,494 4,848 0 — — —
15 Basic 2,480 4,848 0 — — —
16 Diverge 1,500 4,848 1 960 377 2,380
17 Basic 1,000 3,888 0 — — —
18 Merge 1,500 3,888 −2.2 148 747 2,200
19 Basic 3,760 4,036 0 — — —
Additional input parameters for the freeway facility are as follows:

• Urban area
• 3 travel lanes in each direction
• Base FFS: 75.4 mi/h
• Ramp FFS: 35 mi/h
• Ramp side: right
• Right-side clearance: 10 ft
• Traffic composition: 2% trucks on both freeway and ramps
• Familiar facility users
Urban Street Facility 1: Archer Road Westbound

This facility contains two signalized intersections and two segments, as
shown in Exhibit 38-24(a). The corresponding input data for intersection and
segment analysis are presented in Exhibit 38-27 and Exhibit 38-28 respectively.

Example Problem 1:
Intersection Parameter L T R L T R L T R L T R
Input Data for Intersection
Archer Rd. @ Demand (veh/h) 320 2,064 — — 524 548 104 — 260 — — — Analysis – Archer Rd. WB
I-75 NB Phase split (s) 20 80 — 70 20 30 — — — — — —
Archer Rd. @ Demand (veh/h) 120 2,348 88 36 864 548 60 208 96 36 480 304
SW 40th Blvd. Phase split (s) 20 50 — 20 50 — 20 30 — 20 30 —
Note: L = left, T = through, R = right.

Exhibit 38-28 SW 40th Blvd. to SW 37th Blvd. to

Example Problem 1: Input Parameter I-75 WB SW 40th Blvd.
Input Data for Segment Segment length (ft) 530 1,288
Analysis – Archer Rd. WB Speed limit (mi/h) 45 45
Through lanes 3 3
Restrictive median length (ft) 0 0
Upstream intersection width (ft) 50 50
Curb proportion (%) 70 70
Base FFS (mi/h) 46.42 46.42
Running speed (mi/h) 32.24 41.37
Running time (s) 11.21 21.23
Percent of base FFS 50.84 52.04
Urban Street Facility 2: NW 39th Avenue Eastbound

This facility contains two signalized intersections and one segment, as shown
in Exhibit 38-24(b). The corresponding intersection and segment input data are
presented in Exhibit 38-29 and Exhibit 38-30, respectively.

Example Problem 1: Intersection Parameter L T R L T R L T R L T R
Input Data for Intersection
NW 39th @ Demand (veh/h) 72 1,416 — — 872 76 336 — 624 — — —
Analysis – NW 39th Ave. EB
I-75 NB Phase split (s) 20 80 — — 70 — 20 30 — — — —
NW 39th @ Demand (veh/h) 180 1,772 68 96 640 128 84 160 76 60 228 120
NW 95th Blvd. Phase split (s) 20 50 — 20 50 — 20 30 — 20 30 —
Exhibit 38-30 Input Parameter I-75 NB to NW 95th Blvd.

Example Problem 1: Segment length (ft) 510
Input Data for Segment Speed limit (mi/h) 45
Analysis – NW 39th Ave. EB Through lanes 2
Restrictive median length (ft) 0
Upstream intersection width (ft) 50
Curb proportion (%) 70
Base FFS (mi/h) 46.42
Running speed (mi/h) 31.53
Running time (s) 11.03
Percent of base FFS 58.38
Additional input parameters for Urban Street Facilities 1 and 2 are as follows:
• Base saturation flow rate: 1,900 veh/h/ln
• Traffic composition: 0% heavy vehicles
• Cycle length: 120 s
• Grade: 0%
• Arrival type: 3
• Speed limit: 45 mi/h
• Yellow change interval: 4 s
• Red clearance interval: 0 s
• No pedestrians

Step 3: Balance Demands at the Freeway–Urban Street Interface

After each facility along the O-D route is modeled individually, this step
checks the consistency of traffic flows at the interfaces between the urban street
and freeway facilities.
For each of the four on-ramps along the freeway facility, there are two
movements at the corresponding urban street intersection that contribute to the
on-ramp demand: the eastbound left-turn and the westbound right-turn. Exhibit
38-31 compares the demand volumes v at the intersection with their respective
movement capacities c, with the smaller of the two values added to the total on-
ramp demand vR. No movement operates with v/c > 1; therefore, no adjustments
are required and the on-ramp demands are equal to the sum of the turning
movement demands at the ramp intersection.
Parameter Exhibit 38-31

Demand v Capacity c min (v, c) Merge Demand Example Problem 1:
Intersection Movement (veh/h) (veh/h) v/c (veh/h) vR (veh/h) On-Ramp Demands Along the
Freeway Facility
Williston Rd. @ EBL 160 1,055 0.15 160
580
I-75 NB WBR 420 985 0.43 420
Archer Rd. @ EBL 320 935 0.34 320
868
I-75 NB WBR 548 1,037 0.53 548
Newberry Rd. EBL 216 862 0.25 216
380
@ I-75 NB WBR 164 1,163 0.14 164
NW 39th Ave. EBL 72 501 0.14 72
148
@ I-75 NB WBR 76 1,012 0.075 76
Note: EBL = eastbound left, WBR = westbound right.
Next, the off-ramp volumes are checked against the intersection turning
movement demands. The first check determines whether there are bottlenecks
along the freeway facility that may meter off-ramp demands. Exhibit 38-32
shows the estimated LOS for all 19 segments in the freeway facility. Since no
segment is oversaturated, the off-ramp demand is not metered, and no
adjustments are necessary.
Segment Number Exhibit 38-32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Example Problem 1:
Type B D B M B D B M B B D B M B B D B M B Freeway Segment LOS
LOS D D D D D D D E E E D D C D D D C C C
Note: Segment types: B = basic, M = merge, D = diverge.
The second check compares the off-ramp demands to the respective ramp
roadway capacity, as shown in Exhibit 38-33. Demand does not exceed capacity
for any of the ramps; therefore, no adjustments to the intersection volumes are
performed.
Off-Ramp Demand Ramp Ramp Roadway Ramp Exhibit 38-33

Segment (pc/h) Lanes Capacity (pc/h) v/c Example Problem 1:
2 480 1 2,000 0.24 Off-Ramp Demands Along the
6 364 1 2,000 0.18 Freeway Facility
11 936 1 2,000 0.47
16 960 2 4,000 0.24

Step 4: Check for Queue Spillback

The next step determines whether bottlenecks exist at the on- and off-ramps.
Off-Ramp Spillback Check

The procedure presented in Exhibit 38-10 is applied to each of the four off-
ramps in the freeway facility:
• Ramp roadway capacity: The off-ramp demand was previously compared to
ramp roadway capacity (Exhibit 38-33) and no capacity constraint was
detected.
• Queue length estimation: The back-of-queue length (95th percentile) for the
downstream terminals (signalized intersections) are obtained from the
Chapter 31, Signalized Intersections: Supplemental, methodology and are
presented in Exhibit 38-34. The 95th percentile queues are expected to be
shorter than the available turn bay lengths at the intersections, except for
the left-turn movement at Williston Rd. (freeway segment 2). Here, the
queue length will spill back into the ramp roadway, and the next check
will evaluate whether its storage is adequate.
Exhibit 38-34 Ramp 2: Ramp 6: Ramp 11: Ramp 16:

Example Problem 1: Off-Ramp Williston Rd. Archer Rd. Newberry Rd. NW 39th Ave.
Queue Length Estimation and Parameter LT RT LT RT LT RT LT RT
Queue Storage Checks Ramp length (ft) 900 1650 660 2,380
Number of ramp lanes 1 1 1 2
Upstream ramp lane L1 L1 L1 L2 L1
Turn bay length (ft) 250 210 480 480 800 800 1,260 1,200
Back-of-queue length Q95 (ft/ln) 689 21 120 363 223 363 193 482
Queue spillback length (ft) 439 — — — — — — —
Queue storage ratio (RQ) 0.49 — — — — — — —
Note: LT = left turn, RT = right turn.
• Queue storage ratio: Any queues exceeding the available turn bay length at
the intersection must be checked against the available storage along the
ramp roadway. For single-lane off-ramps, any queues upstream of the
intersection will share the same storage and must be aggregated. If a
ramp has two or more lanes, the analyst must determine how ramp lanes
are channelized relative to intersection approaches, based on the off-ramp
geometry. As shown in Exhibit 38-33, only the off-ramp at segment 16
(NW 39th Ave.) has two lanes—the leftmost ramp lane L2 is connected to
the left-turn movement, while the rightmost ramp lane L1 is connected to
the right-turn movement. In this step, the only movement that must be
evaluated is the left turn at Williston Rd. The queue length upstream of
the intersection is compared to the available ramp length, with a resulting
queue storage ratio RQ = 439 / 900 = 0.49 < 1.0. Therefore, spillback is not
expected to occur along the off-ramps.
On-Ramp Spillback Check

On-ramp queue spillback is expected to occur when a freeway merge segment
operates above capacity or when there is active ramp metering with a rate lower
than demand. As shown in Exhibit 38-32, no merge segments operate at LOS F
and no ramp metering is present; therefore, spillback is not expected to occur.

Step 5A: Obtain Speeds for Individual Lanes in the Freeway Facility
First, the flow distribution among freeway lanes must be determined for the
segments in the freeway facility. Using the estimated flow rates, lane speeds are
computed as shown in Exhibit 38-35. The highlighted rows (8 through 16)
represent the segments included in the O-D route and used to compute the
overall travel time. The rightmost lane is labeled Lane 1.
Segment Segment Lane Flow Ratio Lane Speed (mi/h) Exhibit 38-35
ID LOS Lane 1 Lane 2 Lane 3 Lane 1 Lane 2 Lane 3 Example Problem 1:
1 C 0.286 0.351 0.362 66.5 71.3 77.7 Flow Distribution and Speeds
2 C 0.338 0.319 0.343 56.7 73.1 77.5 for Freeway Segments
3 C 0.279 0.356 0.365 68.3 71.6 77.2
4 C 0.259 0.388 0.353 72.0 72.4 77.3
5 C 0.281 0.348 0.371 66.4 69.9 75.6
6 C 0.336 0.326 0.337 55.3 72.2 77.4
7 C 0.286 0.354 0.360 67.7 70.6 76.5
8 C 0.253 0.387 0.359 71.7 71.5 76.8
9 C 0.294 0.346 0.360 56.3 67.3 74.0
10 C 0.288 0.344 0.368 58.2 67.7 73.9
11 D 0.358 0.290 0.352 41.6 71.7 75.2
12 C 0.286 0.355 0.359 67.8 70.8 76.6
13 B 0.253 0.382 0.365 71.9 72.1 76.8
14 C 0.281 0.349 0.370 66.8 70.0 75.7
15 C 0.281 0.349 0.370 66.8 70.0 75.7
16 C 0.350 0.296 0.354 50.6 74.1 76.8
17 B 0.278 0.362 0.361 68.3 72.6 78.1
18 B 0.252 0.383 0.365 72.0 74.2 77.7
19 B 0.272 0.358 0.370 68.7 72.7 78.1
Note: Bold rows represent the segments included in the O-D route that are used to compute the total travel time.
The step by step calculations
to determine lane-by-lane
Next, the expected speed for each segment is computed as the sum of flows and speeds on segment
products of speeds for each lane and the corresponding probability of lane 16 (diverge) are presented in
Example 1 in Appendix C.
choice, as provided in Equation 38-9. The results are shown in Exhibit 38-36.
Lane Choice Probability Exhibit 38-36

for the Subject O-D Lane Speeds (mi/h) Example Problem 1:
Segment Expected
Estimated Speeds by Segment
ID Lane 1 Lane 2 Lane 3 Lane 1 Lane 2 Lane 3 Speed (mi/h)
Based on Lane Choice
8* 90.0% 5.0% 5.0% 71.7 71.5 76.8 71.9
Probability and Speeds
9 29.4% 34.6% 36.0% 55.2 68.7 75.8 67.3
10 28.8% 34.4% 36.8% 57.3 68.8 75.4 67.9
11 35.8% 29.0% 35.2% 41.6 71.7 75.2 62.2
12 28.6% 35.5% 35.9% 69.3 72.5 78.7 73.8
13 25.3% 38.2% 36.5% 71.9 72.1 76.8 73.8
14 28.1% 34.9% 37.0% 67.0 71.3 77.2 72.3
15 28.1% 34.9% 37.0% 67.0 71.3 77.2 72.3
16* 90.0% 5.0% 5.0% 50.6 74.1 76.8 53.1
Note: *Entry and exit segments require the mandatory use of the rightmost lane.
Even though the remaining segments’ travel times segments are not directly
used in calculating the O-D travel time, the entire facility must be analyzed, as
any existing bottleneck would affect the performance of other segments along the
facility. In this example, no segment operates at LOS F, and no queues develop at
the ramps connecting to urban streets.

Step 5B: Compute Travel Speeds for Urban Street Segments

The travel speeds for the urban street segments are calculated using the
core methodology of Chapter 16, Urban Street Facilities, adjusted to consider the
relevant turning movements. Exhibit 38-37 shows the three urban street
segments analyzed, with their associated movements at the intersection.
Exhibit 38-37 Input Parameters

Example Problem 1: Segment Base Running Segment Downstream Control Travel
Urban Street Segment Speeds Length FFS Speed Running Intersection Delay Speed
Facility Segment (ft) (mi/h) (mi/h) Time (s) Movement (s) (mi/h)
SW 40th Blvd. to
530 46.42 32.24 11.21 Right 7.6 19.21
Archer I-75 NB
Road WB SW 37th Blvd. to
1,288 46.42 41.37 21.23 Through 15.1 24.16
SW 40th Blvd.
NW 39th I-75 NB to
510 46.42 31.53 11.03 Through 26.2 9.34
St. EB NW 95th Blvd.
Note: FFS = free-flow speed, WB = westbound, NB = northbound, EB = eastbound.
Step 6: Compute Travel Times for Each Segment

The travel times for each segment on the urban street and freeway facilities
are computed by dividing the segment length by the travel speed (Exhibit 38-38
and Exhibit 38-39, respectively).
Exhibit 38-38 Travel Speed Length Travel

Example Problem 1: Urban Facility Segment (mi/h) (ft) Time (s)
Streets Segment Travel Times SW 40th Blvd. @ I-75 NB 19.21 530 18.8
Archer Rd. WB
SW 37th Blvd. @ SW 40th Blvd. 24.16 1,288 36.4
NW 39th Ave. EB I-75 NB @ NW 95th Blvd. 9.34 1,040 75.9
Exhibit 38-39 Segment ID Expected Speed (mi/h) Segment Length (ft) Travel Time (s)
Example Problem 1: Freeway 8 68.4 1,500 15.0
Segment Travel Times 9 66.5 6,300 64.6
10 67.2 5,385 54.6
11 64.4 1,500 15.9
12 72.0 2,014 19.1
13 73.8 1,500 13.9
14 71.2 6,494 62.2
15 71.2 2,480 23.7
16 53.1 1,500 19.3
Step 7: Obtain Travel Times for Freeway Ramps

As shown, the ramps to and from the freeway facility operate in a state of
undersaturated flow. Therefore, ramp roadway speeds can be estimated using
Equation 38-11, as shown in Exhibit 38-40. For the off-ramp at segment 16, the
control delay at the downstream ramp terminal is included in the computation of
the ramp total travel time.
Exhibit 38-40 Ramp Ramp

Example Problem 1: Ramp Ramp Ramp Roadway Control Delay Total
Ramp Roadway Travel Times Segment Flow FFS Speed Length Travel at Ramp Travel
ID (pc/h) (mi/h) (mi/h) (ft) Time (s) Terminal (s) Time (s)
8 886 35 31.6 2,250 48.5 — 48.5
16 980 35 31.3 1,200 26.2 96.6 122.8
Note: FFS = free-flow speed.

Step 8: Compute Travel Times for the Network and Each O-D
All segments within the subject O-D route (D-H) are sorted according to their
travel sequence and their respective travel times are listed as shown in Exhibit
38-41. The cumulative travel time for the O-D route must also be computed to
evaluate whether the network analysis is being contained correctly within the
temporal scope defined in Step 1. For this example, a single 15-min analysis
period was considered, for a total time of 900 s. Because the cumulative travel
time does not exceed this boundary value, all travel times obtained from analysis
period 1 are valid for the analysis.
Facility Travel Cumulative Exhibit 38-41

Time Travel Time Analysis Example Problem 1:
Type Name Segment Name or ID (s) (s) Period Cumulative Travel Time
Urban Archer SW 37th Blvd. to SW 40th Blvd. 18.8 40.5 1 Computation
Street 1 Rd. WB SW 40th Blvd. to I-75 NB 36.4 76.8 1
On-ramp to I-75 NB 48.5 125.3 1
8* 15.0 140.3 1
9 63.8 204.1 1
10 54.1 258.2 1
11 16.5 274.6 1
Freeway I-75 NB 12 18.6 293.2 1
13 13.9 307.1 1
14 61.3 368.4 1
15 23.4 391.8 1
16* 19.3 411.0 1
Off-ramp to NW 39th Ave. 122.8 533.8 1
Urban NW 39th
I-75 NB to NW 95th Blvd. 75.9 609.7 1
Street 2 Ave. EB
Total travel time (s) 609.7
Notes: *Segments that are the entry to or exit from the freeway facility for the subject O-D route.
WB = westbound, NB = northbound, EB = eastbound.
Step 9: Compute Performance Measures for Segments

Because no spillback occurred in the subject study period, the performance
measures obtained by the respective methods for each segment type are valid.

EXAMPLE PROBLEM 2: ON-RAMP SPILLBACK ANALYSIS

This example problem illustrates the application of the on-ramp spillback
methodology by evaluating operations at an interchange when queue spillback
originates from the on-ramp. The example problem has three parts, each
analyzing a different intersection type at the ramp terminal: signalized, two-way
STOP-controlled (TWSC), and all-way STOP-controlled (AWSC). The main
objective in each scenario is to determine the new control delay for the
movements affected by queue spillback. All other network parameters (freeway
design and traffic demand, and intersection demand) are kept the same between
scenarios.
Elements Common to All Scenarios

A network in Baton Rouge, LA consists of the following facilities:
• One freeway facility (I-10)
• One urban street facility (Acadian Thruway) with four signalized
intersections:
o Perkins Rd.
o Acadian Center Rd.
o I-10 WB ramps
o I-10 EB ramps
The subject network has three freeway lanes throughout its entire length.
One interchange connects the freeway to an urban street facility (Acadian
Thruway), as illustrated in Exhibit 38-42. The selected origin and destination
points for analysis are H and F, respectively, with the traveled segments
highlighted in red.
Exhibit 38-42
Example Problem 2: Network
Interchanges, Intersections,
and O-D Points

The freeway facility (I-10 EB) will be modeled according to the freeway
facilities core methodology (Chapter 10), while the ramp terminal will be
modeled according to its respective intersection methodology (Chapter 19, 20, or
21). First, a check will be performed to confirm the occurrence of queue spillback.
Next, the respective spillback analysis method will be applied to evaluate the
impacts of queue spillback on the capacity of each movement at the intersection.
These reduced capacities will be used to compute control delay values
considering queue spillback and the results will be compared to the delay values
without queue spillback.
Exhibit 38-43 provides a schematic representation of the freeway facility in
the eastbound direction. Segments 3 (merge) and 5 (diverge) connect the freeway
to the urban street facility (Acadian Thruway).
Exhibit 38-43
Example Problem 2: Freeway
Facility Segmentation and O-D
Entry and Exit Points
The urban street facility consists of four signalized intersections and three
segments, as shown in Exhibit 38-44. The on-ramp terminal being analyzed is the
I-10 EB Ramps intersection.
Exhibit 38-44
Example Problem 2:
Urban Street Facility

Example Problem 2A: Signalized Ramp Terminal

This scenario evaluates the impacts of queue spillback originating from the
I-10 EB on-ramp when the upstream ramp terminal is signalized.
Input Data
The geometry of the intersection connected to the I-10 EB on-ramp is shown
in Exhibit 38-45. Three movements lead onto the on-ramp:
• NB right-turn (NBR): One channelized, unsignalized right-turn lane
• SB left-turn (SBL): One exclusive left turn lane with a protected phase
• EB through (EBT): One through lane
Exhibit 38-45
Example Problem 2A:
Geometry: I-10 EB Ramps
Exhibit 38-46 presents the subject intersection’s phasing. The north–south

direction corresponds to the major street, while the minor streets correspond to
the freeway off- and on-ramps. The intersection has a leading left-turn phase
with a protected left-turn movement (SBL).
Exhibit 38-46
Example Problem 2A: Phasing
Sequence: I-10 EB Ramps
The demand volumes for each analysis period are presented in Exhibit 38-47.
Additional input data are summarized in Exhibit 38-48.

Eastbound Northbound Southbound Exhibit 38-47

Analysis Period L T R T R L T Example Problem 2A:
1 8 8 87 362 315 652 804 Demand Flow Rates (veh/h):
2 16 96 20 1,812 521 586 1,759 I-10 EB Ramps
3 16 96 20 271 630 1,071 717
4 8 24 28 845 80 463 201
Eastbound Northbound Southbound Exhibit 38-48

L T R T R L T Example Problem 2A: Other
General Information Input Data: I-10 EB Ramps
Base saturation flow rate, s0 (veh/h) 1,900 1,900 1,900 1,900 1,900 1,900 1,900
Arrival type, AT 3 3 3 3 3 3 3
Lane width, W (ft) 11 11 11 11 11 11 11
Heavy vehicle percentage, PHV (%) 5 5 5 5 5 5 5
Grade, Pg (%) 0 0 0
Speed limit (mi/h) 35 35 35 35 35 35 35
Phase Information
Maximum green, Gmax (s) 20 20 — 53 — 47 100
Yellow change interval, Y (s) 4.7 4.7 — 4.7 — 4.7 4.7
Red clearance interval, Rc (s) 1 1 — 1 — 1 1
Minimum green, Gmin (s) 5 5 — 15 — 5 15
Start-up lost time, lt (s) 2 2 — 2 — 2 2
Green extension, e (s) 2 2 — 2 — 2 2
Passage time, PT (s) 2 2 — 2 — 2 2
Recall mode Off Off — Off — Off Off
Dual entry No No — Yes — No Yes
Freeway Facility
The freeway facility (I-10 EB) is divided into seven segments, as shown in
Exhibit 38-49, where segment 3 (diverge) and segment 5 (merge) connect to the
subject signalized intersection (Acadian Thruway). Exhibit 38-50 summarizes the
facility’s geometric features.
Exhibit 38-49
Example Problem 2A:
Freeway Facility Segments

Exhibit 38-50 Segment Segment Grade Acceleration/Deceleration Ramp

Example Problem 2A: Freeway ID Type Length (ft) (%) Lane Length (ft) Length (ft)
Facility Geometric Features 1 Basic 5,280 0 — —
2 Diverge 1,500 0 800 1,139
3 Diverge 720 0 0 965
4 Basic 732 0 — —
5 Merge 1,000 0 1,000 924
6 Basic 1,200 0 — —
7 Basic 900 0 — —

The first step in the spillback check analysis is to determine the on-ramp
demand flow rates for each time period, based on the demands at the signalized
intersection. For each time period, the demand v and capacities c are compared
for each movement that flows into the on-ramp (EBT, NBR, and SBL). The
minimum value between demand and capacity for each movement is computed
and the merge demand vR is then computed as the sum of the three movements.
The capacities for protected movements (EBT and SBL) are computed for
each time period. Due to the actuated control operation, the green times for these
movements vary by time period; therefore the method uses the average green
time for each phase and for each time period. The NBR movement is
unsignalized and therefore no capacity estimation is provided. The movement’s
capacity is computed by calculating the maximum throughput during one cycle
and then aggregating to an hourly flow rate. During the phases when no
conflicting movements discharge into the on-ramp, the NBR maximum
throughput is computed as its respective saturation flow rate, considering the
applicable adjustment factors fRT (for right-turn movements) and fHV (for the
presence of heavy vehicles). During the transition time between consecutive
phases, the unsignalized turning movement’s throughput is also assumed to be
equal to its saturation flow rate. Therefore:
Equation 38-25 𝑠𝑁𝐵𝑅,𝐹𝐹 = 𝑠0,𝑁𝐵𝑅 × 𝑓𝑅𝑇 × 𝑓𝐻𝑉𝑔
where
sNBR,FF = saturation flow rate of the NBR movement during the phases with
no conflicting flows (veh/h/ln),
s0,NBR = base saturation flow rate (1,900 pc/h/ln),
fRT = adjustment factor for right-turn vehicle presence in a lane group, and
fHVg = adjustment factor for heavy vehicles and grade.
The adjustment factor for right-turn vehicle presence is computed using
Equation 19-13, reproduced here as Equation 38-26:
1 1
Equation 38-26 𝑓𝑅𝑇 = =
𝐸𝑇 1.18
where ET is the equivalent number of through cars for a protected right-turning
vehicle (1.18). The adjustment factor for heavy vehicles and grade is computed

100 − 0.78𝑃𝐻𝑉 − 0.31𝑃𝑔2

𝑓𝐻𝑉𝑔 =
100
100 − 0.78 × 5 − 0.31 × 02
𝑓𝐻𝑉𝑔 = = 0.961
100
where
PHV = percentage heavy vehicles in the corresponding movement group (%),
and
Pg = approach grade for the corresponding movement group (%).
The saturation flow rate is therefore
1
𝑠𝑁𝐵𝑅,𝐹𝐹 = 1,900 × × 0.961 = 1,547 veh/h
1.18
Because there are conflicting movements discharging into the on-ramp (for
example, a protected left turn), the NBR capacity is constrained as drivers yield
to the higher-priority movement. The estimated discharge flow rate for the NBR
movement with a conflicting protected flow vprot can be obtained from Equation
31-100, reproduced here as Equation 38-27:
𝑠𝑝 = Equation 38-27
1 − 𝑒 −𝑣𝑜 𝑡𝑓ℎ/3,600
where
sp = saturation flow rate of a permitted movement (veh/h/ln),
v0 = opposing demand flow rate (veh/h),
The computation of the permitted saturation flow rates must take into
consideration that the conflicting phase may have two distinct flow rates on
signalized intersection operation, as discussed in Chapter 31, Signalized
Intersections Supplemental:
• During the queue service time (gs) portion of the conflicting phase green,
the opposing movement flow rate is equal to its saturation flow rate; and
• During the green extension time (ge), the opposing movement flow rate is
equal to its arrival flow rate during the effective green (qg).
Exhibit 38-51 illustrates the calculation of the NBR capacity for a single cycle
during analysis period 1. For each active phase, the procedure identifies the
respective conflicting flow to the on-ramp along with its duration and flow rate.
The NBR saturation flow rate is then computed using Equation 38-27. The last
column computes the maximum number of vehicles that can be discharged
during each phase as the product of the NBR saturation flow rate and the phase
duration. Clearance times between consecutive phases are also taken into
consideration assuming that they have no conflicting flow rate to the on-ramp.

Exhibit 38-51 NBR NBR

Example Problem 2A: Conflicting Saturation Discharge
Calculation of NB Right Turn Conflicting Duration Flow Rate Flow Rate Volume
Capacity for a Single Cycle: Active Phase Flow (s) (veh/h) (veh/h) (veh)
Analysis Period 2 Φ1 (SBL) – gs,SBL sSBL 40.2 1,739 282 3.1
Φ 1 (SBL) – ge,SBL qg,SBL 3.7 128 1,282 1.3
Clearance time 1 — 5.7 — 1,547 2.5
Φ 2 (NBT) — 50.7 — 1,547 21.8
Clearance time 2 — 5.7 — 1,547 2.5
Φ 7 (EBT) – gs,EBT sEBT 6.3 1,811 263 0.5
Φ 7 (EBT) – ge,EBT qg,EBT 2.0 97.2 1,319 0.8
Clearance time 7 — 5.7 — 1,547 2.5
Total 120.0 34.8
Note: gs = queue service time, ge = green extension time, qg = arrival flow rate during effective green,
s = saturation flow rate, NBR = northbound right turn, SBL = southbound left turn, EBT = eastbound
through.
As shown, for a 120-s cycle, the capacity of the unsignalized NBR movement
is 34.8 vehicles. Aggregated to an hourly flow rate, the capacity is:
3,600
𝑐𝑁𝐵𝑅 = 34.8 × = 1,045 veh/h
120
Because of the actuated control operation, the discharge rate to the on-ramp
is different during each cycle and during each analysis period. Therefore, this
procedure must be repeated for every analysis period to compute the capacity of
the NBR unsignalized movement cNBR, as shown in Exhibit 38-52.
Exhibit 38-52 Analysis Northbound Right-Turn

Example Problem 2A: NBR Period Capacity (veh/h)
Capacity by Analysis Period 1 1,213
2 1,045
3 978
4 1,182
Exhibit 38-53 summarizes the calculations for this step. During analysis
period 3, the SBL movement operates over capacity (v/c = 1.56, highlighted in
red); therefore, its throughput to the ramp is constrained by its capacity of 685
veh/h. For all other movements and analysis periods, the throughput to the on-
ramp equals the demand because v/c < 1.
The calculated on-ramp demand is then provided as an input to the freeway
facility analysis (Exhibit 38-54). As shown, the ramp flow rates for the merge
segment (segment 5) are obtained from Exhibit 38-53 (highlighted in bold).
The results of the freeway facility analysis are provided in Exhibit 38-55.
Oversaturated conditions (LOS F) occur during analysis periods 2 and 3
(highlighted in red), therefore queueing may occur along the on-ramp.


Analysis Period Parameter EBT NBR SBL Example Problem 2A:
Calculation of the On-Ramp
Demand v (veh/h) 8 315 652
Demand vR Based on the
v/c 0.064 — 0.96 Intersection Operation
1 Capacity c (veh/h) 125 1,213 677
min (v, c) 8 315 652
Merge demand vR (veh/h) 975
v/c 0.768 — 0.93
2 Capacity c (veh/h) 125 1045 630
min (v, c) 96 521 586
Merge demand vR (veh/h) 1,203
Demand v (veh/h) 96 630 1,071
v/c 0.77 — 1.56
3 Capacity c (veh/h) 125 978 685
min (v, c) 96 630 685
v/c 0.39 — 0.62
4 Capacity c (veh/h) 62 1,182 746
min (v, c) 24 80 463
Note: NBR = northbound right turn, SBL = southbound left turn, EBT = eastbound through.
Time Period 1 Time Period 2 Time Period 3 Time Period 4 Exhibit 38-54
Mainline Ramp Mainline Ramp Mainline Ramp Mainline Ramp Example Problem 2A: Freeway
Flow Flow Flow Flow Flow Flow Flow Flow Facility Demand Inputs
Segment Rate Rate Rate Rate Rate Rate Rate Rate
ID (veh/h) (veh/h) (veh/h) (veh/h) (veh/h) (veh/h) (veh/h) (veh/h)
1 5,209 — 6,300 — 5,300 — 5,000 —
2 5,209 348 6,300 450 5,300 1,200 5,000 50
3 4,861 135 5,850 116 4,100 1,000 4,950 96
4 4,726 — 5,734 — 3,100 — 4,854 —
5 4,726 975 5,734 1,203 3,100 1,411 4,854 567
6 5,701 — 6,937 — 4,511 — 5,421 —
7 5,701 — 6,937 — 4,511 — 5,421 —
Segment ID and Type Exhibit 38-55

Analysis 1 2 3 4 5 6 7 Example Problem 2A:
Period Basic Diverge Diverge Basic Merge Basic Basic Freeway Facility LOS
1 D C D C D D D
2 E F F F F F E
3 D D F F F E E
4 D C C B C C C
The next step estimates the on-ramp queue length and compares the result to
the available queue storage length to determine whether spillback is expected to
occur. Exhibit 38-56 shows the expected on-ramp queues from the freeway
facility analysis. For each analysis period, the ramp storage ratio RQ is computed
by dividing the ramp queue by the available storage length (924 ft). During
analysis period 2, a queue is expected on the ramp, but it is not long enough to
cause queue spillback (RQ < 1). During analysis period 3, however, the on-ramp is
expected to have RQ = 2.31 (highlighted in red), which indicates that spillback will
occur at the intersection during this analysis period.

Exhibit 38-56 Analysis Ramp Flow Ramp Queue Ramp Ramp Storage Spillback
Example Problem 2A: Period Rate (veh/h) (veh) Queue (ft) Ratio (RQ) Expected?
Spillback Check: I-10 EB 1 975 0.0 0.0 0.00 No
On-Ramp 2 1,203 15.0 388.6 0.42 No
3 1,411 82.1 2,133.6 2.31 Yes
4 567 0.0 0.0 0.00 No
Because spillback will occur during at least one analysis period, its impacts
on the operation of the signalized intersection must be evaluated. The next
section illustrates the application of the methodology to evaluate spillback effects
at a signalized intersection.
Evaluation of Queue Spillback Impacts

The evaluation of queue spillback impacts on the signalized intersection
follows the procedure detailed in the methodology in Appendix B (Exhibit 38-
B5). Because this is a multiperiod analysis, the procedure must be applied for
every analysis period. In this example, analysis periods 2, 3, and 4 will be
evaluated. Analysis period 1 is not analyzed here since it does not have
oversaturated conditions.
Analysis Period 2
The procedure to evaluate queue spillback into intersections is applied for
analysis period 2, even though spillback is not expected to occur during this
analysis period. The application of the methodology is presented for this analysis
period to facilitate the understanding of the calculations.
Step 7A: Determine intersection throughput to on-ramp. The throughput
of movements into the on-ramp were determined previously as part of the queue
spillback check, as shown in Exhibit 38-53.
Step 7B: Obtain merging capacity with the freeway facilities method.
When the freeway facility operates in oversaturated conditions, the capacity of
the subject merge section may be constrained by the presence of queues along
the mainline. The Oversaturated Segment Evaluation procedure (Chapter 25)
computes the on-ramp queue ONRQ and on-ramp capacity ONRO every 15 s.
The merge capacity cmerge is then obtained by aggregating the ONRO parameter
into an hourly flow rate for each analysis period. Exhibit 38-57 shows the values
of ONRQ and ONRO over the study period (60 min), converted to hourly flow
rates.
Exhibit 38-57(a) compares the on-ramp capacity ONRO to the on-ramp
demand. During the first analysis period, undersaturated conditions exist along
the freeway, thus ONRO equals 2,000 pc/h, corresponding to the ramp roadway
capacity given in Exhibit 14-12, or 1,903 veh/h. During analysis periods 2 and 3,
oversaturated conditions occur and the on-ramp capacity drops to 5 pc per time
step, corresponding to 1,142 veh/h. During analysis period 4, the lower demand
along the freeway allows the mainline queue to clear within 4 time steps (60 s).
Therefore, during the first 60 s, the on-ramp capacity remains 1,142 veh/h. From
the fifth time step to the end of analysis period 4, there is no congestion at the
merge and therefore the on-ramp capacity is again 1,903 veh/h.

