WINPAS User Manual

Windows Pavement
Analysis Software
(WinPAS) Guide
Based on the 1993 AASHTO Guide for the
Design of Pavement Structures
WinPAS12 (SW03)
Windows Pavement Analysis

Software (WinPAS) Guide
Based on the 1993 AASHTO Guide for the
Design of Pavement Structures
This publication is intended SOLELY for use by PROFESSIONAL PERSONNEL who

are competent to evaluate the significance and limitations of the information provided
herein, and who will accept total responsibility for the application of this information. The
American Concrete Pavement Association DISCLAIMS any and all RESPONSIBILITY
and LIABILITY for the accuracy of and the application of the information contained in
this publication to the full extent permitted by law.
All rights reserved. No part of this book may be reproduced in any form without
permission in writing from the publisher, except by a reviewer who wishes to quote brief
passages in a review written for inclusion in a magazine or newspaper.
2012 American Concrete Pavement Association
ACPA is the premier national association representing

concrete pavement contractors, cement companies,
equipment and materials manufacturers and suppliers.
We are organized to address common needs, solve other
problems, and accomplish goals related to research,
promotion, and advancing best practices for design and
construction of concrete pavements.
ii | P a g e
Pavement and Overlay Design Based on the 1993 AASHTO

"Guide for the Design of Pavement Structures"
This publication is to help familiarize
engineers on the basics of concrete
pavement design. It gives the
background information that is essential
to effectively design concrete
pavements and overlays using the
"AASHTO Guide for the Design of
Pavement Structures - 1993"1 design
procedure via the ACPA's WinPAS
software. Still, ACPA encourages every
pavement design engineer to purchase
a copy of the complete 93 AASHTO
Design Guide for a complete reference.
This publication is broken down into four
chapters. The first two describe
concrete pavement thickness design
and overlay design according to Parts II
and III (Chapter 5) of the 93 AASHTO
Design Guide. Part II is entitled
"Pavement Design Procedures for New
Construction or Reconstruction," and
Chapter 5 of Part III is entitled
"Rehabilitation Methods with Overlays."1
The third chapter of this publication
describes life-cycle costing procedures
for a project level analysis. Life-cycle
costing is a procedure that economically
compares two competing design
alternatives considering all significant
costs over the economic life of each
alternative, expressed in equivalent
dollars. It includes initial cost,
rehabilitation costs, maintenance and
operation costs, user costs and residual
value.
AASHTO Guide for the Design of

Pavement Structures - 1993
The final chapter of this publication is
the users guide for the WinPAS
software. The software is capable of
conducting concrete and asphalt
pavement designs and analyses, traffic
conversions, life cycle cost analyses,
and overlay designs and analyses.
It is important to note that thickness
design is only one aspect of good
concrete pavement design. Another is
jointing. Proper jointing is essential to
ensure that a concrete pavement will
perform for its intended design life.
Unfortunately, it is often overlooked. For
more information on jointing of concrete
pavements, please refer to other ACPA
publications, including:
Design and Construction Joints

for Concrete Highways (TB010P)
Intersection Joint Layout
(IS006P)
P a g e | iii
Design and Construction Joints

for Concrete Streets (IS061P)
Concrete Pavement Field
Reference: Pre-Paving (EB237P)
Concrete Pavement Field
Reference: Paving (EB238P)
Concrete Intersections: A Guide
for Design and Construction
(TB019P)
iv | P a g e
Table of Contents
Chapter 1 New Concrete Pavement Design................................................................. 1

Introduction .................................................................................................................. 1
Concrete Pavement Basics ...................................................................................... 1
AASHO Road Test....................................................................................................... 3
AASHTO Rigid Pavement Design Equation ................................................................ 6
Thickness ................................................................................................................. 6
Serviceability ............................................................................................................ 7
Environmental Effects........................................................................................... 8
Traffic (ESALs) ......................................................................................................... 9
Rigid versus Flexible ESALs................................................................................. 9
Load Equivalency Factors .................................................................................. 10
Determining Load Equivalency Factors .............................................................. 11
Asphalt LEFs vs. Concrete LEFs ........................................................................ 12
Load Transfer ......................................................................................................... 13
Load Transfer Coefficient (J) .............................................................................. 14
Concrete Properties ............................................................................................... 15
Flexural Strength, S'C ......................................................................................... 15
Center Point Flexural Strength ........................................................................... 16
Compressive Strength ........................................................................................ 16
The Importance of Using Average Strength ....................................................... 17
Modulus of Elasticity ........................................................................................... 18
Subgrade Support .................................................................................................. 19
Modulus of Subgrade Reaction (k-value) ........................................................... 19
Loss of Support .................................................................................................. 20
Determining Subgrade Support for Design ......................................................... 20
AASHTO Procedure to Determine the k-Value ................................................... 21
Determine Resilient Modulus .......................................................................... 21
Page |v
Convert Resilient Modulus to k-Value ............................................................. 21

Adjust k-Value for Depth to Rigid Foundation ................................................. 23
Adjust k-Value for Loss of Support.................................................................. 23
Seasonal Adjustment to the k-Value ............................................................... 23
Problems with the AASHTO Procedure to Determine Subgrade Support .......... 24
Loss of Support ............................................................................................... 25
CBR and R-Value Relationships to Mr ............................................................ 25
Inconsistencies between Base and No Subbase Conditions .......................... 25
The Resulting k-Values are Unrealistic ........................................................... 26
Recommended Values for the Modulus of Subgrade Reaction .......................... 26
........................................................................................................................... 27
Recommended k-Values for Subbases .............................................................. 28
AASHTO and the Benefits of Subbases ............................................................. 28
Coefficient of Drainage (Cd) ................................................................................... 29
Reliability ................................................................................................................ 30
Reliability (R) ...................................................................................................... 31
Standard Deviation (so)....................................................................................... 31
How Reliability Works ......................................................................................... 32
How ZR Relates to R ........................................................................................... 32
The Iterative Process ............................................................................................. 34
Sensitivity Analysis ................................................................................................. 34
Summary ................................................................................................................... 37
Chapter 2 Concrete Overlay Design........................................................................... 39
Introduction ................................................................................................................ 39
The AASHTO Overlay Design Steps ......................................................................... 41
Step 1. Determine Existing Pavement Information ................................................. 41
Step 2. Predict Future ESALs ................................................................................ 41
Step 3. Perform Condition Survey .......................................................................... 42
Step 4. Perform Deflection Testing ........................................................................ 42
Step 5. Perform Coring/Material Testing ................................................................ 43
Step 6. Determine the Required Structural Capacity for Future Traffic (SCf) ......... 43
vi | P a g e
Step 7. Determine the Existing Structural Capacity (SCeff)..................................... 43

Problems with Remaining Life ............................................................................ 44
Step 8. Determine Required Structural Capacity of the Overlay (SCOL) ................. 45
Bonded Concrete Overlays on Concrete ............................................................ 45
Unbonded Concrete Overlays on Concrete ........................................................ 47
Unbonded Concrete Overlays on Asphalt or Composite .................................... 48
Bonded Concrete Overlays on Asphalt or Composite ........................................ 50
Other Considerations ................................................................................................. 50
Other Concrete Overlay Design Procedures/Software .............................................. 51
Chapter 3. Life-Cycle Cost Analysis (LCCA) ................................................................. 53
Introduction ................................................................................................................ 53
Comparable Sections ................................................................................................ 53
Performing an LCCA.................................................................................................. 54
Step 1 Select Analysis Period ................................................................................. 54
Step 2 Select Discount Rate ................................................................................... 55
Selecting an Interest Rate ...................................................................................... 55
Selecting an Inflation Rate ..................................................................................... 56
Calculating the Real Discount Rate........................................................................ 56
Step 3 Estimate Initial Agency Costs ...................................................................... 56
Step 4 Estimate User Costs .................................................................................... 57
Step 5 Estimate Future Agency Costs .................................................................... 58
Maintenance and Operation Costs ......................................................................... 58
Preservation and Rehabilitation Timing and Costs................................................. 58
Step 6 Estimate Residual Value ............................................................................. 59
Residual Value through Recycling (Salvage Value) ............................................... 59
Residual Value through Remaining Service Life .................................................... 59
Residual Value as a Support Layer ........................................................................ 60
Step 7 Compare Alternatives .................................................................................. 60
Cash Flow Diagrams .............................................................................................. 60
Present Worth Calculations .................................................................................... 61
Annual Worth Calculations ..................................................................................... 62
P a g e | vii
Analysis Methods ................................................................................................... 62

Accounting for Material Inflation ............................................................................. 63
Comparison of Results ........................................................................................... 64
More Information on LCCA ........................................................................................ 64
Chapter 4. WinPAS Users Guide ................................................................................. 65
Introduction ................................................................................................................ 65
Menu Options ............................................................................................................ 65
File Menu ............................................................................................................... 66
Units Menu ............................................................................................................. 66
Help Menu .............................................................................................................. 66
Main Menu ................................................................................................................. 66
Project Tab ................................................................................................................ 67
Estimate ESALs Tab.................................................................................................. 67
Total ESALs by Axle Data ...................................................................................... 68
Total ESALs by Vehicle Type ................................................................................. 69
Total ESALs by Truck Factor ................................................................................. 71
Design/Evaluation Tab............................................................................................... 72
Concrete Pavement Design/Analysis ..................................................................... 72
Asphalt Pavement Design/Analysis........................................................................ 76
Asphalt Layer Determination .............................................................................. 78
Both Concrete and Asphalt Design/Analysis (Side-by-Side) .................................. 79
Overlays Tab ............................................................................................................. 80
Existing Pavement Information............................................................................... 81
Bonded Concrete Overlays on Concrete................................................................ 82
Unbonded Concrete Overlays on Concrete ........................................................... 84
Unbonded Concrete Overlays on Asphalt (Conventional Whitetopping) ................ 85
Non-Destructive Testing (NDT) Backcalculation .................................................... 86
NDT Backcalculation for Concrete Pavements ................................................... 87
NDT Backcalculation for Asphalt Pavements ..................................................... 88
NDT Backcalculation for Composite Pavements ................................................ 89
Life-Cycle Costs Tab ................................................................................................. 89
viii | P a g e
Economic Factors .................................................................................................. 90

