Republic of Iraq
Ministry of Higher Education and Scientific Research
Al-Mustansiriya University
College of Engineering
Highway and Transportation Engineering Department
Modeling of Traffic Queues and Delay at
CBD in Sulaymaniyah City
A THESIS SUBMITTED TO
THE HIGHWAY AND TRANSPORTATION ENGINEERING DEPARTMENT
COLLEGE OF ENGINEERING AL-MUSTANSIRIYA UNIVERSITY
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF MASTER OF SCIENCE IN
HIGHWAY AND TRANSPORTATION ENGINEERING
By:
SAAD MOHSEN KHALIL
(B.Sc. 2010/ Highway and Transportation Engineering)
Supervised by
Asst. Prof. Dr. Gandhi G. Sofia
Dhu al-Qi'dah, 1434
Dr. Ali Jabbar Kadhim
September, 2013
‫ِ‬
‫ِ‬
‫ِ‬
‫ِ‬
‫الرِحيِم‬
‫من‬
‫ح‬
‫الر‬
‫اهلل‬
‫م‬
‫س‬
‫ب‬
‫َّ‬
‫َّ‬
‫ْ‬
‫ْ‬
‫ينِآ َمنهواِمنْ هِْكِ َو َاَّلي َِنِ ُآوتهواِ‬
‫اّللِ َاَّل َِ‬
‫يَ ْرفَعِِ َ هِ‬
‫ونِخَبِيرِِ‬
‫اّللِب َماِتَ ْع َمله َِ‬
‫الْع َِْلِ َد َر َجاتِِ َو َ هِ‬
‫الع ِظيم‬
‫ص َ‬
‫دق اهللُ َ‬
‫َ‬
‫(المجادلة‪)11-‬‬
Dedication
To
My Family
CERTIFICATE OF THE EXAMINING COMMITTEE
We certify that we have read this thesis entitled “Modeling of Traffic Queues and Delay
at CBD in Sulaymaniyah City ” and as an examining committee, examined the student
(Saad Mohsen Khalil) in its content and in what is connected with it, and that in our opinion
it meets the standard as a thesis for the Degree of Master of Science in Highway and
Transportation Engineering.
Signature:
Name: Asst. Prof. Dr. Asma Th. Ibraheem
(Chairman)
Date: /11 / 2013
Signature:
Name: Asst. Prof. Dr. Zainab A. AL-Kaissi
(Member)
Date: /11 / 2013
Signature:
Name: Dr. Mohammed B. Abduljabar
(Member)
Date: /11 / 2013
Signature:
Asst. Prof Dr. Gandhi G. Sofia
(Supervisor)
Date: /11 / 2013
Signature:
Dr. Ali J. Kadhim
(Supervisor)
Date: /11 / 2013
Approved by the Dean of College of Engineering
Signature:
Assist. Prof. Dr. Jamal Saeed Abd AL-Amier
Date: /11 / 2013
Supervisor Certificate
We certify that the preparation of this thesis titled (Modeling of Traffic Queues
and Delay at CBD in Sulaymaniyah City) has been prepared by “Saad Mohsen
Khalil " under our supervision at Al-Mustansiriya University college of Eng.
Highway and Trans. Eng. Dept., in partial fulfillment of the requirement for the
degree of Master of Science in Highway and Transportation Engineering.
Signature:
Name: Asst. Prof. Dr. Gandhi G. Sofia
Date:
/ 9 / 2013
Signature:
Name: Dr. Ali Jabbar Kadhim
Date:
/ 9 / 2013
In view of the available recommendations; I found this thesis adequate for debate
by the examining Committee
Signature:
Name Dr. Ali Jabbar Kadhim
(Head of the Highway and Transportation Department)
Al-Mustansiriya University
Date:
/ 9 / 2013
Acknowledgements
First and foremost, all thanks to GOD for giving me the strength,
determination, and courage to implement this goal. All wisdom and knowledge
comes from his almighty, and I could not have reached this stage without his
divine guidance.
Cordial thanks are due to my supervisors Asst. prof. Dr. Gandhi G. Sofia
and Dr. Ali J. Kadhim for their guidance, patience, and encouragement through
the preparation of this study.
Special thanks are devoted to Traffic Directorate of Sulaymaniyah
Governorate for helping me during this research work.
Thanks are presented to Eng. Ghadah Ghassan, Eng. Mays Rabih and
Asst. Lecturer Israa Saeed at Al-Mustansiriya University for their interest and
support.
Finally, I would like to thank and acknowledge my family. They have always
supported everything I have ever attempted; each gives me the time and space I
needed.
I
Abstract
Abstract
Sulaymaniyah city is suffering from congestion, vehicle delays on most
streets and intersections especially during peak hours, causing a lot of traffic
problems and a negative impact on economy of the country. The increasing
number of vehicles and limited new road construction will increase the problem
in the network. The correct prediction of delay and queue length become
important to make the most efficient use of the existing road system by providing
better traffic operation and control.
The objective of this study is the evaluation of traffic performance at
selected network, through the evaluation of intersections performance and
development of delay time and queue length models with reference to traffic
prevailing conditions at Sulaymaniyah city. This is followed by suggestions of
some improvement proposals including changing timing plan, geometric
improvement and changing intersection type.
To carry out the above objective, traffic data is collected at the network in
CBD area in Sulaymaniyah city. Video camera is used to collect field data because
it provides permanent record of data with minimum manpower. These data are
abstracted from video records using EVENT program, and processed by prepared
EXCEL sheets, while, the spot speed data for each entire link in the network are
collected using pavement marking method.
SYNCHRO/SIMTRAFFIC software V.20 is used for evaluation and
analysis of intersections, best timing plan and coordination. Suggested proposals
are evaluated for the target year 2017 by using mentioned software.
Regression models, developed to estimate delay and queue length, show
good correlation with field values. The ratio of total width of exit roadway of
intersection for traffic departing straight forward to total width of lane groups
II
Abstract
departing to the same exit roadway at
the same phase at stop line (We/Ws)
appear to affect the delay. The ratio has a negative effect on delay time when the
ratio is less than 1.0.
Delay is estimated by HCM 2000 delay model and compared with field
delay. Based on results obtained from analysis, HCM 2000 delay model has been
modified to represent the field delay. It has been observed that, HCM 2000 delay
model consistently over estimates delay at degree of saturation more than 1.0. It
has been suggested from the analysis that, theoretical HCM 2000 delay model
may be reduced by 81 % to better reflect field conditions. Recommended delay
model shows good correlation with field data within the limits of this study.
Improvement strategies applied to the traffic flow in the study area of
Sulaymaniyah city, through the application of the calibrated software
(SYNCHRO/ SimTraffic). The improvement strategies divides into two parts.
The first part includes Signal Timing Optimization and Coordination. The second
part includes the geometric improvement. This improvement has very good effect
on the measure of effectiveness (MOE) for the study network. Finally, the
measure of effectiveness obtained from SYNCHRO/ SimTraffic optimization
runs after the last improvement strategy for the target year 2017.
III
L ist of Contents
L ist of Contents
Acknowledgement ................................................................................................ I
Abstract ................................................................................................................II
List of Contents ................................................................................................. IX
List of Tables ...................................................................................................VIII
List of Figures .....................................................................................................X
List of Notations .............................................................................................. XII
Chapter One
Introduction
1.1 General .......................................................................................................... 1
1.2 Sulaymaniyah city...........................................................................................3
1.3 Objectives of the Study ................................................................................. 4
1.4 Research Methodology .................................................................................. 4
1.5 Scope of the Work ......................................................................................... 4
Chapter T wo
L iterature Review
2.1 General............................................................................................................6
2.2 Urban Streets ……………………………………………………… ……….6
2.3 Intersections....................................................................................................7
2.3.1 Signalized Intersections...............................................................................9
2.3.1.1 Signalized Intersection Flow Characteristics .......................................10
2.3.1.2 Performance Measures...........................................................................11
2.3.1.3 Discharge Headway, Lost time, and Saturation Flow............................11
2.3.1.4 Capacity of Signalized Intersections.....................................................14
2.3.1.5 Level of service of Signalized Intersections.........................................18
IV
L ist of Contents
2.3.1.6 Delay at Signalized intersection..........................................................20
2.3.1.6.1 Development of Delay Models........................................................22
2.3.1.7 Queue at Signalized Intersection.........................................................30
2.3.1.7.1 Existing Queue Models...................................................................34
2.3.2 Two Way Stop Controlled intersection (TWSC intersection) ….............39
2.3.2.1 Delay at Two Way Stop Controlled intersection ..................................39
2.4 Calibration of The Traffic Models ………...…………………………....… 41
2.5 Local Traffic Improvement Studies…………………………………………42
2.5.1 Mosul City……………………………………………………………….42
2.5.2 Baghdad City…………………………………………………………….43
Chapter T hree
M ethodology and Data Collection
3.1 General ........................................................................................................ 44
3.2 Study Area Description ............................................................................... 44
3.3 Data Collection Methodology ..................................................................... 47
3.4 Methods of data collection............................................................................49
3.4.1 Manual methods.....................................................................................49
3.4.2 Video Recording Method.......................................................................49
3.5 Geometric Data ............................................................................................50
3.6 Traffic Data ..................................................................................................50
3.6.1 Traffic Volume Data ............................................................................ 50
3.7 Speed Data ................................................................................................... 54
3.8 Saturation Flow Data ................................................................................... 56
3.9 Traffic Signal Data ...................................................................................... 56
3.10 Minimum Green Time for Pedestrian ....................................................... 57
3.11 Data Abstraction ....................................................................................... 58
V
L ist of Contents
3.12 Queue Length .............................................................................................60
3.13 Data Processing ......................................................................................... 60
3.14 Delay Measurement.................................................................................... 61
3.15 Software Used for Data Analysis............................................................... 64
Chapter Four
Data Presentation and A nalysis
4.1 General ........................................................................................................ 67
4.2 Evaluation of Existing Traffic Flow .............................................................67
4.3 SYNCHRO Model Calibration .................................................................... 71
4.4 Development of Statistical Models for Delay and Queue length .................75
4.4.1 Process of Models Building....................................................................75
4.4.2 Identification of Dependent and Predictor Variables..............................75
4.4.3 Data Analysis..........................................................................................76
4.4.3.1 Selecting Sample Size……………………………………………..76
4.4.3.2 Scatter Plots.....................................................................................76
4.4.3.3 Outliers............................................................................................77
4.4.3.4 Testing of Normality.......................................................................78
4.4.3.5 Multicollinearity.............................................................................80
4.4.4 Regression Modeling.............................................................................82
4.4.4.1 Stepwise Regression Procedure......................................................83
4.4.4.2 Developed Models..........................................................................83
4.4.4.2.1 Delay model.............................................................................84
4.4.4.2.2 Left turn Queue Model............................................................85
4.4.4.2.3 Through Queue Model............................................................86
4.5 Models Validation.........................................................................................87
4.6 Models Analysis............................................................................................90
VI
L ist of Contents
4.7 Compression of Observed Delay with That HCM2000 Delay.................. 91
4.8 Improvements ...............................................................................................93
4.8.1 Signal Timing Optimization and Coordination ........................................ 93
4.8.2 Geometric Improvement .......................................................................... 97
4.9 Forecasted Network ....................................................................................101
Chapter Five
Conclusions and R ecommendations
5.1 Conclusions ............................................................................................... 103
5.2 Recommendations ..................................................................................... 104
5.3 Recommendations for Further Research ....................................................104
R eferences
References ........................................................................................................106
A ppendices
A Sample of Spot Speed Data..............................................................................A
B Work Sheet for Field Measurement of Delay...................................................B
C Sample of EVENT Program Output ................................................................C
D Sample of SYNCHRO Output .........................................................................D
E Traffic and Geometric Condition of Signalized intersections …………....…..E
VII
List of Tables
List of Tables
2-1
Principal Factors Affecting Saturation Flow ……….…………………...18
2-2
Level of Service Criteria for Signalized Intersections ….………………19
2-3
Some Traffic Improvement Studies in Baghdad City Sub-networks….....43
3-1
Geometric Features for the Selected Intersections …………...…………52
3-2
Recommended Trap Length ………………………………..…………..56
3-3
Phase Order, Phase and Cycle Length for Signalized Intersections …...57
3-4
Average Observed Queue Length ………………………..…………….60
3-5
Summary of Abstracted Data ………………………………..…………60
3-6 Acceleration/Deceleration Delay Correction Factor (CF) – second…….…64
4-1 Total Delay and Level of Service for Isolated Intersections Produced by
SYNCHRO for Existing Condition ………………..…......................................70
4-2 Comparison of Actual and SYNCHRO Delay Prediction for Signalized
Intersections………………………………………………………………..…...73
4-3 Case Processing Summary for Outliers …………………..…………….....77
4-4 K-S test and (S-W) test Results.....................................................................79
4-5 Correlation Coefficient Matrix for Delay Model.…………..….……........81
4-6 Correlation
Coefficient
Matrix for
Queue
Length of Left
Turn
Movement…………………………..……………………………………....…81
4-7
Correlation
Coefficient
Matrix
for
Queue
Length
of
Through
Movement …………………………………………………………………….82
4-8 Stepwise Regression Models Summary for Delay Model..……….….…...84
4-9 Data Range and Statistical Characteristics for Delay Model ……...…….85
4-10 Stepwise Regression Models Summary for Queue Length of Left Turn
Movements ...……………………………………………………………….….85
VIII
List of Tables
4-11 Data Range and Statistical Characteristics for Queue Length Model of Left
Turn Movement ………….……………………………………………………86
4-12 Stepwise Regression Models Summary for Queue of Through Turn
Movements …………..…………………………………………………….…..86
4-13 Data Range and Statistical Characteristics for Queue Length Model of
Through Movement …………………...……………………………………….87
4-14
Regression Results for the Three Models ………………………..…….88
4-15 ANOVA Test for the Regression Model ……………..………………….92
4-16
Measure of Effectiveness for the Improved Intersections with the
Optimum Cycle Length Produced by SYNCHRO ……………………………96
4-17 Measure of Effectiveness for the Improved Intersections
Produced by SYNCHRO (Geometric Improvement(………….………………98
4-18 Measure
of Effectiveness
Produced by SYNCHRO after Last
Improvement Strategy for the Base Year 2010, and Target Year 2017…..….102
IX
List of Figures
List of Figures
1-1
The Study Area at Sulaymaniyah City …………………………...…..… 3
1-2
Research Methodology …………………………………………………..5
2-1
Traffic Control Type
2-2
Recommended Traffic Control Type with Different Heavy Vehicles
...………………………………...………………8
Percentage ………………………………………………………………..8
2-3
Fundamental Attributes of Flow at Signalized Intersection…….....……10
2-4
Condition at Traffic Interruption ……………………………………….12
2-5
Headways at Traffic Interruption ……………………………………....13
2-6
Saturation Flow and the Related Signal Timing Parameters …………..17
2-7
Illustration of Delay Measures …………..……………………………..21
2-8
Queuing Diagram for Signalized Intersection …………………………32
2-9
Graphical Representation of the queue accumulation and discharge
process………………………………………………………………………….36
3-1
The Study Area in Sulaymaniyah City …….…………………………..45
3-2
Palace Intersection Layout …………………………..…………………46
3-3
Sara Intersection Layout ………………………………………………46
3-4
Great Mosque Intersection Layout ………………….…………………47
3-5
Traffic and Data Collection Procedure ……………..…………………48
3-6
Hourly Traffic Volume for Average Typical Weekdays ……………....53
3-7
Average Spot Speed for Each Link in the Study Area…….……….…..55
4-1
Network Layout as Printout by SYNCHRO ……………………...…….68
4-2
Lane Settings by SYNCHRO …………………………………….…….69
4-3
SYNCHRO Average Delay Values versus Field Delay Values for All
Approach ……………………………………………………….…...…………74
X
List of Figures
4-4
SYNCHRO Delay Values versus Field Delay Values after Calibrating the
Basic saturation flow rate for All Approach…………………...………………74
4-5
Observed versus Predicted Delay……………………………………….89
4-6
Observed versus Predicted Queue Length for the Model of Left Turn
Movement …………………………………………………………………...…89
4-7
Observed versus Predicted Queue Length for the Models of Through
Movement …………………………………………………………….………..90
4-8
HCM Delay versus Field Delay …………………….…………………..92
4-9
Optimization Steps in SYNCHRO ……………………………………...94
4-10 Improvement Proposal for Sara Intersection……….……………………100
4-11 Improvement Proposal for Great Mosque Intersection…..……………..100
4-12 Improvement Proposal for Palace Intersection…………...……………..101
XI
List of Notations
List of Notations
adj
Adjustment term for model
AWSC
All Way Stop Controlled
AR
Annual rate of traffic increase (%)
b1, b2
Regression coefficients
c
Adjusted lane group capacity (veh/hr)
C
Cycle length (sec)
CBD
Central Business District
CCG
Canadian Capacity Guide
ci
Capacity of lane group I (vph)
CI
Controller interface
CL
Lane group capacity per lane (veh/hr)
cm,x
Capacity of movement x (veh/hr)
d
Average overall delay per vehicle (sec)
d1
Uniform delay (sec/veh)
d2
Incremental, or random, delay (sec/veh)
d3
Residual demand delay(sec/veh)
dM
Average delay per vehicle(sec)
do
Overflow delay (sec)
du
Uniform delay (sec)
fp
Progression adjustment factor
FVS
Fraction of vehicles stopping
g
Effective green time (sec)
G
Green time (sec)
G(t)
The mean queue length at time t (veh)
XII
List of Notations
g/C
Green ratio
Ge
Effective green signal duration (sec)
gi
Effective green time for lane group i (sec)
GIS
A geographic information system
Gp
Minimum green time (sec)
h
Saturation flow rate headway (sec)
HCM
Highway Capacity Manual
I
Interval between vehicle-in-queue counts (sec)
I
Upstream filtering/metering adjustment factor
k
Incremental delay factor
KB
Second-term adjustment factor related to early arrivals
L
The segment length(m),
l1
Start-up lost time (sec)
l2
The clearance lost time (sec)
lamda
Effective green ratio
LOS
Level of service
Lp
Crosswalk length (m),
MLR
Multiple linear regression
MOE
Measure of effectiveness
MSE
Mean square error
n
Traffic analysis period (year)
Nped
Number of pedestrians crossing during an interval (p) in (sec)
P
Proportion of vehicles arriving during the green interval
PF
Adjustment factor for the effect of the quality of progression in
coordinated systems
PF2
Adjustment factor for effects of progression
pnmv
Percentage NMV in traffic
XIII
List of Notations
q
Arrival flow rate (veh/hr)
q
Average number of vehicles in queue (veh)
Q
Average queue length (veh)
Q
Maximum distance in vehicles over which queue extends from stop
line on average signal cycle (veh)
q
Vehicle arrival rate (PCU/sec)
Q1
First-term queued vehicles (veh),
Q1CCG
Estimates of average queue reach (veh)
Q2
Second-term queued vehicles (veh)
Q2CCG
Maximum queue reach during the congestion period (vehicles)
Qb
Initial queue at the start of period T (veh)
QbL
Initial queue at start of analysis period (veh)
QM
Maximum queue length (veh)
QQ
Average queue length while queue is present (veh)
r
Effective red time (sec)
R2
Correlation coefficient
s
Mean service rate (veh/hr)
s
Saturation flow for lane group i (vph)
Si
The measured spot speed of vehicle (km/hr)
Sp
Average speed of pedestrians (m/sec)
T
Duration of analysis period (h),
t
Duration of unmet demand in T (h),
T
Evaluation time (hr)
T
Length of analysis period (hr)
Tc
Length of cycle (sec)
te
Evaluation time (minutes)
TFF
Traffic forecast factor
XIV
List of Notations
ti
Incremental headway for the ith vehicle (sec)
ti
The time required for vehicle (i) to transverse the section (sec)
tQ
Time duration of queue (sec)
TWSC
Two-way stop- controlled intersection
u
Delay parameter
v
Arrival rate (veh/hr)
v
Mean arrival rate (veh/hr)
Viq
Sum of vehicle-in-queue counts, vehicle
VL
Lane group flow rate per lane (veh/hr),
vphg
Vehicles per hour of effective green time
Vstop
Volume counts of stopping vehicle
Vtot
Volume counts of total vehicles
vx
Flow rate for movement x (veh/hr),
w
Width of lane group at stop line (m)
WE
Effective cross walk width (m)
We
Total width of exit roadway of intersection (m)
Ws
Total width of lane groups departing to the same exit roadway (m)
x
Degree of saturation
X
Volume to capacity ratio of lane group
XL
Ratio of flow rate to capacity (vL/cL ratio).
