t L € k
13.9222“ has :25
4 ‘ l‘:<l 1‘
Efi‘zaf‘ai‘ ‘3‘"‘1 7“;
.mh‘sgiwlxwuflj‘é‘
‘ ::2§

nifikwx‘ziiiéé

W )‘
N“
5“ $133337. 2.2
.‘c‘ww. “ .
“SJ-i." ‘V-q
~r9~fi

" 9‘5} $3.:
~ I ‘ ‘2 L3 2.: «mm. 2 firgv 22:1
MN "m, :2 «wifi‘uwfia a?» , 2 x2 2 .21“: 2 ., ~22 ~
‘ #36514” 1211:1335 nggfqii‘ {3.5%.V1‘IHM?‘ & fi‘fl‘s “31.1 " ' 5;?“ ”it‘gé)’ §%J*w%:§&lv a. "‘ . ufig‘ meigcfii‘xfi’tmrx
w W‘fi“ :-°:’>~‘~»fiw“7‘:=5\"‘. $2.213: ”42’,“ $32 .‘WL. @3322 593% «‘2‘ Mn 2 _ . x2222 @222. $3}:
.2 .2 .- -' mi? 'P‘S. . va 5. . .1 ‘
h Wk gm 1212‘} 2~252§22~1~1’l%2‘:1{2_2r2 c' Rm ‘ fiéiwq‘f

3‘ ’ . $1449» 1%gi‘31 ”fix-XXV“ ‘ ‘1 in
(A? azcfififik “H. :‘t. 3.}: $153331 yfiéi “\‘k; 4 s gh-

“W; "W 4%“x$w§§fi 2 [Kr-Lg;

‘0
1}

u ‘ 2 'X - 22 . 'n - ‘l ' 2 ' ‘ >

L\ A. / ~ ‘ 3% . $45. § 3 k a 2‘ , r “2.3. M . I . A V. L «I.- .

3‘ 2: 3% 5““ 2. 2 [:22 :72 .- =§Ek a?» ‘ 2 m. . a 2. cxgg
a qua .2 ’ ‘. ‘ V N‘ x

L w~w
‘~>«‘~"::1“11*..‘:,~{ ,

V ' \\\§ .
s

I
q

t"; Lgfi‘ifr‘ .‘ fgl‘h‘ga
“€32 ~ @223
u . "‘d‘ V | '2 ‘Gfi
L~ ~ -.

‘ 1-2233: "“39: 2 ‘F
ME ~. ‘i‘utg‘figfxtd' \3%_1§
~: #3535
22‘; 12
.x 22L, wgkd; 2a
.231:- {‘sz “$155..

12““:‘3. Rd“ ‘L
.“1. 5’15“}. ‘12:.1: " :
13w x“ “in.“ W 4

W .
$.- ~‘:‘;

. .rr ‘
7113-;
1 vw-V‘An ‘ 2.:

. 5;: . ‘

"-"~’12*Z42..2_

~ , 5:2 1%};{1‘2 2%.“ . :-

2 $395 ~13 2 2
" v.

 

 

THESIS

Dunn-city

 

This is to certify that the
dissertation entitled
Analysis of the Applicability of a
Network Simulation Model to Traffic Performance

in the City of Jeddah, Saudi Arabia

presented by

Hamed O . Albar
has been accepted towards fulfillment

of the requirements for

Ph.D. degree in Civil Engineering

¢.'fic

Major profes or

Date December 19‘ 1984

lust/“mm. ,. .. r m, .. - ‘ 0-12771

 

llll/ll/llllllll/llill/llllll/lllll/I

3 1293 10594 0708

 

PV1ESIej RETURNING MATERIALS:
Place in book drop to
LIBRARIES remove this checkout from
.‘nu-q3-..-_ your record. FINES will

 

 

 

 

be charged if book is
returned after the date
stamped below.

 

 

WEB-M 257

”’1 “"1 '-
Xox vii?