Cost Graphs ........................................................................................................... 90
Pavement Cost Information .................................................................................... 91
Life-Cycle Cost Analysis Results............................................................................ 93
Reports Tab ............................................................................................................... 94
Problems or Questions .............................................................................................. 94
References .................................................................................................................... 95
P a g e | ix
This Page Left Intentionally Blank
x|Page
Chapter 1 New Concrete Pavement Design
Introduction
Over the last 100 years, the science of
concrete pavement design has fallen
into two basic categories: mechanistic
and empirical.
Mechanistic pavement design is based
upon a fundamental understanding of
the materials (i.e., the concrete and
soils). It is a true attempt to describe
how the pavement responds to loads.
Unfortunately, until very recently,
mechanistic equations did not consider
a number of practical factors relating to
pavement performance and have only
given an estimate of what could be
expected in the field.
Empirical models are based on known
field pavement performance. Empirical
models started being used in the 1920's
when engineers began to examine the
adequacy of their pavement design
methods. The search for answers to
many of their questions led to the
development of controlled experiments
or "road tests" of actual in-place
pavements. The most complete road
test to date is the AASHO (American
Association of State Highway Officials)
Road Test.1
In recent years, mechanistic and
empirical design methods have been
combined in various design methods,
including ACPAs StreetPave software
and AASHTOs DARWinMETM.
Concrete Pavement Basics

There are three basic types of concrete
pavements built in the United States:
jointed plain concrete pavement (JPCP),
jointed reinforced concrete pavement
(JRCP), and continuously reinforced
concrete pavement (CRCP). The
primary design detail that distinguishes
each concrete pavement type from each
other is the jointing system used to
control natural crack development
(Figure 1 on next page).
Jointed plain concrete pavements
contain enough joints so that the natural
cracks occur at the joints and not
elsewhere in the slab. The spacing
between transverse joints for highways
is typically about 15 ft (4.5 m). JPCP
typically has deformed steel tie bars at
the longitudinal joints to hold the lanes
together, but they do not contain any
other mesh-steel reinforcement.
Depending on the slab thickness, JPCP
may contain smooth steel dowel bars at
transverse joints to improve load
transfer (load transfer is a slab's ability
to share part of its load with its
neighboring slab). For highways, dowels
should be used in pavements that are
greater than 8.0 in. (200 mm) thick
because of the large amounts of truck
traffic such pavements typically carry.
Page |1
dowels should be based on other

criteria, such as whether or not the road
will be in a residential or industrialized
area.
Jointed Plain Concrete Pavement (JPCP)
Jointed Reinforced Concrete Pavement (JRCP)
Continuously Reinforced Concrete Pavement

(CRCP)
Figure 1. Concrete Pavement Types

JPCP under 7.0 in. (150 mm) is usually
built without dowels and depend upon
"aggregate interlock" for load transfer.
Aggregate interlock is the mechanical
locking that forms between the fractured
surfaces along the crack below the joint
saw cut. Undoweled JPCP is generally
used for low-volume and secondary
roads. For pavements between 7.0 to
8.0 in. (150 and 200 mm), the use of
2|Page
JRCP contain steel mesh reinforcement

(sometimes called distributed steel).
With JRCP, designers purposely
increase the joint spacing, and use
reinforcing steel to hold the mid-panel
cracks that will develop together. The
spacing between transverse joints is
typically about 30 ft (9 m). In the past,
some agencies used spacing as great
as 100 ft (30 m), but this was found to
be excessive. For JRCP to perform, the
amount of distributed steel within the
pavement needs to be between 0.10%
and 0.25% of the cross-sectional area. If
there is not at least this amount of steel,
the steel can corrode or rupture and the
cracks can start to open, move, and
deteriorate.4 For this reason, ACPA
does not recommend building JRCP.
CRCP does not have transverse joints.
Rather, it is designed with high amounts
of steel reinforcement to hold the
transverse cracks that do develop tightly
together. The cracks usually develop at
intervals of 3-5 ft (1-1.5 m). Determining
the appropriate amount of steel to
control the crack spacing is part of the
design process for the pavement type.
This type of pavement was not
evaluated at the AASHTO Road Test.
Today, the majority of U.S. state
agencies build JPCP. CRCP is
common in some states for high traffic
applications. Very few states still
employ JRCP designs.
AASHO Road Test

The AASHO Road Test took place in
Ottawa, Illinois (approximately 80 miles
[130 km] southwest of Chicago)
between 1956 and 1960 (Figure 2). The
primary purpose of the Road Test was
to determine a relationship between axle
loading and pavement performance.
Other purposes were to determine a
relationship between the performance of
concrete and asphalt pavements and
the pavement design variables (base
courses, thickness, shoulders, etc.) and
to establish a more equitable taxation
basis for the vehicles that use the
roadways.2
3 through 6 were the main test loops

and carried the heaviest traffic. After the
Road Test was complete, these loops
were reconstructed into what is now
Interstate 80.
Figure 3 shows the layout for the loops
3 through 6. Each loop had a test
tangent of 6,800 ft (2,070 m). The south
tangents and west turnarounds were
concrete designs and the north tangents
and east turnarounds were asphalt
designs. The centerlines divided the
pavements into inside and outside
lanes. Each lane carried a different
vehicle type and so was a different test
section.
Figure 3. AASHO Loop Layout for Loops

3 through 6
Figure 2. The AASHO Road Test

(Ottawa, IL)
The Road Test itself consisted of six
loops. Each loop was constructed as a
parallel segment of a four-lane divided
highway with a turnaround at each end.
Loop 1 was the environmental loop and
was not trafficked. Loop 2 was the light
traffic loop. Both of these loops were
smaller than the main test loops. Loops
In the asphalt pavements, every 100 ft

(30 m) was a different design and
therefore a new structural section. For
the concrete pavements, the design
sections changed every 120 ft (36.5 m)
or 240 ft (73 m), depending on the type
of rigid pavement. Any design could be
located at any place in its test track to
provide randomization and certain
designs were duplicated in the same
test track to provide replication.
Page |3
In total, there were 368 concrete test

sections and 468 asphalt test sections.
The design variables for the concrete
and asphalt pavements are shown in
Table 1.
Figure 4 shows the typical test traffic on
the pavement test sections. All test
vehicles were trucks. The single axles
loads ranged from 2,000 to 30,000 Ibs
(900 to 13,600 kg) and tandem axles
loads ranged from 24,000 to 48,000 Ibs
(10,890 to 21,780 kg). It is important to
note that front axles were not
considered load axles except in loop 2.
Traffic ran on the test loops from
November 1958 to December 1960 (25
months). The test vehicles operated for
18 hours 40 minutes per day for 6 days
a week.2
Figure 4. AASHO Test Traffic and

Loading
Table 1: Design Variables at the AASHO Road Test2

Concrete Pavement Variables
Surface Thickness,
in. (mm)
Asphalt Pavement Variables
2.5 (63), 3.5 (89),

5 (127), 6.5 (165),
8 (203), 9.5 (241),
11 (279), 12.5 (318)
0 (0), 3 (76), 6 (152),
9 (229)
Surface Thickness,
in. (mm)
1 (25), 2 (51), 3 (76),

4 (102), 5 (127),
6 (152)
Base Thickness
0 (0), 3 (76), 6 (152),

9 (229), 19 (483)
Sandy-Gravel
Materials
Base Type
Wire
Reinforcement
Yes or No
Subbase
Thickness, in.
(mm)
Paved Shoulders
Yes or No
Paved Shoulder
Crushed Stone,
Gravel, AsphaltTreated, CementTreated
0 (0), 4 (102), 8
(203), 12 (305),
16 (406) All
Sandy-Gravel
Materials
Yes or No
Subbase
Thickness, in.
(mm)
Subbase Type
4|Page
The average speed on the test loops

was 35 mph (56 km/hr). In total, there
were 1,114,000 load applications during
the 25 months of testing with over 17
million miles (27 million km) driven.2
Figure 5 summarizes the results for
loops 3 through 6 at the Road Test.
These plots show the number of
sections remaining above a given
present serviceability index (PSI) plotted
against load applications. The PSI is a
rating from 0 (very poor) to 5 (very
good) that describes the condition of the
pavement. At the Road Test, pavements
were considered to have failed when the
PSI dropped below 1.5. Sections with a
PSI above 2.5 at the end of the test
were considered to have performed
"good.3
Association of Highway and

Transportation Officials] in the early
seventies.) It was this document which
underwent the largest distribution and
use by highway engineers. In 1981, the
concrete pavement portion of the guide
again received some minor revisions.
In 1986, the guide was extensively
revised into the "AASHTO Guide for the
Design of Pavement Structures." The
1986 version included many changes,
such as the way subgrade support is
characterized. It also introduced many
new concepts, such as reliability, lifecycle cost analysis (LCCA), and
pavement management.
From the tremendous amount of data

collected during the AASHO Road Test,
the engineers and statisticians working
on the project developed a series of
equations relating axle loads to
pavement performance. The equations
represent the predicted performance for
the conditions at the Road Test for
concrete and asphalt pavements.
After the Road Test, AASHO published
the prediction equations in the "AASHO
Interim Guide for the Design of Rigid
Pavement Structures" and "AASHO
Interim Guide for the Design of Flexible
Pavement Structures." In 1972,
AASHTO consolidated and updated
these documents into the "AASHTO
Interim Guide for the Design of
Pavement Structures." (AASHO's name
was changed to AASHTO [American
Figure 5. Present Serviceability Index

Trends for the AASHO Road Test
Page |5
The 1986 guide was also the first guide

to contain an overlay design procedure.
Unfortunately, the overlay procedure
was deficient, complicated, and
incomplete. This made it difficult to
understand and use.
po = Initial serviceability
In 1993, the overlay design procedure

was completely revised. This revised
procedure addressed the deficiencies in
the 1986 overlay design procedure, is
more comprehensive and adaptable to
local agency calibration, and is much
easier to use and understand.
J = Load transfer coefficient
AASHTO Rigid Pavement Design

Equation
The current AASHTO Rigid Design
Equation as published in the 1986 and
1993 guide is as follows:
()
= ZR s + 7.35 Log(D + 1) 0.06
PSI
Log
4.5 1.5 + (4.22 0.32 p )