Exhibit 38-57(b) provides the on-ramp queue estimated by the Chapter 25

Oversaturated Segment Evaluation procedure. Because spillback is expected to
occur, an adjustment to the freeway facility oversaturated segment procedure is
necessary to account for the maximum ramp storage (35.5 veh). This value is the
upper boundary of the on-ramp queue length. At the end of analysis period 3,
the predicted on-ramp queue length would be 82 veh if there were no storage
constraints (shown as the black curve). The red curve represents the adjusted
queue profile for the on-ramp considering the maximum storage capacity. At the
start of analysis period 4, having an on-ramp queue of 35.5 veh instead of 82 veh
results in a shorter queue clearance time, with a slight positive impact on the
freeway performance. In other words, the intersection has a metering effect,
which may improve freeway operations. Exhibit 38-58 compares the performance
results of the freeway segments downstream of the merge (see Exhibit 38-49)
with and without consideration of the maximum storage constraint.
Exhibit 38-57
Example Problem 2A:
Freeway Segment 5 Merge
Capacity and Queue Lengths
(a) Merge Capacity
(b) Queue Length
Segment 5 (Merge) Segment 6 (Basic) Segment 7 (Basic) Exhibit 38-58

Without With Without With Without With Example Problem 2A: Freeway
Storage Storage Storage Storage Storage Storage Performance During Analysis
Constraint Constraint Constraint Constraint Constraint Constraint Period 4 with and without the
Speed (mi/h) 67.2 67.4 67.7 67.8 72.2 72.5 Queue Storage Constraint
Density (pc/mi/ln) 20.9 19.9 20.8 19.7 19.5 18.4

Step 7C: Plot queue accumulation polygons (QAPs) for the on-ramp and
unsignalized movements. In this step, a QAP is plotted for the on-ramp as a
function of all protected and permitted movements entering the on-ramp, on a
cycle-by-cycle basis. Because an unsignalized movement (NBR) also discharges
into the on-ramp, a QAP must be developed for this movement as well. The
latter QAP is required to: (a) determine the discharge pattern of the unsignalized
movement throughout the cycle, and (b) allow the estimation of control delay for
this movement.
Exhibit 38-59 presents the QAPs for analysis period 2 for both the on-ramp
and the NBR movement.
Exhibit 38-59
Example Problem 2A:
Estimated Queue Lengths and
Merge Capacities During
Analysis Period 2
The cycle starts with a permitted left-turn movement (Φ1: SBL) discharging
into the on-ramp with a green time g1 = 43.9 s, divided in a queue service time gs1
= 40.2 s and a queue extension time ge1 = 3.7s (as defined in Chapter 31). During
the green interval for SBL, the capacity of the NBR movement is constrained
because drivers must yield to the protected left-turn vehicles. The saturation flow
rate for the NBR movement with a conflicting flow vSBL can be estimated from
Equation 38-27:
𝑠𝑝 =
1 − 𝑒 −𝑣𝑜 𝑡𝑓ℎ/3,600

𝜆𝑆𝐵𝐿 𝑒 −𝜆𝑆𝐵𝐿 𝑡𝑐𝑔/3,600

𝑠𝑁𝐵𝑅,𝑝𝑒𝑟𝑚 =
1 − 𝑒 −𝜆𝑆𝐵𝐿 𝑡𝑓ℎ/3,600
where
sNBR,perm = saturation flow rate of the NBR movement (veh/h/ln),
λSBL = throughput of the opposing SBL movement (veh/h),
tcg = critical headway = 4.5 s, and
tfh = follow-up headway = 2.5 s.
The saturation flow rates of the NBR movement during Φ1 are determined
next. During the SBL queue service time, the saturation flow rate is:
𝜆𝑆𝐵𝐿 = 𝑠𝑆𝐵𝐿 = 1,739 𝑣𝑒ℎ/ℎ/𝑙𝑛 → 𝑠𝑁𝐵𝑅,𝑝𝑒𝑟𝑚1 = 282 veh/h/ln
where sSBL is the saturation flow rate of the SBL movement (veh/h/ln) and sNBR,perm1
is the saturation flow rate of the NBR movement during the SBL queue service
time (veh/h/ln).
The throughput for the NBR movement is obtained as the minimum of the
demand and the saturation flow rate. Because the demand flow rate is greater
than the saturation flow rate, a queue will develop for the NBR movement:
𝜆𝑁𝐵𝑅,1 = min(𝑠𝑁𝐵𝑅,𝑝𝑒𝑟𝑚1 , 𝑣𝑁𝐵𝑅 ) = min(282, 521)
𝜆𝑁𝐵𝑅,1 = 282 veh/h
where λNBR,1 is the throughput for the NBR movement during the SBL queue
service time (veh/h/ln) and vNBR is the demand flow rate of the NBR movement
(veh/h).
During the SBL green extension time ge, the SBL throughput λSBL is equal to
the arrival flow rate during the effective green (qg,SBL, from Equation 19-32):
𝑣𝑆𝐵𝐿 𝐶
𝜆𝑆𝐵𝐿 = 𝑞𝑔,𝑆𝐵𝐿 = 𝑃𝑆𝐵𝐿 × ×
3,600 𝑔𝑆𝐵𝐿
586 120
𝜆𝑆𝐵𝐿 = 0.08 × × = 0.0356 veh/s/ln = 128 veh/h/ln
3,600 43.9
where
VSBL = SBL demand flow rate (veh/h),
C = cycle time (s), and
gSBL = SBL effective green time (s).
For this conflicting flow, therefore, the NBR saturation flow rate sNBR,perm2 is
obtained using Equation 38-27:
𝜆𝑆𝐵𝐿 𝑒 −𝜆𝑆𝐵𝐿 𝑡𝑐𝑔/3,600
𝑠𝑁𝐵𝑅,𝑝𝑒𝑟𝑚2 =
1 − 𝑒 −𝜆𝑆𝐵𝐿 𝑡𝑓ℎ/3,600
586 120
𝜆𝑆𝐵𝐿 = 0.08 × × = 0.0356 veh/s/ln = 128 veh/h/ln
3600 43.9
with all variables previously defined.

Because a queue is present in the NBR movement, the throughput for the
NBR movement is equal to its saturation flow rate:
𝜆𝑁𝐵𝑅,2 = 𝑠𝑁𝐵𝑅,𝑝𝑒𝑟𝑚2 = 1,282 veh/h
where λNBR,2 is the throughput for the NBR movement during the SBL green
extension (veh/h/ln) and sNBR,perm2 is the saturation flow rate of the NBR
movement during the SBL green extension time (veh/h/ln).
With the discharge patterns for the NBR determined, the on-ramp’s queue
profile during Φ1 can be determined. During the SBL queue service time (cycle
time t = 0 to t = 40.2 s), the throughput to the on-ramp is given by:
𝜆𝑂𝑁𝑅 = 𝜆𝑆𝐵𝐿 + 𝜆𝑁𝐵𝑅,1 = 1,739 + 282 = 2,021 veh/h = 0.561 veh/s
Given that the merge capacity cmerge is 1,142 veh/h for the current analysis
period, the on-ramp queue will grow at the following rate during the SBL queue
service time:
𝜆𝑂𝑁𝑅 − 𝑐merge = 2,021 − 1,142 = 879 veh/h = 0.244 veh/s
Therefore, at the end of the SBL queue service time (t = 40.2s), the queue at
the on-ramp will be 0.244 x 40.2 = 9.8 vehicles (Exhibit 38-59a).
This process is then repeated for all phases throughout the cycle. The results
for a single cycle (120 s) are presented in Exhibit 38-60, where the maximum on-
ramp queue occurs at t = 50.48 s, with 10.82 veh. The expected on-ramp queue at
the end of the cycle is 2.02 veh. The remaining cycles within analysis period 2
show the same pattern, where the on-ramp queue at the end of each cycle
becomes the initial queue at the start of the next cycle.
Each row in Exhibit 38-60 describes a portion of the cycle, as follows:
• gs1 is the queue service time for SBL (Φ1), as previously discussed.
• ge1 is the green extension time for SBL (Φ1). The NBR movement
discharges at the permitted saturation flow rate due to the queue that has
developed during gs1 and the on-ramp queue grows at a rate of 0.07 veh/s.
• r1 is the effective red time for SBL (Φ1). There is no throughput from
protected movements and the NBR movement discharges freely at the
saturation flow rate. The on-ramp queue grows at a rate of 0.11 veh/s.
• g2* is the effective green for NBT (Φ2), with no throughput from protected
movements. The duration of 0.88 s is calculated based on the queue
service time of the NBR approach. The on-ramp queue grows at a rate of
0.11 veh/s.
• g2** is the remaining effective green for NBT (Φ2). For this portion, no
queue remains on the NBR approach, therefore the NBR throughput is
equal to its demand flow rate vNBR. The on-ramp queue discharges at a
rate of 0.17 veh/s.
• r2 is the effective red time for NBT (Φ2). There is no throughput from
protected movements and the NBR throughput is equal to its demand
flow rate vNBR. The on-ramp queue discharges at a rate of 0.17 veh/s.

• gs7 is the queue service time for EBT (Φ7). The EBT discharges into the on-
ramp at the saturation flow rate. The throughput of the NBR movement is
restricted to the permitted saturation flow rate, causing queues to develop
in the NBR approach. The on-ramp queue grows at a rate of 0.26 veh/s.
• ge7* is the green extension time for EBT (Φ7). The duration of 0.03 s is
calculated based on the queue service time of the NBR approach. The
NBR movement discharges at the permitted saturation flow rate. The on-
ramp queue grows at a rate of 0.08 veh/s.
• ge7** is the remaining extension time for EBT (Φ7). The EBT movement
discharges at a rate equal to its arrival flow rate during the effective
green. For this portion, no queue remains on the NBR approach, therefore
the NBR throughput is equal to its demand flow rate vNBR. The on-ramp
queue discharges at a rate of 0.15 veh/s.
• r7 is the effective red time for EBT (Φ7). There is no throughput from
protected movements and the NBR throughput is equal to its demand
flow rate vNBR. The on-ramp queue discharges at a rate of 0.17 veh/s.
Protected Exhibit 38-60

Movement Permitted Movement On-Ramp Analysis Example Problem 2A:
λONR − On-Ramp Discharge Flow Rates into the
NBR
On-Ramp for Each Phase
Active Duration λprot vNBR λNBR Queue λONR cmerge Queue
Throughout the Cycle During
Phase t (s) (s) (veh/s) (veh/s) (veh/s) (veh) (veh/s) (veh/s) (veh)
Analysis Period 2
gs1 0.00 40.16 0.483 0.145 0.078 0.00 0.56 0.24 0.00
ge1 40.16 3.74 0.036 0.145 0.356 2.66 0.39 0.07 9.80
r1 43.90 5.70 0.000 0.145 0.430 1.87 0.43 0.11 10.08
g2* 49.60 0.88 0.000 0.145 0.430 0.25 0.43 0.11 10.72
g2** 50.48 49.82 0.000 0.145 0.145 0.00 0.14 −0.17 10.82
r2 100.30 5.70 0.000 0.145 0.145 0.00 0.14 −0.17 2.22
gs7 106.00 6.25 0.503 0.145 0.073 0.00 0.58 0.26 1.24
ge7* 112.25 2.02 0.027 0.145 0.366 0.45 0.39 0.08 2.85
ge7** 114.27 0.03 0.027 0.145 0.145 0.00 0.17 −0.15 3.01
r7 114.30 5.70 0.000 0.145 0.145 0.00 0.14 −0.17 3.01
Cycle
120 2.02
end
At the end of the analysis period, a residual queue of 23.32 veh is expected
along the on-ramp, and this value is carried to the start of the next analysis
period. The analysis period length of 900 s does not correspond to an exact
number of signal cycles, and the last cycle is interrupted at t = 60 s. Therefore, the
next analysis period will start the analysis from the same timestamp to maintain
consistency.
Step 7D: Calculate equivalent capacities for the affected movements.
Because spillback does not occur during analysis period 2, no adjustment to the
intersection capacity is necessary.
Analysis Period 3
The same analysis steps performed for analysis period 2 are applied again
for analysis period 3.


for movements that discharge into the on-ramp were previously determined as
part of the queue spillback check and were shown in Exhibit 38-53.
Step 7B: Obtain merging capacity with the freeway facilities method. As in
the analysis of the previous analysis period, the merging capacity cmerge is
obtained as an output from the freeway facility method (Exhibit 38-57a). The
merging capacity for analysis period 3 is 1,142 veh/h.
Step 7C: Plot QAPs for the on-ramp and unsignalized movements. The
procedure described earlier is applied again, using an initial on-ramp queue of
23.32 veh, which was the estimated queue at the end of analysis period 2. The
analysis begins at the middle of the cycle (t = 60 s), which was the end of the
previous analysis period. Exhibit 38-61 illustrates the QAPs for both the on-ramp
and the NBR movement.
Exhibit 38-61
Example Problem 2A:
Analysis Period 3
Queue spillback occurs during the third cycle (SBL queue service time),
when the on-ramp queue reaches the maximum storage LONR of 35.5 veh. At this
time, the maximum flow rate that can enter the on-ramp is constrained by the
merge capacity cmerge. In other words, the maximum number of vehicles allowed
to enter the ramp is equal to the number of vehicles that are able to merge to the
freeway mainline. In addition, the queues developed in the NBR are longer
during cycles 3 through 8, causing an increased delay for this movement due to
the queue spillback conditions at the on-ramp.

The on-ramp queue at the start of cycle 3 is 27.9 veh. The cycle starts with the
SBL movement, with an effective green time g1 of 47.3 s. Because this movement
already operates with v/c > 1, the queue service time gs1 is equal to g1, and no
green extension time is available (ge1 = 0). The protected movement then
discharges at a saturation flow rate sSBL of 0.483 veh/s, while the NBR movement
discharges at a permitted saturation flow rate sNBR of 0.078 veh/s. At the same
time, the on-ramp discharges to the freeway at a rate cmerge of 1,142 veh/h, equal
to 0.317 veh/s. Therefore, the on-ramp queue grows at the following rate:
𝜆𝑂𝑁𝑅 − 𝑐merge = (0.483 + 0.078) − 0.317 = 0.244 veh/s
The time remaining until spillback occurs is calculated by dividing the
remaining on-ramp queue storage by the queue growth rate:
35.5 − 27.9
Time to spillback = = 31.2 s
0.244
Spillback is expected to occur within 31.2 s of the onset of g1. The total
effective green g1 value of 47.3 s is then divided into two portions:
• gs1* (31.2 s), discharging at the saturation flow rate; and
• gs1,sp (16.1s), the remaining time that is affected by queue spillback,
limiting the maximum discharge to the on-ramp to the merge capacity
cmerge of 0.317 veh/s. Note that this constraint is shared by two movements
entering the on-ramp (SBL and NBR).
The effect of queue spillback on the intersection capacity during gs1,sp is then
measured by the capacity reduction factor β1,sp, defined as the ratio between the
maximum on-ramp capacity during queue spillback and the throughput from
the intersection movements (SBL and NBR):
𝑐merge 0.317
𝛽1,𝑠𝑝 = = = 0.565
𝜆𝑆𝐵𝐿 + 𝜆𝑁𝐵𝑅 (0.483 + 0.078)
A capacity reduction of 0.565 means that only 56.5% of the expected
intersection throughput is able to enter the on-ramp when queue spillback occurs
during phase gs1,sp. The capacity adjustment factor is applied to each movement
to obtain their adjusted throughputs for this analysis period:
𝜆𝑆𝐵𝐿,𝑎𝑑𝑗 = 𝜆𝑆𝐵𝐿 × 𝛽1,𝑠𝑝 = 0.483 × 0.565 = 0.273 veh/s
𝜆𝑁𝐵𝑅,𝑎𝑑𝑗 = 𝜆𝑁𝐵𝑅 × 𝛽1,𝑠𝑝 = 0.078 × 0.565 = 0.044 veh/s
The procedure is then repeated for the remaining movements of the cycle, as

Exhibit 38-62 Protected

Example Problem 2A: Movement Permitted Movement On-Ramp Analysis
Discharge Flow Rates into the λONR,adj −
On-Ramp for Each Phase Active Duration QONR λprot vNBR λNBR Queue λONR λONR,adj cmerge
Throughout the Cycle During Phase t (s) (s) (veh) (veh/s) (veh/s) (veh/s) (veh) (veh/s) (veh/s) (veh/s) βsp
Analysis Period 3 gs1* 0.0 31.2 27.9 0.483 0.175 0.078 0.00 0.561 0.561 0.244 1
gs1,sp 31.2 16.1 35.5 0.483 0.175 0.078 3.01 0.561 0.317 0.000 0.565
r1 47.3 5.7 35.5 0.000 0.175 0.430 5.12 0.430 0.317 0.000 0.739
g2* 53.0 30.3 35.5 0.000 0.175 0.430 4.31 0.430 0.317 0.000 0.739
g2** 83.3 17.0 35.4 0.000 0.175 0.175 0.00 0.175 0.175 −0.142 1
r2 100.3 5.7 33.1 0.000 0.175 0.175 0.00 0.175 0.175 −0.142 1
gs7 106.0 6.3 32.3 0.503 0.175 0.073 0.00 0.576 0.576 0.259 1
ge7 112.3 2.0 33.9 0.027 0.175 0.366 0.64 0.393 0.393 0.076 1
r7* 114.3 1.0 34.1 0.000 0.175 0.430 0.25 0.430 0.430 0.113 1
r7** 115.3 4.7 34.2 0.000 0.175 0.175 0.00 0.175 0.175 −0.142 1
Cycle
120 33.5
end
As shown, at time t = 31.2 s, the maximum storage length of the on-ramp is

reached and spillback occurs. From this time through t = 83.3 s, the throughput
from intersection movements to the on-ramp λONR is greater than the merge
capacity cmerge. Therefore, the maximum allowed throughput λONR,ajd is
constrained by the on-ramp discharge capacity cmerge of 0.137 veh/s. For these
cases, the spillback capacity reduction factor fsp is computed as the ratio of λONR,ajd
and λONR. Note that during this time range, the on-ramp queue is constant at the
maximum storage of 35.5 veh.
From t = 83.3 s, the on-ramp queue begins to discharge at a rate of 0.142
veh/s, followed by a small increase during the green time of phase 7 (EBT) that is
insufficient to cause spillback. At the end of the cycle, the residual on-ramp
queue is 33.5 veh.
The subsequent cycles follow a recurring pattern, with the on-ramp reaching
maximum storage early in the cycle and the queue diminishing slightly at the
end of the cycle.
Step 7D: Calculate adjusted capacities for the affected movements. The
adjusted capacities of the affected movements are estimated based on the volume
of vehicles that can actually be discharged during each analysis period. Exhibit
38-63 shows the calculation of the SBL movement’s adjusted capacity during
analysis period 3. The table lists all occurrences of green times for the SBL
movement during the analysis time period and their respective durations. For
each row, the expected throughput from the intersection λONR and the actual
throughput λONR,adj are computed. Next, the capacity reduction factor βsp is
computed as the ratio of λONR and λONR,adj. A βsp value < 1.0 indicates the
occurrence of queue spillback in the subject phase. The expected and actual
discharge volumes are obtained by multiplying the values of λONR and λONR,adj,
respectively, by their duration. At the end of the table, the expected and actual
volumes are aggregated and a capacity reduction factor βsp,SBL of 0.704 is obtained
as the ratio of these values.

On-Ramp Analysis Spillback Adjustment Exhibit 38-63

On-Ramp On-Ramp Example Problem 2A:
Expected Actual Calculation of the Spillback
Discharge Discharge Capacity Reduction Factor for
Active Duration λONR λONR,adj Volume Volume the SBL Movement for
Cycle Phase (s) (veh/s) (veh/s) βsp (veh) (veh) Analysis Period 3
2 gs1 47.3 0.561 0.561 1 26.56 26.56
3 gs1* 31.2 0.561 0.561 1 17.51 17.51
3 gs1,sp 16.1 0.561 0.317 0.565 9.04 5.11
4 gs1 8.3 0.561 0.561 1 4.67 4.67
4 gs1,sp 39.0 0.561 0.317 0.565 21.89 12.37
5 gs1 5.1 0.561 0.561 1 2.87 2.87
5 gs1,sp 42.2 0.561 0.317 0.565 23.68 13.39
6 gs1 4.7 0.561 0.561 1 2.62 2.62
6 gs1,sp 42.6 0.561 0.317 0.565 23.93 13.53
7 gs1 4.6 0.561 0.561 1 2.59 2.59
7 gs1,sp 42.7 0.561 0.317 0.565 23.97 13.55
8 gs1 4.6 0.561 0.561 1 2.58 2.58
8 gs1,sp 42.7 0.561 0.317 0.565 23.97 13.55
Total 185.89 130.89
Capacity reduction factor βsp,SBL 0.704
The SBL movement’s capacity without consideration of queue spillback is

685 veh/h (Exhibit 38-53). The adjusted capacity is calculated by applying the
spillback capacity reduction factor βsp calculated in Exhibit 38-63:
𝑐𝑆𝐵𝐿,𝑎𝑑𝑗 = 𝑐𝑆𝐵𝐿 × 𝛽𝑠𝑝,𝑆𝐵𝐿 = 685 × 0.704 = 482.2 veh/h
This step is not required for the EBT movement in this example, because the
EBT movement does not experience effects of queue spillback. As shown in
Exhibit 38-61, the on-ramp queue during the EBT green does not reach the
maximum storage length of 35.5 veh.
Analysis Period 4
The same steps performed for analysis periods 2 and 3 are applied again in
analysis period 4.
for movements that enter the on-ramp was previously determined as part of the
queue spillback check and shown in Exhibit 38-53.
Step 7B: Obtain merging capacity with the freeway facility method. The
merge capacity for analysis period 4 was previously determined, as shown in
Exhibit 38-57a. Because the congestion along the freeway mainline is dissipating
during this analysis period, the merge capacity is not constant: from time steps 1
through 4, the merge capacity is 1,142 veh/h, consistent with oversaturated
conditions from previous time periods. After time step 5, the merge capacity is
equal to the ramp roadway capacity (1,904 veh/h).
Step 7C: Plot QAPs for the on-ramp and unsignalized movements. The
procedure described earlier is applied to plot the QAPs, shown in Exhibit 38-64.
Queue spillback occurs during the first cycle, due to the residual queue from the
previous time period. However, due to low volumes at the intersection and
improvement of performance along the freeway mainline, the on-ramp clears
quickly. The queue has cleared by the end of the second cycle.

Exhibit 38-64
Example Problem 2A:
Analysis Period 4
Step 7D: Calculate adjusted capacities for the affected movements. The
procedure described earlier is used to calculate the capacity reduction factor for
the SBL movement, as shown in Exhibit 38-65. The estimated capacity reduction
is minor, as spillback only occurs during the first cycle. The EBT movement does
not experience queue spillback, therefore no adjustment is necessary.
Exhibit 38-65 On-Ramp Analysis Spillback Adjustment

Example Problem 2A: On-Ramp On-Ramp
Calculation of the Spillback Expected Actual
Capacity Reduction Factor for Active Duration QONR λONR λONR,adj Throughput Throughput
the SBL Movement for Cycle Phase (s) (veh) (veh/s) (veh/s) βsp (veh) (veh)
Analysis Period 4
1 gs1 6.0 34.4 0.505 0.505 1 3.02 3.02
1 gs1,sp 29.9 35.5 0.505 0.317 0.628 15.12 9.50
1 ge1 0.0 35.5 0.388 0.317 0.818 0.00 0.00
2 gs1 31.2 13.2 0.505 0.505 1 15.79 15.79
2 ge1 4.7 19.1 0.095 0.095 1 0.44 0.44
3 gs1 31.2 0.0 0.505 0.505 1 15.79 15.79
3 ge1 4.7 5.9 0.058 0.058 1 0.27 0.27
4 gs1 40.2 0.0 0.561 0.561 1 22.55 22.55
4 ge1 3.7 9.8 0.392 0.392 1 1.46 1.46
5 gs1 40.2 0.0 0.561 0.561 1 22.55 22.55
5 ge1 3.7 1.3 0.392 0.392 1 1.46 1.46
6 gs1 40.2 0.0 0.561 0.561 1 22.55 22.55
6 ge1 3.7 1.3 0.392 0.392 1 1.46 1.46
7 gs1 40.2 0.0 0.561 0.561 1 22.55 22.55
7 ge1 3.7 1.3 0.392 0.392 1 1.46 1.46
8 gs1 40.2 0.0 0.561 0.561 1 22.55 22.55
8 ge1 3.7 1.3 0.392 0.392 1 1.46 1.46
Total 170.49 164.86
Spillback capacity reduction factor 0.967

The adjusted capacity of the SBL movement is calculated by applying the

spillback capacity reduction factor βsp from Exhibit 38-65:
𝑐𝑆𝐵𝐿,𝑎𝑑𝑗 = 𝑐𝑆𝐵𝐿 × 𝛽𝑠𝑝,𝑆𝐵𝐿 = 746 × 0.967 = 721.4 veh/h
Intersection Performance Measures

With the adjusted capacity values obtained, the performance measures for
the intersection can be computed using the remaining steps from the signalized
intersections methodology (Chapter 19): compute the adjusted demand-to-
capacity ratio (Step 8) and compute control delay (Step 9). Exhibit 38-66
compares the performance measures for the affected movement (SBL) for the
cases with and without accounting for spillback effects.
Time SBL Movement capacity (veh/h) SBL Control delay (s/veh) Exhibit 38-66
Period Without spillback With spillback Without spillback With spillback Example Problem 2A:
1 652 652 60.3 60.3 Performance Measure
2 586 586 55.9 55.9 Comparison with and without
3 685 482 293.5 589.2 Consideration of Spillback
4 746 721 575.2 609.5 Effects
There is no change in the performance measures during analysis period 2

even though the on-ramp demand is greater than the merge capacity, because the
queue can be stored within the on-ramp. Analysis period 3 yields a significant
increase in the SBL control delay due to the queue spillback. With consideration
of spillback effects the average SBL control delay is 589.2 s/veh, compared to
293.5 s/veh without consideration of spillback. A small increase in control delay
occurs during analysis period 4, from 575.2 s/veh to 609.5 s/veh. Even though
spillback occurs for only a short time during this time period, the high value of
control delay obtained is due to the initial queue delay d3 resulting from the
unmet demand at the end of analysis period 3.
Example Problem 2B: TWSC Ramp Terminal

This scenario replaces the signalized intersection from Example Problem 2A
with a TWSC intersection, while keeping the freeway facility characteristics
unchanged. Similar to before, the control delay for the intersection movements
will be evaluated and compared with and without the occurrence of queue
spillback
Exhibit 38-67 shows the study intersection’s geometry.

Exhibit 38-67
Example Problem 2B: TWSC
Intersection Geometry: I-10
EB Ramps

demand flow rates for each analysis period, based on the TWSC intersection’s
demand inputs. The demand v and capacity c are compared for each movement
that enters the on-ramp (EBT, NBR, and SBL) during each analysis period. The
minimum value of each movement’s demand and capacity is computed and the
merge demand vR is then computed as the sum of the three values.
The capacities of minor rank movements (EBT and SBL) are computed for
each analyzed period because they change as a function of the conflicting
demand. The NBR movement is uncontrolled; therefore, its capacity is computed
from its saturation flow rate, considering the applicable adjustment factors fRT
(for right-turn movements) and fHV (for the presence of heavy vehicles):
𝑠𝑁𝐵𝑅 = 𝑠0,𝑁𝐵𝑅 × 𝑓𝑅𝑇 × 𝑓𝐻𝑉
1
𝑠𝑁𝐵𝑅 = 1,900 × × 0.961 = 1,547 veh/h
1.18
Unlike the previous signalized intersection scenario, there are no conflicting
flows to the unsignalized right turn because it is a Rank 1 (i.e., highest priority)
movement. Exhibit 38-68 summarizes the calculations for this step.


Analysis Period Parameter EBT NBR SBL Example Problem 2B:
1 c (veh/h) 125 1,547 677
min (v, c) 8 315 652
v/c 2.29 — 1.81
2 c (veh/h) 42 1,547 323
min (v, c) 42 521 323
Demand v (veh/h) 96 630 1,071
v/c 48 — 0.84
3 c (veh/h) 2 1,547 1268
min (v, c) 2 630 1,071
v/c 1.00 — 0.60
4 c (veh/h) 24 1,547 768
min (v, c) 24 80 463
The on-ramp demand estimates are then used as inputs to the freeway
facility analysis. In this example, spillback will occur in analysis period 3.

The evaluation of queue spillback impacts on the TWSC intersection follows
the procedure detailed in Exhibit 38-B11 in Appendix B. Because this is a
multiperiod analysis, the procedure must be applied for each analysis period. In
this example, analysis periods 2, 3, and 4 will be evaluated. Analysis period 1
will be excluded because the freeway operates in an undersaturated state.
Step 9A: Determine Intersection Throughput to the On-ramp

The throughput for movements that enter the on-ramp was previously
determined as part of the queue spillback check; these values are shown in
Exhibit 38-68.
Step 9B: Obtain Merging Capacity Using the Freeway Facility Methodology
This step computes the merging capacity into the freeway cmerge. The freeway
facility inputs are obtained from Exhibit 38-68, yielding the following results:
• Analysis period 2: No queue spillback.
• Analysis period 3: 1,142 veh/h.
• Analysis period 4: 1,142 veh/h during 15 time steps (222 s), and then 1,903
veh/h. This analysis period considers a lower merge capacity while a
mainline queue is present during the first 222 s. For the remainder of the
analysis period, the merge capacity is constrained only by the on-ramp
capacity, similar to the scenario presented in Example Problem 2A.

Step 9C: Determine Proportion of Analysis Period with Queue Spillback

A QAP is developed for the on-ramp to determine the spillback time Tsp.
Exhibit 38-69 shows the calculations for plotting the on-ramp queue. First, the
difference between the on-ramp throughput λONR and the merge capacity cmerge is
calculated for each analysis period. Then, the time to spillback is obtained
considering the queue growth and the available queue storage.
Exhibit 38-69 On-Ramp Queue Initial Final On-

Example Problem 2B: Queue On-Ramp Growth Rate On-Ramp Time to Spillback Ramp
Accumulation Plot Calculations Analysis Duration Demand vR λONR − cmerge Queue Spillback Time Tsp Queue
for the On-Ramp Period (min) (veh/h) (veh/s) (veh) (min) (min) (veh)
2 15 886 — 0.0 — — —
3 15 1,703 0.156 15.2 3.80 11.20 35.5
4 15 567 −0.160 35.5 — — 26.0
The results show that queue spillback occurs only during analysis period 3.
The initial queue of analysis period 3 is 0 and it takes 3.8 min for the on-ramp to
reach its maximum storage capacity. Therefore, the spillback time Tsp is computed
as 15 – 3.8 = 11.2 min. Exhibit 38-70 depicts the on-ramp’s QAP, based on the
table results.
Exhibit 38-70
Example Problem 2B: Queue
Accumulation Polygon for the
On-Ramp

When queue spillback occurs at a TWSC intersection, movements
discharging towards the on-ramp tend to follow a cooperative approach instead
of the priority-based regular operation. Therefore, the merge capacity cmerge is
shared among the three movements that enter the on-ramp:
𝑐𝑠𝑝,𝑆𝐵𝐿 + 𝑐𝑠𝑝,𝑁𝐵𝑅 + 𝑐𝑠𝑝,𝐸𝐵𝑇 = 𝑐merge = 1,142 veh/h
The capacities during spillback conditions are then obtained proportionally
to their demand flow rates (Equation 38-B22):
𝑐merge × 𝑣𝑆𝐵𝐿 1,142 × 1,071
𝑐𝑠𝑝,𝑆𝐵𝐿 = = = 681 veh/h
𝑣𝑆𝐵𝐿 + 𝑣𝑁𝐵𝑅 + 𝑣𝐸𝐵𝑇 1,071 + 630 + 96
𝑐merge × 𝑣𝑁𝐵𝑅 1,142 × 630

𝑐𝑠𝑝,𝑁𝐵𝑅 = = = 400 veh/h
𝑣𝑆𝐵𝐿 + 𝑣𝑁𝐵𝑅 + 𝑣𝐸𝐵𝑇 1,071 + 630 + 96

𝑐merge × 𝑣𝐸𝐵𝑇 × 96
𝑐𝑠𝑝,𝐸𝐵𝑇 = = = 61 veh/h
𝑣𝑆𝐵𝐿 + 𝑣𝑁𝐵𝑅 + 𝑣𝐸𝐵𝑇 1,071 + 630 + 96
The equivalent capacities cEQ,i for each movement i, aggregated for the 15-
min analysis period, are obtained proportionately to the spillback time Tsp:
𝑐𝑠𝑝,𝑆𝐵𝐿 × 𝑇𝑠𝑝 + 𝑐𝑆𝐵𝐿 × (𝑇 − 𝑇𝑠𝑝 )

𝑐𝐸𝑄,𝑆𝐵𝐿 =
𝑇
681 × 11.2 + 1,268 × 3.8
𝑐𝐸𝑄,𝑆𝐵𝐿 = = 830 veh/h
15
𝑐𝑠𝑝,𝑁𝐵𝑅 × 𝑇𝑠𝑝 + 𝑐𝑁𝐵𝑅 × (𝑇 − 𝑇𝑠𝑝 )

𝑐𝐸𝑄,𝑁𝐵𝑅 =
𝑇
401 × 11.2 + 1,547 × 3.8
𝑐𝐸𝑄,𝑁𝐵𝑅 = = 691 veh/h
15
𝑐𝑠𝑝,𝐸𝐵𝑇 × 𝑇𝑠𝑝 + 𝑐𝐸𝐵𝑇 × (𝑇 − 𝑇𝑠𝑝 )

𝑐𝐸𝑄,𝐸𝐵𝑇 =
𝑇
61 × 11.2 + 2 × 3.8
𝑐𝐸𝑄,𝐸𝐵𝑇 = = 46 veh/h
15

With the adjusted capacity values obtained, the intersection’s performance
measures can be computed using Step 11, Compute Movement Control Delay, of
the Chapter 20 TWSC methodology.
Exhibit 38-71 compares the performance measures of the affected intersection
movements for the cases with and without spillback effects during analysis
period 3. All three movements discharging to the on-ramp experienced
significant increase in the control delay.
Capacity (veh/h) Control Delay (s/veh) Exhibit 38-71

Demand Without With Without With Example Problem 2B:
Movement (veh/h) Spillback Spillback Spillback Spillback Performance Measure
Comparison with and without
EBT 96 2 61 3366.8 691.5
Consideration of Spillback
NBR 630 1,547 401 0.0 40.61
Effects—Analysis Period 3
SBL 1071 1,268 161 20.9 157.2

Example Problem 2C: AWSC Intersection Ramp Terminal

This example problem replaces the signalized intersection from Example
Problem 2A with an AWSC intersection, while keeping the freeway facility
characteristics unchanged. Unlike the previous examples, a ramp meter is active
on the freeway on-ramp, with a constant metering rate of 900 veh/h (4 s/veh). In
addition, the NBR movement is not channelized and now conflicts with the other
intersection movements. Exhibit 38-72 shows the intersection’s geometry.
Exhibit 38-72
Example Problem 2C: AWSC
Intersection Geometry: I-10
EB Ramps

demand flow rates for each analysis period, based on the AWSC intersection’s
demand inputs of the AWSC intersection. For each analysis period, the demand v
and capacity c is compared for each movement feeding the on-ramp (EBT, NBR,
and SBL). The minimum value of each movement’s demand and capacity is
computed and the merge demand vR is then computed as the sum of three
movements. Exhibit 38-73 summarizes the calculations for this step.


Analysis Period Parameter EBT NBR SBL Example Problem 2C:
min (v, c) 54 467 313
v/c 0.11 — 0.98
min (v, c) 40 512 432
v/c 0.05 — 1.18
min (v, c) 19 539 462
v/c 0.06 — 0.62
min (v, c) 28 160 316
The estimated on-ramp demand values are provided as inputs to the freeway
facility analysis. The freeway facility is then analyzed and the expected on-ramp
queues are determined as shown in Exhibit 38-74.
On-Ramp Ramp Exhibit 38-74

Analysis Demand vR On-Ramp On-Ramp Storage Ratio Spillback Example Problem 2C:
Period (veh/h) Queue (veh) Queue (ft) RQ Expected? Spillback Occurrence Check
1 834 0 0 0 No
2 984 14.9 21.9 0.62 No
3 1,020 82.1 53.4 1.5 Yes
4 504 0 0 0 No
Because spillback will occur during analysis period 3, its impacts on the
intersection’s operation must be evaluated. The next section illustrates the
application of the queue spillback evaluation methodology at an AWSC
intersection.

The evaluation of queue spillback impacts on the AWSC intersection follows
the procedure detailed in Exhibit 38-B13 in Appendix B. Because this is a
multiperiod analysis, the procedure must be applied for each analysis period. In
this example, analysis periods 2, 3, and 4 will be evaluated. Analysis period 1
will be excluded because oversaturated conditions do not exist on the freeway
during this analysis period.
Step 13A: Determine Intersection Throughput to On-Ramp

The intersection throughput to the on-ramp was previously determined at
the spillback check (Exhibit 38-73).

Step 13B: Obtain Merging Capacity with the Freeway Facilities Method
In this example, the ramp metering rate (900 veh/h) is an additional input to
the freeway facility analysis and acts as a constraint to the merge capacity.
Therefore, the merge capacity for this analysis is kept constant at 900 veh/h.
Step 13C: Determine Fraction of Time Period with Queue Spillback

The procedure to evaluate the spillback time Tsp is similar to the TWSC
procedure. The calculations are provided in Exhibit 38-75.
Exhibit 38-75 On-Ramp

Example Problem 2C: Queue Queue Initial Final On-
Accumulation Plot Calculations On-Ramp Growth Rate On-Ramp Time to Spillback Ramp
for the On-Ramp Analysis Duration Demand vR λONR − cmerge Queue Spillback Time Tsp Queue
Period (min) (veh/h) (veh/s) (veh) (min) (min) (veh)
2 15 984 0.023 0.0 — — 21.0
3 15 1,020 0.033 21.0 7.25 7.75 35.5
4 15 504 −0.110 35.5 — — 0.0
Exhibit 38-76 illustrates the QAP for the on-ramp, based on the results shown
in Exhibit 38-75.
Exhibit 38-76
Example Problem 2C: Queue
Accumulation Polygon for the
On-Ramp
Step 13D: Compute Spillback Departure Headway

This step is similar to the calculation of adjusted capacities in the TWSC
procedure. The same calculations are performed and adjusted capacity values are
converted into headways hsp, as shown in Exhibit 38-77.
Exhibit 38-77 Capacity during Capacity without Equivalent Spillback

Example Problem 2C: Spillback csp Spillback c Capacity cEQ Departure
Equivalent Capacities and Movement (veh/h) (veh/h) (veh/h) Headway hsp (s)
Headways for the On-Ramp: EBT 15 396 212.1 17.0
Analysis Period 3 NBR 439 550 496.5 7.3
SBL 445 462 453.7 7.9
With the adjusted capacity values obtained, the intersection’s performance

measures can be computed using the remaining steps from the Chapter 21
AWSC methodology: Step 13 (compute service times) and Steps 14 and 15
(compute control delay by lane and approach).