Y
Change interval (sec)
β0
Constant term in the model
β1
Coefficient corresponding to q
β2
Coefficient corresponding to c
β3
Coefficient corresponding to x
β4
Coefficient corresponding to lamda
β5
Coefficient corresponding to pnmv
XV
List of Notations
Δ
Minimal time distance between vehicles (sec)
λ
Arrival rate (veh/sec)
λ
Proportion of the cycle that is effective green (g/C)
ρ
Flow ratio (-)
σ2µ
Variance of service time (sec)
µ
Service rate (veh/sec)
XVI
INTRODUCTION
Chapter One
Introduction
Chapter One
Introduction
1.1 General
The continuous traffic growth through developed areas and the difficulties
in building infrastructure have caused many traffic problems. In 1999, the number
of registered vehicles in Sulaymaniyah city was (32468) vehicles. This number
increased to (101,878) vehicles in 2006, corresponding to an annual increment in
the number of vehicles of 17.75 %. After 2006, the growth rate becomes 3.4 %.
(Traffic directorate of Sulaymaniyah governorate, 2013). Therefore, a need for
traffic engineering studies for careful monitoring of operating condition existing
highway facilities, developing a new model for such local conditions, evaluation
of the traffic flow quality are essential for developing improved strategies to
mange the increase in traffic demand.
Over the past years, the quality and accuracy of the simulation models have
been improved. In addition, new concepts and new technologies continue to add
to this field. Simulation models have different characteristics: static or dynamic,
deterministic or stochastic, microscopic or macroscopic. Each simulation model
was developed with different backgrounds and different algorithms. Each model
has its positive and negative points and it is not ideal for every situation.
Therefore, the analyst should select the program that satisfies the project
objectives and the prevailing traffic conditions for the selected study area.
Usually, little information is provided about the default parameters
embedded in the simulation models such as their source, their appropriateness and
1
Chapter One
Introduction
how they should be modified under different traffic conditions. Therefore, the
analyst should ensure that appropriate changes to these parameters are made based
on the field measured traffic data and not only on the engineering judgment
(Turley, 2007).
Before using any analytical or simulation model, there is a need for
calibration of the model in order to obtain a close match between the simulated
results and the field measurements. The parameters or factors that can be
calibrated vary from one model to another. Thus, when the analyst selects a
model; she/he should know which parameters or factors might be used as a yard
stick in the calibration process and which measures of effectiveness (MOEs) must
be considered for such calibration.
Usually, the MOEs are used for calibration including intersection delay,
number of stopped vehicles, platoon dispersion and average or maximum queue
length.
For many years, measurement of the level of performance of signalized
intersections has primarily focused solely on vehicle delay. Beside delay, other
performance measures such as the number of vehicle stops and the spatial extent
of queues on intersection approaches have also been found to play an important
role in the performance evaluation of signalized intersections of developing
countries, where most of the times the intersection remains near or at
oversaturated condition. Hence, stop and go situation is very frequent at
intersections. These measures not only relate to the level of service provided to
the drivers, but also to the level of fuel consumption and air pollution generated
by the vehicles traversing the signalized intersections. In particular, while vehicle
stop estimates play an important role in determining vehicle fuel consumption and
emissions on intersection approaches, queue length estimates are important not
2
Chapter One
Introduction
only for the design of pocket lanes, but also to ensure that traffic signal operations
do not result in vehicle queues that spillback onto upstream intersections.
1.2 Sulaymaniyah City
Sulaymaniyah is a city in Kurdistan Region of Iraq. It lies at 45˚, 26ˋ
longitude and 35˚, 33ˋ latitude, departed at about 355 km north east away from
Baghdad. The Azmer Range, Goyija Range and the Qaiwan Range in the north
east of the city , Baranan Mountain in the south and the Tasluja Hills in the west
surround it.
The study area, as located in the central business district area (CBD)
represents the core area of the city, as shown in Figure (1-1).
Figure (1-1): The Study Area at Sulaymaniyah City.(Google maps,2013)
3
Chapter One
Introduction
The new growth in Sulaymaniyah traffic congestion has been recognized as
a serious problem in the city, which affects the economy, travel time, driver
behavior, and causes discomfort to drivers and visitors. The number of vehicles
increases rapidly without considerable increase in the capacity of the road
network, which leads to increase delay and lower the level of service (LOS).
1.3 Objectives of the Study
The main objectives of this study are, as follows:
1. Evaluating the traffic performance at the studied area, and suggesting the
required traffic and/or geometric solutions to alleviate the congestion
problem at the selected network.
2. Developing statistical model for the prediction of vehicle delay at
signalized intersection.
3. Developing statistical model for the prediction of queue length at signalized
intersection.
4. Evaluating the adequacy of applying selected software on local traffic
conditions, and adjusting models parameters to be applicable with local
conditions.
1.4 The Study Methodology
The study methodology can be presented in Figure (1-2).
1.5 Scope of the Work
The general scope adopted in this study can be summarized, as follows:
1. Chapter one gives a brief idea about the present work.
2. Chapter two reviews the literature which is related to previous work in
delay and queue length. A description of the selected computer program
(SYNCHRO 8) is also presented.
4
Chapter One
Introduction
3. Chapter three describes the study area and the methodology of data
collection and abstraction to be used in the selected traffic program.
4. Chapter four illustrates results analysis, discussion and improvement
strategies.
5. Chapter
five
contains
conclusions,
recommendations,
recommendations for further researches.
Site Selection
Traffic and Geometric Data Collection
Using Video Recording, Satellite Image,
Manual Method and Moving Car
Data Abstraction and Processing Using
EVENT & Prepared EXCEL Sheets
Development of Delay
Model for Signalized
Intersection
Development of Queue
Model for Signalized
Intersection
Applying Software
for Simulation
Compare Observed
with Simulated Delay
Statistical Tests
Improvement
Alternatives
Draw
Recommendations
Figure (1-2): Research Methodology.
5
and
LITERATURE REVIEW
Chapter Two
Literature Review
Chapter Two
Literature Review
2.1 General
In this chapter, a description of urban streets, signalized intersections,
capacity and level of service at signalized intersection, development of delay
and queue models at signalized and unsignalized intersections are presented with
the researches related to this field of study.
2.2 Urban Streets
In the hierarchy of street transportation facilities, urban streets (including
arterials and collectors) are ranked between local streets and multilane suburban
and rural highways. The difference is determined principally by street function,
control conditions and the character and intensity of roadside development
(TRB, 2005).
Urban minor arterials carry large traffic volumes within and through urban
areas. The principal objective of an urban minor arterial should be mobility with
limited or restricted service to local development. Urban arterials are capable of
providing some access to abutting property. Such access service should,
however, be only incidental to the arterial’s primary function of serving major
traffic movements (AASHTO, 2004).
The degree of mobility provided by urban street is assessed in terms of
travel speed which affects highway capacity (HCM) (TRB, 2005), as follows:
6
Chapter Two
Literature Review
1. The street environment includes the through-traffic stream. The speed is
influenced by three main factors, the geometric characteristics of the facility, the
character of roadside activity and adjacent land uses.
2. The interaction among vehicles is determined by traffic density, the
proportion of trucks and buses and turning movements.
3. Traffic control (including signals and signs) forces a portion of all vehicles to
slow or stop. As a result, these factors also affect quality of service.
2.3 Intersections
An intersection is a general area where two or more highways join or cross
at grade, within which are included the roadway and roadside facilities for
traffic movements in that area. Each highway radiating from an intersection and
forming part of it, is an intersection leg. (SORB, 2005)
(Garber and Hoel, 2010) defined the intersection as an area, shared by two
or more roads whose main function is to provide for the change of route
direction. Intersections vary in complexity from a simple intersection, which has
only two roads crossing at a right angle to each other, and more complex
intersection, where three roads or more cross within the same area .
The intersection is required to control conflicting and merging streams of
traffic so that delay is minimized. This is achieved through choice of geometric
parameters that control and regulate the vehicle paths through the intersection.
These determine priority so that all movements take place with safety (Rogers,
2008).
Each intersection type aims at providing vehicle drivers with a road layout
that will minimize confusion. The need for flexibility dictates the choice of most
suitable junction type. The selection process requires the
economic,
environmental and operational effects of each proposed option to be evaluated
7
Chapter Two
Literature Review
(Rogers, 2008). The determination of intersection control type is illustrated in
Figure (2-1).
Figure (2-1): Traffic Control Type (Rogers, 2008).
Polus and Vlahos (2005) recommended ranges of major road one-way
traffic volumes for which the construction of the at-grade intersection control is
preferred, as shown in Figure (2-2).
Figure (2-2): Recommended Traffic Control Type with Different Heavy
Vehicles Percentage (Polus and Vlahos, 2005).
8
Chapter Two
Literature Review
2.3.1 Signalized Intersections
Traffic control signals that are properly designed, located, operated, and
maintained will have one or more of the following advantages (FHWA, 2009):
1. Providing the orderly movement of traffic.
2. Increase the traffic-handling capacity of the intersection if:
A. Proper physical layouts and control measures are used.
B. The signal operational parameters are reviewed and updated (if needed) on
a regular basis (as engineering judgment determines that significant traffic flow
and/or land use changes have occurred) to maximize the ability of the traffic
control signal to satisfy current traffic demands.
3. Reducing the frequency and severity of certain types of crashes, especially
right-angle collisions.
4. Using coordinated signals provide for continuous or nearly continuous
movement of traffic at a definite speed along a given route under favorable
conditions.
5. Using the interrupt heavy traffic at intervals to permit other traffic, vehicular
or pedestrian, to cross.
Traffic control signals, even when justified by traffic and roadway
conditions, can be ill-designed, ineffectively placed, improperly operated, or
poorly maintained. Improper or unjustified traffic control signals can result in
one or more of the following disadvantages (FHWA, 2009):
1. Excessive delay.
2. Excessive disobedience of the signal indications.
3. Increased use of less adequate routes as road users attempt to avoid the traffic
control signals.
4. Significant increases in the frequency of collisions (especially rear-end
collisions).
9
Chapter Two
Literature Review
2.3.1.1 Signalized Intersection Flow Characteristics
For a given approach at signalized intersection, three signal indications are
seen: green, yellow, and red. The indication may include a short period during
which all indications are red, referred to as an all-red interval, which with the
yellow indication forms the change and clearance interval between two green
phases. Figure )2-3( presents some fundamental attributes of flow at signalized
intersection. The diagram represents a simple situation of one-way approach to
signalized intersection having two phases in the cycle (HCM, 2000).
Figure (2-3): Fundamental Attributes of Flow at Signalized Intersection
(Source: HCM 2000)
10
Chapter Two
Literature Review
The diagram is divided into three parts. The first part shows a time-space plot of
vehicles on the northbound approach to the intersection. Intervals for the signal
cycle are indicated in the diagram. The second part repeats the timing interval,
and labels the various time intervals of interest with the symbols. The third part
is an idealized plot of flow rate past the stop line, indicating the saturation flow
rate.
2.3.1.2 Performance Measures
Common measures by which the performance of an intersection may be
evaluated include (1) delay, (2) stops and (3) queue length. Each of these may be
expressed as values, which represent totals or averages for the entire intersection
or for particular approaches or movements within the intersection. Averages are
often expressed on a per vehicle basis. Other measures, which have been used to
characterize the performance, are throughput and total travel time (Mcshane
and Roess, 1990).
Delay, specifically the control delay is the measure used in the signalized
intersection methodology of the 2000 HCM and the primary measure used in the
number of signalization optimization procedure. Performance measures are
critical part of all intersection design methodologies.
2.3.1.3 Discharge Headway, Lost time, and Saturation Flow
Lee and Chen (1986) studied the entering headways in small city
Lawrence, Kansas and six factors were examined. Entering headway values
from total of 1,899 traffic queues were recorded by using video camera
equipment. He found that:
1. Signal type has little influence on entering headway at signalized
intersections.
2. Time of the day (AM or PM) has little influence on entering headways.
11
Chapter Two
Literature Review
3. The inside lane of an approach has slightly lower entering headways than
does outside lane.
4. The entering headways at approaches with speed limits of 20 mph (32
kph) are significantly higher than those at approaches with higher speed
limits. For approaches with speed limits higher than 30 mph (48 kph), the
influence of speed limit on the headway is noticeable.
5. In general, streets that have higher speed limits and less roadside friction
have lower entering headway values.
Figure (2-4) illustrates a queue of vehicles at signalized intersection. When
the signal turns green, the queue begins to move into the intersection. The first
headway is defined as the time between the initiation of the green signal and the
first vehicles front bumper crossing the stop line. This first headway will be
comparatively long, as the first driver must see the light turn green, then react
and accelerate the vehicle into intersection. The second headway, measured as
the time between the first and second vehicles bumper's crossing the stop line,
will be somewhat smaller, as the second driver's reaction to the green overlaps
the first driver's reaction. The third reaction would be smaller than the second,
and so on.
Figure (2-4): Condition at Traffic Interruption (Source: HCM 2000)
12
Chapter Two
Literature Review
Eventually (usually between the fourth and sixth headway), vehicles
entering the intersection have fully accelerated by the time they reach the stop
line, and approximately equal headways would then be observed. Figure (2-5)
represents a plot of average headway of vehicles entering the intersection versus
the position of the vehicle in the queue.
Figure (2-5): Headways at Traffic Interruption (Source: HCM 2000).
The constant headway achieved once a stable moving queue is established,
is called the saturation headway. If it is assumed that, each vehicle entering the
intersection consumes h sec, then the number of vehicles that can enter the
intersection in a lane may be computed as (HCM,2000):
𝑠 =
3600
𝐸𝑞. (2 − 1)
ℎˉ
13
Chapter Two
Literature Review
where:
𝑠 = saturation flow rate (vehicle per hour of green per lane).
ℎˉ = saturation flow rate headway (sec).
The saturation flow rate(𝑠), is the number of vehicles that could enter the
intersection in a single lane if the signal is always green for that lane, and the
vehicles never stop.
Traffic stream at signalized intersection does stop periodically. When the
traffic stream starts, the first several vehicles consume more than h )sec/veh(.
Thus, sum of the incremental headways (above h sec/v) for the first several
vehicles is called start-up lost time (HCM,2000).
𝑬𝒒. (2 − 2)
𝒍𝟏 = ∑ 𝒕𝒊
𝒊
where:
𝑙1 = start-up lost time (sec).
𝑡𝑖 = incremental headway for the ith vehicle (sec).
The clearance lost time 𝑙2 , is the time between the last vehicle from one
approach entering the intersection and the initiation of the green signal for
conflicting movements. The capacity at signalized intersection is based on
saturation flow rate; the lost time and the signal timing (HCM, 2000).
2.3.1.4 Capacity of Signalized Intersections
The 2000 Highway Capacity Manual defines the capacity of facility as "
the maximum hourly rate at which persons or vehicles can reasonably be
expected to traverse point or uniform section of a lane or roadway during a
given time period under prevailing roadway, traffic and control conditions.
14
Chapter Two
Literature Review
The capacity of a traffic signal approach may be defined as the maximum
number of vehicles that can pass when the signal turns to green under the
prevailing traffic and weather conditions. Therefore, the capacity of a signal
controlled junction is limited by the capacities of the individual approaches to
the junction (Sofia, 1998).
The factors which affect capacity are (Al-Azzawi, 2003):
1-Physical and operation factor: Includes parking condition, width of
approaches, one -way or two-way operation and number of lanes.
2-Traffic characteristics: Consist of traffic signals (length of cycle time and
green to cycle time ratio for each approach) and marking of approach lanes.
3-Enviromental factors: Include degree of utilization of an individual approach,
variation of demand during the peak hour and the location of the intersection
within the metropolitan area.
4-Control measures: Involve turning movements and vehicles composition
Cars, buses and trucks).
5- Area type: the saturation flow in central business district (CBD), where
intersection geometry, pedestrian flow, and roadside friction are more
restrictive, is less than the saturation in other areas.
Capacity of signalized intersections is based on the concept of saturation
flow rate, which is defined as the maximum rate of flow that can pass through a
given lane group under prevailing traffic and roadway conditions, assuming that
a given lane group has 100% of real time available as an effective green time
and is expressed in units of vehicles per hour of effective green time (vphg)
(TRB, 2005).
Figure (2-6) shows that, the average rate of flow is lower during the first
few seconds (while vehicles are accelerating to normal running speed) and also
during the amber period (as some vehicles decide to stop and others continue to
15
Chapter Two
Literature Review
move on). It is convenient to replace the green and amber periods by an
"effective green" period, during which the flow is assumed to take place at the
saturation rate, combined with a "lost" time during which no flow takes place.
This is a useful concept because capacity is then directly proportional to
effective green time. The capacity of the lane group is (HCM,2000):
𝑐=𝑠∗
𝑔
𝐶
𝐸𝑞. (2 − 3)
where:
c : capacity of lane group (vph),
s: saturation flow for lane group (vph),
𝑔: effective green time for lane group ,
𝐶: cycle length.
16
Chapter Two
Literature Review
Figure (2-6): Saturation Flow and the Related Signal Timing Parameters
(Akcelik, 2009).
There are also several factors affecting saturation flow which can be
summarized in Table (2-1)
17
Chapter Two
Literature Review
Table (2-1): Principal Factors Affecting Saturation Flow (Al-Azzawi, 2003).
Factors
Element Affecting Saturation Flow
• Approach width
• Width of lanes
• Number of lanes
Geometric Condition
• Grade
• Turning Radius
• Length of turn bay
• Signal
timing
and
phasing
arrangements
• Peaking characteristics
Operating Condition
• Parking activities
• Bus stop operations
• Traffic composition
• Turning movements
Traffic Characteristics
• Pedestrian activity
• Weather
• Driver behavior
• Area population
Environmental and other Factors
• Roadway surface conditions
• Adjacent land use
2.3.1.5 Level of Service of Signalized Intersections
Level of service is defined in terms of the average total vehicle delay of all
movements through an intersection. Vehicle delay is a method of quantifying
several intangible factors, including driver discomfort, frustration and lost travel
time. Specifically, levels of service criteria are stated in terms of average delay
per vehicle during a specified time period (for example, the PM peak hour).