+
1.624 10
1+
(D + 1) .
S C (D . 1.132)
Log
18.42
.
215.63 J D
(E /k) .
where:
ESAL = Allowable number of

equivalent 18-kip (80 KN) single
axles
ZR = Standard normal deviate
so = Overall standard deviation
D = Concrete thickness, in. (mm)
PSI = po - pt
6|Page
pt = Terminal serviceability
S'c = Concrete modulus of rupture,
psi (MPa)
Cd = Drainage coefficient
Ec = Concrete modulus of elasticity,

psi (MPa)
k = Modulus of subgrade reaction,
psi/in. (MPa/m)
Though the equation looks long and
complicated, when it is broken down it is
found that there are eight basic
concepts that affect the concrete
pavement design. They are:
1.
2.
3.
4.
5.
6.
7.
8.
Thickness
Serviceability
Traffic
Load transfer
Concrete properties
Subgrade strength
Drainage properties
Reliability
Understanding the importance of each

of these allows the engineer to properly
design concrete pavements. The
remainder of this section will explain
each of the above concepts.
Thickness
The pavement thickness (D) is
expressed in in. (mm). At the Road Test,
concrete pavement thickness ranged
from 2.5 to 12.5 in. (63 to 320 mm) [see
Table 1]. Therefore, the AASHTO rigid
pavement design equation is only valid
within this range. When the resulting
pavement thickness is outside of this

range, it is important to check the design
with another procedure (i.e.,
StreetPave).5
Occasionally the AASHTO design
procedure produces a pavement less
than 4 in. (100 mm) thick for light traffic
streets. Except for overlays, ACPA
recommends a minimum concrete
pavement thickness of 4 in. (100 mm)
for automobiles and 5 in. (125 mm) for
limited truck traffic. Further guidance on
minimum pavement thickness is
available in other ACPA
publications.5,6,7,8,9
Under some conditions, such as a
bridge overpass, the thickness is
constrained, thus it becomes a design
feature. In such cases, the designer can
solve for the allowable traffic, or the
concrete strength required to carry the
estimated traffic.
Serviceability
Serviceability, or the present
serviceability index (PSI), is a
pavement's "ability to serve the type of
traffic that uses the facility (e.g.,
automobiles, trucks, buses, etc)". It is a
scaled index from 0 to 5 that represents
different levels of deterioration (Figure
6). All pavements lie somewhere on this
scale.
Figure 6. The Present Serviceability

Index (PSI) Corresponds to These
Subjective Descriptions of Pavement
Performance
Pavements with PSI rating of 5 are in
perfect condition, while pavements with
a PSI rating of 0 are impassable. For all
practical purposes, there are no
pavements that have a rating of 5 or 0.
The AASHTO Pavement Design is
based on the predicted loss or drop in
serviceability (PSI) that will occur over
the lifetime of the pavement due to
traffic levels, axle loadings, and
environment (Figure 7).
The PSI is the difference between
initial and terminal serviceability (Po Pt). Initial serviceably (Po) is the
condition immediately after construction.
Terminal serviceability (Pt) corresponds
to the condition at which a pavement
requires some type of rehabilitation in
order to remain in service.
Page |7
The AASHO Road Test pavements

were taken out of service when the PSI
reached 1.5. Table 2 provides the
recommended terminal PSI values for
interstates and major highways, primary
and secondary roads, and secondary
routes and rural residential roads.
Figure 7. The 93 AASHTO Design is

Based on the Serviceability Loss over
the Lifetime of the Pavement
Concrete pavements were built to an
initial serviceability of 4.5 at the AASHO
Road Test. Flexible pavements were
built to the initial PSI of about 4.2. If no
other information on the initial
serviceability is available, the designer
should use 4.5 for concrete and 4.2 for
asphalt. With current construction
procedures, modern techniques/
materials, and improved smoothness
specifications, concrete pavements can
be built with an initial serviceability of
4.7 or 4.8.
The smoother a pavement is built, the
higher its initial serviceability. A higher
initial serviceability results in a larger
PSI. Thus, pavements built smoother
will last longer because they extend the
serviceability curve and allow the
pavement to carry more traffic over its
lifetime (see Figure 7).
The terminal serviceability is typically
based on the type of roadway and the
type of traffic it carries. Generally, highspeed highway traffic requires pavement
in better condition than low-volume
county or municipal streets.
8|Page
Table 2: Recommended Terminal

Serviceability (pt) Values for Various
Roadway Classifications
Terminal
Serviceability
2.50
2.25
2.00
Roadway
Classification
Interstate; Major
Highways or Arterials
Prime Secondary
Routes; Industrial and
Commercial Streets
Secondary Routes;
Residential Streets;
Parking Lots
Environmental Effects
According to the 1986 and 1993 guides,
the primary reason for allowing
adjustment to the initial serviceability is
so the designer can consider long-term
environmental effects like expansive or
frost susceptible soils. The Road Test
was an accelerated program that lasted
only two years. Consequently, the
design equation is somewhat limited in
its ability to consider long-term
environmental effects.
To make environmental adjustment, use
the following equation:
where:
PSITR = PSI PSIENV
PSITR = PSI loss due to traffic

PSI = Total PSI loss over the
design life (po pt)
PSIENV = PSI loss due to soil
displacement or other
environmental factors
Appendix G of the 1993 guide provides
further guidance for selecting the value
for PSIENV. Though it is not difficult to
determine the proper PSIENV, it is
complex and time consuming.
Determining the value requires an initial
estimate of the pavement thickness and
design life, some information on soil
permeability, knowledge of the roadbed
soil types, and information on drainage
conditions and freeze-thaw cycles. The
procedure requires several iterations to
recalculate traffic effects and evaluate
the changes these effects have on
PSIENV.
The impact of PSIENV on projected
performance is fairly low over much of
the U.S. Therefore, in most cases, the
value of PSIENV can be set to zero and
PSITR will equal PSI. This represents
the same conditions as at the AASHO
Road Test. Even if you may suspect that
setting PSIENV to zero does not
represent your design conditions, the
range of typical values that you might
expect for PSIENV is only from 0.0 to
0.7. In the worst case scenario, the
resultant increase in calculated
pavement thickness to carry a given
traffic volume will only be about seven
percent.
Traffic (ESALs)
ESALs are the number and weight of all
axle loads from the anticipated vehicles
expected during the pavement design
life expressed in 18,000 lbs or 18 kip (80
kN) equivalent single axle loads.
In actual practice, highway engineers
work with a variety of axle weights and
configurations in a mixed traffic stream.
At the AASHO Road Test, the engineers
theorized that they could compare the
damage to a particular pavement
section by different axle configurations
and loads to the damage caused by a
standard axle. With that idea, they
developed the concept of the Equivalent
Single Axle Load or ESAL.
Simply put, the design ESALs is all the
traffic, with different vehicle types, axle
types, and tire configurations converted
into an equivalent number of 18 kip (80
kN) single axle loads. At the Road Test,
the total number of ESALs ranged from
a few thousand to over 10 million
flexible and 20 million rigid ESALs for
the heaviest trafficked test loop.
Rigid versus Flexible ESALs
Though the concept of ESALs sounds
simple, it can be very confusing
because there is a difference between
rigid ESALs and flexible ESALs. Flexible
ESALs are generally about 1/3 less than
rigid ESALs, though the exact ratio
varies depending on traffic, pavement
thickness, and terminal serviceability.
Page |9
Table 3: Rigid and Flexible ESALs Generated by a Mixed Traffic Stream

Vehicle
Busses
Panel Trucks
Single Unit, 2 Axle Trucks
Semi-Tractor Trailer, 3 Axles
Automobile, Pick-up, Van
TOTAL
Number
Rigid ESALs
Flexible ESALs
5
10
20
10
15
15
425
500
13.55
10.89
6.38
20.06
39.43
57.33
1.88
149.52
8.73
11.11
6.11
13.41
29.88
36.87
2.25
108.36
Typical street design: concrete thickness = 7 in. (175 mm), flexible structural number = 3.5, pt = 2.5
This difference in ESALs can be

misleading, especially when attempting
to compare concrete (rigid) and asphalt
(flexible) designs. It often causes the
misconception that concrete pavements
are over designed when compared to
asphalt pavements. However, this is not
so. The different rigid and flexible
ESALs each describe the same traffic
stream.
An example is shown in Table 3. In this
table, the traffic stream is made up of a
mix of multiple unit trucks, single unit
trucks, panel trucks, buses, cars, etc.
There are 500 vehicles total on the
roadway for that day. When the traffic is
converted to ESALs, there are 149 rigid
ESALs and 108 flexible ESALs. Though
the values are different, they both
describe the same 500 vehicles.
Therefore, the ESAL counts are
equivalent.
To understand why rigid ESALs and
flexible ESALs are different, one needs
to understand how the traffic is
converted to ESALs.
10 | P a g e
Load Equivalency Factors

Traffic is converted to ESALs by
multiplying each load by a load
equivalency factor (LEF) *. LEFs are the
ratio of the damage of a specific axle
load on pavement serviceability to the
damage produced by an 18 kip (80 kN)
single axle load at the AASHO Road
Test.
Basically, LEFs measure damage
(serviceability loss). An 18 kip (80 kN)
single axle load (SAL) has a LEF of 1.0
because it does one unit of damage.
SALs less than 18 kip (80 kN) do less
than one unit of damage and SALs
greater than 18 kip (80 kN) do more
than one unit of damage.
Table 4 shows a typical set of LEFs for
a pavement. From it, some general
observations can be made (note that the
18 kip (80 kN) single axle LEF is 1.0).
*
Load equivalency factors vary depending on

pavement type (rigid or flexible), thickness, and
serviceability. For a listing of load equivalency
factors for different axle loads, configurations,
and pavement types, see Appendix D of the
AASHTO Guide.
Table 4: Typical Load Equivalency

Factors (Flexible Pavement with pt =
2.5)
Axle Type
Axle Load,
kip (kN)
LEF
Single
2 (8.9)
10 (44.5)
14 (62.3)
18 (80.9)
20 (89.0)
30 (133)
0.0003
0.118
0.399
1.00
1.49
7.90
2 (8.9)
10 (44.5)
18 (80.9)
30 (133)
34 (151)
40 (178)
50 (222)
0.0001
0.011
0.042
0.703
1.11
2.06
5.03
Tandem
The first observation is that an 18 kip