Exhibit 38-78 compares the intersection movements’ performance measures
with and without spillback effects during analysis period 3. The three
movements that discharge into the on-ramp (EBT, NBR and SBL) experience
increased delay, while the remaining movements have unchanged performance.
Capacity (veh/h) Control Delay (s/veh) Departure Headway (s) Exhibit 38-78
Demand Without With Without With Without With Example Problem 2C:
Movement (veh/h) Spillback Spillback Spillback Spillback Spillback Spillback Performance Measure
Comparison with and without
EBL 75 359 359 15.6 15.6 10.0 10.0
Consideration of Spillback
EBT 19 396 212 12.6 21.7 9.1 17.0
Effects—Analysis Period 3
NBT 229 497 497 16.3 16.3 7.2 7.2
NBR 539 550 497 58.9 92.3 6.5 7.3
SBL 546 462 454 128.0 136.5 7.8 7.9
SBT 220 494 494 16.0 16.0 7.3 7.3
Note: EBL = eastbound left turn, EBT = eastbound through, NBT = northbound through, NBR = northbound right
turn, SBL = southbound left turn, SBT = southbound through.
EXAMPLE PROBLEM 3: OFF-RAMP QUEUE SPILLBACK ANALYSIS

This example problem illustrates the application of the off-ramp spillback
methodology to a network consisting of two freeway facilities (I-75 SB to SR-826
SB in Miami, Florida), as shown in Exhibit 38-79. Due to congested conditions at
the downstream merge segment (SR-826), spillback is expected to affect the
operations of the upstream freeway facility (I-75). Vehicles traveling from node A
to D are likely to have their travel time severely affected if spillback occurs.
Exhibit 38-79
Example Problem 3:
Study Site

This freeway-to-freeway network is modeled as two separate freeway

facilities. The upstream freeway (Facility 1, I-75) is modeled as a diverge section
connected to the downstream freeway (Facility 2, SR-826). The network’s
geometry is shown in Exhibit 38-80.
Exhibit 38-80
Example Problem 3:
Freeway Facility Geometry
(a) Facility 1: I-75 SB
(b) Facility 2: SR-826 SB
Input Data
Traffic demands for the freeway facilities and ramps are provided in Exhibit
38-81 for each 15-min analysis period.
Exhibit 38-81 Freeway Facility 1 (I-75 SB) Freeway Facility 2 (SR-826 SB)
Example Problem 3: Mainline Demand Diverge Demand Mainline Demand Merge Demand
Traffic Demands Analysis Flow Rate Flow Rate Flow Rate Flow Rate
Period (veh/h) (veh/h) (veh/h) (veh/h)
1 5,400 1,400 4,000 1,400
2 6,200 3,000 5,700 3,000
3 6,000 3,400 5,600 3,400
4 4,500 800 4,500 800
Additional input parameters are as follows:

• Urban area
• Level terrain
• Grade: 0%
• Queue spillback regime 4 is expected
• Base FFS: 65 mi/h (I-75), 67.1 mi/h (SR-826)
• Ramp FFS: 55 mi/h
• Ramp side: right for both facilities
• Right-side clearance: 10 ft

• Traffic composition: 12% trucks on both freeway and ramps

• Ramp length: 3,588 ft
• Acceleration lane length: 1,500 ft
• No shoulder available
• Deceleration lane length: 700 ft
• Number of ramp lanes: 2
• Familiar facility users
Performance Measures for the Individual Facilities

The LOS of the two freeway facilities, when analyzed independently, is
presented in Exhibit 38-82 and Exhibit 38-83. Facility 1 (I-75) is undersaturated
throughout the study period, while Facility 2 (SR-826) experiences congestion
during analysis periods 2 and 3. Ignoring the interactions between the two
facilities would lead to an inaccurate estimation of the performance of the
upstream facility (Facility 1). Facility 2’s merge segment operates at LOS F, and
the on-ramp capacity may therefore be affected, leading to queue formation and
potential spillback.

1-1 1-2 1-3 1-4 Example Problem 3:
Freeway Facility 1 (I-75) LOS
Analysis Period Basic Basic Diverge Basic
1 C C B B
2 C C C A
3 C C C A
4 B B A B

Analysis 2-1 2-2 2-3 2-4 2-5 Example Problem 3: Freeway
Period Basic Merge Basic Diverge Basic Facility 2 (SR-826) LOS
1 B C C B C
2 C F E F E
3 C F F F E
4 C C C C C
Spillback Check
Analyzing SR-826 using Chapter 25’s Freeway Facilities Oversaturated
Segment Evaluation methodology provides the expected on-ramp queue for
every analysis period. The first check compares the off-ramp demand to the
ramp roadway capacity, as shown in Exhibit 38-84. The ramp queue starts to
develop during analysis period 2. At the end of this time period, a ramp queue
length of 1,188 ft is expected, yielding a queue storage ratio of 0.33. Therefore,
spillback onto I-75 is not expected during analysis period 2. However, during
analysis period 3, a ramp queue length of 5,160 ft is expected, yielding a queue
storage ratio of 1.41. Therefore, spillback onto I-75 will occur during analysis
period 3.

Exhibit 38-84 Total

Example Problem 3: Queue Number Number of Average
Length and Storage Ratio at of Queued Vehicle Ramp Queue
the SR-826 On-Ramp Queued Vehicles in Spacing Queue Length Length Storage
Analysis Vehicles each Lane (ft) (ft) (ft) Ratio Spillback
Period [A] [B] = [A]/2 [C] [D] = [B]×[C] [E] [F] = [D]/[E] Occurs?
1 0.0 0.00 — 0 0.00 No
2 38.3 19.15 62 1,188 0.33 No
3,588
3 159.1 79.55 65 5,160 1.44 Yes
4 0.0 0.00 — 0 0.00 Yes
Spillback Analysis
Because spillback is expected to occur, the methodology described in
Appendix A (Exhibit 38-A8) is applied to evaluate how it affects I-75 SB. The
methodology’s application to each analysis period is presented below.
Analysis Period 1
No oversaturated conditions occur; therefore, no additional calculations are
needed for this analysis period.
Analysis Period 2
During analysis period 2, the downstream merge segment operates at LOS F
and the on-ramp capacity is expected to be reduced.
Step 1: Calculate Background Density for Unblocked Lanes on each Segment

in the Case of Queue Spillback
The I-75 diverge segment has 5 lanes and queue spillback Regime 4 (two
blocked lanes) is expected. Therefore, when queue spillback occurs, this segment
is expected to operate with two blocked lanes (lanes 1 and 2, with lane 1 being
the rightmost lane) and three unblocked lanes (lanes 3 through 5).
The per-lane capacity lane at the diverge segment SC(3) is 2,350 pc/h/ln or
11,750 pc/h. For the 15-s time-step-level analysis, this capacity is converted to
48.95 passenger cars per time step (pc/ts). Therefore, the capacity of the
segment’s unblocked portion is given by Equation 38-A12:
𝑆𝐶𝐸𝑄(𝑖, 𝑁, 𝑁𝑄) = 𝑆𝐶(𝑖, 𝑁 − 𝑁𝑄) × 𝐶𝐴𝐹𝐵𝐿
The capacity adjustment factor CAFBL is obtained from Exhibit 38-3. For a
segment with 5 directional lanes and 2 blocked lanes, a CAF of 0.67 is applied.
Therefore, the equivalent capacity of the unblocked portion is given by:
𝑆𝐶𝐸𝑄(3, 5, 3) = 48.95 × 0.67 = 38.8 pc/ts or 7,872 pc/h
The unblocked background density KBUB is calculated next. During analysis
period 2, an expected mainline demand of 4,165.8 pc/h is used in the calculations.
The KBUB parameter for the unblocked lanes is computed as the density of a 3-
lane basic segment with a capacity SCEQ of 7,872 pc/h:
𝐾𝐵𝑈𝐵(3, 5, 3) = 30.4 pc/h/mi

Step 2: Initialize the Freeway Facility

When spillback occurs, the subject freeway facility is analyzed as a link–node
structure similar to the oversaturated procedure for freeway facilities. However,
the facility structure is expanded to consider the ramp segments. Exhibit 38-85
illustrates the structure for this analysis. Node 4.1 represents the interface between
the diverge segment and the ramp roadway, while node 4.2 represents the
interface between the ramp roadway and the merge at the downstream facility.
Exhibit 38-85
Example Problem 3:
Link–Node Structure for
Spillback Analysis: I-75 SB
Step 2C: Determine Queue Influence Area

The queue influence area (QIA) is obtained as function of the segment FFS,
as shown in Exhibit 38-5. Therefore, for a FFS of 65mi/h, the QIA length is equal
to 1,060 ft.
Step 2F: Determine the Initial Number of Vehicles at the Off-Ramp

The ramp speed at the expected demand is obtained as:
𝑣𝑅
𝑆𝑅 = (1 − 0.109 × ) × 𝑆𝐹𝑅
1,000
1,679
𝑆𝑅 = (1 − 0.109 × ) × 55 = 44.9 mi/h
1,000
Next, the ramp background density is obtained:
𝑣𝑅 1,679
𝑅𝐾𝐵 = = = 37.4 pc/mi/ln
𝑆𝑟𝑎𝑚𝑝 44.9
The initial number of vehicles in the ramp is then computed as:
3,588
𝑅𝑁𝑉(3,0,2,1) = 37.4 × × 2 = 50.8 pc
5,280
Step 2G: Determine the Capacity of the Downstream Terminal

The merge capacity is obtained by analyzing the downstream freeway
facility using the Chapter 25 oversaturated segment evaluation procedure and
aggregating the parameter ONRO for an hourly flow rate. During this analysis

period, the merge capacity is constant at 13.4 pc/ts or 3,217 pc/h, while the ramp
demand is 14.0 pc/ts or 3,369 pc/h.
Given the merge’s demand and capacity, the queue in the ramp roadway
increases by 0.6 pc during every time step. Exhibit 38-86 illustrates the ramp
queue and the total number of vehicles in the ramp, considering an initial value
of 50.8 pc in the ramp at the start of the analysis period, as previously computed.
Exhibit 38-86
Example Problem 3: Queued
Vehicles and Total Number of
Vehicles RNV in the Ramp:
Analysis Period 2
Step 9A: Perform Spillback Analysis

The flow RF that can travel across the ramp node 4.1 and enter the ramp
roadway is obtained as the minimum of demand RI, ramp roadway capacity RC,
and constrained capacity due to a downstream queue in the ramp RSTG, as given
by Equation 38-A22:
𝑅𝐹(𝑖, 𝑡, 𝑝) = min(𝑅𝐼(𝑖, 𝑡, 𝑝, 𝑘), 𝑅𝐶(𝑖, 𝑘), 𝑅𝑆𝑇𝐺(𝑖, 𝑡, 𝑝, 𝑘))
The capacity of a 2-lane ramp roadway with a FFS of 55 mi/h is 4,400 pc/h or
18.3 pc/ts. Therefore, ramp roadway’s capacity is not a constraint to ramp flow.
The other potential capacity constraint RSTG is calculated using Equation 38-A24:
𝑅𝑆𝑇𝐺(𝑖, 𝑡, 𝑝, 𝑘) = 𝑅𝐹(𝑖, 𝑡 − 1, 𝑝, 𝑘)
+ 𝑅𝐾𝑄(𝑖, 𝑡, 𝑝, 𝑘)𝑥[𝑅𝐿(𝑘)𝑥 𝑅𝑁(𝑘)]– 𝑅𝑁𝑉(𝑖, 𝑡 − 1, 𝑝, 𝑘)
The constraint RSTG is dependent on the number of vehicles in the ramp
RNV, which increases progressively as the ramp queue grows. Exhibit 38-87
compares the decreasing value of RSTG with the ramp input RI during analysis
period 2. At the end of the analysis period, the ramp capacity is still greater than
demand; therefore, no spillback occurs at the end of this analysis period,
consistent with the queue spillback check previously performed.
Exhibit 38-87
Example Problem 3: Ramp
Capacity RSTG and Ramp
Input RI: Analysis Period 2

Because spillback does not occur, no additional calculations for the mainline
are required.
Step 30: Calculate Segment Performance Measures

Because spillback does not occur during this analysis period, the mainline’s
performance measures do not need to be recalculated. The ramp, however,
experiences queueing. Therefore, the ramp speed during this analysis period is
calculated using Equation 38-A65 through Equation 38-A67:
𝑆
𝑅𝐹(𝑖, 𝑝, 𝑘) = 4 × ∑ 𝑅𝐹(𝑖, 𝑡, 𝑝, 𝑘) = 1679.5 pc/h/ln

𝑡=1
𝑆
1
𝑅𝐾(𝑖, 𝑝, 𝑘) = × ∑ 𝑅𝑁𝑉(𝑖, 𝑡, 𝑝, 𝑘) = 71.6 pc/mi/ln
60
𝑡=1
𝑅𝐹(𝑖, 𝑝, 𝑘) 1,679.5
𝑆𝑅(𝑖, 𝑝, 𝑘) = = = 31.9 mi/h
𝑅𝐾(𝑖, 𝑝, 𝑘) 71.6
Analysis Period 3
Steps 1, 2, and 9: Ramp Analysis
The same steps are repeated for analysis period 3. The ramp analysis is
summarized in Exhibit 38-88. During this analysis period, the ramp demand is
15.4 pc/ts, while the merge capacity is 13.9 pc/ts. Because demand is greater than
capacity, the number of vehicles increases gradually, causing the capacity
constraint RSTG to decrease each time step. At time step 14, the value of RSTG
becomes equal to the merge capacity (13.9 pc/ts), which implies that the ramp has
reached jam density and the maximum flow that can enter the ramp is equal to
the flow departing the ramp. Therefore, queue spillback into the mainline starts
at time step 15.
Exhibit 38-88
Example Problem 3: Ramp
Capacity RSTG and Ramp
Input RI: Analysis Period 3
After the onset of queue spillback, the number of unserved vehicles at the
exit is computed every time step through the parameter OFRUV(i, t, p). Then, the
expected length of the mainline queue OFRLQ(i, t, p) is computed based on the
number of unserved vehicles and the ramp queue density RKQ, as given by
Equation 38-A35:
𝑂𝐹𝑅𝑈𝑉(𝑖, 𝑡, 𝑝)
𝑂𝐹𝑅𝐿𝑄(𝑖, 𝑡, 𝑝) =
𝑅𝐾𝑄 (𝑖, 𝑡, 𝑝)

The ramp queue density RKQ is obtained using Equation 38-A23:

[(𝐾𝐽 – 𝑅𝐾𝐶) × 𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)]
𝑅𝐾𝑄(𝑖, 𝑡, 𝑝, 𝑘) = 𝐾𝐽–
𝑅𝐶(𝑖, 𝑡, 𝑝)
𝑅𝐾𝑄(𝑖, 𝑡, 𝑝, 𝑘) = 190 − [(190 – 46.5) × 13.87)] / 18.33 = 81.4 pc/mi/ln
Exhibit 38-89 illustrates the expected spillback queue length during analysis
period 3.
Exhibit 38-89
Example Problem 3: Spillback
Queue Length on I-75 SB:
Analysis Period 3
The parameter OFRLQ represents the length of the queue if all unserved
vehicles were queued in a single line. Given the segment geometry (Exhibit 38-
90), the operating regimes and flow modes can be obtained as a function of
OFRLQ:
• 0 < OFRLQ ≤ 1,400 ft: Regime 1
• 1,400 ft < OFRLQ ≤ 3,000 ft: Regime 4, with increased turbulence
• 3,000 ft < OFRLQ: Regime 4, with lane blockage (queue extends upstream
beyond the diverge)
Exhibit 38-90
Example Problem 3: Available
Queue Storage on I-75 SB
As previously shown in Exhibit 38-89, the maximum queue length OFRLQ

during analysis period 3 is 4,696 ft. Because queues develop along mainline lanes
1 and 2, at the end of analysis period 3, the back of queue will be located 848 ft
upstream of the boundary of segments 1-2 and 1-3. The length of the QIA is 1,060
ft, and when it is added to the back of the queue it does not reach the upstream
node of segment 1-2. Therefore, segment 1-2’s capacity is not affected by the
turbulence area upstream of the queue, as illustrated in Exhibit 38-91.

Exhibit 38-91
Example Problem 3: Back of
Queue Length, Including
Queue Influence Area, at the
End of Analysis Period 3

The ramp speed is computed similarly to analysis period 2:
𝑆
𝑅𝐹(𝑖, 𝑝, 𝑘) = 4 × ∑ 𝑅𝐹(𝑖, 𝑡, 𝑝, 𝑘) = 1,707 pc/h/ln

𝑡=1
𝑆
1
𝑅𝐾(𝑖, 𝑝, 𝑘) = × ∑ 𝑅𝑁𝑉(𝑖, 𝑡, 𝑝, 𝑘) = 108.4 pc/mi/ln
60
𝑡=1
𝑅𝐹(𝑖, 𝑝, 𝑘) 1,707
𝑆𝑅(𝑖, 𝑝, 𝑘) = = = 21.5 mi/h
𝑅𝐾(𝑖, 𝑝, 𝑘) 108.4
Performance measures are computed for the blocked and unblocked portions
of each segment of the freeway facility.
Segment 1-3 (diverge)—blocked portion. Similar to the ramp, the flow
through the blocked portion is aggregated for this time period:
𝑆
𝑆𝐹𝐵𝐿(𝑖, 𝑝) = 4 × ∑ 𝑆𝐹𝐵𝐿(𝑖, 𝑡, 𝑝) = 3,030 pc/h

𝑡=1
The average density is obtained as the sum of two separate components. The
average number of vehicles in the blocked portion of the segment is computed as:
𝑆
1
𝐾𝐵𝐿(𝑖, 𝑝) = × ∑ 𝑁𝑉(𝑖, 𝑡, 𝑝) = 51 pc/mi/ln
60
𝑡=1
The increase in density due to the lane blockage ΔK is obtained as:
𝑆
1
∆𝐾(𝑖, 𝑝) = ∑ ∆𝑁𝑉(𝑖, 𝑡, 𝑝) = 20.1 pc/mi/ln
𝑆 × 𝑁𝑄
𝑡=1
The total density is then computed as:
𝐾(𝑖, 𝑝) = 𝐾𝐵𝐿(𝑖, 𝑝) + ∆𝐾(𝑖, 𝑝) = 70.1 pc/mi/ln
Finally, the speed in the blocked lanes is obtained through the fundamental
equation:
𝑆𝐹𝐵𝐿(𝑖, 𝑝) 3,030
𝑆𝐵𝐿(𝑖, 𝑝) = = = 21.2 mi/h
𝑁(𝑖, 𝑝) × 𝐾(𝑖, 𝑝) 2 × 70.1

Segment 1-3 (diverge)—unblocked portion. The same process is repeated

for the segment’s unblocked portion, except that the ΔK component is omitted
because no queues occur in those lanes:
𝑆𝑈𝐵(𝑖, 𝑝) = 56.1 mi/h
Analysis Period 4
During analysis period 4, the congestion at the downstream facility (SR-826)
dissipates, which allows the ramp to discharge at the ramp roadway capacity
(4,400 pc/h, or 18.33 pc/ts). Given the low ramp demand during this time period,
the queue clears quickly (9 time steps, or 135 s). After the 10th time step, the
freeway facility returns to undersaturated conditions.
EXAMPLE PROBLEM 4: ON-RAMP QUEUE SPILLBACK ANALYSIS INTO A

SINGLE-LANE ROUNDABOUT
This example problem illustrates the analysis methodology when spillback
occurs from an on-ramp into a single-lane roundabout. The example is based on
a ramp terminal on I-105 at Bellflower Blvd. in Los Angeles, California.
Input Data
The site layout is shown in Exhibit 38-92. The location’s traffic and geometric
characteristics are as follows:
• Single-lane approaches on all roundabout entries
• Adjusted demand flow rates in pc/h for all movements are as shown in
Exhibit 38-92
• No heavy vehicles
• U-turn movements are negligible
• Pedestrian activity limits the exit capacity to the on-ramp to 1,300 pc/h.
• Ramp length = 1,657 ft
• The on-ramp connecting the roundabout to the freeway is metered at a
rate cRM = 800 pc/h.
Exhibit 38-92
Example Problem 4:
Study Interchange Schematic

Step 3: Determine Circulating Flow Rates

This step calculates all circulating flow rates at the roundabout. For example,
for the NB approach, the circulating flow is calculated using Equation 22-11:
𝑣𝑐,𝑁𝐵,𝑝𝑐𝑒 = 𝑣𝑆𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝐿,𝑝𝑐𝑒 = 100 + 500 + 100 = 700 pc/h
Similarly, the conflicting flows for the other approaches are:
𝑣𝑐,𝑆𝐵,𝑝𝑐𝑒 = 0 pc/h
𝑣𝑐,𝐸𝐵,𝑝𝑐𝑒 = 300 pc/h
Step 4: Determine Entry Flow Rates per Approach

The entry flow rate at each approach is calculated by adding the movement
flow rates that enter the roundabout.
The entry flow rates are calculated as follows:
𝑣𝑒,𝑆𝐵,𝑝𝑐𝑒 = 𝑣𝑆𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝑆𝐵𝑇,𝑝𝑐𝑒 = 100 + 200 = 300 pc/h
𝑣𝑒,𝐸𝐵,𝑝𝑐𝑒 = 700 pc/h
𝑣𝑒,𝑁𝐵,𝑝𝑐𝑒 = 1,600 pc/h
Step 5: Determine the Capacity of Each Entry Lane in Passenger Car

Equivalents
Using the single-lane capacity equation (Equation 22-1), the capacity for each
entry lane is calculated as follows:
−3 )𝑣 −3 )(0)
𝑐𝑝𝑐𝑒,𝑆𝐵 = 1,380𝑒 (−1.02×10 𝑐,𝑝𝑐𝑒,𝑆𝐵 = 1,380𝑒 (−1.02×10 = 1,380 pc/h
(−1.02×10−3 )𝑣𝑐,𝑝𝑐𝑒,𝐸𝐵 (−1.02×10−3 )(300)
𝑐𝑝𝑐𝑒,𝐸𝐵 = 1,380𝑒 = 1,380𝑒 = 1,016 pc/h
(−1.02×10−3 )𝑣𝑐,𝑝𝑐𝑒,𝑁𝐵 (−1.02×10−3 )(700)
𝑐𝑝𝑐𝑒,𝑁𝐵 = 1,380𝑒 = 1,380𝑒 = 676 pc/h

The volume-to-capacity ratios for each entry lane are calculated using
300
𝑥𝑆𝐵 = = 0.22
1,380
700
𝑥𝐸𝐵 = = 0.69
1,016
1,600
𝑥𝑁𝐵 = = 2.37
676

The 95th percentile queue is first computed for each lane without
considering spillback effects. For example, the queue for the southbound
approach is given as follows, applying Equation 22-20:
3,600
√(1 ( 𝑐 )𝑥 𝑐
𝑄95,𝑆𝐵 = 900𝑇 [𝑥 − 1 + 2
− 𝑥) + ]( )
150𝑇 3,600

3,600
(1,380) 0.22 1,380
𝑄95,𝑆𝐵 √ 2
= 900(0.25) 0.22 − 1 + (1 − 0.22) + ( ) = 1 veh
150(0.25) 3,600
[ ]
Similarly,
𝑄95,𝐸𝐵 = 6 veh
𝑄95,𝑁𝐵 = 121 veh
These values are rounded to the nearest vehicle.
Exhibit 38-93 provides the flows and resulting queues at the roundabout.
Exhibit 38-93 Circulating Entry Flow Volume-to-

Example Problem 4: Flow Rates Rates Capacity Capacity 95th Percentile
Roundabout Flows and Approach (pc/h) (pc/h) (pc/h) Ratio Queues (veh)
Queues SB 0 300 1,380 0.22 1
EB 300 700 1,016 0.69 6
NB 700 1,600 676 2.37 121
Note: SB = southbound, EB =eastbound, NB = northbound.
Step 13: Maximum Throughput for each O-D Movement

The first task in calculating the maximum throughput per movement is to
define the priority order, as shown in Exhibit 38-94. The SB approach is the Rank
1 leg because it is the upstream approach to the on-ramp. The EB approach is the
Rank 2 leg, while the NB approach is the Rank 3 leg. Each approach’s capacity
(previously calculated in Step 5) is used to determine the maximum throughput
for each approach and O-D pair.
Exhibit 38-94
Example Problem 4:
Roundabout Approach Priority
Order
Starting with the Rank 1 (SB) approach, the maximum throughput for the
movement exiting through the EB leg (the on-ramp) is calculated as follows:
3,600
Equation 38-28 𝜆𝑆𝐵𝐿,𝑝𝑐𝑒 = min (𝑣𝑆𝐵𝐿,𝑝𝑐𝑒 , 𝑐𝑝𝑐𝑒,𝑆𝐵 × 𝑝𝑆𝐵𝐿 , )
ℎ𝑠

where
λSBL,pce = maximum throughput for the SB left-turn movement (pc/h),
vSBL,pce = flow rate for the SB left-turn movement (pc/h),
cpce,SB = entry lane capacity for the SB roundabout approach (pc/h),
pSBL = percent of demand from the SB approach into the on-ramp
= vSBL,pce divided by the total flow rate for the SB approach, and
hs = departure saturation headway into the on-ramp (s/veh).
Then:
100 3,600
𝜆𝑆𝐵𝐿,𝑝𝑐𝑒 = min (100, [1,380 × ], ) = 100 pc/h
300 2.77
Because the Rank 3 (NB) approach is the only one with a volume-to-capacity
ratio over 1, the conflicting flows and capacity values calculated above are valid.
The next calculation is the maximum throughput for the remaining
movements of the approach that contribute to the conflicting flows for
downstream approaches. This calculation is performed as follows:
𝜆𝑆𝐵𝑇,𝑝𝑐𝑒 = min(𝑣𝑆𝐵𝑇,𝑝𝑐𝑒 , 𝑐𝑝𝑐𝑒,𝑆𝐵 × 𝑝𝑆𝐵𝑇 ) Equation 38-29
where
λSBT,pce = maximum throughput for the southbound-through movement (pc/h),
vSBT,pce = flow rate for the southbound-through movement (pc/h),
cpce,SB = entry lane capacity for the southbound roundabout approach (pc/h), and
pSBL = percent of demand from SB approach for through movement
= vSBT,pce divided by the total flow rate for the SB approach.
Then:
200
𝜆𝑆𝐵𝑇,𝑝𝑐𝑒 = min (200, [1,380 × ]) = 200 pc/h
300
The maximum throughput for each approach and O-D pair is calculated
considering the maximum throughput on the on-ramp and accounting for
higher-rank approaches:
3,600
𝜆𝐸𝐵𝑇,𝑝𝑐𝑒 = min (𝑣𝐸𝐵𝑇,𝑝𝑐𝑒 , 𝑐𝑝𝑐𝑒,𝐸𝐵 × 𝑝𝐸𝐵𝑇 , − 𝜆𝑆𝐵𝐿 )
ℎ𝑠
500 3,600
𝜆𝐸𝐵𝑇,𝑝𝑐𝑒 = min (500, [1,016 × ],[ − 100]) = 500 pc/h
700 2.77
𝜆𝐸𝐵𝐿,𝑝𝑐𝑒 = min(𝑣𝐸𝐵𝐿,𝑝𝑐𝑒 , 𝑐𝑝𝑐𝑒,𝐸𝐵 × 𝑝𝐸𝐵𝐿 )

100
𝜆𝐸𝐵𝐿,𝑝𝑐𝑒 = min (100, [1,016 × ]) = 100 pc/h
700
3,600
𝜆𝑁𝐵𝑅,𝑝𝑐𝑒 = min (𝑣𝑁𝐵𝑅,𝑝𝑐𝑒 , 𝑐𝑝𝑐𝑒,𝑁𝐵𝑅 × 𝑝𝑁𝐵𝑅 , − 𝜆𝑆𝐵𝐿 − 𝜆𝐸𝐵𝑇 )
ℎ𝑠
1,500 3,600
𝜆𝑁𝐵𝑅,𝑝𝑐𝑒 = min (1,500, [678 × ],[ − 100 − 500]) = 634 pc/h
1,600 2.77

The maximum throughput to the on-ramp is lower than the exit capacity
(1,300 pc/h), thus the northbound approach flow rate is limited by its own
approach capacity.
Step 14: Maximum Exit Flow Rate into the On-Ramp

The maximum throughput from the roundabout to the on-ramp λR,pce is
calculated by adding up the maximum throughput to the on-ramp from the
higher-rank approaches:
𝜆𝑅,𝑝𝑐𝑒 = 𝜆𝑆𝐵𝐿,𝑝𝑐𝑒 + 𝜆𝐸𝐵𝑇,𝑝𝑐𝑒 + 𝜆𝑁𝐵𝑅,𝑝𝑐𝑒 = 100 + 500 + 634 = 1,234 pc/h
This maximum on-ramp demand flow rate is lower than the exit capacity. It
is also lower than the total O-D demand to the ramp, which has a rate of:
𝑣𝑒𝑥,𝐸𝐵,𝑝𝑐𝑒 = 𝑣𝑆𝐵𝐿,𝑝𝑐𝑒 + 𝑣𝐸𝐵𝑇,𝑝𝑐𝑒 + 𝑣𝑁𝐵𝑅,𝑝𝑐𝑒 = 100 + 500 + 1500 = 2,100 pc/h
Step 15: On-Ramp Metering Capacity

The departure saturation headway into the on-ramp is 3,600 s/h divided by
the exit capacity of 1,300 pc/h, or 2.77 s/pc. The demand flow rate from the on-
ramp onto the freeway is then the smaller of the exit capacity and the metering
rate:
3600 3,600
𝑐𝑅,𝑝𝑐𝑒 = min (𝑐𝑅𝑀 , ) = min (800, ) = 800 pc/h
ℎ𝑠 2.77
Step 16: On-Ramp Storage Ratio and Queue Spillback Length

The on-ramp storage LONR is calculated in passenger cars, considering an
average spacing of 25 ft and given that the total ramp length is 1,657 ft:
1,657
𝐿𝑂𝑁𝑅 = = 66 pc
25
Knowing the ramp’s maximum exit flow rate, the number of passenger car
equivalents that exit the roundabout through the on-ramp during a 15-min
analysis period is obtained from the difference between the on-ramp throughput
λR,pce and the ramp metering rate cRM:
𝜆𝑅,𝑝𝑐𝑒 − 𝑐𝑅𝑀 1,234 − 800
𝑄𝑠𝑝 = = = 108 pc
4 4
The queue storage ratio RQ is then calculated as the ratio between the
expected queue and the on-ramp storage:
𝑄𝑠𝑝 108
𝑅𝑄 = = = 1.63
𝐿𝑂𝑁𝑅 66
Because RQ > 1.0, queues will develop on each roundabout approach due to
spillback.
Step 17: Queue Spillback Distribution per Approach

The number of vehicles queued during the 15-min analysis period
𝑄spillback = 𝑄𝑂𝑁𝑅 − 𝐿𝑂𝑁𝑅 = 42 pc
The queues due to the on-ramp spillback are assumed to be distributed
proportionally to each approach’s demand flow rate to the on-ramp, and are

added to the 95th percentile queues estimated for undersaturated conditions

(Equation 22-20):
𝜆𝑆𝐵𝐿,𝑝𝑐𝑒 100
𝑄𝑠𝑝,𝑆𝐵 = 𝑄spillback × + 𝑄95,𝑆𝐵 = 42 × + 1 = 4 pc
𝑣𝑅 1234
𝜆𝐸𝐵𝑇,𝑝𝑐𝑒
𝑄𝑠𝑝,𝐸𝐵 = 𝑄spillback × + 𝑄95,𝐸𝐵 = 17 + 6 = 23 pc
𝑣𝑅
𝜆𝑁𝐵𝑅,𝑝𝑐𝑒
𝑄𝑠𝑝,𝑁𝐵 = 𝑄spillback × + 𝑄95,𝑁𝐵 = 22 + 121 = 142 pc
𝑣𝑅
Step 18: Average Delay per Approach

To estimate the average delay on each approach, both the approach’s control
delay and the delay due to the on-ramp capacity limitation must be estimated.
The average control delay per approach is given by Equation 22-17.
3,600
3,600 × 0.22
√ 1,380
𝑑𝑆𝐵 = + 900(0.25) 0.22 − 1 + (0.22 − 1)2 +
1,380 450(0.25)
[ ]
+ 5 × min[0.22,1]
𝑑𝑆𝐵 = 4.44 s/veh

Similarly:
𝑑𝐸𝐵 = 14.47 s/veh
𝑑𝑁𝐵 = 635.86 s/veh
The additional delay due to the on-ramp spillback is given by Equation 38-B41.
2 3,600 2,100
3,600 2,100 2,100 ×
𝑑𝑠𝑝 = + 900(0.25) [ √
−1+ ( − 1) + 800 800 ]
800 800 800 450(0.25)
2,100
+ 5 × min [ , 1]
800
𝑑𝑠𝑝 = 747.94 s/veh
Therefore, the total average delay per approach is:
100
𝑑𝑆𝐵,𝑇 = 𝑑𝑆𝐵 + 𝑑𝑠𝑝 × = 65.05 s/veh
1,234
500
𝑑𝐸𝐵.𝐿 = 𝑑𝐸𝐵 + 𝑑𝑠𝑝 × = 317.54 s/veh
1,234
634
𝑑𝑁𝐵,𝑅 = 𝑑𝑁𝐵 + 𝑑𝑠𝑝 × = 1,020.14 s/veh
1,234

5. REFERENCE
This reference can be found in 1. University of Florida Transportation Institute; Cambridge Systematics, Inc.;
the Technical Reference
Library in Volume 4. and A. Skabardonis. NCHRP Web-Only Document 290: Highway Capacity
Manual Methodologies for Corridors Involving Freeways and Surface Streets.

APPENDIX A: OFF-RAMP QUEUE SPILLBACK ANALYSIS
Chapter 10, Freeway Facilities Core Methodology, evaluates the performance

of each segment individually using standard 15-min analysis periods. If any
segment within the facility yields LOS F (v/c > 1), the analysis continues with the
oversaturated procedure, using smaller time steps.
Similarly, when determining whether queue spillback occurs from a freeway
off-ramp, network analysis is first conducted using 15-min analysis periods. If
the analysis shows that any of the ramps are expected to experience queue
spillback, the oversaturated procedure must be used to estimate the spillback
effects on the freeway mainline lanes, even if the segment-wide performance is
not at LOS F.
The methodology’s framework for conducting a spillback check at diverge
points is presented in Exhibit 38-A1 and described in more detail in the
remainder of this appendix.
Exhibit 38-A1
Check Flowchart
Chapter 38/Network Analysis Appendix A: Off-Ramp Queue Spillback Analysis

CAPACITY CHECKS
The procedure first determines whether capacity is exceeded at any of the
critical points along the diverge section.
Case A: Ramp Roadway

Demand at the study diverge ramp (𝑣𝑅 , as defined in Chapter 14, Freeway
Merge and Diverge Segments) is compared against the ramp roadway’s capacity
using Exhibit 14-12, replicated as Exhibit 38-A2.
Exhibit 38-A2 Ramp FFS, SFR (mi/h) Single-Lane Ramps Two-Lane Ramps
Capacity of Ramp Roadways >50 2,200 4,400
(pc/h) >40–50 2,100 4,200
>30–40 2,000 4,000
≥20–30 1,900 3,800
<20 1,800 3,600
Notes: Capacity of a ramp roadway does not ensure an equal capacity at its freeway or other high-speed junction.
Junction capacity must be checked against criteria in Exhibit 14-10 and Exhibit 14-11.
FFS = free-flow speed.
Case B: Ramp Terminal

Demand at the downstream urban street intersection approach is compared
against the approach’s estimated capacity. If the ramp terminal consists of two
interdependent intersections, the analyst must proceed to Chapter 23, Ramp
Terminals and Alternative Intersections. Otherwise, depending on the type of
intersection located at the end of the ramp roadway, the capacity is obtained
from one of the following chapters: Chapter 19, Signalized Intersections; Chapter
20, Two-Way STOP-Controlled Intersections; Chapter 21, All-Way STOP
Controlled Intersections; or Chapter 22, Roundabouts.
The ramp terminal control will generate queues even during undersaturated
operations. The recommended approach for evaluating queues is as follows:
• Signalized intersections: Although an oversaturated approach is expected to
create longer queues that grow over time and are more likely to spill back
into the freeway diverge, it cannot be guaranteed that the queues from an
undersaturated approach will not affect the freeway mainline. Therefore,
in each analysis period, the methodology estimates the queue length and
compares it to the available storage length. The demand arriving at the
intersection may be constrained by the ramp roadway capacity; for this
reason, the ramp roadway capacity check must be conducted first.
• Unsignalized intersections: Intersection operation better than LOS F does
not guarantee that spillback will not occur. The recommended approach
is to proceed to Case B of the queue length estimation methodology.
Case C: Downstream Merge Junction

Queue spillback may also occur on freeway-to-freeway connectors, and this
is a common issue at high-demand urban interchanges. In this case, the
bottleneck is located at the downstream merge segment and occurs when the
discharge rate into the downstream merge is lower than the off-ramp demand.
As a result, the queue may spill back into the upstream freeway lanes. In this
case, the downstream freeway facility’s merge capacity must be modeled using
Appendix A: Off-Ramp Queue Spillback Analysis Chapter 38/Network Analysis

the Chapter 10 method. For oversaturated conditions, the methodology estimates

the queue length at the on-ramp as described in Chapter 25, Freeway Facilities:
Supplemental. This queue length value should be used as the input to the queue
spillback analysis described next.
The arriving demand at a downstream merge may be constrained by the
ramp roadway capacity. Therefore, the entering ramp demand at the merge is
the minimum value of the exiting flow rate at the diverge and the ramp roadway
capacity.
QUEUE LENGTH ESTIMATION

In this stage, the procedure estimates the expected queue length for any
condition where demand exceeds capacity. Three cases may occur.