Vehicle delay is a complex measure based on many variables, including signal
phasing (i.e., progression of movements through the intersection), signal cycle
length and traffic volumes with respect to intersection capacity (TRB, 2000).
18
Chapter Two
Literature Review
Table (2-2) shows level of service criteria for signalized intersections, as
described in the Highway Capacity Manual (TRB, 2000).
Table (2-2): Level of Service Criteria for Signalized Intersections (TRB, 2000).
Level of
Service
A
B
C
Average Control Delay
(sec/veh)
D
>35-55
E
F
>55-80
>80
≤10
>10-20
>20-35
General Description (Signalized Intersections)
Free Flow
Stable Flow (slight delays)
Stable Flow (acceptable delays)
Approaching unstable flow (tolerable delay,
occasionally wait through more than one signal cycle
before proceeding)
Unstable Flow (intolerable delays)
Forced flow (jammed)
Highway Capacity Manual (2000) describes service quality in the
following terms:
• Speed and travel time: One of the most easily perceived measures of service
quality is speed, or travel time. On freeway, speed is very evident measure of
quality, while on surface streets systems, the driver is very sensitive to total
travel time.
• Density: Density is not often used in traffic analysis. Density describes the
proximity of vehicles to each other in the traffic stream and reflects ease of
maneuverability in the traffic stream, as well as psychological comfort of
drivers.
• Delay: Delay can be described in many ways. It represents excess or additional
travel time due to travel time of controls.
• Other measures: A variety of other measures are used to describe service
quality.
19
Chapter Two
Literature Review
In some cases, the used measures are not directly discernible to drivers or
passengers. Such measures generally rely upon volumes or flow rates.
2.3.1.6 Delay at Signalized Intersection
Delay is one of the key parameters that is utilized in the optimization of
traffic signal timings. Furthermore, delay is a key parameter in computing the
level of service provided to motorists at signalized intersections. Delay,
however, is a parameter that is difficult to estimate because it includes the delay
associated with decelerating to a stop, the stopped delay and the delay associated
with accelerating from a stop (Youn, 2000).
Delay can be quantified in many different ways. The most frequently used
forms of delay are defined below:
• Stopped time delay
• Approach delay
• Travel time delay
These delay measures can be quite different, depending on conditions at
the signalized intersection. Figure (2-7) shows the differences among stopped
time, approach and travel time delay for single vehicle traversing a signalized
intersection. The desired path of the vehicle is shown, as well as the actual
progress of the vehicle, which includes a stop at a red signal.
20
Chapter Two
Literature Review
Figure (2-7): Illustration of Delay Measures (McShane, 2004).
The delays experienced on the arterial signalized streets are mainly
associated with the intersections where conflicting movements are separated and
controlled by traffic signals. These traffic signals can operate under an isolated
control strategy, with the signal settings of each signal set independently of the
settings of adjacent signals. The delay is defined as the difference in travel time
when a vehicle is unaffected by the controlled intersection and when a vehicle is
affected by the controlled intersection. This delay includes lost time due to
deceleration and acceleration as well as stopped delay. Thus, intersection delay
estimates are directed toward estimating total delay or simply stopped delay
(Youn-Soo Kang, 2000).
21
Chapter Two
Literature Review
2.3.1.6.1 Development of Delay Models
Webster, (1958) developed a model for estimating the delay incurred by
motorists at under-saturated signalized intersections that become the basis for all
subsequent delay models. The mathematical form of the model is:
𝑑=
C (1−𝜆)2
2(1−𝜆𝑋)
+
X2
2v(1−X)
C
1
3
− 0.65 ( 2 ) [x 2+51]
v
Eq. (2 − 4)
where:
d = average overall delay per vehicle (seconds),
= proportion of the cycle that is effective green (g/C),
C = cycle length (seconds),
v = arrival rate (vehicles/hour),
c = capacity for lane group (vehicles/hour),
g = effective green time (seconds).
X= lane group v/c ratio or degree of saturation
In Equation (2-4), the first term represents the average delay to the
vehicles assuming uniform arrivals. The second term estimates the additional
delay due to the randomness of vehicle arrivals. This additional delay is
attributed to the probability that sudden surges in vehicle arrivals may cause the
temporary oversaturation of the signal operation. The third term, finally, is an
adjustment factor that is introduced in the model to correct the delay estimates
and that develops semi-empirically.
Following this classical work, numerous studies were conducted in the
field of estimating delays at signalized intersections. As a result of these studies,
a number of delay models based on deterministic queuing theory were proposed
to suite different field conditions. Among these, the most notable are the models
22
Chapter Two
Literature Review
developed by Miller (1963) and Akcelik (1981) in Australia, the models
developed for use in HCM 1985 (TRB, 1985), HCM 1994 (TRB, 1994) and
HCM 2000 (TRB, 2000) in United States and the model developed by Teply et
al. (1995) in Canada.
William (1977) presented simple, accurate technique for measuring
vehicular delay on the approach to a signalized intersection. Precise definitions
were established for four measures of performance: stopped delay, time-inqueue delay, approach delay and percentage of vehicles stopping. Approach
delay was selected as being the most representative of intersection efficiency.
The values; thus, obtained were statistically compared with true values from
time-lapse photography. The point sample, stopped delay procedure and the
percentage of vehicle stopping method were selected as the most suitable
methods for practical use and were performed in the field at three sites.
Correction factors were developed to allow the field results to more
accurately estimate the true values of stopped delay and percentage of vehicle
stopping. Interrelationships among the four measures of performance were
established so that approach delay can be estimated from a value of stopped
time.
Lin (1989) evaluated the reliability of the HCM 1984 procedure, based on
field data, and discussed needed modifications. Stopped delay was measured for
single lane movements at seven intersections. To compare the HCM estimates
with observed delays, the cycle lengths, green durations, yellow durations, and
saturation flow rates were also recorded using video cameras with built-in
stopwatches. The evaluation reveals that, the procedure tends to overestimate
stopped delay at reasonably well-timed signal operations. The discrepancies
between the HCM estimate and the observed delays can be very large even
when correct cycle length and green durations are used as inputs. Given actual
23
Chapter Two
Literature Review
cycle lengths and green durations, the procedure's ability to correctly identify the
level of service was found to be good. He found that large discrepancies
between HCM estimates and some of the observed delays could be reduced
significantly if no progression adjustment is applied to the estimates obtained
from HCM delay equation.
Teply (1989) examined two approaches for measuring delay-- a timespace diagram and queuing diagram -- and explained various problems related to
each. He concluded that, while delay cannot be measured precisely, it could be
useful engineering tool if it is calculated properly. Some of his findings were:
• Uniform delay formulas slightly underestimate overall delay because they
neglect a portion of the acceleration delay.
• The fact that uniform delay formulas and delay surveys based on queue counts
do not account for the rate of arrival at the end of the queue, and the rate of
discharge at the front of queue does not resort the resulting delay values.
• Delay surveys based on stopped queue count produce stopped delay values. In
situations with low volumes and short red interval, these techniques may
overestimate stopped delay to the point of exceeding overall delay values.
• The ratio between measured values of overall delay and stopped delay is not
constant.
Burrow (1989) and Akcelik (1990) presented generalized delay
expressions for existing time-dependent delay models due to their similar forms.
In general, they include a uniform delay term and an overflow or an incremental
delay term given by Equation (2-5).
where:
𝑑 = 𝑑 𝑢 + 𝑑𝑜
𝐸𝑞. (2 − 5)
d = average total delay (seconds),
24
Chapter Two
Literature Review
𝑑𝑢 = uniform delay (seconds) and,
𝑑𝑜 = overflow delay (seconds).
In HCM 2000, average delay per vehicle for a lane group is given by
equation (TRB, 1998):
𝑑 = 𝑑1 ∗ 𝑃𝐹 + 𝑑2 + 𝑑3
𝑑1 = 0.5C
𝑔 2
[1 − ]
𝐶
Eq. (2 − 6)
𝐶
𝑑2 = 900 𝑇 [(𝑋 − 1) + √(𝑋 − 1)2 +
𝑃𝐹 =
Eq. (2 − 7)
𝑔
[1 − 𝑀𝑖𝑛 (1, 𝑋) ]
8𝐾𝐼𝑋
]
𝑐𝑇
(1 − 𝑃)𝑓𝑝
where:
1−
Eq. (2 − 8)
Eq. (2 − 9)
𝑔
𝐶
d = average overall delay per vehicle (seconds/vehicles),
d1 = uniform delay (seconds/vehicles),
d2 = incremental, or random, delay (seconds/vehicles),
d3 = residual demand delay to account for over-saturation queues that may have
existed before the analysis period (seconds/vehicles),
PF = adjustment factor for the effect of the quality of progression in coordinated
systems,
C = traffic signal cycle time (seconds),
25
Chapter Two
Literature Review
g = effective green time for lane group (seconds),
X = volume to capacity ratio of lane group,
c = capacity of lane group (vehicles/hour),
k = incremental delay factor dependent on signal controller setting (0.50 for
pretimed signals; vary between 0.04 to 0.50 for actuated controllers),
I = upstream filtering/metering adjustment factor (1.0 for an isolated
intersection),
T = evaluation time (hours),
P = proportion of vehicles arriving during the green interval,
𝑓𝑝 = progression adjustment factor.
In cases where X > 1.0 for a 15-min period, the following period begins
with an initial queue. This initial queue is referred to as Qb, in vehicles. Qb is
observed at the start of the red period and excludes any vehicles in queue due to
random, cycle-by-cycle fluctuations in demand (overflow queue due to cycle
failures). When Qb ≠0, vehicles arriving during the analysis period will
experience an additional delay because of the presence of an initial queue. The
magnitude of this additional delay depends on several factors, including the size
of the initial queue, the length of the analysis period, and the volume to capacity
ratio during the analysis period. The initial queue delay term is designated which
is a term in the delay model given in Equation (2-10).
where:
𝑑3 =
1800 𝑄𝑏 (1+𝑢)𝑡
Eq. (2 − 10)
𝑐𝑇
Qb = initial queue at the start of period T (veh),
c = adjusted lane group capacity (veh/h),
26
Chapter Two
Literature Review
T = duration of analysis period (h),
t = duration of unmet demand in T (h), and
u = delay parameter.
Sofia (1998) studied the driver behavior and traffic flow in Mosul city. He
develops delay model using regression analysis. Equations (2-11),(2-12) show
the obtained regression models
d = 0.39 C – 39 g/C +22.6 v/c -6.8 (R2=0.93) for v/c ≤ 1.0
d = 0.44 C – 83.5 g/C +20.7 v/c -3.0 (R2=0.91) for v/c > 1.0
Eq. (2 − 11)
Eq. (2 − 12)
Shamsul and Asif (2007) modified Webster’s delay formula under non
lane based mixed road traffic condition by adding an empirical adjustment term
with the sum of first and second terms, which has been calibrated based on field
observations of delays. Hence, the general pattern of the modified Webster’s
delay formula will be, as follows:
𝑑=
𝐶(1−𝜆)2
2(1−𝜆𝑋)
+
𝑋2
2𝑞(1−𝑋)
+ 𝑎𝑑𝑗
Eq. (2 − 13)
𝑎𝑑𝑗 = 𝛽𝑜 + 𝛽1 ∗ 𝑞 + 𝛽2 ∗ 𝑐 + 𝛽3 ∗ 𝑥 + 𝛽4 ∗ 𝑙𝑎𝑚𝑑𝑎 + 𝛽5 ∗ 𝑝𝑛𝑚𝑣 Eq. (2 − 14)
where:
adj = Adjustment term for model.
q = Vehicle arrival rate (PCU/sec).
𝑐= Cycle length (seconds).
𝑥= Degree of saturation.
27
Chapter Two
Literature Review
lamda = Effective green ratio.
pnmv = Percentage NMV in traffic.
β0 = Constant term in the model.
β1 = Coefficient corresponding to q.
β2 = Coefficient corresponding to 𝑐.
β3 = Coefficient corresponding to 𝑥.
β4 = Coefficient corresponding to 𝑙𝑎𝑚𝑑𝑎.
β5 = Coefficient corresponding to 𝑝𝑛𝑚𝑣.
In Equation (2-13), the adjustment term (𝑎𝑑𝑗) is to be calculated from
delays observed in field. To make use of this term in the calibration, it must be
based on the field value of delay and calculated from the Equation (2-15).
𝑎𝑑𝑗 = 𝑑 −
𝐶(1−𝜆)2
2(1−𝜆𝑋)
+
𝑋2
Eq. (2 − 15)
2𝑞(1−𝑋)
In Equation (2-15), 𝑑 is the actual delay observed in field. If the left hand
side of the above equation is taken as the dependent variable, it needs to be
regressed against a set of independent variables.
Sierpiński(2007)
proposed model
for signalized intersections with
queueing theory analytical models usage. The model is a sum of the average
waiting time from the Clayton’s model and the average waiting time from the
M+/G+/1queueing model that uses the compressed queueing processes. The
proposed delay model has a form as Equation (2-16).
28
Chapter Two
Literature Review
𝑑𝑀 =
𝐺𝑒
]
𝑇𝑐
𝐺
2[1− 𝑒 𝜌]
𝑇𝑐
𝑇𝑐 [1−
+
1
2
𝜆.𝜎µ2 +𝜆[𝜇−Δ]
2(1−𝜌)
(1 − 𝜇Δ)
Eq. (2 − 16)
where:
𝑑𝑀 = average delay per vehicle(s).
𝑇𝑐 = length of cycle (s).
𝐺𝑒 = effective green signal duration (s).
= flow ratio (-).
= arrival rate (veh/s).
𝜇= service rate (veh/s).
𝜎𝜇2 = variance of service time (s).
Δ= minimal time distance between vehicles (s).
This formula makes a generalization of the Webster’s model (1958).
Webster used the M/D/1 queuing model and his formula has got a form, as
follow:
𝑑=
𝐺
2
𝑇𝑐 [1− 𝑇𝑒 ]
𝑐
𝐺
2[1− 𝑒 𝜌]
𝑇𝑐
+
𝜌2
2.𝜆(1−𝜌)
𝑇𝑐
− 0.65 [ 2 ] . 𝜌
𝜆
where:
𝑑 = average delay per vehicle (s).
𝑇𝑐 = length of cycle (s).
1
3
29
𝐺𝑒
]
𝑇𝑐
[2+5
Eq. (2 − 17)
Chapter Two
Literature Review
𝐺𝑒 = effective green signal duration (s).
 = flow ratio (-).
𝜆= arrival rate (veh/s).
2.3.1.7 Queue at Signalized Intersection:
Queue length is one of the most important performance measures of an
intersection. Using queue length, other arterial performance measures, such as
intersection delay, travel time and level of service can be estimated quite readily.
For traffic engineers, these performance measures to provide indicators for
identifying problems, thereby, helping decision makers improve the level of
service from individual intersections to the entire road network (HCM,2000)
When demand exceeds capacity at an approach to a signalized intersection
at the start of an effective green period, a queue forms. Because of the arrival of
vehicles during the red phases, some vehicles might not clear the intersection
during the given green phase. Back of queue refers to the number of vehicles
queued at an approach to a signalized intersection due to the arrival patterns of
vehicles and to vehicles unable to clear the intersection during a given green
phase (i.e., overflow). Most queuing theories are related to undersaturated
conditions. To predict the characteristics of a queuing system mathematically, it
is necessary to specify the following system characteristics and parameters
(HCM, 2000):
• Arrival pattern characteristics, including the average rate of arrival and the
statistical distribution of time between arrivals;
• Service facility characteristics, including service-time average rates and the
distribution and number of customers that can be served simultaneously or
the number of channels available; and
• Queue discipline characteristics, such as the means of selecting which
customer is next.
30
Chapter Two
Literature Review
In oversaturated queues, the arrival rate is higher than the service rate; in
undersaturated queues, the arrival rate is less than the service rate. The length of
an undersaturated queue can vary but will reach a steady state with the arrival of
vehicles. By contrast, the length of an oversaturated queue will never reach a
steady state, but will increase with the arrival of vehicles (HCM, 2000).
An undersaturated queue at a signalized intersection is shown in Figure
(2-8). The figure assumes queuing on one approach with two signal phases. In
each cycle, the arrival demand is less than the capacity of the approach; where
no vehicles wait longer than one cycle; and there is no overflow from one cycle
to the next. Figure (2-8) (a) specifies the arrival rate, v, in vehicles per hour and
is constant for the study period. The service rate, s, has two states: zero when the
signal is effectively red and up to saturation flow rate when the signal is
effectively green.
Note that, the service rate is equal to the saturation flow rate only when
there is a queue. Figure (2-8) (b) diagrams cumulative vehicles over time. The
horizontal line, v, in Figure (2-8) (a) becomes a solid sloping line in Figure (2-8)
(b), with the slope equal to the flow rate. Thus, the arrival rate goes through the
origin and slopes up to the right with a slope equal to the arrival rate.
Transferring the service rate from Figure (2-8) (a) to Figure (2-8) (b) creates a
different graph. During the red period, the service rate is zero, so the service is
shown as a horizontal line in the lower diagram. At the start of the green period,
a queue is present, and the service rate is equal to the saturation flow rate. This
forms a series of triangles, with the cumulative arrival line as the top side of
each triangle and the cumulative service line form the other two sides. Each
triangle represents one cycle length and can be analyzed to calculate the time
duration of the queue. It starts at the beginning of the red period and continues
31
Chapter Two
Literature Review
until the queue dissipates. Its value varies between the effective red time and the
cycle length, and it is computed using Equation (2-18) (HCM, 2000).
vtQ  s (tQ  r ) or tQ 
sr
s v 
where:
Eq. (2 − 18)
𝑡𝑄 = time duration of queue (s),
𝑣 = mean arrival rate (veh/h),
𝑠 = mean service rate (veh/h), and
𝑟 = effective red time (s).
Figure (2-8): Queuing Diagram for Signalized Intersection.
32
Chapter Two
Literature Review
The queue length is represented by the vertical distance through the
triangle. At the beginning of red, the queue length is zero and increases to its
maximum value at the end of the red period. Then, the queue length decreases
until the arrival line intersects the service line, when the queue length equals
zero. Three queue lengths can be derived using the relationship shown in Figure
(2-8): the maximum queue length, the average queue length while queue is
present and the average queue length. These are shown in Equations (2-19),
(2-20), and (2-21), respectively (HCM, 2000).
𝑄𝑀 =
𝑄𝑄 =
𝑄=
𝑣𝑟
Eq. (2 − 19)
3600
𝑣𝑟
Eq. (2 − 20)
7200
𝑄𝑀 𝑡𝑄
Eq. (2 − 21)
2𝐶
where
𝑄𝑀 = maximum queue length (veh),
𝑄𝑄 = average queue length while queue is present (veh),
Q = average queue length (veh),
𝑣 = mean arrival rate (veh/h),
r = effective red time (s),
𝐶 = cycle length (s), and
𝑡𝑄 = time duration of queue (s).
33
Chapter Two
Literature Review
2.3.1.7.1 Existing Queue Models:
There are three types existing models of queue length:
1. Catling (1977) adapted existing equations of classical queuing theory to
oversaturated traffic conditions and developed a comprehensive queue and
number of stops estimates that can represent the effects of intersections in a
time-dependent traffic assignment model. The time-dependent assignment
model requires division of a peak period into time intervals that are chosen so
that the mean arrival rate remains approximately constant within an interval. It
is; therefore, obviously important to determine the sensitivity of estimated queue
length on oversaturated traffic signal approaches to the method of dividing the
peak period.