(80 kN) SAL does about 3,333 times
more damage that the 2 kip (8.9 kN)
SAL (e.g., 1.0 / 0.003 = 3,333). A similar
comparison shows that a 30 kip (133
kN) SAL does about eight times more
damage than the 18 kip (80 kN) SAL.
Comparing tandem axle loads to the
SAL, it can be seen that the tandem
axles spread out the load and minimize
the damage to the pavement. A 30 kip
(133 kN) tandem axle load does only
about 0.7 times the damage of a 18 kip
(80 kN) single axle load. Compared to
the 30 kip (133 kN) SAL, the tandem
axle load does only about 0.1 the
amount of damage.
Essentially, this analysis illustrates that

cars do very little structural damage and
that pavements must be designed to
carry trucks. Furthermore, it illustrates
that over-loaded vehicles cause much
more damage than an 18 kip (80 kN)
SAL. This can greatly impact our
nation's infrastructure if axle load
limitations are raised.
Determining Load Equivalency Factors
LEFs can be based on any response
that measures the difference between
any two loading conditions. The
AASHTO design procedure bases its
LEFs on equivalent serviceability loss
for a given pavement structure. Though
it is possible to explain LEFs using the
AASHTO procedure, it easier to
understand them when examined
mechanistically.
In a mechanistic procedure, load
equivalency factors can be based on
equivalent stress, strain, or deflection at
a given location. Figures 8A and 8B
show how LEFs are determined based
on given level of stress, strain or
deflection for a concrete pavement and
an asphalt pavement.
In Figure 8A, the concrete pavement is
loaded with an 18 kip (80 kN) SAL. This
produces a stress or strain at the bottom
of the concrete layer or a deflection at
the top of the concrete. Loading the
pavement with another load (X kip [kN]
load on axle type Y) produces a different
stress, strain, or deflection. Dividing the
stress, strain, or deflection of an X kip
[kN] load on axle type Y by the stress,
strain, or deflection of an 18 kip (80 kN)
P a g e | 11
Figure 8A and 8B. Load Equivalency Factor Determination for Concrete and Asphalt
Pavements
SAL produces a LEF for that load.
Doing the same thing with a flexible
pavement produces the comparable
flexible LEF (Figure 8B).
When the asphalt pavement is loaded, it
produces different stresses, strains, or
deflections than does the concrete
pavement. The responses to the same
applied load are different because the
different pavement types respond
differently to the load.
Consequently, the LEF values that are
calculated for the same vehicles on
each pavement type are different. When
the same traffic is multiplied by different
LEFs, the ESALs calculated for each
pavement type are different. The
AASHTO equations are based on the
same principle, except that they use a
given serviceability loss (PSI) as the
measure of damage. The equation to
determine the LEF for concrete
pavement or asphalt pavement is:
N
LEF = N
12 | P a g e
N)
Asphalt LEFs vs. Concrete LEFs

Because AASHTO bases its design and
its LEFs on serviceability loss (Figure 7),
the LEFs can be used to compare the
amount of loads needed to cause the
same amount of damage on a concrete
or asphalt pavement. That is, the LEFs
tell how many loads are necessary to
cause the same amount of serviceability
loss in the two pavement types.
Table 5 shows the LEF for two
approximately equivalent pavement
sections: an asphalt pavement with a
structural number (SN) equal to 4.0 and
a concrete pavement that is 8 in. (200
mm) thick. As shown, the concrete
pavement's LEFs are always higher
than the asphalt pavement's LEFs for
axle loads greater than 18 kips (80 kN)
and always less for axle loads less than
18 kips (80 kN). It takes more trucks on
a concrete pavement to cause the same
damage or loss in serviceability than it
does on an asphalt pavement.
Table 5: LEF for Two Equivalent

Pavement Sections
Axle Load,
kip (kN)
Asphalt
LEF
Concrete
LEF
2 (8.9)
0.0002
0.0002
6 (26.7)
0.013
0.010
10 (44.5)
0.102
0.082
14 (62.3)
0.388
0.347
18 (80.9)
1.00
1.00
22 (97.9)
1.47
1.55
26 (116)
2.89
4.42
30 (133)
5.21
7.79
34 (151)
11.3
12.9
38 (169)
18.1
20.6
For example, on a concrete pavement,

the LEF for a 26 kip (118 kN) SAL is
4.42. This means that it takes 4.42 18
kips (80 kN) single axles to cause the
same damage (serviceability loss) of
one 26 kip (118 kN) single axle. On the
asphalt pavement, the LEF for the 26
kip (118 kN) SAL is 2.82, which means it
takes just 2.82 18 kips (80 kN) single
axles to cause the same damage of one
26 kip (118 kN) single axle.
For axle loads less than 18 kips (80 kN),
it is an inverse relationship, it takes 9.8
(1/0.102 = 9.8) 10 kip (45 kN) loads to
cause the damage of one 18 kips (80
kN) single axle on an asphalt pavement
while it takes 12.2 (1/0.82 = 12.2) 10 kip
(45 kN) loads on a concrete pavement.
This analytical look at the LEFs shows

that each load does more damage to
asphalt pavement than it does to a
concrete pavement. Real world
verification can be found by looking at
the performance curve of any two
similar designed and trafficked concrete
and asphalt pavements (Figure 9).
Figure 9. Typical Serviceability Curves

for Concrete and Asphalt Pavements
Load Transfer
Load transfer is a slab's ability to
transfer part of its load to its neighboring
slab. A slab with 100% load transfer
shares its load (deflection and stress)
equally with its neighboring slab. A slab
with 0% load transfer shares none of its
load (Figure 10). Generally, pavements
with good load transfer have minimized
faulting, less corner breaking, and better
performance. In the AASHTO design
procedure, load transfer is affected by
the type of concrete pavement, by the
presence of dowels, and by the
presence of edge support (e.g., tied
concrete shoulder, tied curb and gutter,
or an extended lane; not asphalt or
granular shoulders).
P a g e | 13
Figure 10. Diagram Showing Slabs with

Excellent and Poor Load Transfer
Figure 11 illustrates how dowels and
edge support improve pavement
performance. The pavement on the right
has good load transfer (doweled and
edge support) and the one on the left
has poor load transfer (undoweled and
no edge support). Loading both these
slabs in the middle with a given load will
produce nearly identical internal
deflections, i.
Loading the undoweled pavement with
the same load at the outside,
unsupported corner would produce a
deflection that is about five times greater
than the internal deflection (e.g., 5*i).
At the inside supported corner (due to
the adjacent longitudinal lane), the
deflection would be 3*i. In the doweled
pavement, the deflection at the outside
pavement edge would be about 3*i and
at the inside, supported corner, it would
be about 2*i.
14 | P a g e
Figure 11. Diagram Showing How

Shoulders and Dowels Effect
Deflections in a Slab
Load Transfer Coefficient (J)
In the AASHTO design procedure, load
transfer is accounted for with the load
transfer coefficient, or J-factor. The Jfactor is based on how stress is
transferred across the joint or crack. It
is used to minimize corner cracking and
it does not control or account for
faulting. Faulting is not a failure criterion
in the AASHTO design procedure; the
only failure criterion is serviceability
loss. This means that the J-factor
cannot be used to control faulting.
Nevertheless, this does not mean that a
better J-factor does not affect faulting. It
simply means that it was not taken into
account in the development of the
design equation. Better J-factors do
decrease deflections (Figure 11) and,
thus, minimize the potential for faulting.
The J-factor is dependent on the type of

pavement, edge support condition and
how load transfer is provided at the joint
(or crack). Jointed pavements with
dowels provide a higher level of load
transfer than those relying strictly on
aggregate interlock. CRCPs generally
provide the highest level of load
transfer.
necessary for uniform application of the

J-factor in design of concrete
pavements. Both JPCP and JRCP
designs were evaluated at the Road
Test and produced roughly equivalent
performance. The J-factor value for the
Road Test conditions was 3.2.
Table 6 shows recommended J-factors

for typical concrete pavement designs
(lower J-factors mean better load
transfer). The designer simply selects a
J-factor that is consistent with the type
of pavement and edge support condition
for the design. The J-factor also varies
slightly with the expected design traffic
and indirectly with pavement
classification (e.g., local streets and
roads, arterials, and heavy highways).
The 93 AASHTO Design Guide actually
provides less specific guidance for
selecting the J-factor than provided in
Table 6. ACPA believes that the more
specific guidance in Table 6 is
Concrete Properties
There are two concrete properties that
influence rigid pavement design in the
AASHTO design procedure. They are:
S'c Concrete flexural strength
determined at 28-days using
third-point loading
Ec Concrete modulus of
elasticity
Flexural Strength, S'C
The concrete strength used in the
design of concrete pavements is based
on AASHTO Test Method T97 or ASTM
C78, Flexural Strength of Concrete
using Simple Beam with Third-Point
Loading (Figure 12).11
Table 6: Load Transfer Coefficients (J-Factors) for Typical Designs10

ESALs (millions)
Up to 0.3
0.3 to 1
1 to 3
3 to 10
10 to 30
Over 30
Doweled
JPCP and
all JRCP
No
3.2
3.2
3.2
3.2
3.2
3.2
Yes
2.7
2.7
2.7
2.7
2.7
2.7
JPCP with
Aggregate
Interlock
Edge Support*
No
Yes
3.2
2.8
3.4
3.0
3.6
3.1
3.8
3.2
4.1
3.4
4.3
3.6
CRCP
No
---2.9
3.0
3.1
Yes
---2.5
2.6
2.6
Pavement
Class
Local Streets
and Roads
Arterials and
Highways
* Tied concrete shoulder, tied or integral curb and gutter, or a widened lane all provide the same
support conditions according to AASHTO. Asphalt or granular shoulders and no shoulders provide no
support and therefore no benefit.
P a g e | 15
Figure 12. Flexural Strength of Concrete

using Third-Point Loading
Figure 13. Flexural Strength of Concrete

using Center-Point Loading
It is important that the third point loading

28-day flexural strength be used in the
AASHTO equation. If the strength value
is measured using some other test
method, it must be converted to the 28day third-point strength.
Compressive Strength
Many agencies use compressive
strength of concrete cylinders (AASHTO
T22 or ASTM C39)13 as an alternative to
flexural strength testing. Several simple
conversion equations, such as the one
below, can convert 28-day compressive
strengths to 28-day third point flexural
strengths.
Center Point Flexural Strength