 

 

 

 

ANALYSIS OF THE APPLICABILITY OF A NETWORK SIMULATION
MODEL TO TRAFFIC PERFORMANCE IN THE CITY OF

JEDDAH. SAUDI ARABIA

By

Hamed O. Albar

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Civil and Sanitary Engineering

1984

332—34C8

 

ABSTRACT

ANALYSIS OF THE APPLICABILITY OF A NETWORK SIMULATION
MODEL TO TRAFFIC PERFORMANCE IN THE CITY OF
JEOOAH, SAUDI ARABIA
BY

Hamed O. Albar

The practicing traffic engineer has long needed a problem-
solving aid to deal with the increasingly sophisticated and complex
urban traffic flow problem. To understand the behavior of an urban
street system and to evaluate various corrective strategies implemented
on such a system. one has to construct a model that best represents the
internal relationship among components and accurately predicts the
system performance. Due to the size of the urban street network and
the random nature among vehicles and drivers, it is impossible to use
an analytical approach to model such a system. On the other hand, a
simulation model becomes appealing in modeling the large urban network.
Furthermore, with the aid of modern digital computer technology, it is
economical and practical to apply digital computer simulation modeling
in solving vehicular movement problems on a large urban street network.

Among all network simulation models, NETSIM is the most widely
used and among the most extensively validated models. This research

was conducted to calibrate the NETSIM model to be used in analyzing the

Hamed O. Albar

traffic performance in Saudi Arabia. A calibration network was
selected in the city of Jeddah. and all the required input data were
collected from the field. Data on four selected measures of perform-
ance were collected. and the program was modified until the model
output matched these data. A validation network was then selected in
the same city. and the model performance was tested. It was found that
the traffic performance in Saudi Arabia can be simulated and analyzed

using a modified NETSIM model.

 

gr: tgc name. 0/04“ t/iz most mmi/Jand t/iz most fiancfécéent

ACKNOWLEDGMENTS

All praise and thanks are due to Allah. Lord of the Universe.
for His merciful divine direction throughout my study.

I wish to acknowledge all those persons who assisted me in the
undertaking and completion of this dissertation. I am indebted to
Professor William Taylor. my advisor and committee Chairman. for his
valuable time. assistance. and encouragement. His kind consideration
and understanding have been an incentive for the completion of this
dissertation.

Sincere appreciation and gratitude are extended to the other
members of my guidance committee. Drs. Thomas Maleck. K. Rajendra. and
it Salehi. for their contributions. advice. and constructive comments
to the study.

Finally. thanks are due to King Abdulaziz University for sup-
porting this research and to the municipality of Jeddah for its assist-

ance in the data-collection phase of this research.

LIST OF

LIST OF

Chapter

1.

TABLE OF CONTENTS

TABLES ...... . . ................
FIGURES . . . . . . ..... . . . . . ..... .
INTRODUCTION TO THE STUDY ..... . . . . . . . .
l. l Introduction . . . . . . . . . . . . . . . .
1.2 The Pr0b1em O O O O O 0 O O O 0 O O 0 0 O
l .3 Objectives of the Study . . . . . . .....
LITERATURE REVIEW

2.1 Why Simulation? . . . . . . . . . . . . . . . .
2.2 Classification of Models . . . . . ..... .
2.3 Traffic Simulation Models . . . . ......
2.4 Conclusions . . . . . . . . . . . . . .

THE NETSIM MODEL . . . . . . . . . . . . .......
3.l Model Structure . . . . . . . . . . . . . . . .
3.2 Input Requirements . . . . . . . . . . . . . . .
3.3 Output Characteristics . . . . .
3.4 User Options . ..... . . . . . . . . . . .
3.5 Limitations of the Model . . . . . . . . . . .
3.6 Computer Requirements . . . . . . . . . . . . . .
3.7 The Model and Traffic Performance in Jeddah . .

METHODOLOGY . . . . . . . . . . . . . . . . . . . . .

Network Selection and Coding . . . . . . .
Measures of Performance . . . . . . . . . .
Data Collection . . ......

Procedure and Schedule of Activities

h¥h$¢~
O C O
U'lwaA

Adaptation of the Model . . . . . . . . . . . . .