A queue forms as a result of demand exceeding capacity at the entrance to
the ramp roadway and is expected to affect operations. To determine the extent
of this effect, the queue growth during each analysis period is estimated as:
𝑄𝑔,𝑖 = (𝑣𝑅 − 𝑐𝑅 ) × 𝑓𝐻𝑉 × 𝑃𝐻𝐹 × 𝑇 Equation 38-A1
where
Qg,i = queue growth during analysis period i (veh),
vR = off‐ramp demand (pc/h),
cR = off-ramp roadway capacity (pc/h),
fHV = adjustment factor for heavy vehicle presence,
PHF = peak hour factor, and
T = analysis time period (0.25 h for a 15-min period) (h).
The ramp queue during the first time period of the analysis must be zero,
otherwise the analysis’ time–space domain boundaries need to be re-evaluated.
The accumulated queue length at the end of analysis period t is the cumulative
value of Qg,i until t:
𝑡
𝑄𝑡 = ∑ 𝑄𝑔,𝑖 Equation 38-A2

𝑖=0
where
Qt = accumulated queue length at the end of analysis period t (veh),
Qg,i = queue growth during analysis period i (veh), and
t = the current analysis period.
Chapter 9, Glossary and Symbols, defines the study period as “the time
interval within a day for which facility performance is evaluated, consisting of
one or more consecutive analysis periods.” Therefore, the study period t refers to
the time boundaries defined in Step A-1 of the freeway facilities methodology. It
is composed of N analysis periods, which typically have 15-min durations.

The maximum queue length Qmax during the entire study period is the
maximum value of Qt obtained using Equation 38-A2 and is used as an input for
the next stage of the spillback check procedure.
Case B: Ramp Terminal

Spillback occurs when the queue from the downstream ramp terminal
intersection exceeds the available ramp storage. For all cases, the procedure
estimates the maximum throughput v at the downstream intersection approach.
That maximum throughput is limited by the ramp roadway’s capacity cR:
Equation 38-A3 𝑣 = min(𝑣𝑅 , 𝑐𝑅 ) × 𝑓𝐻𝑉 × 𝑃𝐻𝐹
where
v = maximum entering flow rate for the intersection approach (veh/h),
vR = off-ramp demand for the period (pc/h),
cR = capacity of the off-ramp roadway (pc/h),
fHV = adjustment factor for heavy vehicle presence, and
PHF = peak hour factor.
If the off-ramp demand exceeds its capacity, the ramp roadway acts as an
upstream bottleneck and limits the demand arriving at the intersection approach.
This step ensures that the incoming demand at the intersection does not exceed
the ramp roadway’s capacity. The throughput calculations for each intersection
type are described below.
The methodology of Chapters 19 and 31 evaluates the performance of
individual lane groups for a subject approach. It also estimates the back-of-queue
length Q (Equation 31-149) or a percentile back-of-queue length Q% (Equation
31-150). In some cases, only one high-demand movement on the intersection
approach is the bottleneck that results in spillback, yielding an unbalanced lane
usage pattern at the ramp. Field observations have shown that urban street
intersection failures may occur for one lane group. As drivers position themselves
in a specific lane at the ramp in anticipation of the downstream signal, the ramp’s
lane usage becomes unbalanced, as illustrated in Equation 38-A3.
Exhibit 38-A3
Examples of Unbalanced
Ramp Lane Usage
(a) Norfolk, VA (b) Tampa, FL

Sources: (a) RITIS CATT Lab, (b) Florida Department of Transportation.

When off-ramps have two or more lanes, the estimated queue length for each
intersection lane group must be associated with specific ramp lanes. Exhibit 38-
A4 illustrates an example of a typical ramp terminal for a two-lane off-ramp.
Drivers that desire to make a left turn at the intersection will position themselves
in the leftmost lane (ramp lane 2), while drivers who intend to turn right will
likely choose the rightmost lane (ramp lane 1). Analyst judgement is required to
define the grouping of intersection lane groups into ramp lanes.
Exhibit 38-A4
Illustrative Assignment of
Intersection Lane Groups to
Ramp Lanes
By using the results of the queue estimation procedure, the number of

queued vehicles in a given ramp lane k is estimated as follows:
𝑄𝑙,𝑘 = ∑ 𝑄𝐿𝐺,𝑚 = 𝑄%,𝐿𝐺𝑛 × 𝑁𝐿𝐺𝑚 Equation 38-A4
where
Ql,k = number of queued vehicles in ramp lane k during a 15-min interval
(veh);
QLG,m = number of queued vehicles from lane group m associated with ramp
lane k during a 15-min interval (veh);
Q%,LGn = estimated back of queue length (nth percentile), from Equation 38-A5
(derived from Equation 31-150) (veh/ln); and
NLG.m = number of approaching lanes for lane group m.
with
𝑄%,𝐿𝐺𝑛 = (𝑄1 + 𝑄2 )𝑓𝐵% + 𝑄3 Equation 38-A5
where
𝑄𝑖 = ith-term back-of-queue size (veh/ln), from Equation 31-141 to Equation
31-143; and
𝑓𝐵% = percentile back-of-queue factor corresponding to the nth percentile,
from Equation 31-151 or 31-153.
Unsignalized Intersections
Each unsignalized intersection type has its own methodology to estimate
queue length. The TWSC methodology estimates the 95th percentile queue
length for minor movements with Equation 20-68, while the 95th percentile
queue length for AWSC approaches is estimated with Equation 21-33. For
roundabouts, the 95th percentile queue length for a given lane is provided by
Equation 22-20. Once the lane group queue(s) have been determined, they are

assigned to ramp lanes following the same procedure described above for
signalized intersections.
Case C: Downstream Merge

For freeway-to-freeway connectors, the estimated queue length at the
downstream merge is estimated using Equation 25-21, from the Chapter 25
oversaturated segment evaluation methodology. For this specific type of
connector, the demand difference between ramp lanes can be considered
negligible for the purposes of this analysis.
QUEUE STORAGE RATIOS AND SPILLBACK CHCECKS

The next stage of the procedure estimates the queue storage ratio RQ for the
ramp roadway queues. If RQ exceeds 1.00, spillback is expected to occur. The
calculations for each of the three possible cases are provided below.

If the demand exceeds the capacity of the ramp roadway, this step estimates
the queue storage ratio RQ for the ramp roadway queues as follows:
𝐿ℎ 𝑄𝑚𝑎𝑥
Equation 38-A6 𝑅𝑄 =
𝐿𝑅 𝑁
Equation 38-A6 assumes all where
lanes provide the same
storage. If that is not the case, RQ = queue storage ratio (decimal),
the analyst should calculate
the total queue storage as a Lh = average vehicle spacing in a stationary queue (ft/veh),
sum of each lane’s storage.
Qmax = maximum number of vehicles queued on the ramp (veh),
LR = available queue storage (ft/ln), and
N = number of ramp lanes.
In Case A, the bottleneck is the entry to the off-ramp, and the ramp itself
would not necessarily have a queue present. This case estimates the impacts of
the queue as it extends along the deceleration lane. The queue length upstream
of the ramp roadway QSP is estimated based on the “leftover” demand not served
by the off-ramp’s available capacity:
Equation 38-A7 𝑄𝑆𝑃 = (𝑣𝑅 − 𝑐𝑅 ) × 𝑓𝐻𝑉 × 𝑃𝐻𝐹 × 𝐿ℎ × 𝑇
where
QSP = length of queue beyond the ramp storage distance (ft),
vR = off-ramp demand (pc/h),
cR = off-ramp capacity (pc/h),
fHV = adjustment factor for heavy vehicle presence,
PHF = peak hour factor,
Lh = average vehicle spacing in a stationary queue (ft/veh), and
T = analysis time period (0.25 h for a 15-min period (h).

Case B: Downstream Intersection

When demand exceeds capacity at the intersection, the methodology
considers the queues for all lanes from the ramp gore to the stop bar, as well as
the channelization at the stop bar. The total ramp storage length LR can be
estimated as the sum of lane lengths for i different sections, with a section
defined as a uniform length of ramp roadway with a homogenous number of
lanes:
𝐿𝑅 = ∑ 𝑁𝑖 × 𝐿𝑖 Equation 38-A8
𝑖
where
LR = ramp storage length (ft),
Ni = number of lanes in section i, and
Li = length of section 𝑖 (ft).
The individual ramp storage for each of the k lanes in the off-ramp LR.k can be
estimated by assigning the intersection lane groups to ramp lanes, as previously
described:
𝐿𝑅,𝑘 = ∑ 𝑁𝑖,𝑘 × 𝐿𝑖 Equation 38-A9

𝑖
where
LR,k = ramp storage length for lane k (ft),
Ni,k = number of lanes in section 𝑖 associated with ramp lane k, and
Li = length of section 𝑖 (ft).
Finally, the ramp queue ratio for every ramp lane k is obtained as:
𝑄𝐿,𝑘 × 𝐿ℎ
𝑅𝑄,𝑘 = Equation 38-A10
𝐿𝑅,𝑘
where
RQ,k = ramp queue ratio for ramp lane k (decimal),
QL,k = queue length associated with ramp lane k (veh),
LR,k = ramp storage length for lane k (ft).
Next, the total storage length is calculated. The example from Exhibit 38-A4
illustrates a common off-ramp geometry with three different sections from the
stop bar to the gore point:
• Section 1: 4 lanes with length L1; two lanes (LG1) are associated with
ramp lane 1, and two lanes (LG2) are associated with ramp lane 2.
• Section 2: 3 lanes with length L2; one lane (LG1) is associated with ramp
lane 1, and two lanes (LG2) are associated with ramp lane 2.
• Section 3: 2 lanes with length L3; one lane (LG1) is associated with ramp
lane 1, and one lane (LG2) is associated with ramp lane 2.

Therefore, the available ramp storage LR is calculated as:

𝐿𝑅 = (4 × 𝐿1 ) + (3 × 𝐿2 ) + (2 × 𝐿3 )
The ramp storage for each ramp lane is as follows:
𝐿𝑅,1 = (2 × 𝐿1 ) + (1 × 𝐿2 ) + (1 × 𝐿3 )
𝐿𝑅,2 = (2 × 𝐿1 ) + (2 × 𝐿2 ) + (1 × 𝐿3 )
Case C: Downstream Merge

The queue storage ratio for freeway-to-freeway connections is estimated as
follows:
𝑄𝑂𝑁𝑅 × 𝐿ℎ
Equation 38-A11 𝑅𝑄 =
𝐿𝑅 𝑁
where
RQ = queue storage ratio (decimal),
QONR = downstream on-ramp queue length (veh);
LR = ramp storage length (ft),
N = number of ramp lanes.
The queue length at the downstream on-ramp QONR is obtained from the
Chapter 25 oversaturated segment evaluation procedure through the parameter
ONRQ (Equation 25-21). The parameter ONRQ(i, t, p) is defined as the unmet
demand that is stored as a queue on the merge ramp roadway at node i during
time step t in time interval p (veh), and is computed at every 15-s time step. The
on-ramp queue length at the end of a time interval p is obtained by the ONRQ
value at the last time step of time interval p.
OFF-RAMP QUEUE SPILLBACK EVALUATION

Chapter 14 provides three LOS checks for diverge segments, and failure
(LOS F) may occur in either of the following two cases:
• Total demand flow rate on the approaching upstream freeway segment
exceeds the upstream freeway segment’s capacity, or
• Off-ramp demand exceeds the off-ramp capacity.
Chapter 14 also provides a LOS evaluation based on the density of the ramp
influence area (Exhibit 14-3), but it only yields a LOS range of A through E;
failure due to excessive density is not considered in the network analysis
methodology. The first case of LOS F is addressed by the Chapter 25
oversaturated segment evaluation procedure (HCM Chapter 10) and is not the
focus of the network analysis methodology. This section addresses the second
case of LOS F, when the off-ramp demand exceeds the off-ramp capacity, as well
as cases of spillback due to insufficient capacity at the ramp terminal
downstream of an off-ramp.
The methodology described in this section presents the steps applied to
determine whether spillback from an off-ramp is expected to occur during a

standard 15-min analysis period. If spillback is expected to occur, this section

then provides the methodology for evaluating the spillback’s effects on freeway
performance. The approach is based on the Chapter 25 oversaturated segment
evaluation procedure, where performance measures are computed at the 15-s
time step level.
Off-Ramp Operations
To evaluate the interaction between the freeway mainline and the
downstream off-ramp terminal, the link–node approach used by Chapter 25 to
evaluate oversaturated freeway segments is expanded, with additional links and
nodes used to represent the off-ramp segment. As shown in Exhibit 38-A5, the
mainline node for the off-ramp (node 3) is connected to the off-ramp segment,
which has a three-node structure:
• Ramp node 3.1: Interface between the freeway diverge segment (exit lanes)
and the upstream end of the ramp roadway. The volume that flows
through this node is equivalent to the number of vehicles that are able to
leave the freeway.
• Ramp node 3.2: Interface between the ramp roadway and the arterial
intersection approach. The volume that flows through this node is
equivalent to the number of vehicles that are able to leave the ramp
roadway and approach the intersection;
• Ramp node 3.3: Discharge capacity of the arterial intersection approach.
The volume that flows through this node is equivalent to the number of
vehicles that are able to enter the intersection.
Exhibit 38-A5
Expanded Link–Node
Structure to Evaluate Off-
Ramp Segments
An off-ramp is seldom a homogenous road segment, and additional lanes are

frequently added closer to the arterial intersection approach. Exhibit 38-A6
illustrates an example off-ramp, considering its entire length from the
deceleration lane to the stop bar at the downstream signalized intersection. The
ramp roadway is the uniform ramp segment with a downstream boundary

defined by the point where additional lanes are provided. When modeling the
off-ramp geometry, the method considers the channelization at the intersection
approach, because imbalances in the turning movement demands may cause
queues on a subset of lanes. Exhibit 38-A6 shows a typical queue formation
resulting from a left-turn movement that operates with insufficient capacity. In
this scenario, the approaching left-turn vehicles are positioned in the leftmost
lane and spillback may occur even when some approach lanes are undersaturated.
Exhibit 38-A6
Example Off-Ramp Geometry
with Heavy Left-Turn Demand
at a Signalized Intersection
The type of ramp terminal is an important input into the analysis. Signalized
intersections operate in cyclical patterns, and therefore have fluctuating queue
lengths. For certain demand scenarios, this pattern can result in queues backing
up into the freeway and then discharging multiples times within a 15-min
analysis period.
STOP-controlled intersections and downstream merge segments (in the case
of a freeway-to-freeway connection) have more uniform discharging rates. For
cases other than signalized intersections, off-ramp queues are assumed to develop
or discharge linearly based on the relationship between demand and capacity.
Freeway Mainline Operations
Spillback Regimes
The impact of queue spillback on the freeway mainline varies as a function of
the queue length and the lanes blocked. Five spillback regimes are defined (A-1)
and are illustrated in Exhibit 38-A7.
Regime 0
Under this regime, shown in Exhibit 38-2(a), there are no queues in the ramp
roadway or the queue, if it exists, is contained within the ramp roadway
boundaries. There are no operational effects in the ramp influence area.
Regime 1
The queue ends within the deceleration lane and does not spill back into the
freeway mainline, as shown in Exhibit 38-A7(b). During undersaturated
conditions, the deceleration lane serves as a transition zone between speeds on
the mainline (typically 55–75 mi/h) and the advisory speed posted for the off-
ramp (typically 20–50 mi/h). When queues begin to form on the deceleration
lane, the available deceleration distance is reduced and speeds along the
rightmost lane are affected.

Exhibit 38-A7
Regimes
(a) Regine 0: No queue or queue contained within the ramp roadway
(b) Regime 1: Queue at the deceleration lane (c) Regime 2: Queue along the shoulder
(d) Regime 3: Queue along the rightmost lane (e) Regime 4: Queue blockage of the adjacent lane
Regime 2
The queue extends upstream beyond the deceleration lane, but sufficient
lateral clearance on the right-hand shoulder provides additional queue storage.
As shown in Exhibit 38-A7(c), there is no transition zone within the deceleration
lane. Drivers decelerate and join the back of the queue more abruptly, resulting
in turbulence and reduced speeds in the rightmost lane. If no lateral clearance
exists immediately upstream of the deceleration lane, Regime 2 conditions are
not possible. In some cases, this regime does not occur even when storage is
available, depending on site-specific driver behavior.
Regime 3
The queue extends to the rightmost freeway mainline lane, as shown in
Exhibit 38-A7(d). This regime may occur when no shoulder is available for
additional queue storage or when drivers choose to queue in the rightmost lane
once the deceleration lane is entirely occupied. Non-exiting vehicles on the
rightmost lane are delayed or change lanes, which causes increased turbulence
and reduced speeds in the two rightmost lanes.
Regime 4
The queue blocks the rightmost lane, and drivers occasionally or often use
the next freeway mainline lane to the left to force their way into the queue, thus
blocking an additional lane, as shown in Exhibit 38-A7(e). During this regime,
freeway speed and capacity are significantly reduced. The effects of spillback
vary by site and time interval due to differences in driver behavior and site
geometry. Data collection at locations around the United States has shown that
drivers block the adjacent lane at some sites, but do not at other sites, regardless
of the queue spillback length at a given site.

Glossary of Variable Definitions

This glossary defines internal variables used by the off-ramp queue spillback
evaluation methodology. The structure of the variables is similar to that used in
Chapter 25.
Facility Variable
• QIA(i, p)—Length of the queue influence area (ft) for segment i during
analysis period p, measured from the back of the queue.
Segment Variables
• ΔK(i, p)— additional density in the queued mainline lines in segment i
during analysis period p (pc/mi/ln).
• ΔNV (i, t, p) — additional number of passenger cars in the congested
portion of segment i due to an off-ramp queue during time step t in
analysis period p (pc),
• KBBL(i, j)—background density (pc/mi/ln) at the blocked lanes in segment
i, when queue spillback occurs at a downstream segment j.
• KBUB(i, j)—background density (pc/mi/ln) at the unblocked lanes in
segment i, when queue spillback occurs at a downstream segment j.
• KQBL(i, t, p)— queue density (pc/mi/ln) of the blocked portion of segment
i during time step t in analysis period p.
• KQUB(i, t, p)— queue density in the unblocked portion of segment i
during time step t in analysis period p (pc/mi/ln).
• L(i) —length of segment i (ft).
• LCR(i, t, p)—rate of lane change maneuvers in the queue influence area
upstream of a queue from an off-ramp, for segment i during time step t in
analysis period p.
• LD(i, p)—available deceleration lane length (ft) for segment i during
analysis period p. This variable is used to calculate performance measures
for ramp segments.
• MO2UB(i, t, p) —maximum number of passenger cars that can enter the
unblocked portion of segment i during time step t in analysis period p due
to the presence of a queue in the downstream ramp segment.
• MO2BL(i, t, p) —maximum number of passenger cars that can enter the
blocked portion of segment i during time step t in analysis period p due to
the presence of a queue in the downstream ramp segment.
• MQ1(i, t, p)—queue length of off-ramp unserved vehicles in the rightmost
mainline lane for segment i during time step t in analysis period p.
• MQ2(i, t, p)—queue length of off-ramp unserved vehicles in the second-to-
the-right mainline lane, for segment i during time step t in analysis period
p. If Regime 4 is not expected to occur, this parameter value is set to zero.
• NQ(i)—number of blocked lanes if the off-ramp queue backs up into the
freeway mainline. This parameter is a function of the prevailing spillback
regime at segment i as provided by the analyst. The value for this

parameter is an input and can be either 1 (Regime 3, one blocked lane) or 2

(Regime 4 , two blocked lanes).
• OFRFUP(i, t, p)—flow that can exit at the closest off-ramp downstream of
segment i during time step t in analysis period p.
• OFRLQ(i, t, p)—queue length of off-ramp unserved vehicles for diverge
segment i during time step t in analysis period p.
• OFRUV(i, t, p)—number of off-ramp unserved passenger cars for segment
i during time step t in analysis period p.
• SBKQ(i, t, p)—spillback queue density for segment i during time step t in
analysis period p.
• SBLQ(i, t, p)—queue length (ft) within segment i during time step t in
analysis period p, caused by a downstream off-ramp bottleneck.
• SBQS(i, p)—total available off-ramp queue storage (ft) for a diverge
segment i during analysis period p, if the subject segment has an off-ramp
bottleneck. It is calculated as a function of the available storage lengths in
the deceleration lane and shoulder and the prevailing spillback regime.
• SCEQ(i, N, NQ)—equivalent capacity of the unblocked portion of a
segment i with N total lanes and NQ blocked lanes.
• SF(i, t, p)—segment flow out of segment i during time step t in analysis
period p (veh/ts).
• SL(i, p)—available shoulder length (ft) for segment i during analysis period p.
If the value of SL is greater than zero, any off-ramp queues that exceed the
deceleration lane will occupy the shoulder before blocking mainline lanes.
• TIA(i, t, p)—total influence area (ft) upstream of a queue from an off-ramp
bottleneck on segment i during time step t in analysis period p. It is
calculated as the sum of parameters QIA(i, t, p) and MQ(i, t, p).
Node Variables
• CAFBL(i, t, p)—capacity adjustment when one or more lanes of segment i
are entirely blocked during time step t in analysis period p. This variable
is used to calculate friction effects that cause through vehicles to slow
down due to the presence of a queue in the rightmost lanes.
• CAFUP(i, t, p)—capacity adjustment factor for node i during time step t in
analysis period p. This variable affects approaching vehicles within the
queue influence area (QIA) upstream of an off-ramp queue. It accounts for
the turbulence caused by intense lane changing within the QIA as vehicles
adjust their position when there is a downstream off-ramp queue.
• MFBL(i, t, p)—mainline flow rate that can cross the blocked portion of
node i during time step t in analysis period p.
• MFUB(i, t, p)—mainline flow rate that can cross the unblocked portion of
• MIBL(i, t, p)—maximum flow desiring to enter the blocked portion of

• MIUB(i, t, p)—maximum flow desiring to enter the unblocked portion of

• MO2BL(i, t, p)—maximum number of passenger cars that can enter the
blocked portion of segment i during time step t and analysis period p due
to the presence of a queue in the downstream ramp segment.
• MO2UB(i, t, p)—maximum number of passenger cars that can enter the
unblocked portion of segment i during time step t and analysis period p
due to the presence of a queue in the downstream ramp segment.
• NEXTOFR(i)—index of the nearest downstream diverge segment relative
to subject node i.
• OFRDIST(i)—distance (ft) from node i to the start of the deceleration lane
of the nearest downstream off-ramp.
• OFRPCT(i, j)—percent of the off-ramp demand at segment j over the
mainline entering volume at segment i.
• SBLC(i, t, p)—number of lane change maneuvers within the QIA at node i
during time step t in analysis period p.
Ramp Variables
• RC(i, t, p)—capacity of the ramp roadway (pc/ts) from segment i during
time step t in analysis period p. Obtained by dividing the capacity values
for the ramp roadway (pc/h) provided in Exhibit 14-12 in Chapter 14 by
the number of time steps in one hour (240).
• RF(i, t, p, k)—flow (pc/ts) that can enter the ramp roadway at segment i
during time step t in analysis period p using branch k.
• RI(i, t, p, k)—maximum flow (pc/ts) desiring to enter the off-ramp on
segment i during time step t in analysis period p using branch k, including
queues accumulated from previous time periods.
• RKC—ramp density at capacity (pc/mi/ln).
• RKQ(i, t, p, k)—ramp roadway queue density (pc/mi/ln) for segment i
during time step t in analysis period p using branch k.
• RL(i)—ramp roadway length (ft) for segment i.
• RN(i, p, k)—number of ramp lanes for branch k of segment i in analysis
period p. Similar to the number of mainline lanes, it could vary by time
interval if a temporary lane closure is in effect.
• RNV(i, t, p, k)—maximum number of passenger cars within the ramp of
segment i at the end of time step t during analysis period p using branch
k. The number of passenger cars is based initially on the calculations of
Chapters 12, 13, and 14, but as queues grow and dissipate, input–output
analysis updates these values during each time step.
• RO(i, t, p, k)—maximum flow (pc/ts) allowed to leave the ramp roadway
on segment i during time step t in analysis period p using branch k, due to
limited available storage at the downstream ramp terminal.

• RSTG(i, t, p, k)—maximum number of passenger cars that can enter

branch k of segment i during time step t in analysis period p, due to the
presence of a queue in the downstream ramp segment.
• RUV(i, t, p, k)—number of unserved passenger cars at the entrance of the
ramp roadway of segment i at the end of time step t during analysis
period p desiring to use branch k. Any values of RUV greater than zero
indicate the occurrence of queue spillback from an off-ramp.
• SR(i, p, k)—average speed on ramp roadway (mi/h) on branch k of
segment i during analysis period p.
Intersection (Ramp Terminal) Variables

• fHV(i, p)—adjustment factor for heavy vehicle presence for segment i
during analysis period p.
• GT(i, t, p, k)—green time (s) for lane group l from segment i during time
step t in analysis period p (s).
• ID(i, t, p, k)—discharge capacity (veh/ts) from branch k in segment i
during time step t in analysis period p.
• IF(i, t, k, p)—flow (veh/ts) that can enter the intersection on segment i
from branch k during time step t in analysis period p.
• II(i, t, p, k)—maximum flow (veh/ts) desiring to enter the intersection on
segment i using branch k during time step t in analysis period p, including
queues accumulated from previous time periods.
• IL(i, k)—storage length (ft) of movements at the intersection of segment i
using branch k.
• IN(i, k)—number of lanes serving the branch k of segment i,
• INV(i, t, p, k)—number of vehicles at the intersection of segment i using
branch k at the end of time step t in analysis period p.
• IO(i, t, p)—flow (veh/ts) that can be discharged from the intersection on
segment i using branch k during time step t in analysis period p.
• ISTG(i, k)—total available storage length (ft) from branch k at the
intersection of segment i.
• IUV(i, t, p, k)—number of unserved vehicles at the entrance of the
intersection of segment i using branch k at the end of time step t during
analysis period p.

Methodology
The methodology for evaluating off-ramp queue spillback is integrated with
the freeway facility oversaturated segment procedure given in Chapter 25.
Exhibit 38-A8 depicts the methodology, highlighting additions and changes to
the Chapter 25 methodology to address off-ramp queue spillback.
Exhibit 38-A8
Freeway Facility
Procedure, Adapted for
Evaluation

Exhibit 38-A8 (cont’d.)

Freeway Facility
Procedure, Adapted for Off-
Ramp Queue Spillback
Evaluation


Freeway Facility
Evaluation


Freeway Facility
Evaluation

Step 1: Calculate Background Density for Unblocked Lanes on each Segment

in the Case of Queue Spillback
The first step in the oversaturated segment evaluation procedure computes a
background density KB for each segment at the start of each analysis period,
defined as the expected density when there is no queueing on the segment. It
estimates how many vehicles occupy a given segment during undersaturated
conditions, creating an initial reference point for oversaturated analyses.
When Regime 3 or Regime 4 occur, one or more freeway lanes in the affected
segments are blocked, and through vehicles try to move to the unblocked lanes.
The capacity of the unblocked lanes must be calculated at the initialization step,
to be used as a reference value.
For a segment i with N lanes, a subset NQ of lanes will be blocked when
spillback occurs (NQ = 1 for Regime 3 and NQ = 2 for Regime 4). The capacity of
the unblocked lanes is equivalent to that of a similar segment with (N − NQ)
lanes, adjusted for the blockage impact of the blockage using a capacity
adjustment factor CAFBL. The CAFBL values are equal to the incident CAFs given
in Chapter 11, Freeway Reliability Analysis (Exhibit 11-23), because no data are
available currently to accurately assess the impacts of blockage due to spillback.
These CAF values may be conservative, because capacities may be further
reduced during incidents due to the presence of police vehicles. Exhibit 38-A9
presents the recommended values for CAFBL.
Exhibit 38-A9 Directional Lanes 1 Blocked Lane 2 Blocked Lanes

Capacity Adjustment Factors 2 0.70 N/A
for Lane Blockage CAFBL 3 0.74 0.51
4 0.77 0.50
5 0.81 0.67
6 0.85 0.75
7 0.88 0.80
8 0.89 0.84
The equivalent capacity SCEQ (in pc/h) of segment i with N lanes and NQ
blocked lanes is estimated as:
Equation 38-A12 𝑆𝐶𝐸𝑄(𝑖, 𝑁, 𝑁𝑄) = 𝑆𝐶(𝑖, 𝑁 − 𝑁𝑄) × 𝐶𝐴𝐹𝐵𝐿
Exhibit 38-A10 presents an example of a basic 4-lane directional segment
operating in Regime 4 (2 blocked lanes). The capacity of the unblocked lanes will
be equivalent to the capacity of a 2-lane basic segment with a capacity
adjustment factor CAFBL of 0.50 (4 directional lanes with 2 blocked lanes).
Exhibit 38-A10
Equivalent Segment Capacity
for Unblocked Lanes When
Lane Blockage Occurs
For the segment of Exhibit 38-A10, capacity under ideal conditions is:
• c = 2,400 pc/h (capacity per lane), or
• SC = 9,600 pc/h (segment capacity).

When Regime 4 occurs (2 blocked lanes), the equivalent capacity SCEQ is

obtained as the equivalent capacity of a 2-lane segment multiplied by a CAFBL of
0.50 from Exhibit 38-A9:
𝑆𝐶𝐸𝑄 = 2 × 2,400 × 0.5 = 2,400 pc/h
Next, the unblocked queue density KBUB is calculated. This parameter
estimates the queue density of the uncongested portion of a segment operating
under a two-pipe regime due to queue spillback from a downstream off-ramp.
To estimate this value, the method first determines the ratio of the expected
demand ED that will move to the uncongested side of the segment. When queue
spillback occurs in a diverge segment j, the parameter OFRPCT(j) is defined as
the off-ramp demand divided by the mainline entering volume:
𝑣𝑅 (𝑗)
𝑂𝐹𝑅𝑃𝐶𝑇(𝑗) = Equation 38-A13
𝑣𝐹 (𝑗)
For any segment i upstream of segment j and affected by off-ramp spillback
from segment j, the proportion of vehicles traveling towards the off-ramp at
segment i is given by OFRPCT(j), while the proportion of vehicles continuing
through in the unblocked lanes is 1 − OFRPCT(j). Therefore, the unblocked queue
density KBUB at any segment i upstream of off-ramp spillback in segment j is
given by:
𝐾𝐵𝑈𝐵(𝑖, 𝑗) = 𝐾𝐵[𝐸𝐷(𝑖) × (1 − 𝑂𝐹𝑅𝑃𝐶𝑇(𝑖)), 𝑆𝐶𝐸𝑄(𝑗)] Equation 38-A14
where
KBUB(i, j) = background density at the unblocked lanes in segment i, when
queue spillback occurs at the downstream segment;
ED(i) = expected demand at segment i, as defined in Chapter 25;
OFRPCT(i) = rate of off-ramp flow and mainline flow at segment i; and
KB[v, c] = density of a segment with demand flow rate v and capacity c, as
provided by Chapter 12 (basic segments), Chapter 13 (weaving
segments), or Chapter 14 (merge and diverge segments).
Step 2: Initialize the Freeway Facility

These calculations are performed at the start of the analysis, to prepare the
flow calculations for the first time step and to specify return points, such as
background density KB, for later time steps. This subsection presents the
additional parameters required for queue spillback analysis.
Number of mainline blocked lanes. The number of mainline blocked lanes
is stored in the parameter NQ(i) and is determined by the prevailing queue
spillback regime provided by the analyst. If the back of an off-ramp queue is
calculated to reach the freeway mainline, two possible spillback regimes may
occur:
• Regime 3: one lane blocked on the freeway mainline—set NQ(i) = 1; or
• Regime 4: two lanes blocked on the freeway mainline—set NQ(i) = 2.
The analyst should select one of these two regimes based on the prevailing
driver behavior at the site and in the vicinity of the site.

Shoulder length. The available shoulder length must be input by the analyst
for queue spillback analysis and is stored in the parameter SL(i) for
oversaturated calculations.
Deceleration lane length. The deceleration lane length is provided by the
analyst for the analysis of diverge segments. It is stored in the parameter LD(i)
for oversaturated calculations.
Spillback queue storage length. The maximum storage length for off-ramp
queues on segment i is computed as a function of the segment length L(i), the
deceleration lane length LD(i), and the number of queued lanes NQ(i). Exhibit 38-
A11 provides guidance on measuring each of the components required for
Regimes 3 and 4.
Exhibit 38-A11
Maximum Off-Ramp Queue
Storage Length at Diverge
Segments with Regime 3 or 4
Queue Spillback and
No Shoulder Available
(a) Regime 3 (b) Regime 4
Exhibit 38-A12 illustrates queue length measurements for special cases of

queue spillback when a shoulder is present, but its storage length is not sufficient
to accommodate the unserved vehicles. Regime 3A, shown in Exhibit 38-A12(a),
occurs when one mainline lane is blocked in addition to the shoulder. Regime
4A, shown in Exhibit 38-A12 (b) occurs when two mainline lanes are blocked in
addition to the shoulder.
Exhibit 38-A12
Maximum Off-Ramp Queue
Storage Length at Diverge
Segments with Regime 3 or 4
Queue Spillback and
Shoulder Available
(a) Regime 3A (b) Regime 4A
Step 2A: Model Off-Ramp Geometry

The three-branch node structure for the off-ramp shown in Exhibit 38-A5
must be modeled to reflect the site’s geometric characteristics, as illustrated in
Exhibit 38-A6. This is accomplished by setting a branch structure where a node
can connect to multiple downstream links. If a node is connected to more than
one downstream link, the flow through the node will be constrained by the
downstream link with the highest queue storage ratio.

The ramp structure must be modeled from the downstream end towards the
upstream end:
• For the location farthest downstream, provide one node for each lane
group or movement on the approach.
• At each subsequent upstream change in alignment, provide one node for
each ramp roadway lane connecting to a distinct lane group downstream.
The data structure used in the computations should be adjusted according to
this branch structure. Most parameters in the Chapter 25 oversaturated segment
evaluation methodology are computed as three-dimensional arrays (i, t, p),
where i is the segment’s index within the freeway facility and t refers to a specific
time step within a given analysis period p. In the case of two-lane ramps that
need to be evaluated independently, an extra dimension k is added to the ramp
parameter arrays to account for the specific lane under analysis. Lanes are
numbered right from left; therefore, k = 1 indicates the right ramp lane and k = 2
indicates the left ramp lane.
Example 1. In this example, shown in Exhibit 38-A13, only one lane connects
the freeway exit to the entry leg of the downstream roundabout. Therefore, only
one node is required at each location (i.e., a single branch structure, with k = 1 at
all nodes).
Exhibit 38-A13
Node Structure for Example 1
Example 2. A single-lane ramp connects to a STOP-controlled T-intersection

ramp terminal, as shown in Exhibit 38-A14. The intersection node has two lanes
serving different turning movements and therefore has two branches (k = 2). Each
turning movement (left turn and right turn) is represented by a node, and when
a queue develops for either movement, the longer queue will constrain the flow
of vehicles from the ramp roadway. The ramp roadway itself has only one lane
and therefore only one branch (k = 1).