Consider an approach to a traffic signal which has cycle time C. When the
length of queue is increasing with time increasing, Catling's formula for
computing the mean queue length is described by Equation (2-22).
where:
𝐺 (𝑡) = [(𝛽 2 + 2. 𝑥 2. 𝑐 2. 𝑡 2 . 𝛼. 𝑧)1/2 − 𝛽]/𝛼
G(t)= the mean queue length at time t (vehicles),
α= 2(c 𝑡 𝑧),
β= c t [c 𝑡 (1𝑥) 2 𝑧 x],
𝑐 = s g /C, capacity (vehicles/hour),
𝑧 = 0.55, g = effective green time (seconds),
𝑞 = average arrival rate of traffic (vehicles/hour),
𝑠 = saturation flow rate (vehicle/hour),
𝑥 = qc/gs, the degree of saturation.
34
Eq (2 − 22)
Chapter Two
Literature Review
In fact, when x < 1, the steady-state is represented by a limiting value of
G(t) as t . This limiting value is denoted by G1(x) and can be expressed by
Equation (2-23) :
𝐺1(𝑥 ) = lim 𝐺 (𝑡) =
.
𝐶𝑥 2
Eq (2 − 23)
1−𝑥
These expressions can be applied only for an interval with stationary mean
arrival rate and starting with zero queue length. If there is a different demand
during subsequent time intervals, the formulas cannot be applied since the
queue, formed at the end of the first interval, will cause extra delay to the
following arrivals.
2.The Canadian Capacity Guide for Signalized Intersections, published in
1995, contains evaluation criteria for estimating queue lengths. These criteria
recognize all three queue definitions: Uniform Average Queue Accumulation,
Uniform Maximum Queue Reach and Uniform Conservative Approach, as
illustrated in Figure (2-9). The Uniform Maximum Queue Reach is represented
as a liberal approach to determine the required storage space, and the Uniform
Conservative Approach is represented as a conservative approach. These two
parameters converge at high levels of saturation.
35
Chapter Two
Literature Review
Figure (2-9): Graphical Representation of the queue accumulation and
discharge process (Canadian Capacity Guide, 1995).
The CCG model deals with the random and overflow effects in a unique
manner. Most models treat the probability of spillover in a mathematically
rigorous manner, as shown in Equation (2-24) (Canadian Capacity Guide,
1995).
1
𝑃 (𝑛𝑜 𝑠𝑝𝑖𝑙𝑙𝑜𝑣𝑒𝑟) = (𝑠𝑝𝑖𝑙𝑙𝑜𝑣𝑒𝑟)
𝑃
Eq (2 − 24)
However, the CCG model formulates this probability using Equation (2-25).
𝑃 (𝑛𝑜 𝑠𝑝𝑖𝑙𝑙𝑜𝑣𝑒𝑟) = [1 − 𝑃 (𝑠𝑝𝑖𝑙𝑙𝑜𝑣𝑒𝑟)]2
36
Eq (2 − 25)
Chapter Two
Literature Review
This formulation, which expresses the probability of no spillover on either
of two consecutive cycles, is an appropriate surrogate for the explicit treatment
of the random and overflow phenomena. The method of the CCG model also
includes a separate procedure for treating oversaturated conditions. The
equations of queue length for undersaturated conditions and oversaturated
conditions in the CCG model are described by Equations (2-26) and (2-27)
respectively (Canadian Capacity Guide, 1995).
𝑄1𝐶𝐶𝐺 =
where:
𝑄2𝐶𝐶𝐺 =
𝑞𝐶
3600
[𝑡𝑒 (𝑞−𝑐)]
60
+ 𝑄1𝐶𝐶𝐺
Eq. (2 − 26)
Eq. (2 − 27)
𝑄1𝐶𝐶𝐺 = estimates of average queue reach (vehicles),
𝑞 = arrival flow rate (vehicle/hour),
𝐶 = cycle length (seconds),
Q2CCG = maximum queue reach during the congestion period (vehicles),
𝑡𝑒 = evaluation time (minutes),
c = capacity (vehicles/hour).
3. HCM 2000 Queue Model: The queue length definition used in this model is
the back of queue. A relationship for the back of queue is developed, as
described in the next sections. The back of queue is the number of vehicles that
are queued depending on arrival patterns of vehicles and vehicles that do not
clear the intersection during a given green phase (overflow). The model predicts
the average back of queue.
37
Chapter Two
Literature Review
The average back-of-queue measure is the basis to calculate percentile
back of queue. Equation (2-27) shows average back-of-queue characteristics at
signalized intersections.
𝑄 = 𝑄1 + 𝑄2
Eq. (2 − 27)
where:
𝑄 = maximum distance in vehicles over which queue extends from
stop line on average signal cycle (veh),
𝑄1 = first-term queued vehicles (veh), and
𝑄2 = second-term queued vehicles (veh).
The first term, 𝑄1, is the average back of queue, determined first by
assuming a uniform arrival pattern and then adjusting for the effects of
progression for a given lane group. The first term is calculated using Equation
(2-29).
𝑄1 = 𝑃𝐹2
𝑉𝐿 𝐶
3600
𝑔
[1 − ]
𝐶
𝑔
1 − [𝑀𝐼𝑁 (1, 𝑋𝐿 ) ]
𝐶
where:
𝑄1 = first-term queued vehicles (veh),
PF2 = adjustment factor for effects of progression,
𝑉𝐿 = lane group flow rate per lane (veh/h),
C = cycle length (s),
𝑔 = effective green time (s), and
XL = ratio of flow rate to capacity (vL/cL ratio).
38
Eq. (2 − 29)
Chapter Two
Literature Review
𝑄1 represents the number of vehicles that arrive during the red phases and
during the green phase until the queue has dissipated. The second term,𝑄2, is an
incremental term associated with randomness of flow and overflow queues that
may result because of temporary failures, which can occur even when demand is
below capacity. This value can be an approximate cycle overflow queue when
there is no initial queue at the start of the analysis period. Initial queue at the
start of the analysis period is also accounted for in the second term, 𝑄2.
Equation (2-30) is used to compute the second term of the average back of
queue.
𝑄2 = 0.25𝑐𝐿 𝑇 [(𝑥𝐿 − 1) + √(𝑥𝐿 − 1)2 +
8𝐾𝐵 𝑋𝐿 16𝐾𝐵 𝑄𝑏𝐿
+
]
(𝑐𝐿 𝑇)2
𝑐𝐿 𝑇
Eq. (2 − 30)
where:
𝑄2= second term of queued vehicles, estimate for average overflow queue
(veh);
𝑐𝐿 = lane group capacity per lane (veh/h);
T = length of analysis period (h);
𝑥𝐿 = vL/cL ratio;
𝐾𝐵 = second-term adjustment factor related to early arrivals;
𝑄𝑏𝐿 = initial queue at start of analysis period (veh); and
C = cycle length (s).
39
Chapter Two
Literature Review
2.3.2 Two-Way Stop- Controlled Intersection (TWSC intersection)
Unsignalized intersections are the most common intersection type.
Although their capacities may be lower than other intersection types, they do
play an important part in the control of traffic in a network. A poorly operating
unsignalized intersection may affect a signalized network or the operation of an
Intelligent Transportation System (Rod J. Troutbeck; Werner Brilon).
There are three type of unsignalized intersection (TWSC, AWSC, and
roundabout).TWSC intersections, in which stop signs are used to assign the
right-of-way, are the most prevalent type of intersections within the United
States and abroad. At TWSC intersections, the stop-controlled approaches are
referred to as the minor-street approaches and can be either public streets or
private driveways. The intersection approaches that are not controlled by stop
signs are referred to as the major-street approaches. A three-leg intersection is
considered to be a standard type of TWSC intersection as long as the single
minor-street approach (i.e., the stem of the T-configuration) is controlled by a
stop sign (HCM, 1998).
2.3.2.1 Delay at Two-Way Stop- Controlled Intersection
The delay experienced by a motorist is made up of a number of factors that
relate to control, geometrics, traffic, and incidents. Total delay is the difference
between the travel time actually experienced and the reference travel time that
would result during base conditions, in the absence of incident, control, traffic,
or geometric delay. Average control delay for any particular minor movement is
a function of the capacity of the approach and the degree of saturation. The
analytical model used to estimate control delay (Equation 2-31) assumes that,
the demand is less than capacity for the period of analysis. If the degree of
saturation is greater than about 0.9, average control delay is significantly
40
Chapter Two
Literature Review
affected by the length of the analysis period. In most cases, the recommended
analysis period is 15 min. If demand exceeds capacity during a 15-min period,
the delay results calculated by the procedure may not be accurate. In this case,
the period of analysis should be lengthened to include the period of
oversaturation (HCM, 2000).
3600
𝑣
[
][ 𝑥 ]
2
3600
𝑣𝑥
𝑣𝑥
𝑐𝑚.𝑥 𝑐𝑚.𝑥
√
𝑑=
+ 900𝑇 [
−1+ [
− 1] +
] + 5 Eq. (2 − 31)
𝑐𝑚.𝑥
𝑐𝑚.𝑥
𝑐𝑚.𝑥
450𝑇
where:
d = control delay (s/veh),
vx= flow rate for movement x (veh/h),
cm,x= capacity of movement x (veh/h), and
T = analysis time period (h) (T = 0.25 for a 15-min period).
A constant value of 5.0 seconds is used to reflect delay during deceleration.
2.4 Calibration of The Traffic Models
Calibration is the process of comparing model parameters with real-world
data to ensure that the model realistically represents the traffic environment. The
objective is to minimize the discrepancy between model results and
measurements or observations (TRB, 2005).
Microscopic simulation model must be calibrated first against field data or
against other validated analytical models to ensure accuracy. This crucial step of
calibration is unfortunately often neglected.
Sargeant and Christie (2002) compared observed values of intersection
queue lengths with the simulation results of SYNCHRO/SimTraffic 5.0 model.
41
Chapter Two
Literature Review
Adjustments were made to vehicle/driver behavior to reduce the differences
between observed and modeled queues. R-squared analyses were performed on
the maximum and average queues to obtain estimates of the percentage variation
explained by the simulation model. As both of these values indicate that, over 90
percent of the variation has been explained, it can be stated that the
SYNCHRO/SimTraffic model is good.
Al-Jaman (2007) successfully calibrated and validated the SimTraffic
model for the signalized intersections of Riyadh, Saudi Arabia. The calibrated
parameters are travel speed, turning speed, headway factor (a measure of
saturation flow rate), and driver type. The study discovered that, the intersection
approaches in Riyadh seem to have high saturation flow rates, which supports
the results from other studies indicating that, driving behavior in Riyadh is
aggressive.
2.5 Local Traffic Improvement Studies
The overviews of traffic improvement studies use computer program tools
and experimental studies. The type of network improvements might range from
simple striping or parking changes to complete versions of geometry and
signalization. This section provides a summary of the local evaluation
comparisons, and experimental studies of the computer programs in some of
Iraqi cities.
2.5.1 Mosul City
Sofia (1998) studied the driver behavior and traffic flow in Mosul city. He
used TRANSYT-7F program in performance analysis, evaluation, and
coordination of seventeen signalized intersections. These intersections form two
sub-networks, one in central business district (CBD) area, and the other in
suburban area.
42
Chapter Two
Literature Review
Two proposals were made in his study to the CBD network and four
proposals for the suburban network, after the development of statistical models
for the estimation of passenger car unit (PCU), saturation flow, and delay.
2.5.2 Baghdad City
Baghdad city is the capital of Iraq, where the governmental buildings,
ministries, colleges, shopping centers, and so on are located in different sectors
in the city. Each sector has its own traffic problems, and congestions, so several
studies have been adopted to alleviate these problems. Both Al-Karkh and AlRusafa sides were involved in these studies. Each study, subnetwork, and
software used is summarized in Table (2-3).
Table (2-3) Some Traffic Improvement Studies in Baghdad City Sub-networks
Researcher
Area Studied
Software Used
Al-Hummadawee
(2002)
Al-Adhamiya district
(from Bab-Al-Muadam to Ras AlHawash)
Al-Adhamiya sector
HCS-2000,and
TRANSYT-7F
Warid (2004)
Al-Karrada network
HCS-2000,and
TRANSYT-7F
Abdul-Ghani
(2004)
AL-Rusafa side (AL-Rashid,
AL-khulafa, and AL-Kifah Street)
CYCOPT,
TRANSYT-7F,
and HCS-2000
Al-Hemdany
(2003)
43
TRANSYT-7F
METHODOLOGY AND
DATA COLLECTION
Chapter Three
Methodology and Data Collection
Chapter Three
Methodology and Data Collection
3.1 General
The availability of highway transportation has provided several advantages
that contribute to high standards of living. However, several problems related to
the highway mode of transportation, exist. These problems include highway
related parking difficulties, congestion, and delay. To reduce the negative
impact of highways, it is necessary to adequately collect information that
describes the extent of the traffic problems and identifies their locations. This
chapter presents description of the study area (CBD) in Sulaymaniyah city. Brief
description of the methods used for data collection and abstraction is also
presented, with description of the computer programs used for traffic data
abstraction and analysis.
3.2 Study Area Description
The study area has three intersections within CBD area in Sulaymaniyah
city, as shown in Figure (3-1). A brief description of selected intersections is
presented below:
 Palace Intersection (1): It is four leg signalized intersection. Mawlawi
approach is one way for exit only. Other approaches consist of 2-3 lanes
of through, left and right movements. The Mamostayan3 approach is for
right turn only. The geometric layout is illustrated in Figure (3-2).
44
Chapter Three
Methodology and Data Collection
 Sara Intersection (2): It is five leg signalized intersection, three of which
are one way; Mawlawi approach consists of two lanes of left turn only.
Sabunkaran consists of three lanes of through and right turn movement
and Kawa approach consists of three lanes of through
and left turn
movement. Bekas approach is two lanes for exit only and Pirmered
approach is two ways which consists of lanes for U-turn movement only.
The geometric layout is illustrated in Figure (3-3).
 Great Mosque Intersection (3): It is five leg unsignalized intersection.
The Court approach consists of two lanes of right turn movement. Bekas
approach consists of two lanes of through movement. Other approaches
are for exit movement only. The geometric layout is illustrated in Figure
(3-4).
Figure (3-1): The Study Area in Sulaymaniyah City.
45
Chapter Three
Methodology and Data Collection
Figure
Sar
(3-3):
Figure (3-2): Palace Intersection Layout.
a
Int
ers
ecti
on
La
you
t.
Figure (3-3): Sara Intersection Layout.
46
Chapter Three
Methodology and Data Collection
Figure (3-4): Great Mosque Intersection Layout.
3.3 Data Collection
The data collection phase aims at assembling all data required to model
the traffic flow condition at the study area. The required data are collected in
January, February/2012at morning and evening peaks. All required traffic
data are collected during good weather, because adverse weather condition
may cause variation in the normal traffic flow pattern. Figure (3-5) shows the
Procedure for traffic and geometric data collection. The methods of data
collection are described in the following paragraphs.
47
Chapter Three
Methodology and Data Collection
From Video Records
&Field Observations
Peak Hour
Selection
Study Area
selection
Traffic Data
Collection
Field
Queue
Length
Actual Phase
Length and
Sequence
From
Video
Records
Traffic
Signal
control
Actual Cycle
Length and
Timing
Traffic &
Geometric
Data Survey
Traffic
Volume
Survey
Locations
Identification
From Manual
Method
Speed Data
From Manual
Method
Intersection
layout and
Approach
Dimensions
& Satellite Image
Geometric Data
collection
From Personal
Observations and
Field
Measurements
Figure (3-5): Traffic and Data Collection Procedure.
48
No. of Lanes
for each
approach
Chapter Three
Methodology and Data Collection
3.4 Methods of Data Collection
3.4.1 Manual Methods
The following data are collected manually from field.
1. Queue length: two observers have collected Queue length data during the
peak hours. The observation is made at some approaches since it is not
possible to cover the full queue lengths due to obstruction of line of sight
of video camera.
2. Geometric characteristics: Lane widths are measured at the stop line using
measuring tape to obtain the width. All approach slopes for all
intersections are obtained using the level instrument. Other data are
collected by observation (e.g., number of lane per approach, lane
designation and so on).
3. Spot speed: Spot speed data are also collected manually by two observers
using a stop watch, marker to mark the pavement, and tape or measuring
wheel for measuring section distance.
3.4.2 Video Recording Method
The video recording method allows for large number of events to be
recorded at the same time. It also has the advantage that any incident which
might affect the observed data will also be recorded. The incidents can be
reviewed at a later stage to resolve any apparent abnormalities in the data.
All other traffic data, which are not collected manually, are collected by
video camera based technique, which overcomes many of the difficulties in
traffic data collection at field.
49
Chapter Three
Methodology and Data Collection
The video camera is mounted at the roof of a building located nearby the
intersection and is focused to cover each leg of the intersection. Care is taken to
cover full queue formed on the approach and the camera should be hidden from
the drivers to avoid its effect on their behavior.
The digital video camera, which is used in data collection, is SONY
HANDYCAM, digital 8,990x digital zoom.
3.5 Geometric Data
Geometric data, like intersection spacing and right of way, have been
collected by using GIS tools in map measurements depending on the available
satellite images with resolution of 0.6m, updated to 2007 and available at
Sulaymaniyah municipality. Field measurements are conducted for approaches
width, number of lanes per approach at each intersection, lane width, and splitter
island width at all approaches. Field survey is used to obtain the geometric
features that cannot be drawn from satellite image due to unavailability of
updated one, and the main geometric features and layouts of the selected
intersections are shown in Table (3-1).
3.6 Traffic Data
The traffic data have been collected, as described below:
3.6.1 Traffic Volume Data
The traffic volume data are collected for seven days by video recording
technique for each intersection from 3:00 to 5:00 pm after several personal
observations, and pilot survey have been made in the study area. Also, personal
interviews are made with interested people like traffic policemen at the study
area and different road users in order to determine the peak hour period for
traffic data collection. The network average traffic volumes (veh/hr) are shown
in Figure (3-6) for typical weekdays.
50
Chapter Three
Methodology and Data Collection
Predicted traffic volume is defined as "the current (existing) traffic volume
multiplied by the ratio of future traffic volume to current traffic volume"
(SORB, 2005). This ratio is named "Traffic Forecast Factor (TFF)" and can be
calculated from Equation (3-1).
where:
𝑇𝐹𝐹 = (1 + 𝐴𝑅 )𝑛
𝐸𝑞. (3 − 1)
𝑇𝐹𝐹: Traffic forecast factor (composed growth factor)
𝐴𝑅: Annual rate of traffic increase (%).
𝑛: Traffic analysis period (year).
The annual growth rate of traffic volume increases in Sulaymaniyah city
which is estimated to be (3.4%) of growth collected from the census department
of Sulaymaniyah city for the previous 5 years (Traffic directorate of
Sulaymaniyah governorate, 2013).
51
Chapter Three
Methodology and Data Collection
Table (3-1): Geometric Features for the Selected Intersections.
