Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

CIVE460 Lectures17-18

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

American University of Beirut

CIVE460 - Unit C: Traffic Control and Analysis at Signalized


Intersections
Spring 2023-2024
Lectures 17-18: Basic Concepts and Definitions (cont.); Analysis Using
D/D/1 Queuing

March 26 and 28, 2024

Required Reading: MW 7.3, 7.5.1


Supplemental Reading: HCM (2010) Ch. 18

1. Basic Concepts and Definitions

Saturation Flow Rate ( s ): “Maximum hourly volume that can pass through an
intersection, from a given lane or group of lanes, if that lane (or lanes) were allocated
constant green over the course of an hour.” (MW p. 228)

3600
s (1),
h
where s is in veh/h, and h is the saturation headway in s/veh. (Saturation headway is
“the minimum headway that a lane group can sustain across the stop line as the vehicles
depart the intersection.” (HCM, 2000)).

Some observations:

 Typical maximum (base) saturation flow rate = 1900 pc/h/ln  corresponds to a


saturation headway of 1.9 seconds
 Factors affecting saturation flow rate include:
o Number of lanes
o Lane width
o Heavy vehicles
o Grades
o Curbside parking
o Bus stops
o Pedestrians
o Bicycles
 Saturation flow rate is obtained by:
o Application of adjustment factors, or
o Direct measurement in the field

1
In this course, the given saturation flow rates will be assumed to be adjusted for the given
conditions.

 Effect of turning movements on saturation flow rate:


o Lanes that allow left or right turns usually have lower saturation flow
rates because vehicles reduce speed to make a turning maneuver.
o If a turning movement is permitted rather than protected, its saturation
flow rate will be reduced as a result of the turning vehicles yielding to
conflicting through and right-turning vehicles (for left turns only), bicycles,
and/or pedestrians.

Lost Time (for a Lane Group)

Lost time is time that is not effectively serving any movement of traffic. It consists of
start-up lost time and clearance lost time. A lost time is applied each time a movement is
started.

 Start-Up Lost Time ( t sl ): Time used by the first few vehicles in a queue while
reacting to the initiation of the green phase and accelerating. Drivers in the queue
do not instantly start moving at the saturation flow rate. 2 seconds is typical for
start-up lost time. In the figure below, it is the summation of the amount of the
headway time that is greater than the saturation headway, for the first five
vehicles.

Figure 1. Concept of saturation headway and start-up lost time (Shawky and Ghafli, 2016)

2
 Clearance Lost Time ( tcl ): Time between signal phases during which an
intersection is not used by traffic (usually, latter part of yellow interval + all-red
interval).

Therefore, the total lost time t L per phase is:


t L  t sl  tcl (2)

Total Lost Time (L)

 This is the total lost time per cycle during which the intersection is not used by
any movement.
 This amount is fixed regardless of phase or cycle length.
o For shorter cycle length, lost time is a larger % of cycle length, resulting in
a larger lost time over the course of a day than for longer cycle lengths.
o However, longer cycles have more phases which may result in similar %
of lost time.

Effective Green Time (g ) : Time during which a traffic movement is effectively


utilizing the intersection. It is given by:

g  G  Y  AR  t L (3)

Where:

g = effective green time for a traffic movement in seconds,


G = displayed green time for a traffic movement in seconds,
Y = displayed yellow time for a traffic movement in seconds,
AR = displayed all-red time in seconds, and
tL = total lost time for a movement during a cycle in seconds

Effective Red Time (r ) : Time during which a traffic movement is not effectively
utilizing the intersection. It is given by:

r  R  tL (4)

We also have:

r Cg (5)

Capacity (c) (for a lane group): maximum hourly rate (veh/h) at which vehicles can
reasonably be expected to pass through the intersection under prevailing traffic, roadway,

3
and signalization conditions. It is expressed as the saturation flow rate times the ratio of
effective green to cycle length, as follows:

g
c  s (6)
C

2. D/D/1 Queuing

We will study 2 cases of D/D/1 queuing:


 Case 1: Approach Arrivals < Approach Capacity for all cycles
 Case 2: Approach Arrivals > Approach Capacity for some cycles

2.1 Notation

 v = arrival rate (veh/s) [e.g. if flow rate at an approach is 500 veh/h,


500
v  0.139 veh/s]
3600
 s = departure rate during the effective green time (veh/s) [e.g. if saturation flow
2400
rate for an approach is 2400 veh/h, s   0.667 veh/s]
3600
 g = effective green (s)
 r = effective red (s)
 tc = time from start of effective green till queue dissipation (s)

Per cycle, we have:


 Approach arrivals = vC
 Approach capacity (maximum departures) = sg

2.2 Case 1: Approach Arrivals < Approach Capacity for all cycles

Figure 2 below shows the arrivals and departures assuming that the approach capacity
exceeds approach arrivals for all cycles (i.e. no queues exist at the beginning or end of a
cycle).