Exhibit 38-A14
Example 3. A two-lane ramp connects to a signalized intersection ramp

terminal (Exhibit 38-A15). Both the intersection and the ramp roadway nodes
have two branches each (k = 2). At the downstream end of the ramp, one node is
defined for each lane group at the intersection (left turn and right turn). Based on
the ramp geometry, left-turning vehicles will use ramp lane 2, while right-
turning vehicles will use ramp lane 1. Therefore, two nodes are also defined at
the upstream location. If the queue storage ratio for any of the ramp lanes
reaches 1, vehicle flow in the respective upstream node will be constrained,
resulting in queue spillback on the freeway mainline.
Exhibit 38-A15
Step 2B: Determine Spillback Regime for each Diverge Segment

Field observations (A-2) have shown that locations that experience recurring
queue spillback always have the same type of spillback regime when the queue
extends beyond the deceleration lane (i.e., Regime 3 or 4). Regime 4 occurs often
at ramp junctions with a lane drop. At these locations, the exiting traffic can access
the off-ramp with a single lane change. Therefore, drivers are more likely to wait
until they are closer to the exit to change lanes and block the adjacent through

lane when spillback occurs. However, not all lane drop exits experience Regime 4
queue spillback. Regime 4 also occurs more frequently in locations with more
aggressive driver behavior. Local information and driver behavior should be
taken into consideration in determining the prevailing regime at a given site.
For operational analyses of existing locations, it is recommended that the
analyst provide the expected spillback regime based on observed field
conditions. For planning-level purposes where no field data are available, Exhibit
38-A16 provides the expected queue spillback regime as a function of the number
of exiting lanes and driver aggressiveness.
Driver Aggressiveness Exhibit 38-A16

Ramp Geometry Low Medium High Default Spillback Regime as a
Function of Ramp Geometry
Diverge Regime 3 Regime 3 Regime 3
and Driver Aggressiveness
Lane Drop Regime 3 Regime 4 Regime 4
Step 2C: Determine Queue Influence Area

Chapter 14 defines the ramp influence area of right-hand off-ramps
operating under steady conditions as the deceleration lane(s) and Lanes 1 and 2
of the freeway for a distance of 1,500 ft upstream of the diverge point. When
queue spillback occurs in one or more freeway lanes, drivers react to the
presence of the queue further upstream of the ramp, resulting in increasing lane
changes and additional turbulence upstream of the ramp influence area, as
illustrated in Exhibit 38-A17. This step estimates the QIA length, measured
upstream from the back of queue.
Exhibit 38-A17
Queue Influence Area with
Increased Turbulence
The QIA length is based on the time needed by arriving drivers to react to
partial lane blockage and to adjust their speeds and positions. Research (A-1) has
shown that traffic speeds upstream of the back of queue are negatively affected
at a headway distance of 10.95 s. Therefore, the influence area represents the
distance traversed by a vehicle during 10.95 s with a speed consistent with the
traffic stream.
The length is estimated as a function of the segment free-flow speed (FFS), as
shown in Exhibit 38-A18. The exact location of the QIA varies as a function of the
queue length. QIA lengths are shorter than the ramp influence distance of 1,500
ft. However, the two concepts are very different and are used differently in
analyzing ramp operations: the ramp influence area is used to analyze
undersaturated conditions, while the QIA is used to analyze oversaturated
conditions. Because drivers can only detect a downstream queue visually, they
have shorter reaction times compared to arriving at undersaturated off-ramps,
where signing and navigation information is provided in advance and allows
drivers to adjust their position earlier.

Exhibit 38-A18 Segment FFS (mi/h) Queue Influence Area (ft)

Queue Influence Area as 50 810
Function of the Segment Free- 55 900
Flow Speed 60 980
65 1,060
70 1,140
75 1,220
When Regimes 3 or 4 occur and lane blockage occurs on the mainline, the
QIA is added to the queue length to determine the extent of spillback effects. If
an upstream node is located within the combined length of the queue and QIA, a
capacity adjustment factor CAFUP (Equation 38-A39) must be applied to account
for the spillback effects.
Step 2D: Determine Ramp Roadway Capacity and Speed

The first off-ramp parameter to be determined is its capacity RC. This value
is a function of the ramp FFS and is obtained from Exhibit 14-12 in Chapter 14,
reproduced below as Exhibit 38-A19. The ramp capacity is compared to the off-
ramp demand. If the demand-to-capacity ratio is greater than 1.0, queue
spillback is expected to occur.
Exhibit 38-A19 Ramp FFS, SFR (mi/h) Single-Lane Ramps Two-Lane Ramps
Capacity of Ramp Roadways >50 2,200 4,400
(pc/h) >40–50 2,100 4,200
>30–40 2,000 4,000
≥20–30 1,900 3,800
<20 1,800 3,600
Determining the ramp roadway’s speed–flow relationship is also required

for the analysis. Ramp speeds can be obtained through the following equation:
𝑣𝑅
Equation 38-A15 𝑆ramp = (1 − 0.109 × ) × 𝑆𝐹𝑅
1000
where
Sramp = ramp speed (mi/h),
vR = ramp demand flow rate (pc/h), and
SFR = ramp free-flow speed (mi/h).
The ramp speed–flow relationship is linear, with speed decreasing at higher
ramp flows, as shown in Exhibit 38-A20. The maximum value of vR is governed
by the ramp capacity given in Exhibit 38-A19.

Exhibit 38-A20
Freeway Ramp Speed–Flow
Curves
The ramp density at capacity RKC is not necessarily the same as the 45
pc/mi/ln value used for freeway mainline lanes. This parameter is used to
evaluate the queue density at the ramp roadway during oversaturated conditions.
The ramp density at capacity is found by dividing the capacity by the speed.
Exhibit 38-A21 lists RKC values as a function of the ramp FFS.
Ramp Free-Flow Speed Ramp Capacity Ramp Density at Capacity Exhibit 38-A21
(mi/h) (pc/h/ln) (pc/mi/ln) Ramp Density at Capacity as a
Function of Ramp FFS
55 2,200 40.0
50 2,100 42.0
45 2,100 46.7
40 2,000 50.0
35 2,000 57.1
30 1,900 63.3
25 1,900 76.0
20 1,900 90.0
15 1,800 120.0
Step 2E: Determine Intersection Storage Capacity

The intersection storage capacity ISTG is obtained as the sum of the available
storage of every lane group, multiplied by the number of lanes. If the off-ramp
has multiple branches at the intersection (i.e., k > 1), the available storage capacity
must be computed for each branch k individually. This distinction is necessary to
evaluate cases with unbalanced demands at the intersection, when the queues
developed in one oversaturated movement may extend upstream and block the
throughput of all off-ramp movements. ISTG is estimated as:
𝑀
𝐼𝑆𝑇𝐺(𝑖, 𝑘) = (∑ 𝑁𝑚 × 𝐿𝑚 ) × 𝐿ℎ Equation 38-A16
𝑚
where
ISTG(i, k) = total available storage length from branch k at the intersection of
segment i (ft),
Nm = number of lanes serving movement m at the intersection,
Lm = storage length for movement m at the intersection (ft),
M = number of movements at the approach, and
Lh = average vehicle spacing in a stationary queue (ft/veh), from
Equation 31-155.

Step 2F: Determine Initial Number of Vehicles at the Off-Ramp

The computation of the number of vehicles in the off-ramp at every time step
is required to derive performance measures during oversaturated conditions.
Similar to the Chapter 25 Oversaturated Segment Evaluation methodology,
estimating the number of vehicles in the off-ramp under oversaturated
conditions first requires a reference value for undersaturated conditions to be
computed during the initialization steps.
The density of an off-ramp segment is the off-ramp flow rate vR divided by
its speed Sramp. The total number of vehicles RNV is obtained next by multiplying
the ramp density by the ramp length and number of lanes, as follows:
𝑣𝑅,0
Equation 38-A17 𝑅𝑁𝑉(𝑖, 0,0, 𝑘) = × 𝑅𝐿(𝑖) × 𝑅𝑁(𝑖, 𝑝, 𝑘)
𝑆ramp,0
where
RNV(i, 0, 0, k) = number of vehicles in the ramp roadway at the initialization
step (veh),
vR,0 = off-ramp demand during the first time interval (pc/h),
Sramp,0 = off-ramp speed during the first time interval (mi/h),
RL(i) = ramp roadway length for segment i (ft), and
RN(i, p, k) = number of ramp lanes for branch k of segment i in analysis
period p.
The initial number of vehicles in the intersection approach are also
determined as an initial reference point, as follows:
Equation 38-A18 𝐼𝑁𝑉(𝑖, 0, 𝑝, 𝑘) = 𝐼𝑁(𝑖, 𝑘) × 𝑄𝑘
where
INV(i, t, p, k) = number of vehicles at the intersection of segment i using branch
k at the end of time step t during analysis period p (veh),
IN(i, k) = number of lanes serving branch k of segment i, and
Qk = back-of-queue length for branch k (veh).
The back-of-queue length Qk is estimated from the appropriate equation for
the intersection type at the ramp terminal, as shown in Exhibit 38-A22.
Exhibit 38-A22 Intersection Type Reference Equation

Reference Equations for Back-
Signalized 31-149
of-Queue Length Estimation
Two-way STOP-controlled 20-68
All-way STOP-controlled 21-33
Roundabout 22-20
At signalized intersections, due to their cyclic nature, queues form and

discharge at different times for different movements. Therefore, a reference point
within the cycle must be selected as a starting point. The methodology assumes
pretimed control or converts actuated control to the equivalent pretimed pattern.
Typical signalized intersections at ramp terminals have the off-ramp approach as
the minor movement, with a start of green on the right side of the barrier, as

shown in Exhibit 38-A23. It is recommended setting a reference point at the onset

of green for phases 3 and 7, because the back-of-queue lengths at this time can be
readily estimated using the methodology of Chapter 31, Section 4.
Exhibit 38-A23
Selection of a Cycle Reference
Point to Determine the Initial
Number of Vehicles Within the
Approach
(a) Phase numbers corresponding to movements (b) Phase sequence
Step 2G: Determine the Capacity of the Downstream Terminal

The methodology for evaluating the capacity of the ramp terminal is specific
to each intersection type and relies mostly on the methodology of the chapter
corresponding to the intersection type (Chapters 19 through 23). However, due
to the cyclic nature of signalized intersections, additional analysis steps are
required when performing a time step analysis, as described in the remainder of
this step.
The capacity of each movement of a signalized intersection approach at each
time step is a function of the signal phase sequence and the capacities of the
individual movements at the intersection. Equation 38-A24 illustrates a sample
signalized intersection approach from an off-ramp, with two lane groups: left
turn (Phase 3) and right turn (Phase 8).
Exhibit 38-A24
Example Signalized
Intersection Approach from an
Off-Ramp
Input Parameters. The required parameters for evaluating ramp terminal

capacity are generally the same as those listed in Exhibit 19-11 for standard
signalized intersection analyses.
Arrival type. Exhibit 19-14 in Chapter 19 provides guidance for selecting the
appropriate arrival types based on the arterial’s operations characteristics, such as
quality of progression and coordination. Vehicle arrivals on an off-ramp approach
can be considered to be random. Therefore, Arrival Type 3 (random arrivals) is
recommended to analyze the off-ramp approach at a signalized ramp terminal.

Phase duration and effective green time. The duration of each phase at the signal
can be fixed (pre-timed control), or variable (semi-actuated or actuated control).
For the former case, phase duration is known. For the latter, an average phase
duration is estimated as described in Chapter 31, Section 2. The effective green
time g for each phase can then be computed according to Equation 19-3,
reproduced here as:
Equation 38-A19 𝑔 = 𝐷𝑝 − 𝑙1 − 𝑙2
where
l2 = clearance lost time = Y + Rc – e (s),
Y = yellow change interval (s),
Rc = red clearance interval (s), and
e = extension of effective green = 2.0 (s).
Converting approach capacity from analysis periods to time steps. The
standard signalized intersection analysis is performed in 15-min analysis
periods, while the queue spillback evaluation requires a 15-s time step approach
compatible with the freeway facilities oversaturated methodology. Therefore, an
adjustment is necessary to calculate each movement’s capacity in 15-s intervals.
The cycle length C can be divided into n time steps, with a duration of 15 s
each, as seen in Exhibit 38-A25. If an integer number of time steps is not
obtained, the difference is included in the first time-step of the next cycle. Next,
green times for each time step from 1 to n are computed. This procedure must be
repeated for every time step within the 15-min analysis period, resulting in a
total of 900 / 15 = 60 time steps.
Exhibit 38-A25
Assignment of Green Times to
Time Steps

The capacity ID for each movement, corresponding to a branch, for each time
step is obtained by multiplying the movement’s green time by its capacity:
𝐼𝐷(𝑖, 𝑡, 𝑝, 𝑘) = 𝑁𝑘 𝑠𝑘 𝐺𝑇(𝑖, 𝑡, 𝑝, 𝑘) × 𝑓𝐻𝑉 Equation 38-A20
where
ID(i, t, p, k) = discharge capacity from branch k in segment i during time step t
in analysis period p (veh/ts),
Nk = number of lanes serving movement k,
sk = saturation flow rate for movement k (veh/h/ln),
GT(i, t, p, l) = green time for lane group l from segment i during time step t in
analysis period p (s), and
fHV = adjustment factor for heavy vehicle presence (decimal).
The green time parameter GT(i, t, p, k) can range from 0 s (when the movement
has red through the entire time step length) to 15 s (when the movement has
green through the entire time step length). The heavy vehicle factor fHV is applied
to make the units used for intersection capacity (veh/h) consistent with the flow
rates used by the uninterrupted flow methods (pc/h).
Step 2H: Determine Reference Index for the Next Downstream Off-Ramp
This step is required to build this procedure’s computational engine, but it is
unimportant for understanding the overall methodology. The freeway facilities
methodology uses the parameter OFRF(i, t, p) to store the off-ramp flow rate at
diverge segment i. When a segment upstream of an off-ramp is evaluated for
queue spillback, the off-ramp flow rate must be referenced to estimate the
incoming flows for the blocked and non-blocked lanes. Therefore, a new variable
NEXTOFR(i) is introduced to reference the index of the closest diverge segment
downstream of segment i. This process is illustrated in Exhibit 38-A26, where the
node (i + 2) represents a diverge segment with an off-ramp flow vR. When the
queue extends upstream to node i, the approaching flow vf is split into two
groups: the exiting vehicles that will join the back of the queue, and the through
vehicles that will use the non-blocked lanes.
Exhibit 38-A26
Illustration of Mainline Flow
Rate Split into Blocked and
Unblocked Lanes
For nodes i and i + 1, the closest downstream off-ramp is located at node i + 2,

therefore the following parameter is computed:
𝑁𝐸𝑋𝑇𝑂𝐹𝑅(𝑖) = 𝑖 + 2

The parameter NEXTOFR facilitates referencing diverge segments

downstream of a given segment i. It will be used for the spillback analysis
procedure described in the next section.
Step 9A: Perform Spillback Analysis

This is a new step in the freeway facilities method (Exhibit 38-A8). In this
step, spillback effects in a diverge segment are determined after the off-ramp
flow OFRF is determined in steps 7 and 8.
Determine ramp input. The ramp input RI represents demand and it is the
number of passenger cars that wish to travel through the ramp roadway node
during a given time step. It takes into account the off-ramp demand OFRF (as
defined in the freeway facilities oversaturated methodology) and the number of
off-ramp unserved vehicles from the previous time step RUV. The OFRF
parameter takes into consideration any bottleneck segments upstream of the
diverge that may meter the off-ramp demand (Equations 25-23 through 25-25 in
Chapter 25). The ramp input is calculated as:
Equation 38-A21 𝑅𝐼(𝑖, 𝑡, 𝑝) = 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝) + 𝑅𝑈𝑉(𝑖, 𝑡 − 1, 𝑝)
where
RI(i, t, p) = maximum flow desiring to enter the off-ramp on segment i
during time step t in analysis period p, including queues
accumulated from previous time periods (pc/ts);
OFRF(i, t, p) = flow that can exit the off-ramp i during time step t in time period p
(pc/ts);
RUV(i, t, p) = number of unserved passenger cars at the entrance of the ramp
roadway of segment i at the end of time step t during analysis
period p.
Calculate flow to the off-ramp and number of unserved vehicles. The ramp
maximum flow RF represents capacity, i.e., the number of vehicles that are able
to enter the ramp roadway by crossing the boundary node between the diverge
segment and the ramp roadway. It is calculated as the minimum of three
variables: RI, RC, and RSTG.
Equation 38-A22 𝑅𝐹(𝑖, 𝑡, 𝑝) = min[𝑅𝐼(𝑖, 𝑡, 𝑝, 𝑘), 𝑅𝐶(𝑖, 𝑡, 𝑝), 𝑅𝑆𝑇𝐺(𝑖, 𝑡, 𝑝, 𝑘)]
where
RF(i, t, p) = flow that can enter the ramp roadway at segment i during time
step t in analysis period p (pc/ts);
RI(i, t, p, k) = maximum flow desiring to enter the off-ramp using branch k
of segment i during time step t in analysis period p, including
queues accumulated from previous time periods (pc/ts);
RC(i, t, p) = capacity of the ramp roadway from segment i during time step
t in analysis period p (pc/ts); and
RSTG(i, t, p, k) = maximum number of passenger cars that can enter branch k of
segment i during time step t in analysis period p, due to the
presence of a queue in the downstream ramp segment (pc).

The calculation of RSTG starts by calculating the maximum number of

passenger cars allowed on the ramp at a given ramp queue density RKQ. The
calculation of RKQ, in turn, takes an approach similar to the calculation of the
mainline queue density KQ (Equation 25-10), with the following remarks on the
inputs:
• The jam density parameter KJ uses the same value adopted for the
mainline calculations.
• The ramp density at capacity RKC is determined from the ramp FFS, as
given in Exhibit 38-A21.
• The parameters SF (segment flow) and SC (segment capacity) from
Equation 25-10 are replaced with RF (ramp flow) and RC (ramp capacity).
[(𝐾𝐽 – 𝑅𝐾𝐶) × 𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)]
𝑅𝐾𝑄(𝑖, 𝑡, 𝑝, 𝑘) = 𝐾𝐽– Equation 38-A23
𝑅𝐶(𝑖, 𝑝)
where
RKQ(i, t, p, k) = ramp roadway queue density for segment i during time step t
in analysis period p using branch k (pc/mi/ln),
KJ = facilitywide jam density (pc/mi/ln),
RKC = ramp density at capacity (pc/mi/ln),
RF(i, t, p) = flow that can enter the ramp roadway at segment i during time
step t in analysis period p (pc/ts), and
RC(i, p) = capacity of the ramp roadway from segment i in analysis
period p (pc/h).
The parameter RSTG is then calculated using a similar approach to that
taken by the mainline output 2 parameter MO2 (Equation 25-11 in Chapter 25):
𝑅𝑆𝑇𝐺(𝑖, 𝑡, 𝑝, 𝑘) = 𝑅𝐹(𝑖, 𝑡 − 1, 𝑝, 𝑘) + 𝑅𝐾𝑄(𝑖, 𝑡, 𝑝, 𝑘) × [𝑅𝐿(𝑖) × 𝑅𝑁(𝑖, 𝑘)]
Equation 38-A24
− 𝑅𝑁𝑉(𝑖, 𝑡 − 1, 𝑝, 𝑘)
where
RSTG(i, t, p, k) = maximum number of passenger cars that can enter branch k of
segment i during time step t in analysis period p, due to the
presence of a queue in the downstream ramp segment (pc);
RF(i, t, p, k) = flow that can enter the ramp roadway at segment i using
branch k during time step t in analysis period p (pc/ts);
RKQ(i, t, p, k) = ramp roadway queue density for segment i during time step t
in analysis period p using branch k (pc/mi/ln);
RL(i) = ramp roadway length for segment i (ft);
RN(i, k) = number of ramp lanes for branch k of segment i; and
RNV(i, t, p, k) = maximum number of passenger cars within the ramp of
segment i at the end of time step t during analysis period p
using branch k (pc).
Next, the number of unserved passenger cars at the ramp entrance RUV is
calculated. For each time step, the number of unserved passenger cars is

computed as the value from the previous time step, plus the difference between
demand RI and throughput RF at the ramp node. RUV is calculated as:
Equation 38-A25 𝑅𝑈𝑉(𝑖, 𝑡, 𝑝, 𝑘) = 𝑅𝑈𝑉(𝑖, 𝑡 − 1, 𝑝 , 𝑘) + 𝑅𝐼(𝑖, 𝑡, 𝑝, 𝑘) − 𝑅𝐹(𝑖, 𝑡, 𝑝, 𝑘)
where
RUV(i, t, p, k) = number of unserved passenger cars at the entrance of the ramp
roadway of segment i at the end of time step t during analysis
period p desiring to use branch k (pc);
RI(i, t, p, k) = maximum flow desiring to enter the off-ramp using branch k
of segment i during time step t in analysis period p, including
queues accumulated from previous time periods (pc/ts); and
branch k during time step t in analysis period p (pc/ts).
If there are multiple branches k at the ramp roadway (e.g., two lane ramps),
RI and RF are compared for each branch k to obtain RUV for each branch k. The
total number of unserved passenger cars at the ramp RUV(i, t, p) is then obtained
as the sum of RUV for each branch:
𝐾
Equation 38-A26 𝑅𝑈𝑉(𝑖, 𝑡, 𝑝) = ∑ 𝑅𝑈𝑉(𝑖, 𝑡, 𝑝, 𝑘)

𝑘
where K is the total number of branches and other variables are as defined
previously.
Calculate intersection approach input. The intersection approach input II is
the number of vehicles that wish to travel through the intersection node during a
given time step, i.e., its demand. It takes into account the off-ramp flow RF and
the number of unserved vehicles on the approach from the previous time step
IUV. The intersection approach input is calculated as:
Equation 38-A27 𝐼𝐼(𝑖, 𝑡, 𝑝, 𝑘) = 𝑓𝐻𝑉(𝑖, 𝑝) × 𝑅𝐹(𝑖, 𝑡, 𝑝, 𝑘) + 𝐼𝑈𝑉(𝑖, 𝑡, 𝑝, 𝑘)
where
II(i, t, p, k) = maximum flow desiring to enter the intersection on segment i
using branch k during time step t in analysis period p, including
queues accumulated from previous analysis periods (veh/ts);
fHV(i, p) = adjustment factor for heavy vehicle presence for segment i
during analysis period p;
branch k during time step t in analysis period p (pc/ts); and
IUV(i, t, p, k) = number of unserved vehicles at the entrance of the intersection
of segment i using branch k at the end of time step t during
analysis period p (veh).
Calculate maximum allowable ramp output. The maximum allowable ramp
output RO is calculated as the available storage space within the intersection
approach, minus the number of vehicles present at the previous time step and
the number of vehicles discharged during the present time period.

𝑅𝑂(𝑖, 𝑡, 𝑝, 𝑘) = 𝐼𝑆𝑇𝐺(𝑖, 𝑘) − 𝐼𝑈𝑉(𝑖, 𝑡 − 1, 𝑝, 𝑘) + 𝐼𝐷(𝑖, 𝑡, 𝑝, 𝑘) Equation 38-A28
where
RO(i, t, p, k) = maximum flow allowed to leave the ramp roadway on
segment i during time step t in analysis period p using branch
k, due to limited available storage at the downstream ramp
terminal (veh/ts);
ISTG(i, k) = total available storage length from branch k at the intersection
of segment i (ft);
IUV(i, t, p, k) = number of unserved vehicles at the entrance of the intersection
of segment i using branch k at the end of time step t during
analysis period p (veh); and
ID(i, t, p, k) = discharge capacity (veh/ts) from branch k in segment i during
time step t in analysis period p (veh/ts).
Calculate intersection approach flow and number of unserved vehicles.
The intersection flow IF represents the number of vehicles that are able to cross
the boundary node between the ramp roadway and the intersection (i.e., its
capacity). It is computed as the smaller of the number of vehicles wishing to
enter the intersection and the maximum number of vehicles allowed to enter the
intersection due to the available queue storage at the intersection:
𝐼𝐹(𝑖, 𝑡, 𝑝, 𝑘) = min[𝐼𝐼(𝑖, 𝑡, 𝑝, 𝑘), 𝑅𝑂(𝑖, 𝑡, 𝑝, 𝑘)] Equation 38-A29
where IF(i, t, p, k) is the flow that can enter the intersection on segment i from
branch k during time step t in analysis period p (veh/ts), and other variables are
as defined previously.
If the number of vehicles trying to enter the intersection exceeds the amount
of vehicles allowed to enter the intersection, the number of total unserved
vehicles must be computed and considered in the intersection input II during the
next time period:
𝐼𝑈𝑉(𝑖, 𝑡, 𝑝, 𝑘) = 𝐼𝑈𝑉(𝑖, 𝑡 − 1, 𝑝, 𝑘) + 𝐼𝐼(𝑖, 𝑡, 𝑝, 𝑘) − 𝑅𝑂(𝑖, 𝑡, 𝑝, 𝑘) Equation 38-A30

Update number of vehicles at the ramp terminal intersection. The number
of vehicles at the intersection INV is updated every time step based on the value
of INV from the previous time step, plus the number of vehicles that enter the
intersection approach, minus the number of vehicles that are discharged. The
maximum allowable total number of vehicles is a function of the available
storage at the intersection ISTG.
𝐼𝑁𝑉(𝑖, 𝑡, 𝑝) = 𝐼𝑁𝑉(𝑖, 𝑡 − 1, 𝑝) + 𝐼𝐹(𝑖, 𝑡, 𝑝, 𝑘) − 𝐼𝐶(𝑖, 𝑡, 𝑝, 𝑘) Equation 38-A31
where INV(i, t, p) is the number of vehicles at the intersection of segment i using

branch k at the end of time step t in analysis period p, and other variables are as
defined previously.
Calculate number of unserved passenger cars at the off-ramp. The number
of unserved passenger cars OFRUV at the entrance of the ramp roadway is
updated every time step as the difference between the number of passenger cars
that wish to enter the ramp roadway RI and the flow through the ramp node RF:

Equation 38-A32 𝑂𝐹𝑅𝑈𝑉(𝑖, 𝑡, 𝑝) = 𝑅𝐼(𝑖, 𝑡, 𝑝) − 𝑅𝐹(𝑖, 𝑡, 𝑝)

where OFRUV(i, t, p) is the number of off-ramp unserved passenger cars for
segment i during time step t in analysis period p, and other variables are as
defined previously.
Calculate intersection approach output. The intersection flow IO represents
the actual number of vehicles discharging from the intersection approach. It is
computed as the smaller of the intersection discharge capacity and the sum of
number of vehicles present in the intersection and the intersection input demand:
Equation 38-A33 𝐼𝑂(𝑖, 𝑡, 𝑝, 𝑘) = min[𝐼𝐷𝐶(𝑖, 𝑡, 𝑝, 𝑘), 𝐼𝑁𝑉(𝑖, 𝑡 − 1, 𝑝, 𝑘) + 𝐼𝐼(𝑖, 𝑡, 𝑝, 𝑘)]
where IO(i, t, p, k) is the flow that can be discharged from the intersection on
segment i using branch k during time step t in analysis period p (veh/ts), and
other variables are as defined previously.
Update number of passenger cars at the ramp roadway. The number of
vehicles at the ramp roadway RNV at the end of each time step is calculated
based on the number of vehicles from the previous time step, plus the number of
vehicles that entered the ramp, minus the number of vehicles that left the ramp:
Equation 38-A34 𝑅𝑁𝑉(𝑖, 𝑡, 𝑝, 𝑘) = 𝑅𝑁𝑉(𝑖, 𝑡 − 1, 𝑝, 𝑘) + 𝑅𝐹(𝑖, 𝑡, 𝑝, 𝑘) − 𝐼𝐹(𝑖, 𝑡, 𝑝, 𝑘)
where RNV(i, t, p, k) is the maximum number of passenger cars within the ramp
of segment i at the end of time step t during analysis period p using branch k, and
other variables are as defined previously.
Determine the back-of-queue length and spillback regime. Field
observations have shown that off-ramp queues blocking mainline lanes are
typically not stationary. These queues usually consist of a platoon of closely
spaced vehicles moving at very low speeds (< 15 mi/h). The spacing between
vehicles is also longer than the average vehicle spacing in stationary queues,
represented by Lh (Equation 31-155). Therefore, the density of the spillback queue
follows the queue density at the ramp RKQ, which allows the queue length
OFRLQ to be estimated. This parameter estimates the total queue length
upstream of the off-ramp if all unserved vehicles formed a single queue.
𝑂𝐹𝑅𝑈𝑉(𝑖, 𝑡, 𝑝, 𝑘)
Equation 38-A35 𝑂𝐹𝑅𝐿𝑄(𝑖, 𝑡, 𝑝) = max ( ) for every branch 𝑘
𝑅𝐾𝑄(𝑖, 𝑡, 𝑝, 𝑘)
where OFRLQ(i, t, p) is the queue length of off-ramp unserved vehicles for
diverge segment i during time step t in analysis period p (ft), and other variables
are as defined previously. If the ramp roadway has two or more lanes, Equation
38-A35 considers the queues from multiple branches and uses the worst case.
Next, the mainline queue length SBLQ from the previous time step is
compared to the available spillback queue storage for the prevalent spillback
regime for the given time step, as follows:

If OFRLQ(i, t − 1, p) = 0 → Regime 0 Equation 38-A36
If 0 < OFRLQ(i, t − 1, p) ≤ LD(i, p) → Regime 1

If OFRLQ(i, t − 1, p) > LD(i, p):
If SL(i, p) > 0:
If OFRLQ(i, t − 1, p) < LD(i, p) + SL(i, p) → Regime 2
Else: Regime 3 or 4
where
OFRLQ(i, t, p) = queue length of off-ramp unserved vehicles for diverge
segment i during time step t in analysis period p (ft),
LD(i, p) = available deceleration lane length on segment i in analysis
period p (ft),
SBLQ(i, t, p) = queue length within segment i during time step t in analysis
period p (ft), and
SL(i, p) = available shoulder length for segment i during analysis period p
(ft).
Finally, the queue length in the mainline lanes MQ1 (lane 1) and MQ2 (lane
2) are obtained as a function of the expected spillback regime. The total queue
length OFRLQ minus the available storage lengths of the deceleration lane and
shoulder produces the queue length associated with the blockage.
If the site experiences Regime 3:
𝑀𝑄1(𝑖, 𝑡, 𝑝, 𝑘) = 𝑂𝐹𝑅𝐿𝑄(𝑖, 𝑡, 𝑝, 𝑘) − 𝐿𝐷 (𝑖, 𝑝) − 𝑆𝐿(𝑖, 𝑝) Equation 38-A37
𝑀𝑄2(𝑖, 𝑡, 𝑝, 𝑘) = 0
If the site experiences Regime 4:

𝑀𝑄1(𝑖, 𝑡, 𝑝) = 𝑀𝑄2(𝑖, 𝑡, 𝑝) = [𝑂𝐹𝑅𝐿𝑄(𝑖, 𝑡, 𝑝) − 𝐿𝐷(𝑖)– 𝑆𝐿(𝑖) ] / 2 Equation 38-A38
where MQ1(i, t, p) is the queue length of off-ramp unserved vehicles in the

rightmost mainline lane for segment i during time step t in analysis period p (ft),
MQ2(i, t, p) is the queue length of off-ramp unserved vehicles in the second-to-to-
the-right mainline lane for segment i during time step t in analysis period p (ft),
and other variables are as defined previously.
Check for effects on upstream nodes. The freeway nodes upstream of a
congested off-ramp may be affected by spillback as queues grow. When this
event occurs, the methodology calculates the length of the queue in the upstream
segment. The length of the queue within the subject segment will then be used to
evaluate whether the capacity of any upstream node is affected by the queue.
For upstream segments that may be affected by spillback, the queue length
within the segment (measured from its downstream end) is computed and stored
in the parameter SBLQ. This check is performed for every node upstream of a
congested off-ramp, as shown in Exhibit 38-A27.

Exhibit 38-A27
Procedure for Evaluating the
Impact of Queue Spillback on
Upstream Nodes and
Determining the Queue
Length within Upstream
Segments
When queue spillback occurs at a downstream off-ramp, the length of the

mainline queue measured from the start of the deceleration lane is known from
the previous step. If a given segment has a queue blocking one or more lanes,
three possible scenarios may occur at the node, as shown in Exhibit 38-A28:
1. Lane blockage—Queues extend through the entire segment and reach the
upstream node, causing the subject node to operate in a two-pipe regime.
The blocked lanes operate in a congested regime, with their capacity
constrained by the off-ramp capacity. The unblocked lanes operate in an
uncongested state, with a small reduction in capacity due to the friction of
through vehicles passing next to congested lanes. An adjustment factor
CAFBL is applied to the through lanes. This condition occurs when the
spillback queue length SBLQ(i) is greater than or equal to the segment
length L(i).
2. Increased turbulence—Queues extend partially through the segment; the
upstream node is located within the QIA. This region is characterized by
intense turbulence as drivers quickly perform lane changes to adjust their
position in reaction to the queue ahead. All lanes in node i have their
capacity reduced by an adjustment factor CAFUP. This condition occurs
when the sum of the spillback queue length SBLQ(i) and the queue
influence area length QIA(i) is greater than or equal to the segment length
L(i).

3. No effect—Queues extend partially through the segment, but the upstream

node is located within the QIA. No capacity adjustment factors are
applied to the node i for this condition, which occurs when the sum of the
spillback queue length SBLQ(i) and the queue influence area length QIA(i)
is less than the segment length L(i).
Exhibit 38-A28
Potential Effects of an Off-
Ramp Queue on Node i
(a) Lane Blockage (b) Increased Turbulence (c) No Effect
Calculate capacity adjustment factors. Depending on how the upstream

node is affected as described above (lane blockage, increased turbulence, or no
effect), this step computes the corresponding effects on capacity.
Lane blockage adjustment factor. When one or more lanes are blocked, the
subject node is analyzed as a two-pipe operation, with congested flow in one or
more lanes on the ramp side and uncongested flow in the remaining lanes. The
capacity of these lanes is equal to the number of queued vehicles discharged at
the downstream segment. The flow rate attempting to cross the node through the
congested lanes is equal to the off-ramp flow rate OFRF at the closest downstream
off-ramp.
Increased turbulence adjustment factor. When a node experiences the increased
turbulence case shown in Exhibit 38-A28(b), all lanes are affected by the
turbulence caused by the intense lane changing. In this case, an adjustment factor
CAFUP is applied uniformly to the node capacity:
𝐶𝐴𝐹UP (𝑖, 𝑡, 𝑝) = 1 − 0.52 × 𝐿𝐶𝑅(𝑖, 𝑡, 𝑝)0.81 Equation 38-A39
with
𝑆𝐵𝐿𝐶(𝑖, 𝑡, 𝑝)
𝐿𝐶𝑅(𝑖, 𝑡, 𝑝) = Equation 38-A40
𝑆𝐹(𝑖, 𝑡, 𝑝)
where
CAFUP(i, t, p) = capacity adjustment factor for node i during time step t in
analysis period p;
LCR(i, t, p) = rate of lane change maneuvers in the QIA upstream of a queue
from an off-ramp, for segment i during time step t in analysis
period p;
SBLC(i, t, p) = number of lane change maneuvers within the QIA at node i
during time step t in analysis period p; and
SF(i, t, p) = segment flow out of segment i during time step t in analysis
period p (veh/ts).