Lane
Approach
Lane
Name
Name
Group
Palace
Salim
1
LT+UT
3.2
1
Two way
Salim
2
TH
3.2
1
Two way
Salim
3
RT
4.5
1
Two way
Baban
4
LT
2.4
1
Two way
Baban
5
TH
2.4
1
Two way
Baban
6
RT
2.4
1
Two way
Mamostayan3
7
RT
3.7
2
Two way
Mawlawi
1
LT
4.1
2
One way
Kawa
2
TH
3.2
2
One way
Kawa
3
LT
4.5
1
One way
Sabunkaran
4
TH
4.8
1
One way
Sabunkaran
5
RT
5.0
1
One way
Piramerd
6
LT+UT
2.35
2
Two way
Piramerd
7
RT
4.2
2
Two way
Bekas
1
TH
12.2
2
One way
Court
2
RT
10.5
1
One way
Sara
Great
Mosque
52
Movement
Width
(m)
No. of
One Way
Intersection
Lanes
or Two
Way
Chapter Three
Methodology and Data Collection
Figure (3-6): Hourly Traffic Volume for Average Typical Weekdays.
53
Chapter Three
Methodology and Data Collection
3.7 Speed Data
Speed measurement is carried out by video recording technique by
measuring time taken by vehicles to pass trap length. The travel time can be
converted to the speed, as illustrated in Figure (3-7), with a sample of speed
calculation in Appendix(A).The average spot speed can be computed from the
following Equation (Pignataro, 1973):
𝑆 = 3.6 ×
𝐿
𝐸𝑞. (3 − 2)
𝛴 𝑡𝑖 /𝑛
where:
𝑆 : the average measured spot speed of vehicle (km/hr),
𝐿 : the segment length(m),
𝑡𝑖 : the time required for vehicle (𝑖) to transverse the section (sec), and
𝑛: sample size.
The recommended trap length for calculating spot speed is shown in Table
(3-2). This method involves the use of transverse pavement markings, which are
placed at each end of the 25 m course at the midblock location and the observer
starts and stops the watch as the vehicle passes the markings.
54
Chapter Three
Methodology and Data Collection
Figure (3-7): Average Spot Speed for Each Link in the Study Area.
55
Chapter Three
Methodology and Data Collection
Table (3-2): Recommended Trap Length (Kennedy, N. kell J.H. and
Hombuger W.S.).
Recommended
Average speed of traffic stream (km/hr)
Trap Length
(m)
Below 40
25
Between 40 and 64
50
Over 64
75
3.8 Saturation Flow Data
Saturation flow rate is defined as the departure flow rate per lane at which
vehicles can pass through a signalized intersection. Average saturation flow rate
has been calculated– as recommended by HCM(2000) by Equation (3-3):
𝑠=
3600
𝐸𝑞. (3 − 3)
ℎˉ
where:
𝑠 = saturation flow rate (veh/h), and
ℎˉ = saturation headway (s).
3.9 Traffic Signal Data
For signalized intersection (Palace and Sara), the cycle length, phase
length, green time and red time are measured from video films, as shown in
Table (3-3).While, the value of lost time is taken as the default value (2.5 sec)
recommended by SYNCHRO software program.
56
Chapter Three
Methodology and Data Collection
Table (3-3): Phase order, Phase and Cycle Length for Signalized Intersection.
Intersection
Phase length (sec)
Palace
Cycle length
(sec)
300
178
118
Phase length (sec)
Sara
196
96
96
3.10 Minimum Green Time for Pedestrian
Pedestrian is one of the important elements which requires the attention by
the traffic engineer, particularly in urban and CBD locations. Moreover,
pedestrians also represent an element of sharp conflict with vehicular traffic
which results in high accident rates and traffic delay. Therefore, minimum green
time for pedestrian must be calculated and provided for each phase within each
intersection. This study adopted the Highway Capacity Manual 2000 to estimate
the minimum green time for all phases, as it can be found in the following
Equations (TRB, 2005):
𝐺𝑝 = 3.2 +
𝐺𝑝 = 3.2 +
where:
𝐿𝑝
𝑁𝑝𝑒𝑑
𝑆𝑝
+ (0.81
𝑆𝑝
+ ( 0.27 𝑁𝑝𝑒𝑑)
𝐿𝑝
𝑊𝐸
)
for 𝑊𝐸 > 3.0 m
for 𝑊𝐸 ≤ 3.0 m
57
𝐸𝑞 (3 − 4)
Chapter Three
Methodology and Data Collection
𝐺𝑝: minimum green time (sec),
𝐿𝑝 : crosswalk length (m),
𝑆𝑝 : average speed of pedestrians (m/sec),
𝑊𝐸 : effective cross walk width (m),
3.2 : pedestrian start-up time (sec), and
𝑁𝑝𝑒𝑑 : number of pedestrians crossing during an interval (p) in sec.
3.11 Data Abstraction
Data abstraction is based on sessions of 15 minute periods of recorded
data. The selection of a 15 minute period is based on these considerations:
1. To ensure that, the sample is sufficient to provide meaningful results.
2. The traffic flow should be effectively constant during the period without
incidents.
3. Minimization of observer fatigue and subsequent errors which might
result from continuously watching video recordings.
The video camera takes continuous picture of the traffic and the pictures are
recorded on a digital camera. The video recording approach has a number of
advantages:
1. It is unobtrusive and requires little labor power.
2. It produces permanent, complete record of the traffic scene.
3. Recording may be reanalyzed at any stage.
4. It provides account of each traffic event and maneuvers.
The disadvantage of video technique is the large amount of time effort
needed for data abstraction. The data extraction process is mainly achieved with
the aid of a computer program named EVENT, which was developed by
Al-Neami (2000). Many studies, used
58
this program
(Sofia, 1998),
(Al-
Chapter Three
Methodology and Data Collection
Azzawi, 2003),
(Khalaf, 2003), (Razzaq, 2005), (Al-Zubaidi, 2006), (Al-
Dulaimi, 2006(, stated that, EVENT program is a necessary and complementary
tool for traffic studies, which uses the video recording technique. It is worth
mentioning that, the accuracy of the recorded data using EVENT is 0.01 second.
It may be important here to describe the procedure of how this program works.
The recorded film is replayed on a large screen TV monitor to extract the
desired information. Different types of data are retrieved. At the same time,
EVENT program runs- it is better that the screen of the computer on which
EVENT program is operated, be near the TV screen in order to see them at the
same time and each function bottom (Fi) is allocated for movement maneuver.
See Appendix C for operation illustration.
EVENT program output is accumulative time for successive vehicles
departure, using Microsoft Excel program to enable the calculation of the
following traffic parameters (Talabany, 2000):
1. The time headway between successive arriving vehicles;
2. The time headway between successive departing vehicles in queue.
3. The travel time for successive departing vehicles in queue.
4. Saturation flow data calculated by taking the reciprocal of the average
saturation headway.
5. The frequency distribution of the observed departure and arrival
headway.
6. The average observed delay of an approach.
7. Duration of observed data session.
8. Vehicle arrival and departure flow.
Sample of EVENT program output is shown in Appendix C .
59
Chapter Three
Methodology and Data Collection
3.12 Queue Length
The observation were made from replaying video films on each lane for
all approaches to collect the queue length during peak hour periods. The
observed data for the average queue length during red phase, are as shown in
Table (3-4).
Table (3-4): Average Observed Queue Length.
Palace
Intersection
Average Q.L.
Sara
Intersection
Average Q.L.
Sailm
(TH)
12(veh)
Mawlawi
(LT)
16(veh)
Salim
(LT+UT)
13(veh)
Kawa
(TH)
15(veh)
Baban
(LT)
25(veh)
Kawa
(LT)
13(veh)
Baban
(TH)
23(veh)
Sabunkaran Piramerd
(TH)
(LT+UT)
20(veh)
13(veh)
3.13 Data Processing
The abstracted data is then processed using Microsoft Excel program and
output data files from EVENT program. These programs are used to measure
the main traffic parameters that are illustrated in Table (3-5).
Table (3-5): Summary of Abstracted Data.
Major Category
Data Type
• Through and turning traffic
Traffic volume data
volume counts
• Headways and saturation flow rate
Driver performance characteristics
• Gaps and follow-up time
data
• Cycle length, green time , red time
Signal timing control
•
Phase length and sequence
• Delay
Data used in comparison with
• Queue length
simulated results
60
Chapter Three
Methodology and Data Collection
The output file from EVENT program is opened with the aid of EXCEL
program. This program has an option which splits text data to columns.
This method separates the recording time in column and the associate character
in a next column. This arrangement makes dealing with the data easier by
sorting them into digits and characters.
For traffic volume data calculation, Pivot table is used to calculate the
number of each character presented in the selected column. Other time-related
parameters (saturation headway, phase and cycle length, and delay times) are
calculated by arranging EVENT output file in a set of columns according to
their characters. Then, the subtraction of one digit from another is done to find
the value (time) for the subjected parameters. Samples of each parameter
calculation are given in Appendix (C).
3.14 Delay Measurement
Method suggested by HCM 2000 is based on direct observation of
vehicles-in-queue counts at the intersection. This method does not directly
measure delay during deceleration and during part of acceleration, which are
very difficult to measure without sophisticated tracking instrument. However,
this method has been seen to yield a reasonable estimate of control delay. The
method includes an adjustment for error which may occur when this type of
sampling technique is used, as well as an acceleration/deceleration delay
correction factor.
A sample worksheet, used for recording retrieved data, is included in
Appendix (B). The survey period should begin at the start of the red phase of the
approach, ideally when there is no cycle failure (no overflow queue) from the
61
Chapter Three
Methodology and Data Collection
previous green period. Recorded films are replayed to retrieve data for vehicle
delay time.
The procedure adopted to retrieve data is summarized below.
• The moment signal turns to red, cassette is paused and VCP timer is set to
zero. The overflow queue has been excluded from queue counts. This is
due to the need for consistency with the analytical delay equation, which
is based on delay to vehicles that arrive during the survey period. This
time period may differ from analysis period which is typically considered
as 15 minutes as per HCM(2000), because all the vehicles that join the
queue within this analysis period should be included in queue count until
they cross the stop line.
• Cassette is played and the number of vehicles in queue is recorded at
regular interval of 10 to 20 seconds (As per HCM 2000). The regular
interval should not be an integral divisor of the cycle length. Meanwhile,
it is necessary to keep track of end of standing queue by observing the last
vehicle in those stops because of signal. This includes vehicles arriving
when the signal is actually green, but stopped because vehicles in front
have not yet started moving. The vehicles in queue counts often include
some vehicles that have regained speed, but have not yet exited the
intersection.
• Count of vehicles-in-queue is done at regular interval for analysis period
of about 15 minutes. End of the survey period must be clearly defined,
since the last arriving vehicle (s) that stop in the period must be clearly
defined and counted until they exit the intersection, per the next step.
Stopping vehicles that arrive after the end of the analysis period are not
included in the final vehicle-in-queue counts.
62
Chapter Three
Methodology and Data Collection
• Volume counts of total vehicles (𝑉𝑡𝑜𝑡 ) arriving during the survey period,
and total vehicles arriving during the survey period that stop one or more
times. Vehicles stopping multiple times are counted only once as a
•
stopping vehicle (𝑉𝑠𝑡𝑜𝑝 ) as per HCM(2000) delay measurement guideline.
The average time-in-queue per vehicle arriving in the survey period is
estimated as (HCM,2000) :
where:
Delay time-in-queue per vehicle, 𝑑𝑣𝑞 = (𝐼 ∗
∑𝑉𝑖𝑞
𝑉𝑡𝑜𝑡
) * 0.9
𝐸𝑞. (3 − 5)
𝐼 = interval between vehicle-in-queue counts, (s)
𝑉𝑖𝑞 = sum of vehicle-in-queue counts, (veh)
𝑉𝑡𝑜𝑡 = total number of vehicles arriving during the survey period, (veh)
0.9 = an empirical adjustment factor accounts for the errors that may occur when
this type of sampling technique is used to derive actual delay values, which
normally results in an overestimate of delay (As per HCM 2000).
• Next, the fraction of vehicles stopping and the average number of vehicles
stopping in a queue in each cycle are computed.
𝐹𝑉𝑆 = Fraction of vehicles stopping =
𝑉𝑠𝑡𝑜𝑝
𝑉𝑡𝑜𝑡
𝐸𝑞. (3 − 6)
• Correction factor given by HCM is selected based on average free flow
speed (that is measured at the upstream of the selected approaches) and
average number of vehicles stopping per queue in each cycle. The values
of correction factor are given Table (3-6).
63
Chapter Three
Methodology and Data Collection
• The fraction of vehicles stopping is multiplied by correction factor and the
product is added to the time-in-queue value to obtain the final estimate of
control delay.
Acceleration and deceleration delay:
𝑑𝑎𝑑 = 𝐹𝑉𝑆 * 𝐶𝐹
𝐸𝑞. (3 − 7)
Control Delay/vehicle, 𝑑 = 𝑑𝑉𝑞 + 𝑑𝑎𝑑
𝐸𝑞. (3 − 8)
Control delay includes initial deceleration delay, queue move-up time, stopped
delay, and final acceleration delay.
Table (3-6) Acceleration/Deceleration Delay Correction Factor (CF) – second
(HCM, 2000).
Free-Flow
≤7 vehicles
8-19 vehicles
20-30 vehicles
≤ 60km/h
+5
+2
+1
>60-71 km/h
+7
+4
+2
>71 km/h
+9
7
+5
Speed
3.15 SYNCHRO / SimTraffic
SYNCHRO V.8 is a complete software package for modeling and
optimizing traffic signal timings. This software includes:
 SYNCHRO : a macroscopic analysis and optimization program;
 SimTraffic:
a
powerful,
easy-to-use
traffic
simulation
software
application. It is designed to model networks of signalized and
unsignalized intersections, including roundabouts. The primary purpose of
SimTraffic is to check and fine tune traffic signal operations.
64
Chapter Three
Methodology and Data Collection
 3D Viewer, a three-dimensional view of SimTraffic simulations;
 SimTraffic CI, an application that interacts with a controller interface (CI)
device connected to a controller to simulate the operation of the controller
with simulated traffic.
The key features of SYNCHRO include:
• Capacity Analysis. SYNCHRO provides a complete implementation of
the 2010 Highway Capacity Manual, Chapter 18.
• Coordination: SYNCHRO allows quickly generating optimum timing
plans to minimize delays.
• Actuated Signals: SYNCHRO is the only interactive software package
to model actuated signals. SYNCHRO can model skipping and gap-out
behavior and apply this information to delay modeling
• Time-Space Diagram: SYNCHRO has colorful, informative TimeSpace Diagrams. Splits and offsets can be changed directly on the
diagram.
• Integration with SimTraffic, CORSIM and HCS: SYNCHRO
features are preprocessors to these software analysis packages. Data are
entered once with easy-to-use SYNCHRO, and then analyses are
performed with these software packages.
SimTraffic V.8 is microscopic stochastic simulation model used in
conjunction with SYNCHRO.
The following points briefly provide an overview of SimTraffic:
• It incorporates vehicle and driver performance characteristics developed by
FHWA.
65
Chapter Three
Methodology and Data Collection
• It is capable of simulating different types of intersection controls, various
street geometries including lane drops and turning pockets, and a wide range
of traffic flow conditions.
• It uses “car following” logic for simulation.
• It can be calibrated for local driver/traffic behavior by headway factor as
input in SYNCHRO, and vehicle\driver behavior adjustment in SimTraffic.
• It is not capable of modeling ramp metering, bus stops, bus routes, bus
and carpool lanes, light rail, on-street parking and short-term events.
66
DATA PRESENTATION
AND ANALYSIS
Chapter Four
Data Presentation and Analysis
Chapter Four
Data Presentation and Analysis
4.1 General
This chapter presents the analysis of the collected data, calibration of the
output results of the software used for data analysis, the development of delay
and queue statistical models for signalized intersections to evaluate traffic
performance and existing geometric design. Improvement of geometric features
is also proposed and the results are evaluated using traffic software.
4.2 Evaluation of Existing Traffic Flow
To investigate the existing conditions of traffic operation schemes and
geometry of studied intersections, simulation of the actual movement is required
to evaluate traffic performance.
SYNCHRO Plus 8 is applied to simulate the existing traffic flow for three
intersections at the selected network. It is a complete software package for
modeling, optimizing, managing and simulating traffic systems. The initial
release of this software occurred in June 2011.
SYNCHRO models streets and intersections as links and nodes. These
links and nodes are created in the map view. Every intersection to be analyzed in
the study area is represented by a node. These links and nodes are created by
using GIS tools in map measurements depending on the available satellite
images with resolution of 0.6m, updated to 2007 and available at Sulaymaniyah
municipality, like intersection spacing. See Figure (4-1).
67
Chapter Four
Data Presentation and Analysis
Figure (4-1): Network Layout as Printout by SYNCHRO.
The data like geometric characteristics of approaches (number of lanes per
approach at each intersection, lane width and splitter island width) and traffic
characteristics (traffic volume, cycle length, link speed, phase sequence,
pedestrian volume and grade) are required inputs to the program. All values are
entered in easy-to-use forms. Calculations and intermediate results are shown on
the same forms. Figure (4-2) shows process of entering lane setting (See
Appendix D for more details).
68
Chapter Four
Data Presentation and Analysis
Figure (4-2): Lane Settings by SYNCHRO.
Table (4-1) indicates the output results of simulation runs which include
the degree of saturation, total delay and level of service for each intersection in
the study area depending on existing traffic conditions.
69
Chapter Four
Data Presentation and Analysis
Volume
(veh/hr)
Movement
Street
Average Delay for
(sec/veh)
Level of
Service
1069
0.95
71.5
TH
809
1701
0.95
71.7
RT
)permitted)
656
1265
0.56
1.8
A
LT
680
1057
1.27
204
F
TH
440
1642
0.95
116.3
RT
)permitted)
260
1173
0.24
0.5
RT
)permitted)
680
2070
0.34
0.5
0.5
A
A
LT
614
824
0.68
41.1
41.1
D
D
LT
179
713
0.49
40.1
Sabunkaran
Piramerd
Great
Mosque*
*
E
E
135.7
71.9
52.9
D
F
F
D
34.4
TH
312
2052
0.32
30.2
TH
130
1080
0.56
42.9
E
A
C
C
31.9
Baban
Salim
805
Mamostayan3
v/c
LT+UT
Kawa
Palace*
Sara*
Saturation
Flow Rate
(veh/hr)
Mawlawi
Intersection
Table (4-1): Total Delay and Level of Service for Isolated Intersections
Produced by SYNCHRO for Existing Condition.
C
D
22.3
C
RT
)permitted)
224
803
0.34
1.2
A
LT+UT
368
1293
0.44
36.7
D
RT
)permitted)
100
772
0.17
0.5
27.5
C
A
Max. v/c
Average Delay for (sec/veh)
Level of Service
1.90
158.8
F
*Signalized Intersection
**Unsignalized Intersection (Two Way Stop Control)
70
Chapter Four
Data Presentation and Analysis
The results in Table (4-1) show that, most of the intersections suffer from
congestion with high total delay values. All the output results are categorized by
medium to high total delays, thus the level of service for most of the
intersections is (E, F) according to the LOS categories of the software based on
HCM method.
4.3 SYNCHRO Model Calibration
Calibration is the process by which the individual components of the
model are refined and adjusted so that the model accurately represents field
measured or observed traffic conditions (Turley, 2007). With regards to
calibration, traffic models contain numerous variables to define and replicate
traffic control operations, traffic flow characteristics and driver behavior. The
model contains default values for each variable, but also allows a range of userapplied values for each variable.