Some agencies use the center-point
flexural test (AASHTO T177 or ASTM
C293) to determine their concrete
strength (Figure 13).12 Center-point
loading forces the beam to fail directly
under the center of the loading. This
may not be the weakest point in the
beam. In third point loading, the entire
middle one-third of the beam is stressed
uniformly and thus the beam fails at its
weakest point in the middle one-third of
the beam. By forcing the beam to fail at
the center, the center point flexural test
results are somewhat higher than the
third-point test results. Typically, center
point results are about 15% greater.
Though this relationship is not exact, it
does provide a reasonable estimate of
the concrete's average strength.
16 | P a g e
where:
S = C
S'c = Average 28-day thirdpoint flexural strength, psi (MPa)
f'c = Average 28-day
compressive strength, psi (MPa)
C = Constant assumed to be
between 8 and 10 for U.S.
standard units (0.7 to 0.8 for
metric units) for typical paving
concrete; for U.S. units, the value
of 9 (0.75) typically produces
reasonable results for most
designs
Table 7 shows typical value ranges for

compressive strength, third point loading
flexural strength, and center point
loading flexural strengths for
conventional concrete paving mixtures.
A free strength converter app is
available at http://apps.acpa.org.
strength will cause the pavement to be

too overdesigned. Therefore, it is
necessary to adjust the specified
minimum strength to the design strength
using the equation below:
Table 7: Typical Comparison Values

for Compressive Strength and Third
Point and Center Point Flexural
Strengths
where:
Comp,
psi (MPa)
Third Point
Flex,
psi (MPa)
Center
Point Flex,
psi (MPa)
2,000 (13.8)
402 (2.78)
463 (3.19)
2,500 (17.2)
450 (3.10)
518 (3.57)
3,000 (20.7)
493 (3.40)
567 (3.91)
3,500 (24.1)
532 (3.67)
612 (4.22)
4,000 (27.6)
569 (3.92)
655 (4.51)
4,500 (31.0)
604 (4.16)
694 (4.79)
5,000 (34.5)
636 (4.39)
732 (5.05)
5,500 (37.9)
667 (4.60)
768 (5.29)
6,000 (41.4)
697 (4.81)
802 (5.53)
6,500 (44.8)
726 (5.00)
834 (5.75)
7,000 (48.3)
753 (5.19)
866 (5.97)
The Importance of Using Average

Strength
Because of the way the 93 AASHTO
Design Procedure uses reliability, it is
strongly recommended that the
expected average, in-field 28-day
flexural strength (S'c) of the concrete
be used in the design procedure
(AASHTO T97 or ASTM C78). Using
the specified minimum construction
S = S + z
S'c = Estimated average in-field
flexural strength
Sc = Specified minimum flexural
strength
= Estimated standard
deviation of the strength
z = Standard normal deviate
corresponding to the percent of
results which can be below the
specified strength
To use this equation, the designer must

know or have estimate values of:
1.
2.
The percent of strength tests

permitted below the specified
level.
The standard deviation of the
strength tests.
The values for z are derived from basic

statistics and are shown in Table 8.
The standard deviation () of the
strength test results depends upon the
variability of the concrete and accuracy
of the testing. Contractors generally use
either central-mix or ready-mix plants to
produce concrete. These plants are
capable of providing very uniform
concrete.
P a g e | 17
Historically, the standard deviation for

ready-mixed concrete is about 7 to 13
percent of the average strength. The
standard deviation for central-mixed
concrete is from 5 to 12 percent of the
average strength. Generally, records of
the standard deviation from past plant
operations are available.
Table 8: Values of the Standard
Normal Deviate (z) corresponding to
the Percent of Tests below the
Specified Strength (Sc)
Percent of Specimens
Below the Specified
Value
0.841
20
1.037
15
1.282
10
1.645
2.327
Example:
Suppose that you want to design a
small street project. You know that
several local operators supply most
of the concrete in your area using
ready-mixed concrete. You also know
that you will specify concrete with a
minimum 28-day flexural strength of
550 psi (3.79 MPa) and your
specification will permit 10 percent of
tests to fall below that level. What
strength do you use in the AASHTO
design equation?
Step 1: Estimate the strength as 9
percent of the flexible strength or call
several ready mix operators to
determine the value. Since you do
not know the actual average strength,
use the specified value for S'c (it will
be fairly close). The value for then
becomes:
= 0.09*550 psi
The example to the right demonstrates

the above procedure to account for the
average in-field 28-day flexural strength.
Modulus of Elasticity
The other concrete property in the
AASHTO design procedure is the
modulus of elasticity, Ec. Ec indicates
how much the concrete will compress
under load. Concretes with a very high
Ec are very rigid and do not compress
much. Concrete with a lower Ec
compresses more under load. In the
concrete pavement equation, Ec is the
most insensitive parameter and has only
a minor impact on thickness design or
projected performance.
18 | P a g e
= 49.5 psi
Step 2: Estimate the design strength
to use in the equation. Apply the
correction for a 10 percent failure rate
(z = 1.282 from Table 8):
Sc = 550 + 1.282*49.5
Sc =613 psi (4.22 MPa)
Thus, 613 psi (4.22 MPa) is used in
the design equations.
Note: The same principle applies if compressive
strengths are used. The corrected compressive
strength would be converted to third-point flexural
strength using the relationship previously shown.
Although Ec can be tested using ASTM

Test Method C469,14 or an equivalent, it
is rarely done in practice. It is usually
estimated from either the flexural or the
compressive strength. The following two
equations are from the American
Concrete Institute (ACI) and provide
reasonable estimates.15
To get Ec from flexural strength:
E () = 6,750 S ()
To get Ec from compressive strength:

E () = 57,000 ()
The free strength converter app

available at http://apps.acpa.org also
can convert to modulus of elasticity in
both U.S. and metric units.
As was the case with the previous
correlation equations for strength, the
above correlation is not exact. It is a
close estimate and can be relied on to
evaluate projected performance within a
reasonable margin of error.
The ranges of values for Ec that are
reasonable depend largely on the
strength of the concrete. Typical values
are from 3.5 to 5 million psi (21,400 to
34,500 MPa). The average Ec obtained
at the Road Test was 4.2 million psi
(29,000 MPa) and is an acceptable
value for design.
Subgrade Support
In all pavements, the load is eventually
transmitted to the subgrade. Though
bases, subbases, and soil modifications
are used to increase the support
strength and protect the subgrade, it is
the natural subgrade that must be used
as the starting point for support
characterization.
For concrete pavements, the primary
requirement of the subgrade is that it be
uniform. This is the fundamental reason
for specifications on subgrade
compaction. A good quality subgrade
will improve the performance of the
pavement.
In the AASHTO design procedure for
concrete pavements, the strength of the
soil is described by two subgrade
properties:
1.
2.
Modulus of subgrade reaction,

or k-value
Loss of support factor (LOS)
Modulus of Subgrade Reaction (k-value)

The modulus of subgrade reaction is
determined by the plate load test
(AASHTO T235 or ASTM D1194).16
The plate load test models the subgrade
as a bed of springs and the k-value is
analogous to a spring constant. In fact,
the k-value is sometimes referred to as
the subgrade "spring constant."
P a g e | 19
The test involves placing a 30 in. (762

mm) diameter plate on the subgrade
and loading it with a very heavy load.
The plate distributes the load to the
subgrade via the pressure of the plate
(Figure 14). The k-value is found by
dividing the plate pressure by plate
deflection under the load. The units for
k-value are psi/in. (MPa/m).
Subbases can be made from either

unbound granular materials or stabilized
materials. Stabilized subbases include
lean concrete (econocrete) subbase,
cement-treated subbase, asphalttreated subbase, and lime-fly ash
stabilized subbase. When a subbase is
used, the k-value for design is increased
to a "composite k" (kc) to account for the
additional support provided by the
subbase.
Loss of Support
The loss of support (LOS) accounts for
the expected erosion and differential
movements of the subbase or subgrade
over the life of the pavement.
Essentially, it reduces the k-value based
on the size of the void that may develop
beneath the slab. A LOS factor of 0 is
equivalent to the conditions at the
AASHO Road Test and the predominant
JPCP and JRCP failure mode at the
Road Test was pumping/faulting due to
the clay soils at the site; thus, using a
LOS of 0 is conservative.
Figure 14. Photo of a Plate Load Test

In all but low volume applications, the
pavement is constructed on some type
of subbase placed over the subgrade.
The subbase material is used to provide
a uniform support layer and a strong
construction platform. Typically it is less
erodible than the subgrade, which limits
the pumping of material from beneath
the slabs.
20 | P a g e
Determining Subgrade Support for

Design
Though the k-value is determined by the
plate load test, this test is rarely
performed in the field. It is an expensive
and very slow test. Furthermore, it does
not give much information. It only tells
what the k-value is at a single point. A
change in soil type, depth to bedrock,
moisture conditions, location along the
grade, etc. will change the results.
The 93 AASHTO Design Guide

recognized this and developed the
following multi-step procedure, detailed
in Part II of the 93 Guide, to estimate kvalue. It is based on the soil resilient
modulus, Mr, used in the asphalt
pavement design. The procedure is:
1. Determine Mr
a. AASHTO T29417
b. Correlate to CBR18 or Rvalue19
2. Convert Mr to k-value
3. Adjust for effects of a rigid
foundation
4. Adjust for LOS
ACPA does not recommend following
this procedure because it produces
unrealistic results. If followed, the
procedure increases the k-value to
unreasonably high values and then
reduces it back to unreasonably low
values using the loss of support.
Still, it is important to understand the
basic procedure and issues with it. The
next section will describe how to
determine the k-value using the above
procedure as well as the errors in it.
Following it, we will show what we
consider to be more realistic values for
the design of concrete pavements.
subgrade reaction (k-value), Mr

indicates the stiffness of the layer
immediately under the pavement.
The Mr is determined from AASHTO
T294, Resilient Modulus of Unbound
Base/Subbase Materials and Subgrade
Soils. However, it is recognized that
many agencies do not have the
equipment to perform this test.
Therefore, AASHTO recommends the
following correlation equations to relate
the resilient modulus to the California
Bearing Ratio (CBR) or the R-value.
Mr(psi) = 1,500*(CBR)
Mr(psi) = 1000 + 555*(R-value)
Convert Resilient Modulus to k-Value
Once Mr is determined or estimated, it is
converted to a k-value by one of two
ways, depending on whether a subbase
is present or not. If there is no subbase
the k-value is calculated as:
k-value = Mr/19.4
If there is a subbase, Figure 3.3 from
part 2 of the AASHTO guide is used
(reproduced here as Figure 15). This
figure estimates the "composite k-value"
(kc), which represents the additional
strength provided by the subbase.
AASHTO Procedure to Determine the kValue

Determine Resilient Modulus
The resilient modulus measures the
recoverable deformation of a
dynamically loaded test specimen at any
stress level. Like the modulus of
Lime treated soils should be considered as a base course.