The parameter LCR estimates the rate of lane change maneuvers performed
by drivers within the QIA trying to adjust their position when spillback occurs.
The parameter SBLC estimates the number of lane change maneuvers performed.
Exiting vehicles move to the shoulder lane to attempt to join the back of the queue,
while through vehicles move toward the median lanes to avoid the queue.
The computation of SBLC for a given node requires an estimate of the
number of vehicles in each freeway lane that plan to exit at the off-ramp. For
each lane 𝑖, the parameter pi represents the percentage of the off-ramp demand vR
traveling in the subject lane. The value of pi is a function of the distance from the
off-ramp to the subject node, as follows:
1. Within the ramp influence area (1,500 ft upstream from the diverge
point), the off-ramp demand flow rate vR is entirely positioned in the two
rightmost lanes, based on the Chapter 14 diverge segment methodology.
Therefore, the sum of the off-ramp flow rate percentages in the ramp
influence area p1,R and p2,R is equal to 1. The Appendix C methodology to
estimate lane-by-lane flow distribution in freeway segments is used to
estimate the lane flow ratio LFR for lanes 1 and 2. The values of p1,R and
p2,R are then estimated as follows:
𝐿𝐹𝑅1
𝑝1,𝑅 =
𝐿𝐹𝑅1 + 𝐿𝐹𝑅2
Equation 38-A41
𝐿𝐹𝑅2
𝑝2,𝑅 =
𝐿𝐹𝑅1 + 𝐿𝐹𝑅2
2. Beyond 8,000 ft upstream from the diverge point, the off-ramp has
negligible influence, again based on the Chapter 14 methodology.
Therefore, for any nodes located more than 8,000 ft from the off-ramp, pi
is assumed to be equally distributed among all N freeway lanes:
1
Equation 38-A42 𝑝𝑖 =
𝑁
3. Between 1,500 and 8,000 ft upstream from the diverge point, pi can be
obtained through linear interpolation between the first two cases as a
function of the distance from the diverge point dOFR in feet, as given by
Equation 38-A43. Exhibit 38-A29 illustrates the distribution of pi for a 3-
lane freeway segment.
Equation 38-A43 1
(𝑁 − 𝑝1,𝑅 ) × (𝑑𝑂𝐹𝑅 − 1,500)
𝑝𝑖 = 𝑝𝑖,𝑅 +
6500
Exhibit 38-A29
Distribution of pi as Function
of Distance from the Diverge
Point, for a 3-Lane Segment

Once the lane-by-lane distribution of the off-ramp flow is known, the

number of lane change maneuvers SBLC can be estimated. For Regime 3 cases
(one blocked lane), the number of lane changes is:
𝑁
𝑆𝐵𝐿𝐶(𝑖, 𝑡, 𝑝) = 𝑣1 (1 − 𝑝𝑖 ) + ∑(𝑖 − 1) × 𝑣𝑖 𝑝𝑖 Equation 38-A44

𝑖=2
where
SBLC(i, t, p) = number of lane change maneuvers within the QIA at node i
during time step t in analysis period p;
vi = demand flow rate in lane i (pc/h), from Appendix C;
pi = percentage of the off-ramp demand traveling in lane i
(decimal); and
N = number of freeway lanes.
Equation 38-A44 adds the number of through vehicles in lane 1 that move to
lane 2 to avoid the queue and the number of exiting vehicles in the remaining
lanes that adjust their position to join the back of the queue, multiplied by the
necessary number of lane changes. Equation 38-A30 illustrates Equation 38-A44
applied to a 4-lane segment.
Exhibit 38-A30
Illustration of Lane-Change
Maneuvers within the Queue
Influence Area in a 4-Lane
Segment under Regime 3
For Regime 4 cases, the following equation is applied to obtain SBLC:

𝑁
𝑆𝐵𝐿𝐶(𝑖, 𝑡, 𝑝) = 2 × 𝑣1 (1 − 𝑝𝑖 ) + 𝑣2 (1 − 𝑝2 ) + ∑(𝑖 − 2) × 𝑣𝑖 𝑝𝑖 Equation 38-A45

3
where all variables are as defined previously. Exhibit 38-A31 illustrates Equation
38-A45 applied to a 4-lane segment.
Exhibit 38-A31
Illustration of Lane-Change
Maneuvers within the Queue
Influence Area in a 4-Lane
Segment under Regime 4
Step 9: Calculate Mainline Input

The Chapter 25 oversaturated segment procedure computes the mainline
input MI for each node in every time step. It is defined as the maximum flow
desiring to enter the subject node during the current time step.
An adjustment is necessary when the subject node operates in a two-pipe
regime, as the blocked and unblocked portions experience different input

demands. Because exiting and through drivers segregate when approaching a

queue, the mainline input demand on the blocked side consists of the off-ramp
demand, while the remaining demand moves to the unblocked side.
When node i operates in a two-pipe regime, the MI parameter is split into
two components: MIUB, representing the mainline input in the unblocked lanes,
and MIBL, representing the mainline input joining the back of the queue. These
parameters are computed as follows:
Equation 38-A46 𝑀𝐼𝐵𝐿(𝑖, 𝑡, 𝑝) = 𝑂𝐹𝑅𝐹(𝑁𝐸𝑋𝑇𝑂𝐹𝑅(𝑖), 𝑡, 𝑝)
Equation 38-A47 𝑀𝐼𝑈𝐵(𝑖, 𝑡, 𝑝) = 𝑀𝐼(𝑖, 𝑡, 𝑝) − 𝑀𝐼𝐵𝐿(𝑖, 𝑡, 𝑝)
where
MIBL(i, t, p) = maximum flow desiring to enter the blocked portion of node i
during time step t in analysis period p,
OFRF(i, t, p) = actual flow that can exit at off-ramp i during time step t in
analysis period p,
NEXTOFR(i) = index of the nearest downstream diverge segment relative to
subject node i,
MIUB(i, t, p) = maximum flow desiring to enter the unblocked portion of
node i during time step t in analysis period p, and
MO3(i, t, p) = maximum flow desiring to enter node i during time step t in
analysis period p.
Step 12: Calculate On-Ramp Maximum Output

If a merge segment is located upstream of an off-ramp bottleneck, the
capacity of the on-ramp output may be affected by the blockage caused by the
spillback queue. The on-ramp maximum output is calculated by Equation 25-18
in Chapter 25, based on a series of potential constraints that include ramp
metering, the on-ramp capacity, the merge capacity, and the presence of
downstream queues. At high demands on both the freeway and the on-ramp,
zipper merging (one-to-one) is expected to occur. Therefore, a new capacity
constraint is added to Equation 25-18, shown in bold font in Equation 38-A48 and
illustrated in Exhibit 38-A32:
Equation 38-A48 𝑂𝑁𝑅𝑂(𝑖, 𝑡, 𝑝)
𝑅𝑀(𝑖, 𝑡, 𝑝)
𝑂𝑁𝑅𝐶(𝑖, 𝑡, 𝑝)
min {𝑀𝐹(𝑖 + 1, 𝑡 − 1, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)} − 𝑀𝐼(𝑖, 𝑡, 𝑝)
𝑀𝑂3 (𝑖, 𝑡 − 1, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)
= min
max
min {𝑀𝐹(𝑖 + 1, 𝑡 − 1, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)} /2𝑁(𝑖, 𝑝)
𝑀𝑂3 (𝑖, 𝑡 − 1, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖, 𝑡 − 1, 𝑝)
𝑹𝑭(𝑶𝑭𝑹𝑵𝑬𝑿𝑻(𝒊), 𝒕, 𝒑))
{ { 𝟐 × 𝑵𝑸(𝑶𝑭𝑹𝑵𝑬𝑿𝑻(𝒊)) }}

where
ONRO(i, t, p) = maximum output flow rate that can enter the merge point
from on-ramp i during time step t in analysis period p;
RM(i, t, p) = maximum allowable rate of an on-ramp meter at the on-ramp
at node i during time step t in analysis period p;
ONRC(i, t, p) = geometric carrying capacity of the on-ramp at node i during
time step t in analysis period p;
MF(i, t, p) = actual mainline flow rate that can cross node i during time step
t in analysis period p;
ONRF(i, t, p) = actual ramp flow rate that can cross on-ramp node i during
time step t in time interval p;
MO3(i, t, p) = maximum allowable mainline flow rate across node i during
time step t in time interval p, limited by the presence of
queued vehicles at the upstream end of segment i while the
queue clears from the downstream end of segment i;
SC(i, t, p) = maximum flow rate that can pass through segment i at the end
of time step t in analysis period p based strictly on traffic and
geometric properties;
N(i, p) = number of lanes on segment i in analysis period p; and
all other variables are as defined previously.
Exhibit 38-A32
Effect of Queue Spillback on
the Discharge Capacity of an
Upstream On-Ramp
If one or more lanes are blocked due to a downstream off-ramp bottleneck,

the throughput in lane 1 will be equal to the maximum exit throughput in the
congested off-ramp if the site operates in Regime 3, or 50% of the maximum exit
throughput, if it operates in Regime 4. It is assumed that the on-ramp and the
flow arriving from upstream on Lane 1 contribute equally to the downstream
Lane 1 flow and thus the on-ramp maximum output. In this case, it is assumed to
be half of the downstream throughput in Lane 1.
Step 21: Calculate Mainline Output 2

The Chapter 25 oversaturated segment procedure calculates the maximum
number of vehicles MO that can exit a node, constrained by a downstream
bottleneck or by merging on-ramp traffic. Among the potential constraints used
in calculating MO, the Mainline Output 2 parameter MO2 accounts for queue

The freeway facilities

oversaturated methodology growth on a downstream segment, eventually limiting the maximum number of
calculates the number of
vehicles in a given segment vehicles that can enter it.
during every 15-s timestep and
aggregates them to a 15-min When a queue exists in a downstream segment caused by a downstream off-
analysis period to calculate the ramp bottleneck, the segment is expected to operate under two distinct densities,
average segment density.
Similarly, the parameters RKQ as illustrated in Exhibit 38-A33. Therefore, the total number of vehicles in the
and SBLQ are used to compute downstream segment takes into account two different density values: the ramp
the number of vehicles in the
congested portion of the queue density RKQ prevailing within the queued area (shown in red), and the
segment in order to aggregate background density KB prevailing in the remaining portion of the segment
density at the congested
portion of the segment. (shown in blue).
Exhibit 38-A33
Illustration of Different
Density Values within One
Diverge Segment
If there are no spillback effects, the segment operates with a uniform density.
In this case, the constraints for the unblocked and blocked portions (MO2UB and
MO2BL, respectively) are calculated proportionately to the number of unblocked
and blocked lanes:
(1 − 𝑁𝑄(𝑖))
Equation 38-A49 𝑀𝑂2𝑈𝐵(𝑖, 𝑡, 𝑝) = 𝑀𝑂2(𝑖, 𝑡, 𝑝) ×
𝑁(𝑖)
𝑁𝑄(𝑖)
Equation 38-A50 𝑀𝑂2𝐵𝐿(𝑖, 𝑡, 𝑝) = 𝑀𝑂2(𝑖, 𝑡, 𝑝) ×
𝑁(𝑖)
where
MO2UB(i, t, p) = maximum number of passenger cars that can enter the
unblocked portion of segment i during time step t and analysis
period p due to the presence of a queue in the downstream
ramp segment;
MO2(i, t, p) = maximum allowable mainline flow rate across node i during
time step t in time interval p, limited by available storage on
segment i due to a downstream queue;
NQ(i) = number of blocked lanes if the off-ramp queue backs up into
the freeway mainline in segment i;
N(i) = number of lanes in segment i; and
MO2BL(i, t, p) = maximum number of passenger cars that can enter the blocked
portion of segment i during time step t and analysis period p
due to the presence of a queue in the downstream ramp
segment.

If node i operates under increased turbulence (i.e., the node is in the QIA),
the unblocked portion of segment i will operate similar to a regular segment.
Therefore, the component MO2UB is equal to MO2 but proportional to the
number of lanes in the unblocked portion:
(1 − 𝑁𝑄(𝑖))
𝑀𝑂2𝑈𝐵(𝑖, 𝑡, 𝑝) = 𝑀𝑂2(𝑖, 𝑡, 𝑝) × Equation 38-A51
𝑁(𝑖)
For the blocked portion of segment i, MO2BL is calculated as equal to MO2
proportional to the number of lanes in the blocked portion plus an additional
number of vehicles due to the presence of a partial queue. This additional number
of vehicles is obtained by the bold terms in the following equation, which takes
into account the difference between the queue spillback density RKQ and the
segment queue density KQ, multiplied by the queue length:
𝑁𝑄(𝑖)
𝑀𝑂2𝐵𝐿(𝑖, 𝑡, 𝑝) = 𝑀𝑂2(𝑖, 𝑡, 𝑝) × Equation 38-A52
𝑁(𝑖)
+ 𝑺𝑩𝑳𝑸(𝒊, 𝒕 − 𝟏, 𝒑) × 𝑵𝑸(𝒊, 𝒕 − 𝟏, 𝒑)
× [𝑹𝑲𝑸(𝑶𝑭𝑹𝑵𝑬𝑿𝑻(𝒊), 𝒕 − 𝟏, 𝒑) − 𝑲𝑸(𝒊 − 𝟏, 𝒕 − 𝟏, 𝒑)]
where KQ(i, t, p) is the queue density (pc/mi/ln) of segment i during time step t in
analysis period p, and all other variables are as defined previously.
If node i experiences lane blockage, the values of queue density must be
computed for both the unblocked KQUB and blocked KQBL portions of segment
i. For the unblocked portion, KQUB is calculated similarly to Equation 25-10 in
Chapter 25, but the inputs for segment flow SF and segment capacity SC are
replaced by their equivalent parameters SFUB and SCEQ:
𝐾𝑄𝑈𝐵(𝑖, 𝑡, 𝑝) = 𝐾𝐽 − [(𝐾𝐽 − 𝐾𝐶) × 𝑆𝐹𝑈𝐵(𝑖, 𝑡 − 1, 𝑝)]/𝑆𝐶𝐸𝑄(𝑖, 𝑝) Equation 38-A53
where
KQUB(i, t, p) = queue density in the unblocked portion of segment i during
time step t in analysis period p (pc/mi/ln),
KJ = facilitywide jam density (pc/mi/ln),
KC = ideal density at capacity (pc/mi/ln),
SFUB(i, t, p) = segment flow out of the unblocked portion of segment i during
time step t in analysis period p (pc/ts), and
SCEQ(i, p) = equivalent capacity of the unblocked portion of segment i in
analysis period p (pc/ts).
The queue density for the blocked portion is equal to the ramp queue density:
𝐾𝑄𝐵𝐿(𝑖, 𝑡, 𝑝) = 𝑅𝐾𝑄(𝑂𝐹𝑅𝑁𝐸𝑋𝑇(𝑖), 𝑡 − 1, 𝑝) Equation 38-A54
where KQBL(i, t, p) is the queue density (pc/mi/ln) of the blocked portion of

segment i during time step t in analysis period p, and all other variables are as
defined previously.
With the queue density values for both the blocked and unblocked portions
of the segment known, the MO2 components MO2BL and MO2UB can be
computed:

Equation 38-A55 𝑀𝑂2𝑈𝐵(𝑖, 𝑡, 𝑝) = 𝑆𝐹𝑈𝐵(𝑖, 𝑡 − 1, 𝑝) − 𝑂𝑁𝑅𝐹(𝑖, 𝑡, 𝑝)

+ [𝐾𝑄𝑈𝐵(𝑖, 𝑡, 𝑝) × 𝐿(𝑖) × (𝑁(𝑖, 𝑝) − 𝑁𝑄(𝑖, 𝑝))] − 𝑁𝑉(𝑖, 𝑡 − 1, 𝑝)
Equation 38-A56 𝑀𝑂2𝐵𝐿(𝑖, 𝑡, 𝑝) = 𝑆𝐹𝐵𝐿(𝑖, 𝑡 − 1, 𝑝) − 𝑂𝑁𝑅𝐹(𝑖, 𝑡, 𝑝)

+ [𝐾𝑄𝐵𝐿(𝑖, 𝑡, 𝑝) × 𝐿(𝑖) × 𝑁𝑄(𝑖, 𝑝)] − 𝑁𝑉(𝑖, 𝑡 − 1, 𝑝)
where
MO2UB(i, t, p) = maximum number of passenger cars that can enter the
unblocked portion of segment i during time step t in analysis
period p due to the presence of a queue in the downstream
ramp segment,
L(i) = length of segment i (ft),
MO2BL(i, t, p) = maximum number of passenger cars that can enter the blocked
portion of segment i during time step t in analysis period p
due to the presence of a queue in the downstream ramp
segment, and
all other variables as previously defined.
Step 22: Calculate Mainline Flow

The Chapter 25 oversaturated segment evaluation procedure computes the
mainline flow through a subject node as the minimum of several variables, as
presented in Equation 25-16. If the node experiences spillback, the calculation of
mainline flow must consider the flow through both the blocked and the
unblocked portions of the node. Therefore, the Mainline Flow parameter MF is
split into two components in an approach similar to the Mainline Input
parameter: the component MFUB represents flow across the node in the
unblocked lanes, while the component MFBL represents the flow across the node
in the blocked lanes. For both components, the resulting flow is computed as the
minimum value between the input and the maximum allowed flow.
For MFUB, the maximum allowed flow is equal to the capacity of unblocked
lanes in the segment downstream of the node, represented by the parameter
SCEQ as computed in the initialization step:
Equation 38-A57 𝑀𝐹𝑈𝐵(𝑖, 𝑡, 𝑝) = min[𝑀𝐼𝑈𝐵(𝑖, 𝑡, 𝑝), 𝑆𝐶𝐸𝑄(𝑖, 𝑡, 𝑝), 𝑀𝑂2𝑈𝐵(𝑖, 𝑡, 𝑝)]
where MFUB(i, t, p) is the mainline flow rate (pc/h) that can cross the unblocked
portion of node i during time step t in analysis period p, and all other variables
For MFBL, the maximum allowed flow is equal to the flow allowed to enter the
nearest downstream off-ramp RF:
Equation 38-A58 𝑀𝐹𝐵𝐿(𝑖, 𝑡, 𝑝) = min[𝑀𝐼𝐵𝐿(𝑖, 𝑡, 𝑝), 𝑅𝐹(𝑁𝐸𝑋𝑇𝑂𝐹𝑅(𝑖), 𝑡, 𝑝), 𝑀𝑂2𝐵𝐿(𝑖, 𝑡, 𝑝)]
where MFBL(i) is the mainline flow rate (pc/h) that can cross the unblocked
portion of node i during time step t in analysis period p, and all other variables
Finally, the mainline flow through node i is computed as the sum of the
blocked and unblocked portions, as follows:
Equation 38-A59 𝑀𝐹(𝑖, 𝑡, 𝑝) = 𝑀𝐹𝑈𝐵(𝑖, 𝑡, 𝑝) + 𝑀𝐹𝐵𝐿(𝑖, 𝑡, 𝑝)

where MF(i, t, p) is the actual mainline flow rate (pc/h) that can cross node i
during time step t in analysis period p, and other variables are as previously
defined.
Step 25: Update Number of Passenger Cars in the Blocked Portion of the
Segment
The number of passenger cars in the blocked portion NVBL during increased
turbulence is updated based on the number of vehicles in the previous time step
and considers the number of passenger cars that are able to leave the current and
upstream segments:
𝑁𝑉𝐵𝐿(𝑖, 𝑡, 𝑝) = 𝑁𝑉𝐵𝐿(𝑖, 𝑡 − 1, 𝑝) + 𝑀𝐹𝐵𝐿(𝑖 − 1, 𝑡, 𝑝) + 𝑂𝑁𝑅𝐹(𝑖 − 1, 𝑡, 𝑝) Equation 38-A60
− 𝑀𝐹𝐵𝐿(𝑖, 𝑡, 𝑝) − 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝)
where NVBL(i, t, p) is the number of passenger cars present on the blocked
portion of segment i at the end of time step t during analysis period p, and all

The aggregated segment flow for a 15-min analysis period is obtained as the
sum of flows for every time step based on Equation 25-30, reproduced here as
Equation 38-A61:
𝑆
𝑇
𝑆𝐹(𝑖, 𝑝) = ∑ 𝑆𝐹(𝑖, 𝑡, 𝑝) Equation 38-A61
𝑆
𝑡=1
where
SF(i, p) = segment flow out of segment i during analysis period p (pc/h),
T = number of time steps in 1 h (integer),
S = number of time steps in an analysis period (integer), and
SF(i, t, p) = segment flow out of segment i during time step t in analysis
period p (pc/ts).
Similarly, the aggregated off-ramp flow is aggregated to 15-min analysis
periods:
𝑆
𝑇
𝑂𝐹𝑅𝐹(𝑖, 𝑝) = ∑ 𝑂𝐹𝑅𝐹(𝑖, 𝑡, 𝑝) Equation 38-A62
𝑆
𝑡=1
where OFRF(i, p) is the flow (pc/h) that can exit off-ramp i during analysis period p,
and all other variables are as previously defined.
The additional density in the queued lanes is obtained by aggregating the
additional number of vehicles NV(i, t, p) in the off-ramp queue:
𝑆
1
∆𝐾(𝑖, 𝑝) = ∑ ∆𝑁𝑉(𝑖, 𝑡, 𝑝) Equation 38-A63
𝑆 × 𝑁𝑄(𝑖) × 𝐿(𝑖)
𝑡=1
with
𝑂𝐹𝑅𝐿𝑄(𝑖, 𝑡, 𝑝) × [𝑆𝐵𝐾𝑄(𝑖, 𝑡, 𝑝) − 𝐾𝐵(𝑖, 𝑝)/𝑓𝐻𝑉(𝑖, 𝑝)]
∆𝑁𝑉(𝑖, 𝑡, 𝑝) = Equation 38-A64
5,280

where
ΔK(i, p) = additional density in the queued mainline lines in segment i
during analysis period p (pc/mi/ln);
S = number of time steps in an analysis period (integer);
NQ(i) = number of blocked lanes in segment i (integer);
L(i) = length of segment i (mi);
ΔNV (i, t, p) = additional number of passenger cars in the congested portion
of a segment i due to an off-ramp queue during time step t in
analysis period p (pc);
OFRLQ(i, t, p) = queue length of off-ramp unserved vehicles for diverge
segment i during time step t in analysis period p (ft);
SBKQ(i, t, p) = spillback queue density for segment i during time step t in
analysis period p (pc/mi/ln), defined as equal to the ramp
roadway density RKQ(i, t, p) of the downstream off-ramp
segment experiencing queue spillback during the same time
step t in analysis period p;
KB(i, t, p) = background density (veh/h/ln), from Chapter 25; and
fHV(i, p) = adjustment factor for heavy vehicle presence for segment i
during analysis period p.
Similar to the mainline, the flow in the ramp roadway is also aggregated:
𝑆
𝑇
Equation 38-A65 𝑅𝐹(𝑖, 𝑝, 𝑘) = ∑ 𝑅𝐹(𝑖, 𝑡, 𝑝, 𝑘)
𝑆
𝑡=1
where RF(i, p, k) is the flow that can enter the ramp roadway at segment i using
branch k in analysis period p (pc/h), and all other variables are as previously
defined.
The aggregated density at the ramp is calculated as the average of the
number of vehicles inside the ramp during the analysis period:
𝑆
1
Equation 38-A66 𝑅𝐾(𝑖, 𝑝, 𝑘) = ∑ 𝑅𝑁𝑉(𝑖, 𝑡, 𝑝, 𝑘)
𝑆
𝑡=1
where RK(i, p, k) is the ramp roadway density for segment i using branch k in
analysis period p (pc/mi/ln), and all other variables are as previously defined.
Finally, the average speed SR(i, p, k) on branch k of the ramp roadway of
segment i during analysis period p (mi/h) is obtained by dividing the total ramp
flow using the branch in the analysis period by its average density:
𝑅𝐹(𝑖, 𝑝, 𝑘)
Equation 38-A67 𝑆𝑅(𝑖, 𝑝, 𝑘) =
𝑅𝐾(𝑖, 𝑝, 𝑘)
where other variables are as previously defined.

REFERENCES
A-1. Elefteriadou, L., M. Armstrong, Y. Zheng and G. Riente. Highway These references can be found
in the Technical Reference
Capacity Manual (HCM) Systems Analysis Methodology. Federal Highway Library in Volume 4.
Administration, Washington, D.C., 2016.
A-2. University of Florida Transportation Institute; Cambridge Systematics,
Inc.; and A. Skabardonis. NCHRP Web-Only Document 290: Highway
Capacity Manual Methodologies for Corridors Involving Freeways and Surface
Streets. Transportation Research Board, Washington, D.C., 2020.

APPENDIX B: ON-RAMP QUEUE SPILLBACK ANALYSIS
Queue spillback into an urban street intersection may occur when the
freeway merge segment has insufficient capacity to process the ramp’s demand.
Spillback may also occur in cases of ramp metering. This appendix presents the
methodology for determining whether spillback will occur from an on-ramp into
the upstream intersection.
As shown in the framework in Exhibit 38-B1, the methodology considers
signalized intersections, two-way and all-way STOP-controlled intersections, and
roundabouts. The procedure first estimates the demand approaching the on-
ramp (determined based on the upstream intersection’s configuration), and then
estimates the on-ramp’s capacity. The Chapter 10 freeway facilities methodology
for oversaturated conditions can estimate the resulting queue length; however,
the user must input the on-ramp demand flow rate.
DEMAND ESTIMATION
The first step in the methodology calculates the entering demand flow rate vR
at the on-ramp as a function of the upstream intersection configuration and
operations. Under low-demand conditions, the on-ramp demand flow rate is
calculated as the sum of the demands on each of the intersection approaches that
discharge into the ramp. However, if any of these movements operates over
capacity, the total throughput to the ramp will be constrained by the capacity of
these oversaturated movements. Hence, this check ensures that the on-ramp
demand is not overestimated. The analysis approach for each of four intersection
types is presented next.
Case A: Signalized intersections

The throughputs of a signalized intersection are highly dependent on several
parameters, including phasing sequences, actuation, cycle lengths, and use of
permitted–protected phasing, among others. The methodology identifies the
movements that discharge to the on-ramp and their operational characteristics
(permitted or protected). For example, typical diamond interchanges will include
a left-turn movement, a right-turn movement and a through movement (which
will typically have negligible flow).
The on-ramp demand vR is computed as the sum of the throughputs of each
movement that discharges into the on-ramp. The throughput of a given
movement i is obtained as the minimum value of its demand and capacity:
𝑁
Equation 38-B1 𝑣𝑅 = ∑ min(𝑣𝑖 , 𝑐𝑖 )

1
where
vR = on-ramp demand (veh/h),
vi = demand for movement 𝑖 at the intersection (veh/h),
ci = demand for movement 𝑖 at the intersection (veh/h), and
N = number of intersection movements that discharge into the on-ramp.
Appendix B: On-Ramp Queue Spillback Analysis Chapter 38/Network Analysis

Exhibit 38-B1
Procedure for Detecting
Spillback Occurrence at an
On-Ramp
Chapter 38/Network Analysis Appendix B: On-Ramp Queue Spillback Analysis

If all movements operate below capacity, the on-ramp demand is the sum of
the movement demands. However, if any movement that discharges into the on-
ramp operates over capacity, the total throughput to the on-ramp will be lower
than the sum of the corresponding intersection movements.
In the case of movements not controlled by the traffic signal discharging into
the on-ramp, those movements’ demands must also be compared to their
capacities. The potential capacity cp,i of an unsignalized movement can be
computed by aggregating its saturation flow rates during different phases of a
cycle.
If the unsignalized movement is free-flowing and there are no other
conflicting movements discharging to the on-ramp, its saturation flow rate sFF is
obtained from Equation 19-8, reproduced below as Equation 38-B2, applying the
applicable adjustment factors:
Equation 38-B2 𝑠𝐹𝐹 = 𝑠0 𝑓𝑤 𝑓𝐻𝑉𝑔 𝑓𝑔 𝑓𝑏𝑏 𝑓𝑎 𝑓𝐿𝑈 𝑓𝐿𝑡 𝑓𝑅𝑇 𝑓𝐿𝑝𝑏 𝑓𝑅𝑝𝑏 𝑓𝑤𝑧 𝑓𝑚𝑠 𝑓𝑠𝑝
where
sFF = saturation flow rate for the unsignalized movement during free-flow
(veh/h/ln),
s0 = base saturation flow rate (pc/h/ln), and
all other adjustment factors are as described with Equation 19-8.
If the unsignalized movement must yield to a conflicting movement
discharging to the on-ramp, the permitted saturation flow rate sp is calculated
based on Equation 31-100, reproduced below as Equation 38-B3:
𝜆0 𝑒 −𝜆0 𝑡𝑐𝑔/3,600
Equation 38-B3 𝑠𝑝 =
1 − 𝑒 −𝜆0 𝑡𝑓ℎ/3,600
where
sp = permitted saturation flow rate for unsignalized movement (veh/h/ln),
λ0 = throughput of the conflicting movement (veh/h/ln),
The throughput of the conflicting movement λ0 is determined as a function of
the flow profile of the respective conflicting movement. The effective green g of
the conflicting movement is divided into a queue service time gs and a green
extension time ge, each with a specific flow profile:
• If the conflicting movement occurs during the queue service time gs, λ0 is
equal to the saturation flow rate s of the conflicting movement.
• If the conflicting movement occurs during the green extension time ge, λ0
is equal to the arrival flow rate during the green qg (Equation 19-32) of the
conflicting movement.

Case B: Two-Way STOP-Controlled Intersections

The two-way STOP-controlled (TWSC) intersection analysis is based on the
calculation of each movement’s potential capacity, based on such factors as
priority order, conflicting flow, and critical gap. Estimating the on-ramp
throughput from this intersection type is a relatively straightforward task,
involving few adjustments.
The procedure first identifies the movements that discharge to the on-ramp
and their respective ranks (priority orders). The evaluation of freeway–arterial
interactions assumes that the arterial will always be the major street at TWSC
interchanges.
Exhibit 38-B2 illustrates a typical TWSC intersection at a freeway
interchange, where movements discharging into the on-ramp are numbered
according to their ranks, using the default movement numbering of Chapter 20
(Exhibit 20-1).
Exhibit 38-B2
Schematic of Movements
Turning to an On-Ramp from
a TWSC Intersection
Similar to signalized intersections, three movements turn into the ramp.

Their respective flows are discussed below.
Rank 1 Movement (Right Turn from the Major Street)

This movement is considered unimpeded, experiencing zero delay. The only
physical constraint able to limit the throughput of this movement is its saturation
flow rate, if demand is very high. Therefore, the maximum throughput λRT for
this movement is given by:
𝜆𝑅𝑇 = min(𝑣𝑅𝑇 , 𝑠𝑅𝑇 ) Equation 38-B4
where
λRT = departure rate from major street right turn into the on-ramp (veh/h),
vRT = demand flow rate for the major street right turn (veh/h), and
sRT = saturation flow rate for a right-turn movement (veh/h).

Rank 2 Movement (Left Turn from the Major Street)

This movement’s maximum throughput is limited by its potential capacity
cp,j, as defined in Equation 20-36. Therefore, the movement’s maximum
throughput is given by:
Equation 38-B5 𝜆𝐿𝑇 = min(𝑣𝐿𝑇 , 𝑐𝑝,𝑗 )
where
λLT = departure rate from the major-street left-turn into the on-ramp (veh/h),
vLT = demand flow rate for the major-street left turn (veh/h), and
cp,j = potential capacity for the major-street left turn (veh/h).
Rank 3 Movement (Through Movement from the Minor Street)

Similar to rank 2 movements, this movement’s maximum throughput is
limited by its potential capacity cm,k, as defined in Equation 20-47. Therefore, the
movement’s maximum throughput is given by:
Equation 38-B6 𝜆𝑡ℎ = min(𝑣𝑡ℎ , 𝑐𝑚,𝑘 )
where
λth = departure rate from the minor-street through into the on-ramp (veh/h),
vth = demand flow rate for the minor-street through (veh/h), and
cm,k = potential capacity for the minor-street through (veh/h).
Total On-Ramp Demand Flow Rate

The total on-ramp demand flow rate vR is estimated as follows:
Equation 38-B7 𝑣𝑅 = 𝜆𝑅𝑇 + 𝜆𝐿𝑇 + 𝜆𝑡ℎ
Case C: All-Way STOP-Controlled Intersections

The all-way STOP-controlled intersection methodology uses departure
headways hd for each approach, making the calculation of the on-ramp flow
straightforward. The on-ramp demand flow rate can be obtained directly from
the departure headways of the three movements combined:
3,600 3,600 3,600
Equation 38-B8 𝑣𝑅 = + +
ℎ𝑑,𝑅𝑇 ℎ𝑑,𝐿𝑇 ℎ𝑑,𝑡ℎ
where
vR = on-ramp flow rate (veh/h),
hd,RT = departure headway for the major-street right turn (s),
hd,LT = departure headway for the major-street left turn (s), and
hd,th = departure headway for the minor-street through (s).
Exhibit 38-B3 illustrates the movements discharging into an on-ramp from an
AWSC intersection.

Exhibit 38-B3
an AWSC Intersection
Case D: Roundabouts
The roundabouts methodology is based on calculating the potential capacity
of each approach, based on three main variables: the critical headway, the
follow-up headway, and the circulating flow (Equation 22-21 through Equation
22-23). Critical and follow-up headway values can be obtained from Chapter 33,
Roundabouts: Supplemental. The methodology considers each approach
independently. To analyze roundabouts within a network, it is first necessary to
estimate the on-ramp throughput from a roundabout.
The procedure first identifies the movements that discharge to the on-ramp
and their respective ranks (priority orders). Exhibit 38-B4 illustrates a typical
roundabout, where movements discharging into the on-ramp are numbered
according to their ranks. In contrast to other types of intersections, the approach
furthest from the on-ramp has priority as it enters the circulating stream without
any significant conflicting traffic (other than occasional U-turns).
Exhibit 38-B4
a Roundabout

Rank 1 Movement (Left Turn from the Third Upstream Approach from the
On-Ramp)
This movement has priority over the other movements because it enters the
circulating stream first. In addition, because the on-ramp does not have an
approach into the roundabout, this movement is most often unopposed by the
circulating stream (except for occasional U-turns in the intersection). Therefore,
the maximum throughput 𝜆1−4 (veh/h) for this left-turn movement is given by:
Equation 38-B9 𝜆1−4 = min(𝑣1−4 , 𝑐1 )
where
λ1–4 = departure rate from the third upstream approach into the on-ramp
(veh/h),
v1–4 = demand flow rate for the third upstream approach into the on-ramp
(veh/h), and
c1 = potential capacity for the third upstream (rank 1) approach (veh/h).
Rank 2 Movement (Through from the Second Upstream Approach, Most Likely an
Off-Ramp):
This movement’s maximum throughput is limited by the upstream approach
departure rate and its own potential lane capacity c2, as defined in Equations 22-
21 through 22-23. Therefore, its maximum throughput is given by:
where
λ2–4 = departure rate from the second upstream approach into the on-ramp
(veh/h),
v2–4 = demand flow rate for the second upstream approach into the on-ramp
(veh/h), and
c2 = potential capacity for the second upstream (rank 2) approach (veh/h).
Rank 3 Movement (Right-Turn for the First Upstream Approach):

Similar to rank 2 movements, this movement’s maximum throughput is
limited by the upstream approach and its own potential capacity, as defined in
Equation 22-21 through Equation 22-23. Therefore, its maximum throughput is
given by:
where
λ3–4 = departure rate from the first upstream approach into the on-ramp
(veh/h),
v3–4 = demand flow rate for the first upstream approach into the on-ramp,
and
c3 = potential capacity for the first upstream (rank 3) approach (veh/h).

Total On-Ramp Demand Flow Rate

Finally, the total on-ramp demand flow rate is estimated as follows:
𝑣𝑅 = 𝜆1−4 + 𝜆2−4 + 𝜆3−4 Equation 38-B12

The total on-ramp demand flow rate can be determined using a similar
method for roundabouts with more than three upstream approaches.
CAPACITY ESTIMATION
The on-ramp’s capacity is estimated in order to predict the occurrence of
queue spillback. The maximum output flow rate ONRO(i, t, p) that can enter the
merge point from on-ramp i during time step t in analysis period p can be
constrained by (a) the ramp metering rate, if ramp metering is active, or (b)
oversaturated conditions in the downstream merge segment.
Case 1: Ramp Metering is Active

In this case, the metering rate in veh/h is a required user input and is stored RM(i,t,p) is defined in Chapter
25 as the maximum allowable
in the parameter RM defined in Chapter 25. The maximum output flow rate rate of an on-ramp meter at
ONRO that can enter the merge point is adjusted when ramp metering is a the on-ramp node i during time
interval p, measured in veh/h.
constraining factor to the on-ramp discharge. RM(i,t,p) is also one of the
inputs used in calculating the
maximum on-ramp output
Case 2: No Ramp Metering, Oversaturated Merge Segment (Equation 25-18).
In this case, the ramp merge capacity is computed by aggregating the ONRO
parameter to a 15-min analysis period and then converting the result into an
hourly flow rate.
Case 3: No Ramp Metering, Undersaturated Merge Segment

This case does not require any adjustments to the Chapter 10 freeway
facilities methodology.
EVALUATION OF ON-RAMP QUEUE SPILLBACK IMPACTS

This section describes the methodological modifications required to address
the occurrence of queue spillback from an on-ramp. The occurrence of queue
spillback affects each type of intersection differently. The methods outlined here
address signalized intersections, TWSC intersections, AWSC intersections, and
roundabouts.
Exhibit 38-B5 presents the core methodology for evaluating the performance
of signalized intersections, modified to address the effects of on-ramp queue
spillback. The new and modified steps to the methodology are described below.