The measures of effectiveness estimated by the models such as queue
length and delay are used for comparison with local traffic conditions in order to
ensure that the model accurately represents the measured traffic parameters. The
delay time estimated by models is mostly used by traffic engineers to evaluate
intersection performance. For this study, the delay has been selected to be
compared with existing traffic conditions and the values of the selected
parameters have been manipulated until an acceptable convergence between the
observed and simulated delay values has been reached.
Unfortunately, the user manual of SYNCHRO does not provide substantial
guidance on how the user should modify these parameters for different types of
conditions. Therefore, the user has a great responsibility for ensuring that the
appropriate changes made are based on field measured data and not on
engineering judgment.
71
Chapter Four
Data Presentation and Analysis
The main parameter at signalized intersection that depends on driver
behavior is the saturation headway. This parameter affects mainly the saturation
flow rate. Saturation flow rate is the maximum departure (queue discharge) flow
rate achieved during the green period at traffic signals.
SYNCHRO saturation flow estimation is affected by many factors
including lane width, grade, turning radius, parking, buses, pedestrian
interference, percentage of heavy vehicles, shared lanes, opposed turns and short
lanes. Furthermore, saturation flows estimated by SYNCHRO are dependent on
signal timings.
The default value of the basic saturation flow rate used by SYNCHRO
model is 1900 pcuphgpl. This is applied at the signalized intersections in the
study area. The delay time produced by SYNCHRO for each approach is
compared with the average delay times measured from the field.
The results of the comparison between measured and estimated delays are
shown in Figure (4-3). This Figure explains that, for high values of
v/c ,
SYNCHRO highly over estimates the delay time. The approaches field delay are
regressed against the predicted ones producing the results shown in Table (4-2).
The R² value indicates that, SYNCHRO explains about 77 % of the variability in
the delay times.
A comparison is made by using the paired samples t-test. The paired
samples t-test is made by using SPSS V.20 program. The paired t-test results, as
in Table (4-2), show that, the mean of delay differences is 14.7 seconds with
a P-value of 0.281, so the hypothesis that the deviations are equal to zero is
accepted at 95% confidence. In order to reduce standard error mean of delay, the
model`s parameters should be adjusted.
The calibration process is to adjust the basic saturation flow rate to be
2200 pcuphgpl in lane groups with high values of v/c, as recommended by
72
Chapter Four
Data Presentation and Analysis
HCM. The results show 93.3 % variability in delay times, See Figure (4-4). The
paired t-test results show that, the mean of delay differences of 5.322 seconds is
lower than the difference obtained by using the default value of saturation flow,
with a P-value of 0.455. So, the hypothesis is that there is no significant
difference accepted at 95% confidence.
Table (4-2) Comparison of Actual and SYNCHRO Delay Prediction for
Signalized Intersections.
Default Values of
Saturation flow rate
(1900 veh/hr)
Values after adjustment of
Saturation flow rate
(2200 veh/hr)
Field delay
0.501* SYNCHRO Delay
0.808*SYNCHRO Delay
R2
0.772
0.933
Adj.R2
0.744
0.924
Sig.
0.001
0.000
Paired Sample t-test
Mean Difference in
Delay (sec)
14.713
5.322
t
1.157
0.784
p
0.281
0.455
All p-value greater than 0.05, so that there is no statistically significant difference
73
Chapter Four
Data Presentation and Analysis
Figure (4-3): SYNCHRO Average Delay Values versus Field Delay Values for
All Approach.
Figure (4-4): SYNCHRO Delay Values versus Field Delay Values after
Calibrating the Basic saturation flow rate for All Approach.
74
Chapter Four
Data Presentation and Analysis
4.4 Development of Statistical Models for Delay and Queue
Length
The traffic data, which represent many variables are extracted from video
records to be used in the process of building statistical models. These data
include; traffic and geometric characteristics. The next paragraphs describe the
statistical techniques used for the models development are required for the
prediction of delay and queue length.
4.4.1 Process of Models Building
The following steps, which are recommended by many researchers (Keller
and Warrack, 2000), are followed in this study;
a. Identifying the dependent variables.
b. Listing potential predictors.
c. Gathering the required observations for the potential models.
d. Using statistical software to estimate the models.
e. Determining whether the required conditions are satisfied.
f. Using the engineering judgment and the statistical output to select the best
models.
4.4.2 Identification of Dependent and Predictor Variables
The variables, which are used to simulate traffic and geometric
characteristics, are as follows:
1. dependent variables:
g = Effective green time for movement or lane group (sec),
c = Capacity of lane group (veh/h),
C = Cycle length (sec),
v = Vehicular flow rate (veh/h),
w =Width of lane group at stop line (m).
75
Chapter Four
Data Presentation and Analysis
Ws = Total width of lane groups departing to the same exit roadway, in
the same phase at stop line (m).
We = Total width of exit roadway of intersection for traffic departing
straight forward (m),
2. potential predictors:
q = Average number of vehicles in queue (veh),
d = Delay for lane group (sec/veh).
These data are shown in Appendix E.
4.4.3 Data Analysis
The data analysis includes the following steps:
4.4.3.1 Selecting Sample Size
To determine the required sample size, (Kennedy and Neville, 1986)
presented the following equation to calculate percent of error according to
sample size:
E=VT/ (n) 0.5 …………………Equation (5-1)
Where:
E=Error of mean,
V=Coefficient of variation (c.o.v),
t=t-statistics, and
n= Sample size
Therefore, it is suggested to use the whole data for the delay model
building with a (0.026133) percentage of error, and the queue model building
with a (0.050) percentage of error.
4.4.3.2 Scatter Plots
Scatter plot is carried out between the dependent and independent
variables for the requirements of the modeling process. The nature of relation
76
Chapter Four
Data Presentation and Analysis
between dependent and independent variables can be expected and the best
relations are selected depending on resulted plots.
4.4.3.3 Outliers
If one or more of observations is different significantly from all others, it is
called “outlier“. The cause of a faulty observation may be a mistake(Kennedy
and Neville, 1975). Outliers and influential observations are checked by using
SPSS software (V.20).The results of this test can be found in Table (4-3).It can
be seen that there is no outliers because there are no extreme values.
Table (4-3) (a): Case Processing Summary for Outliers for Field Delay
(SPSS, V.20).
Cases
Field delay
Valid %
Missing %
Total %
100.0%
0.0%
100.0%
Table (4-3) (b): Case Processing Summary for Outliers for Field Queue
of Left Turn Movement (SPSS, V20).
Cases
Field queue
Valid %
Missing %
Total %
100.0%
0.0%
100.0%
77
Chapter Four
Data Presentation and Analysis
Table (4-3) (c): Case Processing Summary for Outliers for Field Delay of
Left Turn Movement (SPSS, V.20).
Cases
Field queue
Valid %
Missing %
Total %
100.0%
0.0%
100.0%
4.4.3.4 Testing of Normality
There are several methods of assessing whether data are normally
distributed or not. The Kolmogorov-Smirnov test (K-S) and Shapiro-Wilk (SW) test are designed to test normality by comparing the data to a normal
distribution with the same mean and standard deviation of the sample. If the
test is not significant, then the data are normal, so any value above 0.05
indicates normality. If the test is significant (less than 0.05), then the data are
non-normal. Thus, results of tests in Table (4-4) show that all results above
0.05 follow normal distribution.
78
Chapter Four
Data Presentation and Analysis
Table (4-4) (a): K-S test and (S-W) test Results for Field Delay
(SPSS, V.20).
Kolmogorov-Smirnov
Field delay
Shapiro-Wilk
Statistic
Sig.
Statistic
Sig.
0.199
0.122
0.928
0.258
Table (4-4) (b): K-S test and (S-W) test Results for Field Queue of
Left Turn Movement of All Approaches (SPSS, V20).
Kolmogorov-Smirnov
Field
Queue
Shapiro-Wilk
Statistic
Sig.
Statistic
Sig.
0.117
0.200
0.967
0.804
Table (4-4) (c): K-S test and (S-W) test Results for Field Queue of
Through Movement of All Approaches (SPSS, V20).
Kolmogorov-
Shapiro-Wilk
Smirnov
Field Queue
Statistic
Sig.
Statistic
Sig.
0.228
0.151
0.895
0.194
79
Chapter Four
Data Presentation and Analysis
4.4.3.5 Multicollinearity
Multicollinearity (collinearity and intercorrelation) is a statistical
procedure to find the correlation between independent variables. The adverse
effect of multicollinearity is that the estimated regression coefficients (b1, b2,
etc.) tend to have large sampling variability.
The basic methodology followed in this procedure is that, predictor
variables are eliminated one – by – one based on multicollinearity analysis.
With multicollinearity control, the process is repeated eliminating predictor
variables remaining. At that point, interactions among the variables are
considered; again the process is iterated discarding terms based on significance.
A correlation matrix is produced to determine the correlation coefficients
for the variables. In many cases, the correlation matrix will not be able to
identify whether multicollinearity is a serious problem because there are many
ways for variables to be related (Keller and Warrack, 2000).
SPSS software (Ver.20) is employed for the development of the models.
A confidence level of 95 percent is adopted, thus a significant level (α = 0.05)
is employed. By using SPSS software, the correlation coefficients between all
of the variables are calculated and the correlation matrix is setup. This matrix
can be seen in Tables (4-5), (4-6) and (4-7) for delay and queue models. Then,
the independent variables having the highest correlation coefficient with the
designated dependent variables are selected and calculated, and the regression
equation is formulated.
80
Chapter Four
Data Presentation and Analysis
Table (4-5): Correlation Coefficient Matrix for Delay Model.
Field
delay
Cycle
(sec)
g/C
v/c
Field delay
1.0
Cycle (sec)
0.854
1.0
g/C
-0.022
0.351
1.0
v/c
0.759
0.643
-0.016
1.0
We/Ws
0.659
0.697
0.171
0.719
We/Ws
1.0
Table (4-6): Correlation Coefficient Matrix for Queue Length of Left Turn
Movement.
Field Queue
(sec)
g/C
v/c
Width
(m)
Field Queue
(sec)
1.000
g/C
0.05
1.000
v/c
0.870
-0.118
1.000
Width (m)
-0.389
-0.127
-0.285
1.000
Cycle (sec)
0.829
0.351
0.666
-0.326
81
Cycle
(sec)
1.000
Chapter Four
Data Presentation and Analysis
Table (4-7): Correlation Coefficient Matrix for Queue Length of Through
Movement.
Field Queue
v/c
g/C
Width(m) Cycle(sec)
Field Queue
1.000
v/c
0.831
1.000
g/C
-0.144
-0.256
1.000
Width(m)
-0.624
-0.520
-0.053
1.000
Cycle)sec)
0.854
0.664
0.292
-0.544
1.000
4.4.4 Regression Modeling
Regression analysis is a statistical method that uses the relationships
between two or more variable quantities to generate a model that may predict
one variable from the others. The term multiple linear regression (MLR) is
employed when a model is a function of more than one dependent variable.
The objective behind (MLR) is to obtain adequate models, at a selected
confidence level, using the variable data, while at the same time satisfying the
basic assumptions of regression analysis, which are:
• Severe multicollinearity does not exist among predictor variable
• Influential observations or outliers do not exist in the data.
• The distribution of error is normal.
• The mean of error distribution is zero.
The objective is accomplished by selecting the model, which provides the
highest adjusted coefficient of determination (R²) and the lowest mean square
error (MSE) (Keller and Warrack, 2000).
82
Chapter Four
Data Presentation and Analysis
4.4.4.1 Stepwise Regression Procedure
This procedure begins by computing the simple regression model for each
independent variable. The independent variable with the largest f- statistic is
chosen as the first entering variable. SPSS software uses the f- statistics and the
standard is usually set at F = 3.8 which is chosen because the significant level
is about 5%. The standard is called the f- to enter.
If at least one variable exceeds the standard, the procedure continues, then
the software will examine if the variables verify the condition of dependent
variables remaining in the model or not. It then considers whether the model
would be improved by adding a second independent variable. It examines all
such models to determine which is the best and whether the f- statistic of the
second variable (with the first variable already in the equation) is greater than fto enter (Ahmed, 2002).
If two independent variables are highly correlated, only one of them will
enter the equation. Once the first variable is included, the added explanatory
power of the second variable will be minimal and its f-statistic will not be large
enough to enter the model. In this way, multicollinearity is reduced. The
procedure continues by deciding whether to add another independent variable
to each step. The p-values of all variables are computed (at each step) and
compared with the f- to be removed. If a variable f – statistic falls below this
standard, it is removed from the equation. These steps are repeated until no
more variables are added or removed.
4.4.4.2 Developed Models
The summary of the stepwise regression and possible developed models
is shown in the following articles for delay and queue length models.
83
Chapter Four
Data Presentation and Analysis
4.4.4.2.1 Delay Model
The summary of stepwise regression delay model can be
seen in
Table (4-8).
Tables (4-8): Stepwise Regression Models Summary for Delay Model.
Unstandardized
Coefficients
Model
1
Standardized
Coefficients
t
Sig.
R2
B
Std. Error
Beta
Cycle
0.102
0.020
0.457
5.192 0.000
(1-We/Ws)
19.594
3.622
0.140
5.409 0.000 0.995
v/c
30.190
5.648
0.462
5.345 0.000
Adj.R2
SEE
0.993
4.175
The results of regression technique and selected developed delay model are
shown in Equation (4-1)
d = 0.102 C +30.19 v/c +19.59 (1-We/Ws)
𝐸𝑞. (4 − 1)
The variables that have significant impact on total delay are explained in
Table (4-9) for the linear regression model, with statistical characteristics. The
minimum and maximum traffic parameters limits represent the range at which
the model will be applicable.
84
Chapter Four
Data Presentation and Analysis
Table (4-9): Data Range and Statistical Characteristics for Delay Model.
Minimum Maximum
Mean
Std. Deviation
g/C
0.370
0.610
0.48253
0.068641
v/c
0.50
1.15
0.7395
0.230800
Cycle (sec)
97.0
300.0
215.067
70.1310
We/Ws
0.458
1.000
0.7298462
0.271017
4.4.4.2.2 Left Turn Queue Model
The summary of stepwise regression model for queue length of left turn
movements can be seen in Table (4-10).
Table (4-10): Stepwise Regression Models Summary for Queue Length of Left
Turn Movements.
Unstandardized Standardized
Coefficients
Coefficients
Model
t
B
2
Std.
Error
Sig.
R2
Adj.R2
SEE
Beta
(Constant)
-9.482
2.912
-3.256 0.007
v/c value
18.223
4.488
0.571
4.061 0.002 0.868 0.846
Cycle
0.051
0.016
0.449
3.195 0.008
3.1479
The results of regression technique and selected developed model are
shown in Equation ( 4-2 ).
q =0.051 C+18.223 v/c -9.48
85
𝐸𝑞. (4 − 2)
Chapter Four
Data Presentation and Analysis
The variables that have significant impact on total delay are explained in
Table (4-11) for the linear regression model, with statistical characteristics. The
minimum and maximum traffic parameters limits represent the range at which
the model will be applicable.
Table (4-11): Data Range and Statistical Characteristics for Queue
Length Model of Left Turn Movement.
Variable
Minimum Maximum
Mean
Std. Deviation
v/c value
0.500
1.172
0.75427
0.251457
Cycle (sec)
97.0
300.0
215.067
70.1310
4.4.4.2.3 Through Queue Model
The summary of stepwise regression model for queue length of through
movements can be seen in Table (4-12).
Tables (4-12): Stepwise Regression Models Summary for Queue of Through
Turn Movements.
Model
Unstandardized
Coefficients
B
(Constant) -11.618
3
Std. Error
Standardized
Coefficients
t
Sig.
-2.755
0.025
R2
Adj.R2
SEE
0.853
0.816
4.10304
Beta
4.218
Cycle
0.071
0.024
0.540
2.978
0.018
v/c value
13.330
5.124
0.472
2.601
0.032
86
Chapter Four
Data Presentation and Analysis
The results of regression technique and selected developed models can be
seen in Equation (4-3).
q = 0.071 C+13.33 v/c -11.618
𝐸𝑞. (4 − 3)
The variables that have significant impact on queue length of through
movements are explained in Table (4-13) for the linear regression model, with
statistical characteristics. The minimum and maximum traffic parameters limits
represent the range at which the model will be applicable.
Table (4-13): Data Range and Statistical Characteristics for Queue
Length Model of Through Movement.
Variable
Minimum
Maximum
Mean
Std. Deviation
v/c value
0.15
1.30
0.7607
0.33875
Cycle (sec)
97
300
225.73
72.874
4.5 Models Validation
Validation process is determining whether the selected model is
appropriate for the given conditions and for the given task; it compares model
prediction with measurements or observations (TRB, 2005). The objective of
validation is to assess the adequacy of the proposed prediction models, and
measure the error or accuracy of the prediction for the validation period.
There are several methods used for models validation. One of these
methods is to compare the model with another data set that is not included in
model building.
87
Chapter Four
Data Presentation and Analysis
The data used for this purpose is one hour data abstracted from video
recording films at different times for the same intersections in the network. The
average field delay from one hour is regressed with the delay time predicted by
the model. The regression results are shown in Figures (4-5), (4-6) and (4-7)
and Table (4-14). It can be concluded from the models values of R 2 that, the
predicted values from models can represent the actual field values of delay and
queue length.
Table (4-14): Regression Results for the Three Models.
Model
1- Delay Model
2- Left Queue
Model
3-ThroughQueue
Model
Model Fit
R 2value
2
Adj.
R -value
Sig.
0.958 * Field Delay + 3.36
0.922
0.911
0.00
1.139 * Field Queue – 2.24
0.906
0.874
0.013
1.058 * Field Queue – 1.944
0.956
0.942
0.004
88
Chapter Four
Data Presentation and Analysis
Figure (4-5): Observed versus Predicted Delay.
Figure (4-6): Observed versus Predicted Queue Length for the Model of
Left Turn Movement.
89
Chapter Four
Data Presentation and Analysis
Figure (4-7) :Observed versus Predicted Queue Length for the Models of
Through Movement.
4.6 Models Analysis
The delay model depends on many variables such as cycle length(C),
degree of saturation(v/c) and Total width of exit roadway of intersection for
traffic departing straight on divided by total width of lane groups departing to
the same exit roadway, in the same phase (We/Ws) while queue length depends
on cycle length and degree of saturation. From the previous models of delay and
queue length, it is obvious that, delay time and queue length increase as cycle
time and v/c value increase. The increase in delay is attributed mainly to the
increase in the effective red and hence in waiting time for vehicles before the
start of green. The delay time has a proportional relationship with the degree of
90
Chapter Four
Data Presentation and Analysis
saturation (v/c) which is based on flow rate and proportion of cycle time
effectively green (g/c).The minimum of g/c and large flow rate cause high v/c
value.
Delay time has an inverse relationship with (We/Ws). It is obvious that
delay time increases when (We/Ws) is less than 1.0. The limit of (We/Ws)
adopted in the model should be ≤ 1.0, as in Table (4-9) for lane group width not
for the whole approach.
4.7 Comparison of Observed Delay with HCM 2000 Delay Values
The 2000 HCM delay method suggested the given below equation for the
purpose of calculation of vehicle delay.