The subgrade k-value used for design is a "composite kvalue" starting with the k-value of the actual subgrade and
not the modified soil. This applies to asphalt pavements as
well, although resilient modulus is used instead of k-value.
In no case should the subgrade soil be ignored in design.
P a g e | 21
Figure 15. Chart for Estimating Composite Modulus of Subgrade Reaction, Assuming
a Subgrade Depth Greater than 10 ft (3 m) [Figure 3.3, part 2 of the 1993 AASHTO
Pavement Design Guide]
22 | P a g e
Adjust k-Value for Depth to Rigid

Foundation
This step accounts for the proximity of
the pavement to bedrock. When a
pavement is within 10 ft (3 m) of the
bedrock, the confining pressure of the
bedrock causes the subgrade support to
increase. This step is disregarded when
the depth to the rigid foundation is
greater than 10 ft (3 m).
To adjust for the depth to a rigid
foundation, Figure 3.4 from part 2 of the
AASHTO guide is used (reproduced
here as Figure 16)
Adjust k-Value for Loss of Support

After the k-value is calculated, it is
adjusted for LOS using Figure 3.6 from
part 2 of the AASHTO guide
(reproduced here as Figure 17). A LOS
of 0 represents Road Test conditions.
Seasonal Adjustment to the k-Value
AASHTO also recommends doing this
procedure for each month of the year to
reflect seasonal changes. However,
because the Road Test ran year round
for several seasons, impacts of
seasonal changes of the k-value are
inherent in the AASHTO equations.
Figure 16. Chart to Modify k-Value to Consider Effects if Foundation is within 10 ft (3

m) of the Surface [Figure 3.4, part 2 of the 1993 AASHTO Pavement Design Guide]
P a g e | 23
Furthermore, for concrete pavement,

these adjustments have very little effect
on the final results. For this reason,
seasonal adjustments have not been
included in WinPAS. For further
information, consult the AASHTO guide.
Problems with the AASHTO Procedure
to Determine Subgrade Support
As mentioned, there are several
problems with the current AASHTO
procedure to determine the subgrade
support values. The most glaring errors

deal with:
1.
2.
3.
4.
The LOS factor,

The accuracy of the CBR and
R-value relationships to Mr,
Inconsistencies with the
relationships between k-value
and Mr for base and no
subbase that occur with high
in-situ Mr values, and
Unrealistic resulting k-values.
Figure 17. Correction of Effective k-Value for Potential Loss off Support [Figure 3.6,
part 2 of the 1993 AASHTO Pavement Design Guide]
24 | P a g e
Loss of Support
This factor reduces the k-value for an
expected loss of support by subgrade
erosion. A LOS = 0 models the soil
conditions at the AASHO Road Test. A
closer look at the soils at the Road Test
show that it consisted of three feet of
embankment with these properties:
AASHO A-6 (clay)

Group Index = 9-13
Plastic Index = 11-15
Liquid Limit = 27-32
80-85% passing the #200 (75
m) sieve
Loss of support was the primary failure

mode of concrete pavements at the
AASHO road test. The pumping of
subbase fines from underneath the
slabs preceded all cracking. Therefore,
the effects of LOS are inherent in the
equation that predicts concrete
pavement performance or serviceability
loss. As a result, this additional factor is
unnecessary.
ACPA recommends setting the loss of
support factor to 0 for all designs.
Necessary improvements in subgrade or
subbase support should be made using
an improved subbase material or
improved drainage design.
CBR and R-Value Relationships to Mr
The AASHTO CBR and R-value to Mr
relationships are considered reasonable
only for fine-grained soils (e.g., CBR
less than 10 or R-value less than 20).
These two equations greatly over
estimate Mr values at higher CBRs and
R-values (Figure 18).
Figure 18. Relationships between

Resilient Modulus and CBR or R-value
[from ACPAs MC016P, WinPAS
Pavement Analysis Software]
ACPA has developed two non-linear
relationships, based on NCHRP Report
128, Evaluation of the AASHO Interim
Guide for the Design of Pavement
Structures,20 to more accurately
estimate Mr from CBR or R-value:
M () = 1941.488 CBR
M () = 2165.935 e
.
R
As shown in Figure 18, these equations

better follow the actual values, leading
to better prediction or correlation.
Inconsistencies between Base and No
Subbase Conditions
The relationships between k and Mr for
the base and no subbase can give
inconsistent values, especially with high
in-situ Mr values.
P a g e | 25
For example, assume that a soil has

been tested and has a Mr of 12,000 psi
(82.7 MPa). In the case with no base,
the subgrade k-value is calculated in the
AASHTO method by:
k-value = Mr /19.4
= 12,000 psi/19.4
= 619 psi/in.
When a 6 in. (150 mm) unstabilized
(granular) subbase with a resilient
modulus of 25,000 psi (173 MPa) is
placed over this soil, the composite kvalue decreases:
k = 572 psi/in. (from Fig 15)
This is obviously not correct. The
subbase course is stronger, provides
better support, and will improve the
pavement's performance. This
inconsistency in the AASHTO method
becomes greater as the Mr value
increases.
The Resulting k-Values are Unrealistic
Probably the most compelling reason
not to follow the AASHTO procedure to
estimate k-values is that it does not
produce realistic results. Historical and
theoretical values for the types of
subgrades and subbases found under
concrete pavements typically are in the
range of 50 to 550 psi/in. (13 to 150
MPa/m). Using the AASHTO procedure,
it is possible to get values as high as
2,000 psi/in. (540 MPa/m). Furthermore,
the ranges that can be derived can be
extremely large.
26 | P a g e
Table 9 shows a comparison of

historical and AASHTO derived values
for different types of subgrades and
subbases. As can be inferred, this
disparity can affect designs greatly.
Table 9: Comparison of Historical
and AASHTO derived k-Values
Layer
Type
Historical
k-value,
psi/in.
(MPa/m)
AASHTO
k-value,
psi/in.
(MPa/m)
Silts &
Clays
60-100
(16-30)
10-20
(2.7-5.4)
Granular
150-250
(40-68)
12-73
(3.5-20)
AsphaltTreated
300-400
(80-108)
95-128
(25-35)
CementTreated
405-550
(110-150)
128-400
(35-110)
Recommended Values for the Modulus

of Subgrade Reaction
Though the AASHTO procedure does not
produce reasonable k-values, the basic
premise of relating it to other soil
properties is reasonable. Furthermore,
an exact value is not required. Normal
variations from an estimated value will
not appreciably affect pavement
thickness (e.g., an error in the k-value of
100 percent only increases or decreases
a typical pavement thickness by about
0.4 in. [10 mm]). Figure 19 shows
relationships that are satisfactory for
design purposes.
Figure 19. Approximate Interrelationships of Soil Classifications and Bearing Values

[from ACPA EB109P, Thickness Design for Concrete Highways and Street
Pavements]
P a g e | 27
Because Figure 19 does not contain Mr,

ACPA has developed a correlation
equation for Mr to k-value that, when
used in conjunction with the previously
mentioned ACPA correlations from CBR
or R-value to Mr, will result in k-values
that match those in Figure 19. The Mr to
k-value correlation equation is:
If Mr 15,089:
k = M 0.0000001155 M
0.0004683533 M
+ 41.1348117373
If Mr > 15,089:
k = M 0.0000000106 M
0.0007608054 M
+ 69.4602909796
See http://apps.acpa.org for free apps to
easily convert CBR or R-value to Mr and
then to k-value.
Recommended k-Values for Subbases
When a subbase is used, there is an
increase in the k-value. The magnitude of
the increase depends on whether the
subbase is stabilized (treated) or
unstabilized (untreated). Table 10 shows
an approximate increase of k-value
based on the type of subbase and its
thickness. A composite k-value
calculator also is available at
http://apps.acpa.org.
AASHTO and the Benefits of Subbases
It is not economical to use a base or
subbases for the sole purpose of
increasing the k-value. An increase of kvalue from 90 psi/in. (25 MPa/m) to 500
psi/in. (135 MPa/m) will only decrease
thickness by about 10 percent.
28 | P a g e
Table 10: Approximate Composite kValues (kc) for Various Subbase