Exhibit 38-B5
Methodology With
Adjustments to Address On-

The volume of vehicles that enters a freeway on-ramp is a function of the
demands and capacities of the individual intersection movements that discharge
into the ramp. A typical signalized intersection within a diamond interchange is
shown in Exhibit 38-B6, with three movements discharging into the on-ramp,
southbound left turn (SBL), eastbound through (EBT), and northbound right turn
(NBR).

Exhibit 38-B6
Typical Signalized Intersection
Ramp Terminal in a Diamond
Interchange
The total throughput from the intersection into the on-ramp λONR in veh/h is
the sum of the throughput from each of the contributing movements:
𝜆𝑂𝑁𝑅 = 𝜆𝑆𝐵𝐿 + 𝜆𝐸𝐵𝑇 + 𝜆𝑁𝐵𝑅 Equation 38-B13
The throughput for each movement i is the minimum value of its demand
and capacity:
𝜆𝑖 = min(𝑣𝑖 , 𝑐𝑖 ) Equation 38-B14
where
vi = demand flow rate for intersection movement i (veh/h), and
ci = capacity for intersection movement i (veh/h), from Equation 19-16.
Unsignalized movements, which are common for right-turn movements to
an on-ramp, are unrestricted. The capacity of these movements can be estimated
as the saturation flow rate (Equation 19-8) multiplied by the adjustment factor for
right turns fRT (Equation 19-13).
If all movements at the intersection are undersaturated, (i.e., vi ≤ ci for every
movement i), then Equation 38-B13 is simplified and the total on-ramp demand
throughput λONR is as follows:
𝜆𝑂𝑁𝑅 = ∑ 𝑣𝑖 Equation 38-B15

𝑖
Step 7B: Obtain Merging Capacity Using the Freeway Facilities Methodology
This step computes the merging capacity into the freeway cmerge. Three
potential bottlenecks can limit the on-ramp discharge into the freeway:
• The on-ramp capacity (Exhibit 14-12 or Exhibit 38-A2);
• The merge segment capacity, when the freeway facility is oversaturated; or
• The ramp metering rate, when ramp metering is active.
The procedure to obtain cmerge is presented in Exhibit 38-B7. The freeway
facility must be analyzed using the Chapter 10 methodology to evaluate whether

the merging capacity is constrained by oversaturated conditions on the mainline.

If the freeway facility is undersaturated (LOS A-E), the merging capacity takes
the minimum value between the on-ramp capacity and the ramp metering rate, if
ramp metering is active.
Exhibit 38-B7
Estimation of Freeway On-
Ramp Merging Capacity
If the freeway facility is oversaturated (LOS F), the Chapter 25 oversaturated

segment procedure provides the maximum on-ramp output parameter ONRO,
computed at a time-step level (15 s). The merging capacity cmerge is then computed
by aggregating ONRO to an hourly flow rate:
𝑆
𝑇
Equation 38-B16 𝑐merge = ∑ 𝑂𝑁𝑅𝑂(𝑖, 𝑡, 𝑝)
𝑆
𝑡=1
where
cmerge = capacity of the freeway merge section (pc/h),
T = number of time steps in 1 h (integer),
S = number of time steps in an analysis period (integer), and
ONRO(i, t, p) = maximum output flow rate that can enter the merge point
from on-ramp 𝑖 during time step t in analysis period p.

Step 7C: Plot Queue Accumulation Polygon for the On-Ramp

In this step, a Queue Accumulation Polygon (QAP) is built for the on-ramp,
considering the throughput from all contributing movements during the cycle.
Exhibit 38-B8 shows a sample intersection that will be used in describing this step.
Exhibit 38-B8
Sample Intersection for
Calculation of a QAP for the
On-Ramp
The methodology requires that the first analysis period be undersaturated.

Based on this requirement, the QAP starts with no vehicles inside the on-ramp.
The on-ramp QAP for this example is provided in Exhibit 38-B9.
Exhibit 38-B9
On-Ramp Queue
Accumulation Polygon During
Queue Spillback

The cycle starts with the SBL green discharging into the on-ramp at a
throughput rate λSBL, while the on-ramp discharges to the freeway merge at a
rate cmerge. Therefore, the number of vehicles within the on-ramp grows at a rate
equal to (λSBL − cmerge). When the number of vehicles along the on-ramp reaches
the maximum ramp storage length LONR, vehicles from the intersection can only
be discharged to the on-ramp at the same the rate they are discharged from the
on-ramp into the freeway. The number of vehicles within the on-ramp is then
maintained and it is equal to LONR until the end of the green for the SBL
movement. At the end of the SBL green, the vertical difference between the
projected number of vehicles (dashed line) and the actual number of vehicles
inside the on-ramp represent the number of unserved vehicles for the SBL
approach. This additional queue can be considered in a multiperiod analysis for
the signalized intersection or interchange, using the methods provided in
Chapter 23, Ramp Terminals and Alternative Intersections.
The slope of the red line connecting the number of vehicles at the end and at
the start of green represent the reduced capacity of the SBL movement due to
queue spillback. For the remainder of the cycle, the NBR movement discharges at
a constant rate into the on-ramp, as this is an unsignalized movement. Given that
the discharge capacity cmerge is greater than the on-ramp demand λNBR, the vehicles
along the on-ramp are discharged to the freeway until the on-ramp is cleared.
Therefore, the NBR movement’s capacity is not affected by queue spillback.
This procedure can be applied for both pretimed and actuated control types,
because the core methodology can address both controller types. If the signal is
actuated, the average phase durations are applied, as obtained in Step 6.
Step 7D: Calculate Adjusted Capacities for the Affected Movements

Based on the on-ramp QAP developed in the previous step, the adjusted
capacity cSP must be calculated for every movement affected by the queue
spillback. For the example of Exhibit 38-B9, the adjusted capacity for the SBL
movement cSBL,SP can be obtained from the QAP as the slope of the red line (cSBL,SP
− cmerge) as follows:
𝑁(𝑔1 ) − 𝑁(0)
Equation 38-B17 𝑐𝑆𝐵𝐿,𝑠𝑝 − 𝑐merge −
𝑔1
where
N(g1) = number of queued vehicles along the on-ramp at t = g1 (end of green
for phase 1),
N(0) = number of queued vehicles along the on-ramp at t = 0 (start of the
cycle),
g1 = effective green time for phase 1, and
other variables as previously defined.
The adjusted capacity of the SBL movement cSBL,SP is then computed as:
𝑁(𝑔1 ) − 𝑁(0)
Equation 38-B18 𝑐𝑆𝐵𝐿,𝑠𝑝 = 𝑐merge +
𝑔1


If the queue develops and fully discharges during every cycle, then
subsequent cycles will have the same discharge pattern. However, if there are
residual queues on the on-ramp at the end of the cycle, the QAP must be plotted
again for the following cycle with an initial queue equal to the number of queued
vehicles at the end of the present cycle. This process is repeated for a number of
cycles N = 900 / C sufficient to analyze the entire 15-min period. The adjusted
capacity for each movement is estimated as the average of the discharge rates
during each cycle.
Step 8: Determine Delay

The calculations for obtaining delay at the intersection approaches do not
need to be modified. The only change needed is to replace the input value for the
demand-to-capacity ratio X (Equation 19-17) with the adjusted value Xsp,
estimated using the adjusted capacity due to spillback:
𝑣
𝑋𝑠𝑝 = Equation 38-B19
𝑐𝑠𝑝
Two-Way STOP-Controlled Intersections

The operation of TWSC intersections is based on the priorities of movements
arriving at the intersection. Minor-street movements have lower priority and
must stop before entering the intersection. Left-turning drivers from the major
street must yield to oncoming major-street through or right turning traffic, but
are not required to stop in the absence of conflicting traffic or pedestrians.
The methodology for evaluating TWSC intersection operation is based on
gap-acceptance theory. Drivers from lower-priority movements must select a
suitable gap to proceed through the intersection. During oversaturated
conditions and when queue spillback occurs, drivers show cooperative behavior,
with higher-priority vehicles often yielding to those with lower priority (B-1), as
illustrated in Exhibit 38-B10. In such cases, the gap-acceptance model is no longer
valid, and a different approach must be used to evaluate the intersection
performance.
Exhibit 38-B10
Illustration of Cooperative
Behavior in Unsignalized
Intersections with Queue
Spillback

When queue spillback occurs at a TWSC intersection, the maximum

throughput to the on-ramp (exit capacity) is constrained by the discharge capacity
of the freeway merge. It is assumed that during oversaturated conditions, the
intersection movements that discharge to the on-ramp share the exit capacity
proportionately to their demands.
Exhibit 38-B11 presents the core methodology for evaluating the
performance of TWSC intersections, with modifications to address the effects of
on-ramp queue spillback. New and modified methodological steps are described
in the following paragraphs.
Exhibit 38-B11
TWSC intersections Core
Methodology with
Adjustments to Address On-

The throughput to the on-ramp is calculated using the approach described in
Step 7A of the queue spillback analysis for signalized intersections (Exhibit 38-
B5). The total throughput from the intersection into the on-ramp λONR is the sum
of the throughput from each of the contributing movements. For each movement

i discharging into the on-ramp, the throughput is the minimum value of its
demand and the movement capacity:
𝜆𝑖 = min(𝑣𝑖 , 𝑐𝑚,𝑖 ) Equation 38-B20
where
vi = demand flow rate for movement i, and
cm,i = movement capacity for movement i (Equations 20-36, 20-37, and 20-40).
Step 9B: Obtain Merging Capacity Using the Freeway Facilities Methodology
This step computes the merging capacity into the freeway cmerge. The
procedure described in Step 7B of the queue spillback analysis for signalized
intersections (Exhibit 38-B5) is applied.
Step 9C: Determine Fraction of Analysis Period with Queue Spillback

While signalized intersections operate in a cyclical pattern, STOP-controlled
intersections have relatively uniform patterns of demand and capacity within a
given analysis period. Therefore, the 15-min aggregated demand and capacity
values are assumed to be constant, and the growth and discharge of queues are
assumed to be linear.
The QAP is used to illustrate the development of queues along the on-ramp,
as illustrated in Exhibit 38-B12. For a given time interval of T minutes (typically
T = 15 min), the intersection discharges a throughput λONR to the ramp (Step 5B),
while the merge has capacity cmerge. If λONR > cmerge, queues will develop along the
on-ramp until the number of vehicles reaches the maximum ramp storage LONR,
when queue spillback begins. When that occurs, the maximum rate of vehicles
that can enter the on-ramp is limited by the merging capacity cmerge for the rest of
the time period.
Exhibit 38-B12
On-Ramp Queue
Accumulation Polygon: TWSC
Intersection

From the relationship shown in Exhibit 38-B12, the spillback time Tsp is
defined as the amount of time within an analysis period when spillback occurs:
𝐿𝑂𝑁𝑅 − 𝑁(0)
Equation 38-B21 𝑇𝑠𝑝 = 𝑇 −
𝜆𝑂𝑁𝑅 − 𝑐merge
where
Tsp = amount of time with active spillback (min),
T = analysis period duration (min),
LONR = available queue storage at the on-ramp (veh),
N(0) = number of queued vehicles along the on-ramp at t = 0 (start of the
cycle),
λONR = discharge from the intersection into the on-ramp (veh/h), and
cmerge = merging capacity of the on-ramp (veh/h).
Estimating the spillback time Tsp is a key element of the methodology,
because the aggregated calculations of capacity for each movement depend on
the amount of time that the intersection operates under queue spillback.

In this step, the capacities of the movements affected by spillback are
obtained and then aggregated to an analysis period level. When on-ramp queue
spillback occurs at an intersection, movements discharging to the on-ramp
switch to a cooperative approach instead of the priority-based regular operation.
When there is queue spillback, the maximum throughput to the on-ramp is
equal to the merging capacity cmerge. This capacity is then used by all movements
traveling into the on-ramp. The capacity of each affected movement i during
spillback csp,i is obtained proportionally to its demand flow rate:
𝑐merge × 𝑣𝑖
Equation 38-B22 𝑐𝑠𝑝,𝑖 =
∑𝑁
𝑖
𝑂𝑁𝑅
𝑣𝑖
where
csp,i = capacity during spillback for movement i (veh/h),
vi = demand flow rate for movement i (veh/h),
cmerge = merging capacity of the on-ramp (veh/h), and
NONR = number of movements at the intersection discharging to the on-ramp.
Finally, the adjusted capacity of each affected movement cEQ,,i is obtained as a
function of the amount of time within the analysis period when spillback was
present. The adjusted capacity considers the proportion of time that blockage
occurs during queue spillback. It consists of the aggregation, at an analysis
period level, of the movement capacities cm,i observed during undersaturated
conditions and the spillback capacities csp,i observed during oversaturated
conditions:

𝑐𝑠𝑝,𝑖 × 𝑇𝑠𝑝 + 𝑐𝑚,𝑖 × (𝑇 − 𝑇𝑠𝑝 )

𝑐𝐸𝑄,𝑖 = Equation 38-B23
𝑇
where
cEQ,,i = adjusted capacity for movement i (veh/h);
csp,i = capacity during spillback for movement i (veh/h);
Tsp = duration of active spillback during the analysis period (min);
cm,i = capacity for movement i without spillback (veh/h), from Chapter 20; and
T = analysis period duration (min), usually 15 min.
When queue spillback lasts for the entire analysis period T (for example, in a
multi-period analysis), the spillback time Tsp is equal to T, and the capacity of
each movement i is obtained as the capacity during spillback. In this case,
Equation 38-B23 simplifies to:
𝑐𝐸𝑄,𝑖 = 𝑐𝑠𝑝,𝑖 Equation 38-B24

The average control delay considering the effects of queue spillback dsp
(s/veh) is obtained based on Equation 20-64, but replacing the movement
capacity cm,i by the adjusted capacity cEQ,,i:
3,600 𝜆𝑂𝑁𝑅
𝑐𝐸𝑄,𝑖 × 𝑐𝐸𝑄,𝑖
2
3,600 𝑣𝑖 √ 𝑣𝑖
𝑑𝑠𝑝 = + 900𝑇 −1+ ( − 1) + +5 Equation 38-B25
𝑐𝐸𝑄,𝑖 𝑐𝐸𝑄,𝑖 𝑐𝐸𝑄,𝑖 450𝑇
[ ]
All-Way STOP-Controlled Intersections

The methodology to evaluate queue spillback into AWSC intersections
follows the approach presented above for TWSC intersections. As shown in
Exhibit 38-B13, after the capacities of individual movements during
undersaturated conditions are computed (Step 12), the process described above
for TWSC intersections is performed by new steps 13A through 13D.
The only step that differs from the TWSC method is Step 13D, Compute
Spillback Departure Headway. The AWSC methodology calculates the delay for
each approach based on its departure headway instead of its capacity. The
estimated spillback capacity csp,i is converted to a spillback headway hsp (in
seconds) as follows:
3,600
ℎ𝑠𝑝 = Equation 38-B26
𝑐𝑠𝑝,𝑖

Exhibit 38-B13
AWSC Intersection Core
Methodology with
Adjustments to Address
Roundabouts
The methodology presented in Chapter 22, Roundabouts, is shown in Exhibit
38-B14. The steps added to the methodology to evaluate queue spillback effects
are shown in red. Each of the new steps is discussed in the subsections below.
This methodology is applicable only to single-lane roundabouts. Exhibit 22-9
provides the required input data and potential data sources for the core
roundabout methodology. Exhibit 38-B15 lists the additional input data required
for queue spillback analysis.

Exhibit 38-B14
Roundabouts Methodology
With Adjustments to Address
Required Data and Units Potential Data Source(s) Suggested Default Exhibit 38-B15
On-Ramp Data Required Data and Potential
Data Sources for Roundabout
On-ramp metering rate (veh/h) Design plans, field data Must be provided
Queue Spillback Evaluation
On-ramp storage length LONR (ft) Field data Must be provided
Roundabout Data
Departure saturation headway into
Field data 3 s/veh
the on-ramp hs (s/veh)
Step 13: Compute the Maximum Throughput Into the On-Ramp for Each O-D
Movement
The maximum throughput into the on-ramp for each movement is calculated
using the roundabout priority order, starting with the most upstream approach
from the on-ramp exit leg and proceeding counterclockwise. The Rank 1
approach (Exhibit 38-B16) is the one whose flow has the highest priority, given
that it enters the circulating stream upstream of all other approaches. The next-
highest priority movement is the Rank 2 approach, and the lowest-priority
movement is the Rank 3 approach.

Exhibit 38-B16
Example Priority Order for a
Roundabout Upstream of an
On-Ramp
Next, the methodology calculates the capacity of the roundabout’s exit lane
into the on-ramp. Research (B-2, B-3) suggests that the capacity of an exit lane,
accounting for pedestrian and bicycle traffic in a typical urban area, is in the
range of 1,200 to 1,300 veh/h. Starting from the Rank 1 approach, and proceeding
counterclockwise with the remaining approaches, each approach’s capacity is
used to determine the maximum throughput for every movement discharging to
the on-ramp.
Rank 1 Approach
The Rank 1 approach (the SB approach in the example in Exhibit 38-B16) has
priority over the other movements connecting to the on-ramp because it enters
the circulating stream first. In addition, because the on-ramp leg usually does not
have an approach into the roundabout, the Rank 1 movement is most often
unopposed by the circulating stream (except for occasional U-turns along the
arterial). Therefore, the maximum throughput λSB-ONR for this left-turn movement
is limited by its own lane capacity cSB and the maximum throughput to the on-
ramp:
3,600
Equation 38-B27 𝜆𝑆𝐵−𝑂𝑁𝑅 = min (𝑣𝑆𝐵−𝑂𝑁𝑅 , 𝑐𝑆𝐵 × 𝑝𝑆𝐵−𝑂𝑁𝑅 , )
ℎ𝑠
with
𝑣𝑆𝐵−𝑂𝑁𝑅
Equation 38-B28 𝑝𝑆𝐵−𝑂𝑁𝑅 =
𝑣𝑆𝐵𝑈 + 𝑣𝑆𝐵−𝑂𝑁𝑅 + 𝑣𝑆𝐵𝑇
where
λSB-ONR = departure rate from the SB approach into the on-ramp (veh/h);
vSB-ONR = demand flow rate for the SB approach into the on-ramp (veh/h);
cSB = lane capacity for the SB approach, from Equation 22-21 (veh/h);
pSB-ONR = percentage of on-ramp demand from the SB approach, from Equation

38-B28 (decimal); and
hs = departure saturation headway into the on-ramp (s/veh).

Rank 2 Approach
The maximum throughput for the Rank 2 movement (the EB approach in the
example in Exhibit 38-B16) is limited by its own lane capacity cEB, as defined in
Equations 22-21 through Equation 22-23, and the maximum throughput after
considering the departure rate of the upstream Leg 1. Therefore, the maximum
throughput λEB-ONR for this movement is given by:
3,600
𝜆𝐸𝐵−𝑂𝑁𝑅 = min (𝑣𝐸𝐵−𝑂𝑁𝑅 , 𝑐𝐸𝐵 × 𝑝𝐸𝐵−𝑂𝑁𝑅 , − 𝜆𝑆𝐵−𝑂𝑁𝑅 ) Equation 38-B29
ℎ𝑠
with
𝑣𝐸𝐵−𝑂𝑁𝑅
𝑝𝐸𝐵−𝑂𝑁𝑅 = Equation 38-B30
𝑣𝐸𝐵𝐿 + 𝑣𝐸𝐵−𝑂𝑁𝑅 + 𝑣𝐸𝐵𝑅
where
λEB-ONR = departure rate from the EB approach into the on-ramp (veh/h);
vEB-ONR = demand flow rate for the EB approach into the on-ramp (veh/h);
cEB = lane capacity for the EB approach, from Equation 22-21 (veh/h);
pEB-ONR = percentage of on-ramp demand from the EB approach, from Equation

Rank 3 Approach
Similar to Rank 2 movements, the maximum throughput for the Rank 3
movement (the NB approach in the example in Exhibit 38-B16) is limited by its
own lane capacity cNB, as defined in Equation 22-21 through Equation 22-23, and
the maximum throughput to the on-ramp after considering departure rates from
the upstream approaches. Therefore, the maximum throughput λNB-ONR for this
right-turn movement is given by:
3,600
𝜆𝑁𝐵−𝑂𝑁𝑅 = min (𝑣𝑁𝐵−𝑂𝑁𝑅 , 𝑐𝑁𝐵 × 𝑝𝑁𝐵−𝑂𝑁𝑅 , − 𝜆𝑆𝐵−𝑂𝑁𝑅 − 𝜆𝐸𝐵−𝑂𝑁𝑅 ) Equation 38-B31
ℎ𝑠
with
𝑣𝑁𝐵−𝑂𝑁𝑅
𝑝𝑁𝐵−𝑂𝑁𝑅 = Equation 38-B32
𝑣𝑁𝐵𝑈 + 𝑣𝑁𝐵𝑇 + 𝑣𝑁𝐵−𝑂𝑁𝑅
where
λNB-ONR = departure rate from the NB approach into the on-ramp (veh/h);
vNB-ONR = demand flow rate for the NB approach into the on-ramp (veh/h);
cNB = lane capacity for the NB approach, from Equation 22-21 (veh/h);
pNB-ONR = percentage of on-ramp demand from the NB approach, from Equation

The total on-ramp demand flow rate can be calculated in a similar manner if
the roundabout has more than three approaches.

Step 14: Calculate the Maximum Throughput into the On-Ramp

The maximum throughput from the roundabout to the on-ramp λONR in
veh/h is calculated as:
Equation 38-B33 𝜆𝑂𝑁𝑅 = 𝜆𝑆𝐵−𝑂𝑁𝑅 + 𝜆𝐸𝐵−𝑂𝑁𝑅 + 𝜆𝑁𝐵−𝑂𝑁𝑅
Step 15: Compute On-Ramp Merging Capacity and Compare to the Maximum
Throughput to the On-Ramp
The on-ramp merging capacity is calculated exactly the same as in Step 7B of
the queue spillback methodology for signalized intersections (Exhibit 38-B5). The
maximum number of vehicles that can merge into the on-ramp cmerge (from
Equation 25-18) is compared to the maximum throughput from the roundabout
to the on-ramp λONR. If cmerge > λONR, then spillback is not expected to occur, and
no adjustments are necessary. If cmerge ≤ λONR, queues will develop along the on-
ramp, and spillback may occur if the queue storage is insufficient. In this case,
the analyst proceeds to Step 16 to evaluate the on-ramp queue storage ratio to
evaluate whether spillback will occur.
Step 16: Determine the On-Ramp Storage Ratio and Queue Spillback Length
Given the throughput from the roundabout into the on-ramp λONR, the queue
length QONR (in veh) along the on-ramp during a 15-min analysis period is:
𝜆𝑂𝑁𝑅 − 𝑐merge
Equation 38-B34 𝑄𝑂𝑁𝑅 =
4
If a multi-period analysis is performed, the queue length for the current analysis
period p must be added to the queue length from the previous analysis period:
𝜆𝑂𝑁𝑅,𝑝 − 𝑐merge,𝑝
Equation 38-B35 𝑄𝑂𝑁𝑅,𝑝 = 𝑄𝑂𝑁𝑅,𝑝−1 +
4
The on-ramp storage ratio is calculated by dividing the available on-ramp
storage length by the average vehicle spacing:
𝐿ℎ × 𝑄𝑂𝑁𝑅
Equation 38-B36 𝑅𝑄 =
𝐿𝑂𝑁𝑅
where
RQ = on-ramp storage ratio (decimal);
Lh = average vehicle spacing (ft/veh), from Equation 31-155;
QONR = on-ramp queue length (veh); and
LONR = on-ramp length (ft).
If the on-ramp storage ratio RQ is greater than 1, queues will form along each
roundabout approach due to spillback. The value of RQ is specific to a given
analysis period. If congestion is expected, but RQ < 1 for a single analysis period,
multi-period analysis may have to be conducted.

Step 17: Compute the Queue Spillback Distribution by Approach

When spillback occurs, the total number of vehicles queued Qsp during a 15-
min analysis period is:
𝑄𝑠𝑝 = 𝑄𝑂𝑁𝑅 − 𝐿ℎ × 𝐿𝑂𝑁𝑅 Equation 38-B37
These queues are assumed to be distributed in proportional to the demand

flow rates from each approach to the on-ramp, and are added to the 95th
percentile queues estimated for undersaturated conditions. For the example
roundabout given in Exhibit 38-B16, these queues are:
𝜆𝑆𝐵−𝑂𝑁𝑅
𝑄𝑠𝑝,𝑆𝐵 = 𝑄𝑠𝑝 × + 𝑄95,𝑆𝐵 Equation 38-B38
𝜆𝑂𝑁𝑅
𝜆𝐸𝐵−𝑂𝑁𝑅
𝑄𝑠𝑝,𝐸𝐵 = 𝑄𝑠𝑝 × + 𝑄95,𝐸𝐵 Equation 38-B39
𝜆𝑂𝑁𝑅
𝜆𝑁𝐵−𝑂𝑁𝑅
𝑄𝑠𝑝,𝑁𝐵 = 𝑄𝑠𝑝 × + 𝑄95,𝑁𝐵 Equation 38-B40
𝜆𝑂𝑁𝑅
where
Qsp,i = queue on approach i due to on-ramp spillback (veh);
Qsp = total intersection queue due to on-ramp spillback (veh);
λi-ONR = maximum throughput for approach i into the on-ramp (veh);
λONR = maximum throughput into the on-ramp, considering all approaches

(veh); and
Q95,i = 95th percentile queue on approach i (veh), from Equation 22-20.
Step 18: Calculate the Average Delay by Approach and Aggregate to the Average
Control Delay
To estimate the average delay per approach, the delay due to the on-ramp
capacity limitation is estimated and added to the approach control delay
calculated by Equation 22-17. This equation assumes no residual queue at the
start of the analysis period. If queue spillback occurs, the average control delay is
significantly affected by the analysis period length. However, Chapter 22 does
not provide a multiperiod analysis method. Therefore, the delay results may not
be accurate when a queue exists at the start of the analysis period.
As an alternative, an iterative process that carries over queues from one time
period to the next (B-4) may be used. The additional delay (in sec/veh) due to the
on-ramp spillback is calculated as follows:
3,600 𝜆𝑂𝑁𝑅
𝑐merge × 𝑐merge
2
3,600 𝜆𝑂𝑁𝑅 𝜆
√ 𝑂𝑁𝑅
𝑑𝑠𝑝 = + 900𝑇 −1+ ( − 1) + Equation 38-B41
𝑐merge 𝑐merge 𝑐merge 450𝑇
[ ]
𝜆𝑂𝑁𝑅
+ 5 × min [ , 1]
𝑐merge

REFERENCES
Some of these references can B-1. Aakre, E., and A. Aakre. Modeling cooperation in unsignalized
Reference Library in Volume 4. intersections. Procedia Computer Science, Vol. 109, 2017, pp. 875–880.
B-2. Robinson, B., L. Rodegerdts, W. Scarbrough, W. Kittelson, R. Troutbeck,
W. Brilon, L. Bondzio, K. Courage, M. Kyte, J. Mason, A. Flannery, E.
Myers, and J. Bunker. Roundabouts: An Informational Guide. Federal
Highway Administration, Washington, D.C., 2000.
B-3. Rodegerdts, L. A., and G. E. Blackwelder. Analytical Analysis of
Pedestrian Effects on Roundabout Exit Capacity. In Transportation Research
Circular E-C083, Transportation Research Board, Washington, D.C., 2005.
B-4. Kimber, R. M. and E. M. Hollis. Traffic queues and delays at road junctions.
Transport and Road Research Laboratory, Wokingham, Berkshire, U.K.,
1979.

APPENDIX C: LANE-BY-LANE ANALYSIS FOR FREEWAY

FACILITIES
LANE-BY-LANE FLOW MODELS BY SEGMENT TYPE

The lane flow ratio (LFR) model for each lane is estimated as a function of
the logarithm of the segment volume-capacity ratio v/c (C-1). The LFR equation is
applied to each lane in the segment except for the leftmost lane, which is
estimated as the remaining flow, to ensure the sum of the flow shares from each
lane always equals 100%. The equations to estimate LFR are as follows:
Equation 38-C1
𝐿𝐹𝑅𝑖 = 𝑎 × ln(𝑣/𝑐) + 𝑏
𝑁−1
𝐿𝐹𝑅𝑁 = 1 − ∑ 𝐿𝐹𝑅𝑖 Equation 38-C2

𝑖=1
where
LFRi = share of the total flow on lane i, where i ranges from 1 to N−1 (decimal);
N = number of lanes in the segment;
a = multiplicative calibration parameter, from Equation 38-C3, Equation
38-C5, or Equation 38-C7;
v/c = volume-to-capacity ratio (0 < v/c ≤ 1), with volume and capacity given
in veh/h;
b = additive calibration parameter, from Equation 38-C4, Equation 38-C6,
or Equation 38-C8; and
LFRN = share of the total flow on the leftmost lane, lane N (decimal).
The model in Equation 38-C1 and Equation 38-C2 can be applied to basic,
merge, diverge, and weaving segments. For merge and diverge segments, the
share of flow is estimated in the area upstream of the ramp; therefore, the inputs
for demand and capacity must also be measured upstream of the ramp.
For weaving segments, the share of flow is estimated on the mainline
upstream the on-ramp. Flow distribution on auxiliary lanes is not addressed by
this methodology, as the flow in these lanes is not expected to be constant along
their length. The auxiliary lane flow will be similar to the on-ramp flow rate
upstream and to the off-ramp flow downstream. Therefore, the estimation of the
lane flow ratio in a weaving segment addresses the mainline lanes only, with
lane 1 defined as the rightmost mainline lane.
The following subsections show how the model’s calibration parameters are
determined for each freeway segment type.
Chapter 38/Network Analysis Appendix C: Lane-by-Lane Analysis for Freeway Facilities

Basic Segments
The calibration parameters a and b used to analyze basic segments are
Equation 38-C3 𝑎 = 𝑎0 + 𝐺 × 𝑎𝐺 + 𝐻𝑉% × 𝑎𝐻𝑉 + 𝑟𝑑 × 𝑎𝑟𝑑
Equation 38-C4 𝑏 = 𝑏0 + 𝐺 × 𝑏𝐺 + 𝐻𝑉% × 𝑏𝐻𝑉 + 𝑟𝑑 × 𝑏𝑟𝑑
where
a = multiplicative calibration parameter;
a0 = empirical constant, from Exhibit 38-C1;
G = grade (%);
ag = empirical coefficient due to the effect of grade, from Exhibit 38-C1;
HV% = heavy vehicle percentage (%);
aHV = empirical coefficient due to the effect of trucks, from Exhibit 38-C1;
rd = ramp density = total number of ramps 0.5 mi upstream and 0.5 mi
downstream of the segment;
ard = empirical coefficient due to the effect of ramp density, from Exhibit
38-C1;
b = additive calibration parameter;
b0 = empirical constant, from Exhibit 38-C1;
bg = empirical coefficient due to the effect of grade, from Exhibit 38-C1;
bHV = empirical coefficient due to the effect of trucks, from Exhibit 38-C1; and
brd = empirical coefficient due to the effect of ramp density, from Exhibit
38-C1.
Exhibit 38-C1 provides the coefficients used in the LFD model for basic,
merge, and diverge segments. The coefficients are specific for each combination
of lane number, segment type, and total number of lanes in the segment.
Appendix C: Lane-by-Lane Analysis for Freeway Facilities Chapter 38/Network Analysis

Lane Para- Basic Segments Diverge Segments Merge Segments Exhibit 38-C1
# meter 2 lanes 3 lanes 4 lanes 2 lanes 3 lanes 4 lanes 2 lanes 3 lanes 4 lanes Lane Flow Distribution Model
a0 0.18 0.027 0.068 0.0097 −0.075 0.31 0.015 0.0029 −0.077 Coefficients for Basic, Merge,
b0 and Diverge Segments
0.52 0.27 0.22 0.44 0.27 0.25 0.59 0.28 0.24
ag 0.024 0.021 −0.011 0.0097 0.0077 −0.034 0.015 −0.0029 −0.0030
aHV −0.048 −0.0036 −0.0021 −0.0093 0.00080 −0.057 −0.0093 −0.0029 0.011
ard −0.095 −0.0083 −0.059 −0.0097 0.014 −0.028 −0.0047 −0.0029 0.014
L1
bg 0.0030 0.0097 −0.034 −0.0098 −0.0081 −0.00016 0.020 0.031 0.040
bHV 0.008 −0.0029 0.0024 0.0078 0.0014 −0.019 −0.014 −0.0018 −0.027
brd 0.0013 0.032 −0.035 0.00057 0.031 0.0052 −0.040 −0.042 −0.041
avr −0.21 −0.067 −0.0087 −0.035 −0.10 0.026
bvr −0.13 0.013 −0.021 −0.070 −0.030 0.0091
a0 −0.063 −0.025 0.0096 0.29 −0.0082 −0.080
b0 0.31 0.29 0.34 0.25 0.38 0.24
ag −0.0060 0.0015 −0.0096 −0.035 −0.0082 0.00048
aHV 0.0011 0.00027 −0.00054 −0.052 −0.00082 0.013
ard 0.0037 −0.0085 −0.0096 −0.030 −0.0026 0.018
L2
bg −0.017 −0.024 −0.0019 0.0019 0.0079 −0.019
bHV 0.0024 −0.00036 0.00089 −0.0041 −0.00048 −0.0067
brd 0.01 −0.041 0.0052 0.0044 −0.0060 0.0010
avr −0.048 −0.0065 −0.12 −0.033
bvr −0.073 −0.0091 −0.039 −0.013
a0 −0.045 0.27 0.029
b0 0.28 0.25 0.25
ag −0.0017 −0.036 −0.0017
aHV 0.0021 −0.044 −0.0058
ard 0.0081 −0.034 −0.0068
L3
bg 0.011 0.0034 0.00060
bHV −0.0011 0.0092 0.014
brd 0.015 0.0016 0.018
avr 0.021 −0.079
bvr −0.0064 −0.041
Note: Empty cells indicate the factor is not used in the model for the given combination of lane number, segment
type, and number of lanes in the segment (i.e., substitute a value of 0 for the coefficient).
Merge and Diverge Segments

The calibration parameters a and b used to analyze merge and diverge
segments are computed by Equation 38-C5 and Equation 38-C6. Compared to the
basic segment model, two additional coefficients, avr and bvr, are included to
address ramp demand.
𝑣𝑅
𝑎 = 𝑎0 + 𝐺 × 𝑎𝐺 + 𝐻𝑉% × 𝑎𝐻𝑉 + 𝑛 × 𝑎𝑛 + × 𝑎𝑣𝑟 Equation 38-C5
1,000
𝑣𝑅
𝑏 = 𝑏0 + 𝐺 × 𝑏𝐺 + 𝐻𝑉% × 𝑏𝐻𝑉 + 𝑛 × 𝑏𝑛 + × 𝑏𝑣𝑟 Equation 38-C6
1,000
where
avr = empirical coefficient due to effect of ramp flow, from Exhibit 38-C1;
bvr = empirical coefficient due to effect of ramp flow, from Exhibit 38-C1; and

Weaving Segments
The calibration parameters a and b used to analyze weaving segments are
computed by Equation 38-C7 and Equation 38-C8. The coefficients in these
equations include factors addressing the effects of weaving-specific properties.
Equation 38-C7
𝑣𝑅,𝑚 𝑣𝑅,𝑑
𝑎 = 𝑎0 + 𝐺 × 𝑎𝐺 + 𝐻𝑉% × 𝑎𝐻𝑉 + 𝐼𝐷 × 𝑎𝐼 + × 𝑎𝑣𝑚 + × 𝑎𝑣𝑑
1,000 1,000
𝐿𝑠
+ × 𝑎𝐿𝑆 + 𝑉𝑅 × 𝑎𝑉𝑅
1,000
Equation 38-C8 𝑏 = 𝑏0 + 𝐺 × 𝑏𝐺 + 𝐻𝑉% × 𝑏𝐻𝑉 + 𝐼𝐷 × 𝑏𝐼 + × 𝑏𝑣𝑚 + × 𝑏𝑣𝑑
1,000 1,000
𝐿𝑠
+ × 𝑏𝐿𝑆 + 𝑉𝑅 × 𝑏𝑉𝑅
1,000
where
Chapter 13, Freeway Weaving
Segments, defines interchange ID = interchange density (interchanges/mi);
density as the number of
interchanges within 3 mi aI = empirical coefficient due to effect of interchange density, from Exhibit
upstream and downstream of 38-C2;
the center of the subject
weaving segment, divided vR.m = on-ramp flow (veh/h);
by 6.
avm = empirical coefficient for on-ramp flow, from Exhibit 38-C2;
vR.d = off-ramp flow (veh/h);
avd = empirical coefficient for off-ramp flow, from Exhibit 38-C2;
Ls = weaving segment length (ft);
aLS = empirical coefficient for weaving segment length, from Exhibit 38-C2;
VR = volume ratio = weaving volume / total volume (decimal);
aVR = empirical coefficient for volume ratio, from Exhibit 38-C2;
bI = empirical coefficient due to effect of interchange density, from Exhibit
38-C2;
bvm = empirical coefficient for on-ramp flow, from Exhibit 38-C2;
bvd = empirical coefficient for off-ramp flow, from Exhibit 38-C2;
bLS = empirical coefficient for weaving segment length, from Exhibit 38-C2;
bVR = empirical coefficient for volume ratio, from Exhibit 38-C2; and
Exhibit 38-C2 provides the coefficients used in the LFD model for weaving
segments. The coefficients are specific for each combination of lane number,
segment type, and total number of lanes in the segment.