𝑑 = 𝑑1 ∗ 𝑃𝐹 + 𝑑2 + 𝑑3
𝐸𝑞. (4 − 4)
where:
d = average overall delay per vehicle (sec/veh),
d1 = uniform delay (sec/veh),
d2 = incremental, or random delay (sec/veh),
d3 = residual demand delay to account for over-saturation queues that may have
existed before the analysis period (sec/veh),
PF = adjustment factor for the effect of the quality of progression in coordinated
systems.
Figure (4-8) shows the calculated delay by HCM and field delay. The
correlation coefficient obtained is R2=0.88 which suggests a good correlation.
The regression relationship is as given below:
Field delay =0.813 * HCM delay
91
for v/c ≤ 1.2
𝐸𝑞. (4 − 5)
Chapter Four
Data Presentation and Analysis
The form of regression equation indicates that, the HCM equations provide
higher estimate of delay values than the field delay. Table (4-15) lists the results
of the ANOVA test, for the delay model showing that the regression model is
statistically significant (p ≈ 0).
Figure (4-8) HCM Delay versus Field Delay
Table (4-15 ) ANOVA Test for the Regression Model.
Sum of Squares Mean Square
Regression
21773.123
21773.123
Residual
2802.473
311.386
Total
24575.595
92
F
Sig.
69.923
.000
Chapter Four
Data Presentation and Analysis
4.8 Improvements
This section presents the improvement strategies applied to the traffic flow
in the study area of Sulaymaniyah city, through the application of the calibrated
software.
SYNCHRO is used to evaluate the existing traffic operations. It is also
used as a tool to determine the optimum traffic signal timings for a single
intersection or for a series of coordinated or uncoordinated signalized
intersections.
This section divides the improvement strategies into two parts. The first
part includes Signal Timing Optimization and Coordination. The improvement
type is suggested and evaluated by SYNCHRO program. The second part
includes the geometric improvement.
4.8.1 Signal Timing Optimization and Coordination
Traffic signal control is a system for synchronizing the timing of any
number of traffic signals in an area, with the aim of reducing stops and overall
vehicle delay or maximizing throughput.
One of the most common methods for increasing the efficiency of traffic
operation is coordination of consecutive traffic signals. This is because, the
higher increase in the traffic demand will result in higher increase in the need of
coordination. Traffic signal coordination can reduce the traffic congestion in
many areas. Substantial improvements in traffic flow and reduction in delay,
fuel consumption and stops and uniform flow speed links could be achieved by
coordination.
Signal coordination is the process to synchronize the start of the “green
light” along the major roadway, so that vehicles can travel through a group of
signals with minimal or no stopping. There are three key timing parameters to
optimize and are noticeable to the driver. These include the “cycle length”,
93
Chapter Four
Data Presentation and Analysis
intersection “offset,” or progression, and the individual traffic movement “green
+ yellow + red” phase (referred to as a movement “split”).
SYNCHRO contains a number of optimization types. It optimizes cycle
length, split times, offsets and phase sequence to minimize driver stops and
delays. These types are applied to the network in the same order shown in
Figure (4-9) (Husch and Albeck, 2006).
Setup Intersection Timing Plan
Optimize Splits and Cycle Length
Optimize Network Cycle
Length
Optimize Offset and Phase
Order
Figure (4-9): Optimization Steps in SYNCHRO.
The range of cycle length recommended by HCM is 60-120 seconds.
Range of cycle length 100-120 seconds will be recommended to accommodate
an extra 10% capacity. Long cycle lengths more than 120 seconds may have
negative operation aspects including long queues, inefficient use of turning
lanes, and blocking. Lower cycle lengths (30-60) may be better to reduce queues
which in general increase capacity and provide smoother traffic operation
(Sofia, 1998). The optimization results with the HCM recommended cycle range
do not result in optimum cycle length, so the default range of SYNCHRO is
used.
94
Chapter Four
Data Presentation and Analysis
To calculate the best cycle time and the optimal timing of the existing
phase sequences of the observed intersections, SYNCHRO is used. The range of
cycle times used in the optimization process is 50-200 seconds. This range is
used for each observed isolated intersection. Table (4-16) provides a summary
for the optimization results. This is together with the best cycle time for each
intersection in the observed network.
Table (4-16) shows that, the use of optimization functions for the isolated
signalized intersections such as split and offset has good effect on the measure
of effectiveness (MOE) for Palace intersections and has minimized the effect of
(MOE) at Sara intersection.
95
Chapter Four
Data Presentation and Analysis
Sara
palace
sec
sec/veh
Total Fuel
Consumption
Total Uniform
Stop
LOS
Total Delay
Max. v/c at Intersection
Cycle Length
Improvement Type
Intersection Name
Table (4-16): Measure of Effectiveness for the Improved Intersections with the
Optimum Cycle Length Produced by SYNCHRO.
veh/hr
%
l/hr
Base
Condition
300
1.1
61.2
F
2417
0.55
350
Split
300
1.01
63.0
E
2453
0.56
359
Cycle length
120
1.05
45.6
D
2320
0.53
302
Network cycle
length
110
1.05
42.46
D
2310
0.53
291
Phasing
Sequence
110
1.05
42.46
D
2310
0.53
291
Base
Condition
196
0.68
31.9
C
1106
0.57
128
Split
196
0.67
31.9
C
1102
0.57
130
Cycle length
65
0.92
29.7
C
1245
0.64
118
Network cycle
length
65
0.92
29.7
C
1245
0.64
118
Phasing
Sequence
65
0.92
29.7
C
1245
0.64
118
96
Chapter Four
Data Presentation and Analysis
The optimization results show that, although the cycle time increases to
high values, the isolated intersections perform oversaturation and undersaturated
conditions. So, it is necessary to improve the performance by coordination to
reduce cycle length and delay.
After the coordination process for signalized intersections in the network,
the delay times are reduced by 6.9 % for Palace intersection and with no effect
at Sara intersection. The fuel consumption is reduced from 302 l/hr to 291 l/hr,
as shown in Table (4-16).
Although the optimization and coordination runs improve the network
measure of effectiveness, the degree of saturation does not reduce to a value less
than 0.9 thus we should use another trial by improving the geometries of
intersections.
4.8.2 Geometric Improvement
The improvement proposals shall be as follows:
1. Re-marking of pavement and utilization of parking lanes.
2. Widening of approaches.
3. Change the intersection type.
The measure of effectiveness produced for each strategy is illustrated in
Table (4-17) for Signalized intersections analyzed by SYNCHRO. The first step
of the geometric improvement is to apply pavement marking. From the
measurement, it can be noticed that the number of lanes in many approaches of
these intersections can be increased by decreasing the lane width (with keeping
the width of lanes greater than a minimum value of 2.4 m according to HCM)
which is actually observed to be used in field at intersection approaches.
Increasing the number of lanes will increase the capacity of affected
approaches, which leads to improve traffic performance measure
of
effectiveness for some intersections at the selected network. Therefore, in this
97
Chapter Four
Data Presentation and Analysis
step, consideration is given to pavement marking the approaching lanes so that
re-arranging of queuing vehicles at stop line is attained.
Palace
Great Mosque
Sara
sec
Base
Condition
Pavement
Marking and
utilization of
parking lanes
Base
Condition
Change
intersection
type
(Un signalized
to signalized)
Pavement
Marking and
utilization of
parking lanes
sec/veh
Total Fuel
Consumption
Total
Uniform
Stop
LOS
Total Delay
Max. v/c at
Intersection
Cycle Length
Improvement
Type
Intersection Name
Table (4-17): Measure of Effectiveness for the Improved Intersections
Produced by SYNCHRO (Geometric Improvement( .
veh/hr
%
l/hr
65
0.92
29.7
C
1245
0.64
118
60
0.79
23.7
C
1246
0.64
99
---
2.01
112.9
F
---
---
---
60
0.5
11.8
B
433
0.58
27
60
0.3
9.6
A
392
0.52
26
Base
Condition
110 1.05
42.46
D
2310
0.53
291
Pavement
Widening
60
14.0
B
2155
0.5
197
0.8
98
Chapter Four
Data Presentation and Analysis
The pavement marking is applied to Sara intersection to increase number
of lanes with the same approach width. This method reduces the maximum v/c
value from 0.92 to 0.79 and reduces delay from 29.7 seconds to 23.7 seconds.
The fuel consumption reduces from 118 l/hr to 99 l/hr. The improvement is
shown in Figure (4-10).
The Great Mosque is unsignalized intersection (Two Way Stop
Controlled). This intersection suffers from congestion at Court approach, so a
suggestion is made by changing the intersection type to signalized intersection.
Results obtained by SYNCHRO and illustrated in Table (4-17) show an increase
in all measures of effectiveness and reduction in v/c to a maximum value of 0.5
for Court approach . This reduces delay time from 112.9 seconds to 11.8
seconds. Another trial is made by pavement marking and utilization of parking
lanes, as shown in Figure (4-11). This reduces delay time from 11.8 seconds to
9.6 seconds and reduces the maximum v/c value to 0.3.
Another trial is made for Salim approach and Baban approach at Palace
intersection by widening due to the presence of wide sidewalk at this
intersection section taking into consideration pedestrian volume, as shown in
Figure (4-12). The obtained values of v/c are less than 0.8.
Although the optimization and coordination is applied to Palace
Intersection, but the degree of saturation does not reduced to value less than 0.9,
and needs further modification to achieve the objected delay. Another trial is
made by widening approaches due to presence of wide sidewalk at this section
taking into consideration pedestrian volume. The pavement widening at Salim
approach will increase the capacity. This reduces the maximum v/c value to 0.8,
While, The pavement widening at Baban approach reduces the maximum v/c
value to 0.77. Pavement widening at Palace intersection reduces the delay time
from 42.46 seconds to 14 seconds and the fuel consumption from 291 l/hr to 197
99
Chapter Four
Data Presentation and Analysis
l/hr. Figure (4-12) shows the Palace intersection before and after geometric
improvement.
Figure (4-10): Improvement Proposal for Sara Intersection.
Figure (4-11): Improvement Proposal for Great Mosque Intersection.
100
Chapter Four
Data Presentation and Analysis
Figure (4-12): Improvement Proposal for Palace Intersection.
4.9 Forecasted Network
The growth factor of 3.4 % (i.e. traffic forecast factor TFF) is introduced
for the application of forecasting traffic volume on the improvement proposals
for intersections. The growth rate together with the analysis period is applied
with SYNCHRO for the application of forecasting traffic volume on the
improvement proposals.
Table (4-18) shows the measure of effectiveness obtained from
SYNCHRO optimization runs after the last improvement strategy for the target
year 2017.
101
Chapter Four
Data Presentation and Analysis
Table (4-18):Measure of Effectiveness Produced by SYNCHRO after Last
Improvement Strategy for the Base Year 2010, and Target Year 2017.
Intersection
Name
Base Year 2012
Average Delay
(sec/veh)
Target Year 2017
LOS
Average Delay
(sec/veh)
LOS
Palace
14.0
B
17.5
B
Sara
23.7
C
24.7
C
Great mosque
9
A
10.4
B
The LOS shown in Table (4-18) shows that the increase in traffic volume
for
the
target year 2017 will not increase
(B) or (C) which is acceptable.
102
delay so much and LOS remains
CONCLUSIONS AND
RECOMMENDATIONS
Chapter Five
Conclusions and Recommendations
Chapter Five
Conclusions and Recommendations
5.1 Conclusions
Within the limits of the traffic and geometric features of the study area, the
main conclusions that can be drawn from this study are summarized, as follows:
1. Delay time for signalized intersections produced by SYNCHRO has a no
significant difference with the delay time measured at field with 95%
confidence level, SYNCHRO has a good representation of field delay at
low to medium values, but it is overestimated at high delay range.
2. Regression models, developed to estimate delay time and queue length,
show good correlation with field values. Delay model can used to estimate
delay time at any intersections knowing signal timing, traffic volume, ratio
of total width of exit roadway of intersection for traffic departing straight
forward to total width of lane groups departing to the same exit roadway, in
the same phase at stop line (We/Ws). Queue model can be used to estimate
queue length at any intersections knowing signal timing, traffic volume.
These models may not be suitable for extreme range of input variable.
3. Delay estimate by HCM2000 model gives higher estimation of field delay.
From the analysis, it is recommended that, the theoretical delay (due to
random arrival and oversaturated condition) in HCM 2000 delay model can
be reduced by (0.813) to better reflect field delay values.
4. All signalized intersections suffer from long cycle length that causes high
delay time values.
5. Optimization and coordination process could improve the performance of
the network.
103
Chapter Five
Conclusions and Recommendations
6. Pavement remarking and lane designation slightly improve the intersections
performance.
7. The delay increases when the width of exit roadway is less than the width at
stop line, as in the case of bottleneck.
5.2 Recommendations
Based on the study results, the following recommendations may be put
forward:
1. Fixing permanent video recording camera on streets and intersections to
measure traffic volume and speed of vehicle, which helps in the data
collection process and analysis for future studies and for continuous
monitoring of causes of delay and congestion.
2. Converting great mosque from unsignalized intersection (TWSC) to
signalized intersection to improve intersection performance.
3. Locating traffic signals in all intersections and retiming the existing phase
sequence, depending on existing traffic volumes on those intersections.
4. Marking and lightening all streets and intersections.
5.3 Future Studies
1. Few studies on traffic and planning have been conducted in Sulaymaniyah
city; therefore, it is necessary to encourage more studies. This can be
achieved by supporting researchers in terms of modern equipment supply
that requires less manpower and provide data on different parameters of
traffic engineering.
2. The possibility of constructing pedestrian overpass at streets and
signalized intersections to reduce conflict with pedestrian, to provide
safety, and decrease delay time.
104
Chapter Five
Conclusions and Recommendations
3. Network optimization and coordination study in Sulaymaniyah for all
close intersections in the city can be implemented.
105
Reference
References
 AASHTO, American Association of State Highway and Transportation
Officials, (2004): "A Policy on Geometric Design of Highways and
Streets", 5th Edition, Washington D.C.
 Abd Ul-Ghani, M.M.(2004), "Evaluation and Improvement of Traffic
Flow Patterns for a Selected CBD Area in Baghdad City", M.Sc.
thesis, University of Technology, Baghdad, Iraq.
 Ahmed N.G., (2002): "The Development of Models for the prediction
of Thermal Cracking in Flexible Pavements", Ph. D. Thesis, Civil
Engineering Department, College of Engineering, University of
Baghdad, Iraq.
 Akçelik R., (2009): "SIDRA Intersection User Guide", Akçelik&
Associates Pty Ltd.
 Al-Azzawi, Z.A. (2003): "Evaluation of Signalized Intersections
Capacity in Baghdad to Attain the Traffic Flow Performance", M.Sc.
thesis, Al- Mustansirya University, Baghdad, Iraq.
106
Reference
 Al-Dulaimi M. R.,(2006): " Modeling of Pedestrian-Vehicle Conflict
on Arterial Street Using Simulation Approach", M.Sc. thesis, AlMustansiriyah University, Baghdad, Iraq.
 Al-Hemdany H.M.(2003), "Network Traffic Performance within
Adhamiyasector in Baghdad City", M.Sc. thesis, Al-Mustansirya
University, Baghdad, Iraq.
 Al-Hummadawee B. H.(2002), "Evaluation of Performance for
Selected Signalized Intersections in Baghdad City", M.Sc. thesis,
University of Technology, Baghdad, Iraq.
 Al-Jaman, S.M. (2007):"Assessment and Calibration of SYNCHRO
for Riyadh Traffic Conditions", M.Sc. thesis, King Saud University,
Saudi Arabia.
 Al-Neami, A. H., "EVENT – A Computer Program for Abstraction of
Traffic Data". Engineering Journal, University of Tikrit, vol.3, No.2,
2000 (Cited By Reference: Razzaq, 2005).
 Al-Zubaidi Y. K., (2006): "Determination of Passenger Car Unit
Factors for Different Types of Vehicles in Baghdad City", M.Sc.
Thesis, Al-Mustansirya University, Baghdad, Iraq.
107
Reference
 Burrow, I. (1989), "A Note on Traffic Delay Formulas", ITE Journal,
Vol. 59, No. 10, pp: 29-32,UK.
 Catling, I. (1977), "A Time-Dependent Approach to Junction
Delays", Traffic Engineering and Control, Vol. 18, No. 11, pp: 520523, 526.
 Fambro, D. B. and Rouphail, N. M. (1997), "Generalized Delay Model
for Signalized Intersections and Arterial Streets", Transportation
Research Record 1572, TRB, National Research Council, Washington,
D. C., pp: 112-121, USA.
 FHWA Federal Highway Administration, (2009): "Manual on
Uniform Traffic Control Devices for Streets and Highways".
 Garber,N.J.,
and
Hoel,
L.A.,(2009):
"Traffic
and
Highway
and
Highway
Engineering", Thomas Learning, Fourth Edition.
 Garber,N.J.,
and
Hoel,
L.A.,(2010):
"Traffic
Engineering", Thomas Learning, Fourth Edition.
 "Highway Capacity Manual", 4th Edition, Transportation Research
Board, National Research Council, Washington D.C, 2000, USA.
 Husch, D., Albeck, J., (2006): "Traffic Signal-User Guide, Synchro
Studio7".
108
Reference
 Keller, G., and Warrack, B. (2000): "Statistics for Management and
Economies", 4th Edition, USA. (Cited By Reference: Ahmed N.G.,
2002).
 Kennedy , J .B., and Neville, A.M.(1986): “Basic Statistical Methods
for Engineers and Scientists" , 2nd Edition , Thomas Y., Crowell
Company , In. U.S. (Cited By Reference: Ahmed N.G., 2002)
 Khalaf T. M., (2003): "Modified Predictive Models for Vehicle Delay
at Selected Signalized Intersection in Baghdad City", Ph.D. thesis,
Baghdad University, Baghdad, Iraq.
 Lee, J. and Chen, R. L. (1986), "Entering Headway at Signalized
Intersections in Small Metropolitan Area", Transportation Research
Record 1091, TRB, National Research Council, Washington, D.C., pp:
117-126, USA.
 Lin, Feng-Bor (1989), "Application of 1985 Highway Capacity
Manual
for
Estimating
Delay
at
Signalized
Intersections ",
Transportation Research Record 1225, TRB, National Research
Council, Washington, D.C., pp: 18 -23, USA.
 McShane, W. R., Roess, P. R. (1990), "Traffic Engineering", Prentice
Hall, Eaglewood Cliffs, New Jersey, USA.
109
Reference
 Md. Hadiuzzaman, (2008): "Development of saturation flow and
delay models for signalized intersection in Dhaka city", M.Sc. thesis,
Bangladesh University, Dhaka, Bangladesh.
 Miller, A.J. (1968), "Australian Road Capacity Guide", Australian
Road Research Board (ARRB), Bulletin No. 4, Australia.
 Pickrell, S. and L. Neumann, (2000): "Linking Performance Measures
with Decision Making", Presented at the 79th Annual Meeting of the
Transportation Research Board, Washington, D.C, USA.
 Pignataro, L.T., (1973): "Traffic Engineering", Pentice –Hall Inc.,
Third Edition, USA.
 Polus A., Vlahos E., (2005): "Evaluation of Roundabouts versus
Signalized and Unsignalized Intersections in Delaware", Department
of Civil and Environmental Engineering, University of Delaware,
Delaware, USA.
 R. Troutbeck, NCHRP Web Document 5(1997): "Capacity and Level of
Service at Unsignalized Intersections", Final Report, Vol. 1—Two-Way
Stop-Controlled Intersections.