Types and Thickness
Unstabilized (Granular) Subbase
Composite k-value, psi/in. (MPa/m)
Subgrade
k-value,
pci/in.
(MPa/m)
4 in.
(100
mm)
6 in.
(150
mm)
9 in.
(230
mm)
12 in.
(305
mm)
50 (14)
65.2
(17.6)
75.2
(20.3)
85.2
(23.0)
110
(29.7)
100 (27)
130
(35.1)
140
(37.8)
160
(43.2)
190
(51.3)
150 (41)
175
(47.3)
185
(50.0)
215
(58.1)
255
(68.9)
200 (54)
220
(59.4)
230
(62.1)
270
(72.9)
320
(86.4)
Asphalt-Treated Subbase
Subgrade
k-value,
pci/in.
(MPa/m)
4 in.
(100
mm)
6 in.
(150
mm)
9 in.
(230
mm)
12 in.
(305
mm)
50 (14)
85.2
(23.0)
112
(30.2)
155
(41.9)
200
(54.0)
100 (27)
152
(41.0)
194
(52.4)
259
(69.9)
325
(87.8)
150 (41)
217
(58.6)
271
(73.2)
353
(95.3)
437
(118)
200 (54)
280
(75.6)
345
(93.2)
441
(119)
541
(146)
Cement-Treated Subbase
Subgrade
k-value,
pci/in.
(MPa/m)
4 in.
(100
mm)
6 in.
(150
mm)
9 in.
(230
mm)
12 in.
(305
mm)
50 (14)
103
(27.8)
148
(40.0)
222
(59.9)
304
(82.1)
100 (27)
185
(50.0)
257
(69.4)
372
(100)
496
(134)
150 (41)
263
(71.0)
357
(96.4)
506
(137)
664
(179)
200 (54)
348
(94.0)
454
(123)
634
(171)
823
(222)
Subbases and bases are primarily used

to prevent the pumping of fines from
underneath the slab. Secondarily, they
are used to help control frost heave and
swelling soils, provide a drainage layer
when needed, and provide a working
platform for construction.
1.
The current AASHTO design does not

model the contribution of subbases
accurately. At the AASHO Road Test, it
was found that the concrete pavements
with any granular subbase could carry
about 30% more traffic.
Controlling any one of these items will

minimize pumping. Edge drains and
free-draining subbase layers help
minimize the free water between the
slab and subgrade and thus minimize
the amount of pumping. Dowels and
edge support also minimize pumping by
controlling the deflections of heavy
wheel loads. The use of improved
drainage, dowels, and edge support will
definitely lead to improved performance.
The current design procedure allows

concrete pavements built with granular
bases to carry only about 5 to 8% more
traffic. This indicates that concrete
pavements built with granular subbases
should perform better than predicted by
the AASHTO design equations.
Coefficient of Drainage (Cd)
Trapped water within a pavement
structure is one of the primary
contributors to pavement distresses. It
can lead to:
1.
2.
3.
4.
5.
6.
Reduced strength of unbound

granular materials.
Reduced strength of subgrade
soils.
Pumping of fines.
Differential heaving/swelling of
soils.
Loss of structural support.
Pavement settlement and/or
faulting.
For concrete pavements, the major item

that drainage control is the pumping of
fines. The conditions that lead to pumping
are:
2.
3.
4.
Subgrade soil that will go into

suspension.
Free water between slab and
subgrade.
Frequent heavy wheel loads with
large deflections.
Poor load transfer between slabs.
In the AASHTO design procedure,

drainage is accounted for by use of the
drainage coefficient (Cd). The drainage
coefficient accounts for improved or
decreased quality of drainage over
those conditions at the Road Test.
Table 11, taken from the AASHTO
guide, provides recommended Cd
values. The value of Cd depends on the
quality of drainage and percent of time
the pavement is exposed to moisture
levels approaching saturation. Because
the Cd value depends on the saturation
of the subgrade/subbase, it is possible
for a pavement in a dry environment
with poor drainage to perform as well as
a pavement in a wet environment with
excellent drainage.
As mentioned, the subgrade soil at the

AASHTO Road Test was very poor (clay).
Though the pavement was designed with
elevated cross-sections and drainage ditches,
edge drains were not used. Thus, the subgrade
below the Road Test pavements was not welldrained.
P a g e | 29
Table 11: Recommended Values of the Drainage Coefficient (Cd) for Concrete
Pavement Design
Quality of
Drainage
Percent of Time Pavement Structure is Exposed to

Moisture Levels Approaching Saturation
< 1%
1% - 5%
5% - 25%
> 25%
Excellent
1.25 1.20
1.20 1.15
1.15 1.10
1.10
Good
1.20 1.15
1.15 1.10
1.10 1.00
1.00
Fair
1.15 1.10
1.10 1.00
1.00 0.90
0.90
Poor
1.10 1.00
1.00 0.90
0.90 0.80
0.80
Very Poor
1.00 0.90
0.90 0.80
0.80 0.70
0.70
Appendix DD of Volume II of the 1993 guide offers the following definitions for quality of drainage:
Excellent Drainage - Soil drained to 50 percent of saturation in 2 hours.

Good Drainage - Soil drained to 50 percent of saturation in 1 day.
Fair Drainage - Soil drained to 50 percent of saturation in 7 days.
Poor Drainage - Soil drained to 50 percent of saturation in 1 month.
Very Poor Drainage - Soil does not drain.
As a basis for comparison, a Cd value of

1.00 represents conditions at the
AASHTO Road Test. Thus, a Cd value
of 1.00 has no impact on the design.
Lower Cd values increase the required
pavement thickness and higher values
decrease the required thickness.
Caution is recommended when using
drainage coefficients of less than 1.00.
Because the subgrade soils at the Road
Test were very poorly draining soils, the
AASHTO design equations already
account for a large degree of poor
drainage. Values less than 1.00 would
indicate conditions worse than that of
the AASHO Road Test. Open-graded,
free-draining subbases and freedraining soils which can be maintained
can be modeled with a drainage
coefficient greater than 1.00.
30 | P a g e
Reliability
Reliability (R) accounts for the chance
variation in traffic predictions,
performance predictions, concrete
material properties, subgrade support
conditions, etc. It incorporates some
degree of certainty into the design
process to ensure that the pavements
will survive the analysis period for which
they are designed.
In the AASHTO design procedure there
are two basic statistical factors that
make up reliability:
1.
2.
Reliability (R)
Standard deviation (s0)
It is important to note that by treating

reliability and standard deviation as
separate design factors, conservative
estimates for all the other design inputs
should no longer be used. Rather, all
input values should be input as the best
estimate of the average in-place, in-field
values. The selected level of reliability
and overall standard deviation will
account for the combined effect of the
variation of all the other design
variables.
Reliability (R)
Reliability is the statistical probability
that the pavement will meet its design
life. Essentially, reliability tells how
much of the pavement will be operative
at the end of its design life. For
example, a pavement designed with
90% reliability will have 90% of the
pavement in operational condition at the
end of the design period, and only 10%
of the pavement will have "failed."
Evaluating the acceptable level of risk
for the design is necessary when
choosing an appropriate reliability. High
levels of reliability, or low risk, are
appropriate for high traffic volume
pavements in urban areas where future
repairs are difficult and undesirable.
Medium or low levels of reliability are
appropriate for lower levels of traffic or
rural areas where repairs pose little
difficulty and more risk is tolerable. The
lowest reliability level, 50%, corresponds
to local roads.
Table 12 presents recommended levels

of reliability for different roadway
classifications. When comparing two
different pavement sections (e.g., a
new concrete section to a new asphalt
section, two different concrete sections,
or two different asphalt sections), the
same level of reliability for each must
be used. When an existing pavement is
being analyzed, it must be evaluated
with the reliability equal to 50%.
Table 12: Suggested Reliability
Levels for Various Functional
Roadway Classifications
Classification
Recommended
Reliability (R), %
Urban
Rural
85 99.9
80 99.9
80 99
75 99
Collectors
80 95
75 95
Local
50 80
50 80
Interstate & Other

Freeways
Principal Arterials
Standard Deviation (so)

Standard deviation is the amount of
statistical error present in the design
equations due to the variability in the
materials, construction, etc. It
represents the amount of scatter
between predicted performance and
actual performance. To determine the
true value of s0 requires knowledge of
the individual s0 values of each
parameter (strength, elasticity, soil
support, etc.). Obtaining this information
is fairly difficult.
P a g e | 31
A typical range of s0 values for each

pavement type are published in the 93
AASHTO Design Guide:
Concrete Pavements:
0.30 < s0 < 0.40
Asphalt Pavements:
0.40 < s0 < 0.50
The actual s0 value for concrete
pavements at the Road Test, where the
conditions were controlled and exactly
known, was 0.25. AASHTO
recommends increasing the s0 value to
account for error in traffic projections.
When lacking better information on the
value of s0 for a particular situation, 0.35
and 0.45 are appropriate values for
concrete and asphalt pavement design,
respectively.
How Reliability Works
Understanding reliability requires
understanding the design curve and its
relationship to the performance curve. In
Figure 20, the performance curve
represents the average pavement
performance at the AASHO Road Test.
The AASHTO concrete pavement
design equation defines its shape and
its intersection with the various PSl
levels for all combinations of concrete
sections tested at the Road Test. This
curve passes through the average Road
Test PSl values and represents a
reliability level of 50 percent.
32 | P a g e
Figure 20. How Reliability Shifts

Performance Curve to Obtain a Design
Curve
The design curve is offset from the
performance curve based on the
specified reliability chosen by the user.
Essentially, the design curve is shifted
from the performance curve by an
amount equal to s0 multiplied by the
standard normal deviate (ZR) for a given
level of reliability (ZR is the standard
normal deviate for the normal
distribution at a given value of
reliability).
How ZR Relates to R
ZR is the degree of offset from the
average PSl value, as shown in Figure
21. Basically, it describes the area
under the curve, which is the probability
of success or failure. Using statistics
and the standard normal distribution
curve, the ZR value is selected so that
the percentage of the area enclosed by
the curve is the desired level of
reliability, R (the area to the right of the
offset).
Table 13: Standard Normal Deviate

(ZR) Values Corresponding to
Selected Levels of Reliability
Figure 21. Standard Normal Curve

The engineer selects the appropriate ZR
value after choosing the desired level of
reliability. For example, a 50% level of
reliability corresponds to a ZR of 0.0
(50% of the area under the curve is to
the right of the average or mean value).
At a higher level of reliability (e.g., 95%)
the appropriate ZR value is chosen so
that 95 percent of the area under the
curve is to the right of the average
value.
An engineer can choose ZR values for
any desired reliability. ZR values are
available in most statistics textbooks.
Typical values used for pavement
design are shown in Table 13.
Once the R and s0 values are known,
the offset between the design and
performance curve is established.
The end result is basically a load safety
factor or a multiplier of the allowable
ESALs. The predicted design ESALs in
the AASHTO equation are multiplied by
the safety factor in order to ensure the
design performance is met.
Reliability (R), %
Standard Normal
Deviate (ZR)
50
0.000
75
-0.674
80
-0.841
90
-1.282
95
-1.645
97
-1.881
99
-2.327
99.9
-3.090
For example, when a pavement is

designed at 50 percent reliability, the
predicted design ESALs are multiplied
by a safety factor of 1.0 (no safety
factor). When designed at 80%
reliability, the predicted ESALs are
multiplied by a safety factor of 1.97.
Essentially, the pavement will be able to
carry about 100 percent more ESALs
than the predicted design ESALs. A
pavement with a 95 percent level of
reliability will, on average, carry about
3.75 times more ESALs than that for
which it is designed (e.g., the predicted
ESALs using all the same inputs but a
reliability of 50% such that the AASHO
performance curve is followed).
For more information on reliability, see
Chapter 4 in the 1993 AASHTO
Pavement Design Guide.
P a g e | 33
The Iterative Process