2-Lane Exhibit 38-C2

Segments 3-Lane Segments 4-Lane Segments Lane Flow Distribution Model
Parameter L1 L1 L2 L1 L2 L3 Coefficients for Weaving
a0 0.99 0.64 0.48 −0.13 0.0048 0.12 Segments
b0 0.40 0.40 0.33 0.24 0.26 0.27
ag −0.21 −0.28 0.11 0.13 −0.0048 −0.12
aHV −0.12 −0.055 −0.033 −0.012 −0.0048 0.019
aI 0.13 0.0037 −0.035 −0.0025 −0.0048 −0.12
avm 0.022 0.075 −0.090 0.072 −0.031 −0.011
avd −0.19 −0.036 0.017 −0.13 0.030 0.051
aLS −0.20 0.098 −0.031 0.056 0.0020 −0.041
aVR 0.0080 0.024 0.089 −0.11 −0.0045 0.12
bg 0.069 −0.40 0.039 −0.030 0.045 0.041
bHV 0.0032 −0.051 0.0045 −0.0043 −0.011 −0.0043
bI −0.016 0.40 −0.020 −0.0067 −0.0050 −0.0026
bvm −0.048 −0.14 0.0047 0.065 −0.0089 −0.038
bvd 0.040 0.039 −0.047 0.063 −0.015 −0.037
bLS −0.011 0.15 0.0050 −0.030 0.011 0.020
bVR 0.078 0.40 0.018 −0.14 0.040 0.15
Note: The number of lanes reflects the freeway upstream of the weave. Lanes connecting the on-ramp and off-
ramp are not included.
LANE FLOW RATIO DISTRIBUTION AS A FUNCTION OF THE DEMAND-

TO-CAPACITY RATIO
As discussed in the previous section, the LFR is obtained as a function of a
series of operational factors. Of these, the most influencing factor is the demand-
to-capacity ratio, as research (C-2) shows that LFR distributions follow typical
patterns depending on the number of lanes.
Exhibit 38-C3 demonstrates that, for 2-lane segments, flow distribution
follows a “scissors” pattern, with the flow highly concentrated in lane 1 during
free-flow conditions. As the segment’s demand flow rate increases, flow
gradually migrates to lane 2. During oversaturated conditions, flow is more
concentrated in lane 2.
Exhibit 38-C3
LFR Distribution for a Sample
2-Lane Basic Freeway
Segment
Note: I-694 in Minneapolis, MN.
Exhibit 38-C4 illustrates the LFR distribution for a 3-lane segment. At low
demand, most of the flow is concentrated in the center lane (Lane 2), followed by
Lane 1 and Lane 3. As demand increases, the LFR increases in Lane 3 and
decreases in Lanes 1 and 2.

Exhibit 38-C4
Segment
Note: I-4 in Tampa, FL.
Exhibit 38-C5 shows the LFR distribution for a 4-lane segment. Under free-
flow conditions, Lanes 2 and 3 carry the majority of flow. Lane 4 is typically
underused during undersaturated conditions, but at higher demands it carries
the majority of the flow.
Exhibit 38-C5
Segment
Note: I-275 in Tampa, FL.
The flow distribution patterns shown in the exhibits above for basic
segments are also observed in merge, diverge, and weaving segments.
Additional factors such as ramp volume, grade, and truck percentage influence
the boundary values and slopes of the curves, but do not change the typical LFR
distribution as a function of the v/c ratio.
CHECKING FOR NEGATIVE FLOWS AND LANE CAPACITIES

After lane flow ratios are obtained, a two-step check must be performed to
ensure the estimated flow distribution is reasonable. The first check identifies
any estimated negative flows. This issue is more likely to occur in the leftmost
lane, as the flows on this lane are obtained by the difference between the total
segment flow and the sum of estimated flows in the other lanes. Therefore, if
flows on the remaining lanes are overestimated, the estimated flow in the
leftmost lane may be negative. Exhibit 38-C6 illustrates the procedure for this
check.

Exhibit 38-C6
Check for Negative Lane
Flows
The variables in Exhibit 38-C6 are defined as follows:

i = index for the subject lane,
vi = flow rate on lane i (veh/h), and
N = number of lanes in the segment.
The second check compares the estimated flow by lane with the respective
lane capacities to ensure no lane operates with a demand-to-capacity ratio
greater than 1. This procedure is illustrated in Exhibit 38-C7, where ci is the
capacity of lane i (veh/h) and other variables are as previously defined. If any
lane is estimated to operate above its capacity, the flow in this given lane is
constrained by the capacity value and the exceeding demand is moved to the
adjacent lane.

Exhibit 38-C7
Check for Lane Capacity
SPEED–FLOW CURVES BY LANE AND SEGMENT TYPE

This section presents the models used to obtain speed–flow curves for each
lane in a freeway segment as a function of two key inputs: free-flow speed (FFS)
and lane capacity. The first part discusses the estimation of lane FFS, while the
second part presents models for obtaining lane capacities. The last part provides
the speed–flow models as a function of lane FFS and lane capacities.
Lane-by-Lane FFS
Field observations have shown that speeds differ among lanes, with speeds
typically lower in shoulder lanes and higher in median lanes. The model for
estimating individual lane FFS applies a multiplicative factor xFFS to the segment
FFS as follows:
Equation 38-C9 𝐹𝐹𝑆𝑖 = 𝐹𝐹𝑆𝑎𝑑𝑗 × 𝑥𝐹𝐹𝑆
where
FFSi = free-flow speed for lane i (mi/h);
FFSadj = adjusted free-flow speed for the segment (mi/h), from Equation 12-5;
and
xFFS = FFS multiplier, from Exhibit 38-C8.
Exhibit 38-C8 presents the recommended multipliers, which are provided as
a function of the segment type and the number of lanes in the segment. As shown,
when the number of lanes increases, the range of FFS multipliers increases as
well (i.e., lower speeds exist in the shoulder lanes and higher speeds exist in the
median lanes). For 2-lane segments, merge and diverge segments have a higher
FFS differential between the two lanes, compared to basic segments. For 3-lane
segments, basic segments have the greatest FFS range, while merge segments

have more uniform lane FFS. Finally, for 4-lane segments, merge segments have
the greatest FFS range, while basic and merge segments have similar ranges.
FFS Multiplier xFFS Exhibit 38-C8

Segment Type Number of Lanes Lane 1 Lane 2 Lane 3 Lane 4 Multipliers to Estimate Lane
2 0.965 1.032 — — FFS from Segment FFS
Basic 3 0.934 1.010 1.087 —
4 0.924 0.989 1.028 1.079
2 0.964 1.044 — —
Merge 3 0.955 1.015 1.045 —
4 0.935 0.991 1.036 1.091
2 0.961 1.035 — —
Diverge 3 0.943 1.024 1.068 —
4 0.933 0.975 1.018 1.074
2 0.969 1.018 — —
Weaving 3 0.968 1.023 1.062 —
4 0.910 0.988 1.053 1.110
Lane-by-Lane Capacity
Basic, Merge, and Diverge Segments
Similar to the case of free-flow speeds, capacities also differ among lanes, with
capacities typically lower in shoulder lanes and higher in median lanes. Center
lanes typically have values similar to the segment average. The model for
estimating individual lane FFS in basic, merge, and diverge segments applies a
multiplicative factor xc to the segment capacity as follows:
𝑐𝑖 = 𝑐𝑎𝑑𝑗 × 𝑁 × 𝑥𝑐 Equation 38-C10
where
ci = capacity of lane i (pc/h);
cadj = adjusted capacity for the segment (pc/h/ln), from Equation 12-8;
N = number of lanes in the segment; and
xc = capacity multiplier, from Exhibit 38-C9.
Exhibit 38-C9 presents the percent of the total segment capacity distributed
to each lane in the segment, defining a capacity multiplier xc for each
combination of segment type and number of lanes.
Capacity Multiplier xc Exhibit 38-C9

Segment Type Number of Lanes Lane 1 Lane 2 Lane 3 Lane 4 Multipliers to Estimate Lane
2 0.44 0.56 — — Capacity from Segment
Capacity for Basic, Merge, and
Basic 3 0.25 0.35 0.40 —
Diverge Segments
4 0.19 0.25 0.28 0.28
2 0.42 0.58 — —
Merge 3 0.23 0.36 0.41 —
4 0.21 0.24 0.25 0.30
2 0.42 0.58 — —
Diverge 3 0.26 0.34 0.40 —
4 0.21 0.24 0.27 0.28
Segment capacities measured from field data may not equal the estimated
capacities from the Chapter 12 methodology for basic freeway segments. Field
measurements of capacity have been found to be lower than HCM estimates (C-3).

Such differences can result in overestimating a segment’s performance. Therefore,

it is recommended that a calibration capacity adjustment factor (CAF) be applied
using Equation 12-8 to adjust the HCM base capacity to local conditions.
Weaving Segments
Capacity distributions are observed to be significantly more complex in
weaving segments compared to other types of freeway segments, and the
breakdown method does not provide reliable results. Capacity is assumed to be
uniform for all lanes within a weaving segment, and is obtained from Equation
13-5, adapted here as Equation 38-C11, based on a maximum density of 43 pc/h/ln:
Equation 38-C11 𝑐𝐼𝑊𝐿 = 𝑐𝐼𝐹𝐿 − [438.2 (1 + 𝑉𝑅)1.6] + (0.0765 𝐿𝑆 ) + (119.8𝑁𝑊𝐿 )
where
cIWL = per-lane capacity of the weaving segment under equivalent ideal
conditions (pc/h/ln),
cIFL = per-lane capacity of a basic freeway segment with the same FFS as the
weaving segment under equivalent ideal conditions (pc/h/ln),
VR = volume ratio,
LS = weaving segment length (ft), and
NWL = number of lanes from which weaving maneuvers may be made with
either one or no lane changes.
Lane-by-Lane Speed
With flow, capacity, and FFS determined by lane, the speed–flow model for
freeway segments given by Equation 12-1 and Exhibit 12-6 is then adapted to
estimate the speeds in individual lanes. Speed in each lane i is determined as:
𝑆𝑖 = 𝐹𝐹𝑆𝑖 if 𝑣𝑖 ≤ 𝐵𝑃𝑖
Equation 38-C12
𝑐
(𝐹𝐹𝑆𝑖 − 𝑖 ) (𝑣𝑖 − 𝐵𝑃𝑖 )2
𝑆𝑖 = 𝐹𝐹𝑆𝑖 − 45 if 𝑣𝑖 > 𝐵𝑃𝑖
(𝑐𝑖 − 𝐵𝑃𝑖 )2
with
Equation 38-C13 𝑣𝑖 = 𝑣 × 𝐿𝐹𝑅𝑖
where
Si = speed in lane i (mi/h);
FFSi = free-flow speed for lane i (mi/h);
ci = capacity of lane i (pc/h/ln);
vi = demand flow rate for lane i (pc/h/ln);
BPi = breakpoint value for lane i (pc/h/ln), from Equation 38-C14;
v = demand flow rate for the segment (pc/h); and
LFRi = share of the total flow on lane i.

The breakpoint value is also determined for each lane:

𝐵𝑃𝑖 = [1000 + 40 × (75 − 𝐹𝐹𝑆𝑖 )] × 𝐶𝐴𝐹 2 Equation 38-C14
where
BPi = breakpoint value for lane i (pc/h/ln),
FFSi = free-flow speed for lane i (mi/h), and
CAF = capacity adjustment factor, from Exhibit 12-6.
For auxiliary lanes in weaving segments, individual lane speeds cannot be
addressed by this methodology because conditions vary widely along the
auxiliary lane’s length, as discussed previously. For the O-D analysis described
in Section 3 of this chapter, auxiliary lane speeds are only relevant when the
subject weaving segment is the entry or exit point of a freeway facility for a
particular O-D pair. In this case, the expected speed for the segment described by
Equation 38-9 can be replaced by the average speed of weaving vehicles
described in Chapter 13:
𝑆𝑊 = 15 + ( ) Equation 38-C15
1+𝑊
with
𝐿𝐶ALL 0.789
𝑊 = 0.226 + ( ) Equation 38-C16
𝐿𝑆
where
SW = average speed of weaving vehicles within the weaving segment (mi/h),
FFS = free-flow speed of the weaving segment (mi/h),
SAF = speed adjustment factor,
W = weaving intensity factor (unitless),
LCALL = speed adjustment factor (unitless), and
LS = weaving intensity factor (unitless).
APPLICATION EXAMPLES
Example 1: Diverge Segment from Example Problem 1

This section presents an application of the LFR model for a freeway segment
extracted from Example Problem 1. A 3-lane diverge segment (segment 16 of the
freeway facility) was selected for lane-by-lane analysis, with the following input
data:
• Grade G = 1%
• Heavy vehicle percentage HV% = 2%
• PHF = 0.989
• Ramp density rd = 1 adjacent ramp
• Mainline hourly demand volume V = 4,848 veh/h
• Capacity adjustment factor CAF = 1.0

• Off-ramp demand vR = 960 veh/h
• Measured segment capacity c = 2,400 pc/h/ln (7,200 pc/h).

The mainline hourly demand volume (in veh/h) is first converted to a
demand flow rate under equivalent base conditions (in pc/h), by applying the
heavy vehicle adjustment factor fHV of 0.980 determined from Equation 12-10:
𝑉 4,848
𝑣= = = 5,003 pc/h
𝑃𝐻𝐹 × 𝑓𝐻𝑉 0.989 × 0.980
The flow ratio for lane 1 (right lane) is obtained from Equation 38-C1:
𝐿𝐹𝑅1 = 𝑎1 × ln(𝑣/𝑐) + 𝑏1
This is a diverge segment; therefore, the calibration parameters a and b for
lane 1 are obtained from Equation 38-C5, Equation 38-C6, and Exhibit 38-C1 as
follows:
𝑣𝑅
𝑎1 = 𝑎0 + 𝐺 × 𝑎𝐺 + 𝐻𝑉% × 𝑎𝐻𝑉 + 𝑟𝑑 × 𝑎𝑟𝑑 + × 𝑎𝑣𝑟
1,000
960
𝑎1 = −0.075 + 1 × 0.0077 + 2 × 0.0008 + 1 × 0.014 + × (−0.067)
1,000
𝑎1 = −0.116
𝑣𝑅
𝑏1 = 𝑏0 + 𝐺 × 𝑏𝐺 + 𝐻𝑉% × 𝑏𝐻𝑉 + 𝑟𝑑 × 𝑏𝑟𝑑 + × 𝑏𝑣𝑟
1,000
960
𝑏1 = 0.27 + 1 × (−0.00810) + 2 × 0.00140 + 1 × 0.031 + × 0.013
1000
𝑏1 = 0.308
The lane flow ratio for lane 1 is then:
5,003
𝐿𝐹𝑅1 = −0.116 × ln ( ) + 0.308 = 0.350
7,200
The same procedure is applied to obtain the lane flow ratio for lane 2, using
the respective coefficients from Exhibit 38-C1:
𝑣𝑅
𝑎2 = 𝑎0 + 𝐺 × 𝑎𝐺 + 𝐻𝑉% × 𝑎𝐻𝑉 + 𝑟𝑑 × 𝑎𝑟𝑑 + × 𝑎𝑣𝑟
1,000
𝑎2 = 0.0096 + 1 × (−0.00960) + 2 × (−0.00054) + 1 × (−0.0096)
960
+ × (−0.048)
1,000
𝑎2 = −0.0568
𝑣𝑅
𝑏2 = 𝑏0 + 𝐺 × 𝑏𝐺 + 𝐻𝑉% × 𝑏𝐻𝑉 + 𝑟𝑑 × 𝑏𝑟𝑑 + × 𝑏𝑣𝑟
1,000
960
𝑏2 = 0.34 + 1 × (−0.0019) + 2 × (0.00089) + 1 × 0.0052 + × (−0.073)
1,000
𝑏2 = 0.275
5,003
𝐿𝐹𝑅2 = −0.0568 × ln ( ) + 0.275 = 0.296
7,200
Finally, the LFR for the leftmost lane (lane 3) is obtained from Equation 38-C2:
𝐿𝐹𝑅3 = 1 − 𝐿𝐹𝑅2 − 𝐿𝐹𝑅1 = 1 − 0.350 − 0.296 = 0.354

Lane flows can be obtained by multiplying the segment demand by

respective LFR values for each lane, using Equation 38-C13:
𝑣1 = 𝑣 × 𝐿𝐹𝑅1 = 5,003 × 0.350 = 1,751 pc/h/ln
𝑣2 = 𝑣 × 𝐿𝐹𝑅2 = 5,003 × 0.296 = 1,481 pc/h/ln
𝑣3 = 𝑣 × 𝐿𝐹𝑅3 = 5,003 × 0.354 = 1,771 pc/h/ln
Lane-by-lane FFS is obtained by multiplying the segment FFS (75.4 mi/h) by
the appropriate multipliers from Exhibit 38-C8:
𝐹𝐹𝑆1 = 𝐹𝐹𝑆 × 0.943 = 75.4 × 0.943 = 71.10 mi/h
𝐹𝐹𝑆2 = 𝐹𝐹𝑆 × 1.024 = 75.4 × 1.024 = 77.21 mi/h
𝐹𝐹𝑆3 = 𝐹𝐹𝑆 × 1.064 = 75.4 × 1.068 = 80.53 mi/h
Lane-by-lane capacities are obtained by multiplying the segment capacity
(7,200 pc/h) by the appropriate multipliers from Exhibit 38-C9:
𝑐1 = 𝑐 × 0.26 = 7200 × 0.26 = 1,872 pc/h
𝑐2 = 𝑐 × 0.34 = 7200 × 0.34 = 2,448 pc/h
𝑐3 = 𝑐 × 0.40 = 7200 × 0.40 = 2,880 pc/h
Breakpoint values for each lane are obtained from Equation 38-C14:
𝐵𝑃1 = [1,000 + 40 × (75 − 𝐹𝐹𝑆1 ) × 𝐶𝐴𝐹2 ]
𝐵𝑃1 = [1,000 + 40 × (75 − 71.10) × 1] = 1,156 pc/h
𝐵𝑃2 = [1,000 + 40 × (75 − 𝐹𝐹𝑆2) × 𝐶𝐴𝐹2 ]
𝐵𝑃2 = [1,000 + 40 × (75 − 77.21) × 1] = 912 pc/h
𝐵𝑃3 = [1,000 + 40 × (75 − 𝐹𝐹𝑆3) × 𝐶𝐴𝐹2 ]
𝐵𝑃1 = [1,000 + 40 × (75 − 80.53) × 1] = 779 pc/h
The average speed of each lane is obtained by applying Equation 38-C12:
𝑐𝑖
(𝐹𝐹𝑆𝑖 − 45 ) (𝑣𝑖 − 𝐵𝑃𝑖 )2
𝑆𝑖 = 𝐹𝐹𝑆𝑖 −
(𝑐𝑖 − 𝐵𝑃𝑖 )2
1,872
(71.10 − 45 ) (1,751 − 1,156)2
𝑆1 = 71.10 − = 50.7 mi/h
(1,872 − 1,156)2
2,448
(77.21 − 45 ) (1,481 − 912)2
𝑆2 = 77.21 − = 74.1 mi/h
(2,448 − 912)2
2,880
(80.53 − 45 ) (1,771 − 779)2
𝑆3 = 80.53 − = 76.8 mi/h
(2,880 − 779)2
The lane-by-lane speed–flow curves are compared to the overall segment
curve in Exhibit 38-C10.

Exhibit 38-C10
Comparison of Speed–Flow
Curves by Lane and for the
Segment
Example 2: Weaving Segment

This section presents an application of the LFR model for the weaving
segment depicted in Exhibit 38-C11 to estimate the upstream lane flow shares.
Exhibit 38-C11
Example of LFR Calculation for
a Weaving Segment
The following input data are available:

• Number of lanes within the weave N = 5
• Number of upstream lanes NUP = 4
• Grade G = −0.5%
• Level terrain
• Heavy vehicle percentage HV% = 3.3%
• Interchange density ID = 0.67
• Weaving length LS = 3,920 ft
• Upstream mainline demand flow rate vUP = 4,512 veh/h
• On-ramp demand flow rate vR,m = 428 veh/h
• Freeway-to-freeway demand vFF = 3,312 veh/h
• Freeway-to-ramp demand vFR = 1,200 veh/h
• Ramp-to-freeway demand vRF = 404 veh/h
• Ramp-to-ramp demand vRR = 24 veh/h
• Off-ramp flow rate vR,d = 1,224 veh/h
• Number of weaving lanes NWL = 2
• Measured segment FFS = 70 mi/h

The heavy vehicle adjustment factor is calculated from Equation 12-10, using
a passenger car equivalency ET of 2:
1 1
𝑓𝐻𝑉 = = = 0.968
1 + 𝑃𝑇 (𝐸𝑇 − 1) 1 + 0.03(2 − 1)
The weaving and non-weaving demands are adjusted to flow rates under
ideal conditions. Because the demands are estimated based on 15-min intervals,
it is assumed the PHF is set to 1.
𝑉
𝑣=
24
𝑣𝑅𝑅 = = 24.8 pc/h
1 × 0.968
404
𝑣𝑅𝐹 = = 417.3 pc/h
1 × 0.968
1200
𝑣𝐹𝑅 = = 1,239.6 pc/h
1 × 0.968
3312
𝑣𝐹𝐹 = = 3,421.3 pc/h
1 × 0.968
The weaving and non-weaving flows are given by:
𝑣𝑊 = 𝑣𝐹𝑅 + 𝑣𝑅𝐹 = 1,239.6 + 417.3 = 1,656.9 pc/h
𝑣𝑁𝑊 = 𝑣𝑅𝑅 + 𝑣𝐹𝐹 = 24.8 + 3,421.3 = 3,446.1 pc/h
The volume ratio is computed as:
𝑣𝑊 1,656.9
𝑉𝑅 = = = 0.325
𝑣 1,656.9 + 3,446.1
A weaving segment’s capacity is the smaller of the density-based capacity
cIWL from Equation 13-5 and the weaving-demand-based capacity cIW from
Equation 13-7. The segment’s base capacity cIFL is 2,400 pc/h based on the
measured FFS of 70 mi/h. Because the LFR is calculated in the next step using
flows and capacities in veh/h/ln, the capacities calculated in this step are
converted from units of pc/h/ln by applying the heavy vehicle factor.
𝑐′𝐼𝑊𝐿 = 𝑐𝐼𝐹𝐿 − [438.2(1 + 𝑉𝑅)1.6 ] + (0.0765𝐿𝑠 ) + (119.8𝑁𝑊𝐿 )
𝑐′𝐼𝑊𝐿 = 2,400 − [438.2(1 + 0.325)1.6 ] + (0.0765 × 3,920) + (119.8 × 2)
𝑐′𝐼𝑊𝐿 = 2,252.3 pc/h/ln
𝑐𝐼𝑊𝐿 = 𝑐′𝐼𝑊𝐿 × 𝑓𝐻𝑉 = 2,252.3 × 0.968 = 2,180.4 veh/h/ln
2,400 2,400
𝑐′𝐼𝑊 = = = 7,385 pc/h
𝑉𝑅 0.325
𝑐′𝐼𝑊𝐿 × 𝑓𝐻𝑉 7,385 × 0.968
𝑐𝐼𝑊 = = = 1,787 veh/h/ln
𝑁UP 4
𝑐 = min(𝑐𝐼𝑊𝐿 , 𝑐𝐼𝑊 ) = min(2,180.4, 1,787) = 1,787 veh/h/ln

The capacity of the weaving segment upstream the weave is obtained by:
𝑐UP = 𝑐 × 𝑁UP = 1,787 × 4 = 7,148 veh/h
The flow ratio for lane 1 (right lane) is obtained from Equation 38-C1:
𝑣up
𝐿𝐹𝑅1 = 𝑎1 × ln ( ) + 𝑏1
𝑐UP
Because this a weaving segment, the calibration parameters 𝑎 and 𝑏 for lane 1
are obtained using Equation 38-C7, Equation 38-C8, and Exhibit 38-C2 as follows:
𝑎1 = 𝑎0 + 𝐺 × 𝑎𝐺 + 𝐻𝑉% × 𝑎𝐻𝑉 + 𝐼𝐷 × 𝑎𝐼 + × 𝑎𝑣𝑚 + × 𝑎𝑣𝑑
1,000 1,000
𝐿𝑠
1,000
𝑎1 = −0.13 + (−0.5) × 0.13 + 3.3 × (−0.012) + 0.67 × (−0.0025)

428 1,224 3,920
+ × 0.072 + × (−0.13) + × 0.056
1,000 1,000 1,000
+ 0.325 × (−0.11)
𝑎1 = −0.181
𝑏1 = 𝑏0 + 𝐺 × 𝑏𝐺 + 𝐻𝑉% × 𝑏𝐻𝑉 + 𝐼𝐷 × 𝑏𝐼 + × 𝑏𝑣𝑚 + × 𝑏𝑣𝑑
1,000 1,000
𝐿𝑠
1,000
𝑏1 = 0.24 + (−0.5) × (−0.03) + 3.3 × (−0.0043) + 0.67 × (−0.0067)

428 1,224 3,920
+ × 0.065 + × 0.063 + × (−0.03)
1,000 1,000 1,000
+ 0.325 × (−0.14)
𝑏1 = 0.178
The lane flow ratio for lane 1 is then:
4,512
𝐿𝐹𝑅1 = −0.181 × ln ( ) + 0.178
7,148
𝐿𝐹𝑅1 = 0.261
The same procedure is applied to estimate the LFR for lane 2, using the
respective coefficients from Exhibit 38-C2:
1,000 1,000
𝐿𝑠
1,000
𝑎2 = 0.0048 + (−0.5) × (−0.0048) + 3.3 × (−0.0048) + 0.67 × (−0.0048)

428 1224 3920
+ × (−0.031) + × 0.03 + × 0.002
1,000 1,000 1,000
+ 0.325 × (−0.0045)
𝑎2 = 0.01797

1,000 1,000
𝐿𝑠
1,000
𝑏2 = 0.26 + (−0.5) × 0.045 + 3.3 × (−0.011) + 0.67 × (−0.005)

428 1,224 3,920
+ × (−0.0089) + × (−0.015) + × 0.011
1,000 1,000 1,000
+ 0.325 × 0.04
𝑏2 = 0.2318
4,512
𝐿𝐹𝑅2 = 0.01797 × ln ( ) + 0.2318 = 0.224
7,148
The same procedure is applied to obtain the LFR for lane 3, using the
respective coefficients from Exhibit 38-C2:
1,000 1,000
𝐿𝑠
1,000
𝑎3 = 0.12 + (−0.5) × (−0.12) + 3.3 × 0.019 + 0.67 × (−0.12)

428 1,224 3,920
+ × (−0.011) + × 0.051 + × (−0.041)
1,000 1,000 1,000
+ 0.325 × 0.12
𝑎3 = 0.09830
1,000 1,000
𝐿𝑠
1,000
𝑏3 = 0.27 + (−0.5) × 0.041 + 3.3 × (−0.0043) + 0.67 × (−0.0026)

428 1,224 3,920
+ × (−0.038) + × (−0.037) + × 0.02
1,000 1,000 1,000
+ 0.325 × 0.15
𝑏3 = 0.2992
4,512
𝐿𝐹𝑅3 = 0.09830 × ln ( ) + 0.2992 = 0.254
7,148
Finally, the LFR for the leftmost lane (lane 4) is obtained from Equation 38-C2:
𝐿𝐹𝑅4 = 1 − 𝐿𝐹𝑅3 − 𝐿𝐹𝑅2 − 𝐿𝐹𝑅1 = 1 − 0.254 − 0.224 − 0.261 = 0.261
It is worth noting that the methodology predicts flow distribution and
speeds for lanes upstream of the on-ramp, which means the weaving auxiliary
lane is not covered by the scope of this example problem.

Example 3: Basic Segment

In this example, a 2-lane basic segment on northbound Highway 1 in Santa
Cruz, CA is modeled, and the lane-by-lane performance is compared to field
data. Field-measured parameters are as follows:
• Segment FFS = 69.1 mi/h
• Segment capacity = 3,993 veh/h = 1,996.5 veh/h/ln
• % heavy vehicles = 1.7%
• Grade = 3% (rolling)
The FFS multipliers for a 2-lane basic segment given in Exhibit 38-C8 are
applied to the segment FFS to obtain lane-by-lane FFS:
𝐹𝐹𝑆1 = 𝐹𝐹𝑆 × 0.965 = 69.1 × 0.965 = 66.68 mi/h
𝐹𝐹𝑆2 = 𝐹𝐹𝑆 × 1.032 = 69.1 × 1.032 = 71.31 mi/h
Next, individual lane capacities are obtained by applying the capacity
multipliers for a 2-lane basic segment (Exhibit 38-C9) to the segment capacity:
𝑐1 = 𝑐 × 0.44 = 3,993 × 0.44 = 1,757 veh/h
𝑐2 = 𝑐 × 0.56 = 3,993 × 0.56 = 2,236 veh/h
For comparison, the Chapter 12 method produces the following theoretical
capacity:
𝑐 = [2200 + 10 × (𝐹𝐹𝑆 – 50)] × 𝑓𝐻𝑉
𝑐 = [2200 + 10 × (69.1 − 50) )] × 0.967
𝑐 = 2,312 veh/h/ln
Therefore, a calibration CAF for this location can be obtained by dividing the
field-measured capacity by the theoretical capacity:
𝑐𝑎𝑑𝑗 1,996.5
𝐶𝐴𝐹 = = = 0.864
𝑐 2,312
Next, the breakpoint values for each lane are obtained using Equation 38-C14:
𝐵𝑃1 = [1,000 + 40 × (75 − 𝐹𝐹𝑆1)] × 𝐶𝐴𝐹 2
𝐵𝑃1 = [1,000 + 40 × (75 − 66.68)] × 0.8642
𝐵𝑃1 = 995 pc/h
𝐵𝑃2 = [1,000 + 40 × (75 − 𝐹𝐹𝑆2)] × 𝐶𝐴𝐹 2
𝐵𝑃2 = [1,000 + 40 × (75 − 71.31)] × 0.8642
𝐵𝑃2 = 857 pc/h
Lane flow ratios are obtained by applying Equation 38-C1 to the flow rate
entering the segment. Next, speeds on individual lanes are obtained using the
speed–flow relationship described in Equation 38-C12. For this location, a sample
of 14,690 observations (15 min each) was randomly selected. Predicted speeds
were compared to field data as shown in Exhibit 38-C12.

Exhibit 38-C12
Comparison of Predicted and
Field-Measured Lane-by-Lane
Speeds
(a) Lane 1 (Right Lane)
(b) Lane 2 (Left Lane)
As shown in the figures, the lane-by-lane speed–flow models can replicate

field conditions with good accuracy. Naturally, the oversaturated portion of the
speed–flow curve cannot be addressed by the model, because this is a limitation
of the core Chapter 12 method.
REFERENCES
C-1. University of Florida Transportation Institute; Cambridge Systematics, Some of these references can
Inc.; and A. Skabardonis. NCHRP Web-Only Document 290: Highway Reference Library in Volume 4.
Capacity Manual Methodologies for Corridors Involving Freeways and Surface
Streets. Transportation Research Board, Washington, D.C., 2020.
C-2. Sasahara, F., L. Elefteriadou, and S. Dong. Lane-by-Lane Analysis
Framework for Conducting Highway Capacity Analyses at Freeway
Segments. Transportation Research Record: Journal of the Transportation
Research Board, Vol. 2673, Issue 8, 2019, pp. 523–535.
C-3. Sasahara, F., L. Carvalho, T. Chowdhury, Z. Jerome, L. Elefteriadou, and A.
Skabardonis. Predicting Lane-by-Lane Flows and Speeds for Freeway
Segments. Transportation Research Record: Journal of the Transportation
Research Board, Vol. 2674, Issue 9, 2020, pp. 1052–1068.


HCM Vol 4

Uploaded by

Copyright:

Available Formats

HCM Vol 4

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

HCM Vol 4

Uploaded by

Copyright:

Available Formats

HIGHWAY CAPACITY

MANUAL 7th Edition

Copyright 2022 by the National Academy of Sciences. All rights reserved.

Printed in the United States of America

Hardcover International Standard Book Number-13: 978-0-309-08766-7

The three Academies work together as the National Academies of Sciences,

Learn more about the Transportation Research Board at www.TRB.org.

2. GLOSSARY OF VARIABLE DEFINITIONS .................................................... 25-3

3. UNDERSATURATED SEGMENT EVALUATION ........................................ 25-9

4. OVERSATURATED SEGMENT EVALUATION .......................................... 25-11

5. WORK ZONE ANALYSIS DETAILS ............................................................... 25-28

6. PLANNING-LEVEL METHODOLOGY FOR FREEWAY FACILITIES .... 25-34

Chapter 25/Freeway Facilities: Supplemental Contents

7. MIXED-FLOW MODEL FOR COMPOSITE GRADES ................................25-41

8. FREEWAY CALIBRATION METHODOLOGY .............................................25-53

9. FREEWAY SCENARIO GENERATION ..........................................................25-68

10. COMPUTATIONAL ENGINE OVERVIEW ...............................................25-84

11. EXAMPLE PROBLEMS ...................................................................................25-85

APPENDIX A: TRUCK PERFORMANCE CURVES .......................................25-137

Contents Chapter 25/Freeway Facilities: Supplemental

Exhibit 25-1 Node–Segment Representation of a Directional Freeway

Chapter 25/Freeway Facilities: Supplemental Contents

Exhibit 25-25 δ Values for TTs .................................................................................25-48

Contents Chapter 25/Freeway Facilities: Supplemental

Exhibit 25-59 Example Problem 2: Expanded LOS Matrix.................................. 25-96

Chapter 25/Freeway Facilities: Supplemental Contents

Exhibit 25-90 Example Problem 6: Demand Flow Rates (pc/h) on the

Contents Chapter 25/Freeway Facilities: Supplemental

Chapter 25/Freeway Facilities: Supplemental Contents

CHAPTER SCOPE VOLUME 4: APPLICATIONS

LIMITATIONS OF THE METHODOLOGIES

Chapter 25/Freeway Facilities: Supplemental Introduction

Introduction Chapter 25/Freeway Facilities: Supplemental

2. GLOSSARY OF VARIABLE DEFINITIONS

Chapter 25/Freeway Facilities: Supplemental Glossary of Variable Definitions

• KJ—facilitywide jam density (pc/mi/ln).

Glossary of Variable Definitions Chapter 25/Freeway Facilities: Supplemental

• NV(i, t, p)—number of vehicles present on segment i at the end of time

Chapter 25/Freeway Facilities: Supplemental Glossary of Variable Definitions

• MO3(i, t, p)—maximum Mainline Output 3: maximum allowable mainline

Glossary of Variable Definitions Chapter 25/Freeway Facilities: Supplemental

TRAVEL TIME RELIABILITY VARIABLES

• DSP—duration of study period SP (h).

Chapter 25/Freeway Facilities: Supplemental Glossary of Variable Definitions

• VMTSeed—vehicle miles of travel in the seed file.

Glossary of Variable Definitions Chapter 25/Freeway Facilities: Supplemental

3. UNDERSATURATED SEGMENT EVALUATION

FACILITY SPEED CONSTRAINT

FFS = segment free-flow speed (mi/h),

L = distance from midpoints of the upstream segment and the subject

Chapter 25/Freeway Facilities: Supplemental Undersaturated Segment Evaluation

DIRECTIONAL FACILITY MODULE

Average facility density in time interval p is calculated with Equation 25-3:

These performance measures can be compared for different alternatives to

Undersaturated Segment Evaluation Chapter 25/Freeway Facilities: Supplemental

4. OVERSATURATED SEGMENT EVALUATION

Oversaturated flow conditions occur when the demand on one or more

The oversaturated analysis focuses on the computation of segment average

Chapter 25/Freeway Facilities: Supplemental Oversaturated Segment Evaluation

Oversaturated Segment Evaluation Chapter 25/Freeway Facilities: Supplemental