4
Figure 2. Cumulative arrival and departure curves for Case 1 (Figure 7.13 in MW)

Analysis of Figure:

Assuming a D/D/1 model, the equations below are valid only if vC  sg .

Time to Queue Dissipation after Start of Effective Green, tc :

vr  tc   stc  vr  vtc  stc


vr
 tc 
sv

Proportion of Cycle with a Queue, Pq :


r  tc
Pq 
C

5
Proportion of Stopped Vehicles, Ps :

vr  tc  r  tc
Ps    Pq
vr  g  C
Also:
vr  tc  stc t
Ps    c
vr  g  vC v C
s

Maximum Number of Vehicles in Queue, Qmax :

Qmax  vr

Total Vehicular Delay Per Cycle, Dt :

1  vr  1  sr  vr 2
 r  v r  t c   rv  r 
1
Dt    rv  
2 2  sv 2 sv  v
21  
 s

Average Vehicular Delay Per Cycle, d avg :

Dt r2
d avg  
vC  v
2C 1  
 s

Maximum Delay of Any Vehicle (assuming FIFO), d max :

d max  r

2.3 Case 2: Approach Arrivals > Approach Capacity for some cycles

Example:

Consider an approach to a signalized intersection:


 Saturation flow rate s = 1700 veh/h
 Cycle length: C = 60 s
 Effective red: r = 40 s
 During 3 consecutive cycles:
o 15 vehicles arrive in 1 st cycle
o 8 vehicles arrive in 2 nd cycle
o 4 vehicles arrive in 3rd cycle

6
Assuming D/D/1 queuing regime, find the time of queue dissipation and the total vehicle
delay.

Solution:

1700 veh/h
Departure rate for all cycles: s   0.472 veh/s
3600 s/h

Max number of vehicles which may depart during any cycle = sg  0.472  20  9.44 veh.

In 1st cycle

 Arrivals = 15 veh (note that arrivals in any given cycle are spread out uniformly –
because of the D/D/1 assumption – over the length of the cycle)
 Departures = 9.44 veh
 Vehicles remaining at approach at end of 1st cycle = 15 – 10 = 5.56 veh

In 2nd cycle

 New arrivals = 8 veh  8 + 5.56 = 13.56 vehicles in 2nd cycle trying to pass
through the intersection.
 Departures = 9.44 veh
 Vehicles remaining at approach at end of 2nd cycle = 13.56 – 9.44 = 4.12 veh

In 3rd cycle

 New arrivals = 4 veh  4 + 4.12 = 8.12 vehicles in 3rd cycle trying to pass
through the intersection.
 Departures = 8.12 veh
 Vehicles remaining at approach at end of 3rd cycle = 0 veh (i.e. queue dissipates at
some time during 3rd cycle)

This process is shown in Figure 3 below.

7
tc

Figure 3. Cumulative arrival and departure curves for Case 2 (Figure 7.15 in MW)

Condition for queue dissipation in 3rd cycle is:

4.12  v3 40  t c   0.472t c


4 veh
Where v3   0.067 veh/s is the arrival rate in the 3rd cycle and tc is the time
60 s
elapsed after the beginning of effective green in the third cycle until queue dissipation.
4.12  0.067  40
4.12  v3 40  t c   0.472t c  t c   16.8 s
0.472  0.067

Therefore, the queue dissipates at t  60  60  40  16.8  176.8 s (i.e. 40 + 16.8 = 56.8 s


after the start of the third cycle).

8
Total vehicle delay is equal to the shaded area of the figure above.

In the 1st cycle, the vehicle delay is:

1 1
D1   60  15   20  9.44  355.6 veh-s
2 2

In the 2nd cycle, the vehicle delay is:

 60  15  23  40  9.44   20  9.44  18.88  479.2 veh-s


1 1
D2 
2 2

In the 3rd cycle, the vehicle delay is:

 56.8  23  26.8  40  18.88   16.8  18.88  26.8  275.4 veh-s


1 1
D3 
2 2

(In the equation of D3, 26.8 is the cumulative arrivals of vehicles 56.8 s after the start of
the third cycle, i.e. at the time of queue dissipation, computed as: 23 + 0.067 × 56.8 =
26.8).

Therefore, total vehicle delay over all 3 cycles is:


Dt  D1  D2  D3  355.6  479.2  275.4  1110.2 veh-s

You might also like