The AASHTO design procedure is an
iterative process. It requires the
designer to know the volume and types
of axle loads, the desired terminal
serviceability (pt), an estimate of the
required pavement thickness, etc. If you
do not have a "feel" for the probable
range of thicknesses for your design
traffic, start with a concrete surface
course thickness of 9 in. (230 mm).
After determining the design pavement
thickness using the estimated values,
the designer should check the results
against the ESAL calculations. If the
assumed pavement thickness is within
five percent of the design pavement
thickness, the results are reasonable.
However, if the computed pavement
thickness is greater than five percent,
the design ESALs should be
recalculated using the last design
thickness, and the whole design re-run.
In practice, ESAL recalculation will
probably not significantly affect the new
pavement thickness. However, the
iterative process is technically correct.
Sensitivity Analysis
One of the frequent pieces of
information missing in a pavement
design is which variables will most
influence the required thickness. Figure
22 shows the change in thickness over
the typical range of each design
variable. These graphs illustrate the
relative importance of each variable and
how a change in each design input will
affect the final design for this set of
inputs.
34 | P a g e
If a change in a variable produces a

steep slope in the graph, the required
thickness is sensitive to changes in that
variable. If the slope is relatively flat, the
required thickness is not very sensitive
to the variable.
The baseline design for generation of
the plots in Figure 22 was:
Design ESALs: 6,142,000

Reliability (R): 80%
Overall standard deviation (s0):
0.34
Modulus of rupture (Sc): 600 psi
(4.14 MPa)
Modulus of elasticity (Ec):
4,000,000 psi (27.6 MPa)
Load transfer (J): 3.2 (e.g.,
doweled w/o edge support)
Modulus of support (k-value): 200
psi/in. (54 MPa/m)
Drainage Coefficient: 1.0
Initial Serviceability: 4
Terminal Serviceability: 2
These basic inputs resulted in a

required concrete pavement thickness
of 9 in. (230 mm).
It is important to note that, while the
general trends will remain true, the
magnitude in change of required
thickness shown in these sensitivity
plots is unique to this set of inputs.
Note: If a reliability of 50% had instead been

used, ZR would have been zero and the ZR*s0
term will drop from the AASHTO concrete
pavement design equation. In such cases, as is
the case with analysis of an existing pavement,
the thickness is completely unchanged by
changes in the overall standard deviation.
Figure 22. Charts Illustrating the Sensitivity of each Variable in the AASHTO Design
Equation on Design Thickness (in inches)
P a g e | 35
As shown, the variables with the largest

effect on the required thickness are
reliability, load transfer, drainage
coefficient, and flexural strength. It is
critical that proper thought be given to
each of these variables in the design.
Under- or over-estimating their actual
value can impact the design greatly.
The least sensitive variables are the
standard deviation and concrete
modulus of elasticity.
While the designer might investigate the
sensitivity of the design on variables
such as load transfer, drainage
coefficient, standard deviation and
initial/final serviceability, many of these
variables typically are standardized
based on local practice and experience.
Thus, the remaining design variables
oftentimes are the focus of a designer
who is looking to optimize the design.
Most of the effects shown in the
reliability sensitivity chart do occur over
a small range of very high reliability
levels (e.g., 95% to 99.9%); at levels
below 95%, the impact of reliability
drops considerably. For this reason,
caution is recommended when deciding
what reliability factor to use. When
choosing a high reliability, all the design
procedure does is increase the
thickness. However, most concrete
pavements do not fail because of
inadequate thickness; rather, they
typically fail because of poor jointing or
material problems. Therefore, using a
high reliability can cause the design to
be overly conservative.
36 | P a g e
Modulus of rupture (flexural strength)

and modulus of elasticity go hand-inhand. If the designer chooses to
increase the flexural strength to see if
the required thickness can be
decreased, the modulus of elasticity
must also be increased because
stronger concrete mixtures generally
also are more rigid. Regardless, the
designer should consider ancillary
effects of increased strength (e.g.,
fracture toughness typically is
decreased [and, thus, crack propagation
occurs more quickly] in stronger, stiffer
concrete mixtures).
For most designers, the k-value is the
design element in concrete pavement
design that tends to garner the most
focus when the goal is to optimize the
pavement structure. However, as can be
seen on the chart, composite k-value
has relatively little impact on the
required thickness.
From a design perspective **, an
engineer really only needs to know if the
pavement is going to be built on the
natural subgrade (k-value 100 psi/in.
**
This does not mean the condition of the

subgrade is unimportant. For concrete
pavements, the most important objective of the
subgrade support is that it be uniform throughout
the pavement's life. Proper subgrade design
and construction are absolutely necessary if the
pavement is to perform. Likewise, poor
subgrade/subbase preparation cannot be
overcome with thickness increases. Any
concrete pavement, of any thickness, will have
problems on a poorly designed and constructed
subgrade or subbase. For more information on
subgrades and subbases, see ACPAs EB204P,
Subgrades and Subbases for Concrete
Pavements.
[25 MPa/m]), an unstabilized/granular

subbase (k-value 150 psi/in. [40
MPa/m]), an asphalt-treated subbase (kvalue 200 psi/in. [54 MPa/m]), a
cement-treated subbase (k-value 250
psi/in. [68 MPa/m]), or a lean concrete
subbase (k 500 psi/in. [125 MPa/m]).
Any changes in thickness that may
result from a better estimate of the
actual k-value, due to better
information, are most likely not worth
the effort/cost. It is better to
concentrate on other design inputs.
Summary
There have been many welcome
additions in the 1993 Design Guide,
such as the improved overlay and lowvolume road design. However, items
that the concrete industry feels are still
in question include:
Loss of Support Factor was the primary
failure mode of rigid pavement sections
in the Road Test. Many of the failed
sections were the result of the migration
and pumping of subbase fines from
underneath the pavement. Therefore,
loss of support is inherent in the
equation that predicts concrete
pavement performance.
Seasonal Variation of Subgrade Support

are also inherent in the equation that
predicts concrete pavement
performance because the Road Test ran
year round for several seasons. The
pavements and subgrade materials
underwent seasonal variations during
the testing time. Therefore, the 93
AASHTO Design Guide's recommended
procedure to the adjust subgrade
support for seasonal variation in the
concrete pavement design is
unnecessary.
Traffic Equivalency Factors used in the
design of concrete pavements are not
appropriate for modern pavements.
These factors were established only for
unstabilized subbases. No adjustment
is provided for the now more-common
stabilized support layers. This biases
the results of the ESAL determination in
favor of other pavement types.
For these reasons, the industry heartily
endorses the efforts by AASHTO to
move in the direction of improved
mechanistic methods, such as the
recently released mechanistic- and
empirical-based AASHTO
DARWinMETM.
The 1986 and 1993 revisions to the

Guide provide no manner to improve the
support value for non-erodable
subbases. Many modern concrete
pavement designs include such
subbases. It is logical that some factor
should be available for the engineer to
improve the support characterization to
the pavement for a non-erodable base.
P a g e | 37

WINPAS User Manual

Uploaded by

Copyright:

Available Formats

WINPAS User Manual

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

WINPAS User Manual

Uploaded by

Copyright:

Available Formats

Windows Pavement

Windows Pavement Analysis

This publication is intended SOLELY for use by PROFESSIONAL PERSONNEL who

2012 American Concrete Pavement Association

ACPA is the premier national association representing

Pavement and Overlay Design Based on the 1993 AASHTO

AASHTO Guide for the Design of

Design and Construction Joints

Design and Construction Joints

Chapter 1 New Concrete Pavement Design................................................................. 1

Convert Resilient Modulus to k-Value ............................................................. 21

Step 7. Determine the Existing Structural Capacity (SCeff)..................................... 43

Analysis Methods ................................................................................................... 62

Economic Factors .................................................................................................. 90

This Page Left Intentionally Blank

Chapter 1 New Concrete Pavement Design

Concrete Pavement Basics

dowels should be based on other

Jointed Plain Concrete Pavement (JPCP)

Jointed Reinforced Concrete Pavement (JRCP)

Continuously Reinforced Concrete Pavement

Figure 1. Concrete Pavement Types

JRCP contain steel mesh reinforcement

AASHO Road Test

3 through 6 were the main test loops

Figure 3. AASHO Loop Layout for Loops

Figure 2. The AASHO Road Test

In the asphalt pavements, every 100 ft

In total, there were 368 concrete test

Figure 4. AASHO Test Traffic and

Table 1: Design Variables at the AASHO Road Test2

Asphalt Pavement Variables

2.5 (63), 3.5 (89),

1 (25), 2 (51), 3 (76),

0 (0), 3 (76), 6 (152),

The average speed on the test loops

Association of Highway and

From the tremendous amount of data

Figure 5. Present Serviceability Index

The 1986 guide was also the first guide

In 1993, the overlay design procedure

J = Load transfer coefficient

AASHTO Rigid Pavement Design

4.5 1.5 + (4.22 0.32 p )

ESAL = Allowable number of

Ec = Concrete modulus of elasticity,

Understanding the importance of each

pavement thickness is outside of this

Figure 6. The Present Serviceability

The AASHO Road Test pavements

Figure 7. The 93 AASHTO Design is

Table 2: Recommended Terminal

PSITR = PSI PSIENV

PSITR = PSI loss due to traffic

Table 3: Rigid and Flexible ESALs Generated by a Mixed Traffic Stream

This difference in ESALs can be

Load Equivalency Factors

Load equivalency factors vary depending on

Table 4: Typical Load Equivalency

The first observation is that an 18 kip