 Razzaq N. A., (2005):"A Study of Pedestrian Behavior at Random
Crossings in Baghdad City", M.Sc. thesis, Baghdad University,
Baghdad, Iraq.
110
Reference
 Reilly, W. R. and Gardner,(1977): "Technique for Measurement of
Delay at Intersections" ,Transportation Research Record 644, TRB,
National Research Council, C. C. ,Washington, D.C., pp: 1-7, USA.
 Rogers, M., (2008): "Highway Engineering", Second Edition,
Blackwell Publishing, Ireland.
 Sargeant S., Christie J., (2002): "Evaluation of a Series of
Roundabouts Implemented on South Golden Road", Institute of
Transportation Engineers, Annual Meeting, California,USA.
 Shamsul Hoqueand and Asif Imran, (2007): " Modification of
Webster’s delay formula under non-lane based heterogeneous road
traffic condition ", Journal of Civil Engineering (IEB).
 SIERPIŃSKI, (2007): "Proposition of Delay Model for Signalized
Intersection with Queueing Theory Analytical Models Usage ",
Transport Problems, Silesian University of Technology.
 Sofia, G.G. (1998):"Development of Traffic Flow Models at Networks
in Mosul City", Ph.D. thesis, Baghdad University.
 SORB, State Organization of Roads and Bridges, (2005): "Highway
Design Manual", Republic of Iraq, Ministry of Housing and
Construction, Baghdad, Iraq.
 Teply, S. (1989), "Accuracy of Delay Surveys at Signalized
Intersections", Transportation Research Record 1225, TRB, National
111
Reference
Research Council, Washington, D.C., pp: 24-32, USA. (Cited by Md.
Hadiuzzaman, 2008).
 "Traffic
Directorate
of
Sulaymaniyah
Governorate" ,( 2013)
Department of Statistics, Vehicle Registering File.
 Trafficware, Ltd., (2011): "Traffic Signal-User Guide, Synchro
Studio 8".
 TRB, Transportation Research Board, (2005):" Highway Capacity
Manual", Washington D.C, USA.
 Turley C., (2007): "Calibration Procedure for a Microscopic Traffic
Simulation Model", M.Sc. thesis, Brigham Young University, USA.
 Warid, A.K.(2004), "Evaluation and Improvement of Traffic Flow
Patterns at Al-Karrada CBD Area in Baghdad City", M.Sc. thesis,
university of Technology, Baghdad, Iraq.
 Webster, F.V. and Charles worth, G., (1958) "Some Factors Affecting
the Capacity of Intersections Controlled by Traffic Signals ", 4th
International Study Week in Traffic Engineering, Copenhagen,
Denmark. (Cited by Md. Hadiuzzaman, 2008).
 Webster, F.V. and Cobbe, B.M., (1966) "Traffic Signals". Road
Research Technical Paper No. 56, HMSO, London. (Cited by Md.
Hadiuzzaman, 2008).
112
Reference
 Youn-Soo Kang,(2000): " Delay, Stop and Queue Estimation for
Uniform and Random Traffic Arrivals at Fixed-Time Signalized
Intersections ", Ph.D. thesis, Virginia University, USA.
113
Appendix (A)
Appendix (A) Sample of Spot Speed Data
Table (A-1) Speed Data for Mawlawi Street
Vehicle No.
Time (sec)
Length (m)
Spot Speed (kph)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
2.95
3.1
2.8
2.85
2.7
3.05
3.4
3
3.25
3.2
2.85
3.05
3.05
2.7
3.5
3.2
2.85
3.1
3.1
3.8
3.55
2.75
3.95
3.65
3.1
2.85
3
3.2
3.95
3.45
3.6
3.65
3.2
3.3
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
30.51
29.03
32.14
31.58
33.33
29.51
26.47
30.00
27.69
28.13
31.58
29.51
29.51
33.33
25.71
28.13
31.58
29.03
29.03
23.68
25.35
32.73
22.78
24.66
29.03
31.58
30.00
28.13
22.78
26.09
25.00
24.66
28.13
27.27
A-1
Appendix (A)
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
3.25
2.8
3.75
3.3
3.95
3.7
2.8
4.05
3.6
3.45
3
3
3.1
3.35
2.75
3.6
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
Average = 28.09 km/hr
A-2
27.69
32.14
24.00
27.27
22.78
24.32
32.14
22.22
25.00
26.09
30.00
30.00
29.03
26.87
32.73
25.00
Appendix (B)
Appendix (B) W ork Sheet for Field Measurement of Delay
Intersection Control Delay W orksheet
Analyst : Saad Mohsen Khalil
Intersection:
Palace / Baban St.
Area Type :
CBD
Agency or Company : ( ‫)ـــ‬
Analysis Year : 2012
Date Performed : 10/1/2012
Analysis Time Period :
3:00 to
3:17
Input Initial Parameters
Number of Lanes , L : 1
Total vehicle arriving, Vtot = 189
Free – flow speed, FFS ( km / hr ) : 38 km / hr
Stopped vehicle count , V stop = 93
Survey count interval ( s ) : 15 sec
Input Field Data
Clock
Time
3.00
Cycle
Number
Count interval
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1
0
4
10
10
10
16
18
18
11
5
0
0
2
3
8
9
13
21
12
22
20
12
9
6
3
3
3
3
6
6
8
11
12
16
17
4
0
2
2
5
10
17
14
19
19
6
1
7
24
34
47
63
70
71
68
17
18
19
2
0
0
0
0
21
19
12
11
5
0
0
0
0
17
13
13
9
3
0
0
0
0
0
59
45
34
26
16
5
0
0
0
0
3.17
Total
Total vehicles in queue, ƩViq =585
Computations
No. of cycle surveyed, Nc =4
Time- in –queue per vehicle, dvq=(I*ƩViq/VTot)*0.9=41.78 sec
Fraction of vehicle stopping, FVS=Vstop/Vtotal= 0.492
Accel/Decel delay,dad =FVS*CF = - 0.492
Accel/Decel correction factor, CF =-1
No. of vehicles stopping per each cycle = Vstop/NcXL = 23.25
Control delay = dad + dvq =41.288 sec
B-1
Appendix (C)
A- Appendix (C) Sample of EVENT Program Output
This file has been opened with Microsoft Excel for counting traffic volume
for each turning movement, and its composition for Intersection.
Palace Intersection ( Left Turn of Salim Street )
Tuesday
3:00
10/1/2012
TO
3:15 pm
Volume = 206 veh
154.6703343d
157.3626403d
161.0989073d
164.3955993d
167.4725343d
169.6153873d
171.4285743d
174.4505463d
176.9230803d
179.2307743d
181.9780273d
184.9450533d
188.2417603d
190.4395603d
193.2967073d
195.8241733d
198.1868133d
200.1648413d
201.8681343d
204.7252813d
207.9670263d
210.9340673d
214.0109863d
217.2527473d
219.0109863d
221.0439613d
224.5054933d
227.1428533d
231.1538393d
234.2857213d
238.1868133d
239.7252813d
242.3076933d
244.5604403d
246.9230803d
C-1
Appendix (C)
249.0659333d
251.3186803d
255.4395603d
258.1867983d
259.8901063d
263.7362673d
265.9340523d
269.2857063d
272.6373603d
275.2747193d
278.3516543d
280.7142943d
283.5164793d
286.3736273d
289.3406683d
294.3955993d
298.6813053d
302.2527473d
439.7252813d
443.3516543d
445.7692263d
447.6923223d
450.1648253d
452.3626403d
455.6593323d
459.0109863d
462.3626403d
464.7802123d
468.2417603d
470.6044013d
473.6263733d
477.6373603d
479.7802123d
482.9120793d
485.9890143d
489.5054933d
492.5824283d
494.9450683d
497.7472533d
500.3846133d
502.7472533d
506.0439453d
509.6153873d
512.9121093d
515.1648563d
517.6923223d
520.6593633d
523.9011233d
527.5823973d
C-2
Appendix (C)
532.9121093d
544.3956303d
547.8571173d
550.4945073d
552.9670413d
556.2088013d
558.6264043d
561.4285893d
563.5164793d
565.6593633d
568.2417603d
570.4395753d
573.2417603d
575.7692263d
577.8021853d
579.7802123d
583.0219733d
584.3956303d
586.7033083d
589.2857063d
592.0878913d
596.1538703d
598.6264043d
601.4835213d
603.6813353d
607.3076783d
609.7802123d
611.6483763d
614.2857063d
618.0769043d
620.2197883d
622.0329593d
624.2307743d
654.5604253d
727.9121093d
731.4285893d
733.6813353d
737.0878913d
740.0000003d
742.5274663d
744.3956303d
750.2197883d
752.0329593d
754.4505623d
756.6483763d
759.1758423d
761.5384523d
763.8461303d
765.8791503d
C-3
Appendix (C)
769.2307743d
772.6923223d
775.8241583d
778.2417603d
781.2088013d
784.5604253d
787.2527473d
790.1648563d
794.0659183d
796.3735963d
797.1978153d
798.6813353d
801.5384523d
803.7911993d
805.9890143d
809.2857063d
811.7582403d
814.5604253d
816.9780273d
820.6593633d
822.9670413d
826.0439453d
830.1648563d
833.1868293d
835.8241583d
839.3956303d
842.3626103d
844.5604253d
847.0329593d
849.6153563d
852.3076783d
855.1648563d
858.0769043d
860.0000003d
862.0329593d
864.6153563d
866.7033083d
870.1648563d
871.8681033d
873.9560553d
876.8131713d
878.9560553d
881.1538703d
884.2857063d
886.9780273d
892.6923223d
893.5164793d
895.8241583d
898.0219733d
C-4
Appendix (C)
901.4835213d
904.4505623d
906.7582403d
908.4615483d
910.4395753d
912.1978153d
914.2857063d
915.9890143d
918.4066163d
921.9230963d
924.6153563d
926.3735963d
928.1318973d
930.2747193d
932.8021853d
934.8351443d
936.7582403d
941.0439453d
942.8571173d
944.7802123d
948.5714113d
951.4835213d
1014.0109863d
1028.0219733d
C-5
Appendix (C)
B- Procedure Shows Event Program Operation.
Event program
Press any key to continue
Press ESC button to start
Operate video tape on TV
Press Fi* buttons, when the front tire of the vehicles touches the stop line
At end, press ESC button to stop the program
A new window asks for the name of the file
Write a name and press enter to close the window
C-6
Appendix (D)
Appendix (D) SYNCHRO Output
SYNCHRO software provides different output results in graphical and
in tabular form. The main output used in present study is presented.
D-1
Appendix (D)
D-2
Appendix (D)
D-3
Appendix (D)
D-4
Appendix (D)
D-5
Appendix (D)
D-6
Appendix (D)
D-7
Appendix (D)
D-8
Appendix (D)
D-9
Appendix (D)
Time Space Diagram for Palace Intersection
D-10
Appendix (E)
Appendix (E) Traffic and Geometric Condition of Signalized
intersections
Table (E-1) Traffic and Geometric Condition for Delay Model
Cycle(sec)
v/c
g/C
Field
delay
(sec)
270
1.15
0.390
67.263
1.000
3.50
657
300
1.06
0.610
69.108
1.000
3.20
805
266
1.02
0.565
51.070
1.000
3.20
776
300
0.95
0.370
62.482
1.000
3.50
691
266
0.95
0.420
50.482
1.000
3.50
541
270
0.79
0.585
47.622
1.000
3.20
606
97
0.73
0.480
40.870
0.740
4.20
495
225
0.69
0.440
50.383
0.458
4.80
222
97
0.58
0.470
36.650
0.458
4.80
179
196
0.56
0.486
47.340
0.458
4.20
379
97
0.55
0.470
38.559
0.458
2.85
368
196
0.54
0.486
49.852
0.458
2.85
360
225
0.53
0.540
46.750
0.458
4.20
357
196
0.50
0.486
43.390
0.458
4.80
161
225
0.50
0.440
49.370
0.458
2.85
337
E-1
We/Ws Width(m)
Volume
(veh/hr)
Appendix (E)
Table (E-2) Traffic and Geometric Condition for Queue of Left Turn Movements.
Width
(m)
Cycle(sec)
v/c
g/C
Field Queue
Length(veh)
3.2
300
1.06
0.61
31.333
3.2
270
0.79
0.585
15.166
3.2
266
1.022
0.565
20.66
3.5
300
1.172
0.37
24.333
3.5
270
1.154
0.39
23.75
3.5
266
0.95
0.42
23.666
4.2
97
0.73
0.48
11.666
4.2
196
0.56
0.486
9.665
4.2
225
0.526
0.54
13.666
4.8
97
0.577
0.47
2.625
4.8
196
0.5
0.486
6.166
4.8
225
0.691
0.44
16.666
2.85
97
0.547
0.47
6.5
2.85
196
0.535
0.486
12.666
2.85
225
0.5
0.44
11.25
E-2
Appendix (E)
Table (E-3) Traffic and Geometric Condition for Queue of Through Turn
Movements.
g/C
v/c
Field Queue
Length(veh)
Cycle
(sec)
Width
(m)
0.59
1.06
28.33
300
3.2
0.61
0.80
12.75
270
3.2
0.59
0.87
19.25
266
3.2
0.39
1.14
26.67
300
3.5
0.39
1.30
21.25
270
3.5
0.39
0.90
20.67
266
3.5
0.49
0.47
6.00
97
3.2
0.50
0.51
4.40
196
3.2
0.49
0.53
15.67
225
3.2
0.49
0.65
1.88
97
4.8
0.50
0.15
3.00
196
4.8
E-3
‫المستخلص‬
‫( ‪ )% 10‬ليعكس بشكل افضل للقيم الحقلية‪ .‬الموديل المقترح للتأخير يبين عالقة جيدة مع القيم الحقلية ضمن‬
‫حدود هذه الدراسة‪.‬‬
‫تم اقتراح تحسينات لجريان المرور ضمن منطقة الدراسة في مدينة السليمانية من خالل تطبيق‬
‫برنامج ( ‪ )SYNCHRO/ SimTraffic‬المعدل‪ .‬هذه التحسينات تنقسم الى جزئين ‪ ،‬الجزء األول هو‬
‫إعادة ضبط التوقيتات والتنسيق بين التقاطعات اما الجزء الثاني يتضمن تحسينات هندسية‪ .‬هذه التحسينات‬
‫لها تأثير جيد جدا على مقاييس الفعالية للشبكة ضمن منطقة الدراسة‪. .‬واخيرا تم إيجاد مقاييس الفعالية‬
‫بواسطة برنامج (‪ )SYNCHRO/ SimTraffic‬بعد أخر التحسينات وللسنة المستقبلية ‪.7102‬‬
‫المستخلص‬
‫المستخلص‬
‫تعاني مدينة السليمانية من االزدحام المروري‪ ،‬تأخير المركبات على معظم الطرق والتقاطعات‬
‫وخصوصا خالل ساعات الذروة وهذا يسبب الكثير من المشاكل المروية ويؤثر سلبا ً على اقتصاد البلد‪.‬‬
‫زيادة عدد المركبات ومحدودية بناء طرق جديدة ستزيد من مشاكل الشبكة‪ .‬عملية التنبؤ الصحيح لتخمين‬
‫التأخير طوابير السيارات تصبح مهمه لالستخدام األمثل لشبكة الطرق الحالية بوساطة توفير أفضل عميلة‬
‫تشغيل وسيطرة مرورية‪.‬‬
‫الهدف من هذه الدراسة الحالية تقييم أداء حركة المرور للشبكة المختارة من خالل تقييم أداء‬
‫التقاطعات وتطوير نماذج للتأخير وطوابير السيارات في التقاطعات في حركة المرور بالنسبة للظروف‬
‫الحالية في مدينة السليمانية ويعقب ذلك اقتراحات لبعض مقترحات التحسين والتي تتضمن تغير خطة‬
‫التوقيت لإلشارة الضوئية وتحسين هندسة التقاطع وتغير نوع التقاطع‪.‬‬
‫لتحقيق الهدف المذكور اعاله ‪ ،‬تم جمع بيانات حركة المرور للشبكة الواقعة ضمن المنطقة التجارية‬
‫المركزية في مدينة السليمانية بوساطة استخدام كاميرا فيديو لجمع البيانات الحقلية ألنها توفر سجل دائم‬
‫للبيانات مع الحد األدنى للقوى العاملة‪ ,‬هذه البيانات تستخرج من تحليالت الفيديو بواسطة برنامج‬
‫(‪ )EVENT‬وتعالج بعد ذلك بواسطة أوراق (‪ )EXCEL‬معدة لهذا الغرض‪ .‬تم استخراج السرعة اللحظية‬
‫لكل أجزاء الشبكة بطريقة عالمات التبليط‪.‬‬
‫تم استخدام برنامج (‪ )SYNCHRO‬لتقييم وتحليل التقاطعات واقتراح افضل خطة توقيت لإلشارة‬
‫الضوئية والتنسيق بين التقاطعات وتم تقييم االقتراحات لسنة الهدف ‪ 7102‬باستخدام البرنامج المذكور‪.‬‬
‫النماذج اإلحصائية التي طورت لتخمين التأخير و طوابير السيارات بينت عالقة جيدة مع القيم‬
‫الحقلية‪.‬‬
‫نسبة العرض الكلي لمخرج التقاطع والمستخدم للمرور الى العرض الكلي للخطوط التي يتجه فيها‬
‫حركة المرور للمخرج نفسه وفي نفس الطور) ‪ (We/Ws‬أظهرت تأثير على التأخير‪ .‬وهذه النسبة يكون‬
‫لها تأثير سلبي عندما تقل عن ‪. 0‬‬
‫التأخير المخمن اعتمادا على )‪ (HCM 2000‬تم مقارنته مع التأخير الحقلي‪ .‬بناءاً على النتائج التي‬
‫تم الحصول عليها من التحليل فانه من األفضل تعديل نموذج (‪)HCM 2000‬‬
‫ليمثل التأخير الحقلي‬
‫وقد لوحظ ان نموذج )‪ (HCM 2000‬يستمر بحساب قيمة اعلى للتأخير عندما تكون درجة التشبع‬
‫اكبر من ‪ 0‬باالعتماد على نتائج التحليل يجب تخفيض التأخير النظري لل)‪(HCM 2000‬‬
‫بنسبة‬
‫جمهورية العراق‬
‫وزارة التعليم العالي و البحث العلمي‬
‫ألجامعة المستنصرية‬
‫كلية الهندسة‬
‫قسم هندسة الطرق والنقل‬
‫نمذجة طوابير السيارات والتأخير للمنطقة التجارية‬
‫المركزية في مدينة السليمانية‬
‫رسالة مقدمة إلى‬
‫قسم هندسة الطرق و النقل ‪ /‬كلية الهندسة ‪ /‬الجامعة المستنصرية‬
‫كجزء من متطلبات نيل درجة الماجستير في علوم هندسة الطرق و النقل‬
‫من قبل‪:‬‬
‫سعد محسن خليل‬
‫)بكالوريوس ‪ /‬قسم هندسة الطرق و النقل ‪) 0202،‬‬
‫بأشراف‬
‫أ‪.‬م‪.‬د غاندي غانم صوفيا‬
‫م‪.‬د علي جبار كاظم‬
‫أيلول ‪ 0202‬م‬
‫ذي القعدة ‪0323‬هـ‬