Page

vii

—I

TO
ll
l7

T8

18
21
24
25
26
27
27

30

3O
32
33
36
37

Page
5. DATA MALYSIS O O O O O O O O O O O O O O O O O O O O O 39

5.1 Data Reliability . . . . . . . . . . . . . . . . 39

5.2 Performance of the Current NETSIM Model . . . . . 43

5.3 Discussion of Program Subroutines . . . . . . . . 44

5.4 Embedded Data 0 O O O O O O O O O O O O O O O O 47

5 OS The MOde1 O O O O O O O O O O O O O O O O O O 64

5.6 Model Validation . . . . . . . . . . . . . . . . 64

5. Application of the Model . . . . . . . . . . 71

6. CONCLUSIONS AND RECOMMENDATIONS . . . . . . . . . . . . 76

6.1 Conclusions . . . . . . . . . . . . . . . . . . . 76

6.2 Recommendations . . . . ........ . . . . . 77

APPENDICES O O O O O O O O O O O O O O O O O O O O O O 0 O O O O 80
A. STATISTICS ON DRIVER AND TRAFFIC CHARACTERISTICS

IN JEDDAH O O O O O O O O O O O O O O O O O O O 0 O O 81

B. PROGRAM OUTPUT OF CALIBRATION NETWORK . . . . . . . . . 86

C. PROGRAM OUTPUT OF VALIDATION NETWORK . . . . . . . . . . TOO

BIBLIOGRAWY O O O O O O O O O O O O O O O O ........ O O 115

LIST OF TABLES

Table Page
1.1 Vehicle and Accident Statistics in Jeddah . . . . . . . . 3

1.2 Vehicle and Accident Statistics in Kuwait and the
United States . . . . . . . . . . . . . . . . . . . . . 3

1.3 Number of Traffic Accidents in Jeddah (1976-1981) by

Time and Place of Accident . . . . . . . . . . . . . . 5
1.4 Number of Accidents by Type in Jeddah (1978-1981) . . . . 5
5.1 Travel Time (V-sec) in the Calibration Network . . . . . 41
5.2 Average Speed (mph) in the Calibration Network . . . . . 41

5.3 Average Delay Time per Vehicle in the Calibration
Network 0 O O O O I 0 O O O O O O O O O O O O O O O O O 42

5.4 Observed Cycle Failures per One-Half Hour in the

Calibration Network . . . . . . . . . . . . . . . . . . 42
5.5 T-Test of Selected Links Using the Current Model . . . . 45
5.6 Left-Turning and Right-Turning Vehicles in the

Calibration Network . . . . . . . . . . . . . . . . . . 51
5.7 A Comparison of Two Mean Queue Departure Headways . . . . 55

5.8 A Comparison of Lost-Time Delay of First Queued

Vehicles . . . . . . . . . . . . . . . . . . . . . . . 58
5.9 The Simulated MOP Using the Modified Model . . . . . . . 62
5.10 Travel Time (sec) in the Validation Network . . . . . . . 65
5.11 Average Speed (mph) in the Validation Network . . . . . 68
5.12 Average Delay Time per Vehicle in the Validation

Network . . . . . . . . . . . . . . . . . . . . . . . . 68
5.13 Cycle Failure of Second Network . . . . . . . . . . . . . 69

5.14 The Simulated MOP of the Second Network . . . . . .

5.15 A Comparison Between Simulated Values of MOP Using Two

A1

A2

(3

Different Timing Plans . . . . . . . . . . . . . . .

Vehicles and Licenses in Jeddah (1971-1981) . . . .

Some Characteristics of Drivers Involved in Accidents

in Jeddah (1978-1981) 0 o o o o o o o o o o o o 0
Number and Causes of Accidents in Jeddah (1978-1981)

Number and Type of Vehicles Involved in Accidents
1" Jeddah (1978-1981) 0 o o o o o o o o o o o o 0

Number and Type of Accidents in Jeddah (1971-1981)

vi

Page
70

75

82

83

84

84

85

 

LIST OF FIGURES

Figure Page
3.1 NETSIM Model System . . . . . . . . . . . . . . . . . . 20
3.2 Logic Flow for NETSIM Executive Routine . . . . . . . . 22
3.3 Major Features of NETSIM Model . . . . . . . . . . . . 23
4.1 Physical Network #1 . . . . . . . . . . . . . . . . . . 34
4.2 Coded Network #1 . . . . . . . . . . . . . . . . . . . 35
5.1 Logical Structure of NETSIM Network Simulation Model . 46
5.2 Overlay Structure for NETSIM Pre-processor. Simulator.

and Post-Processor . . . . . . . . . . . . . . . . . 48
5.3 Speed-Volume Relationship for the Original Model . . . 60
5.4 Speed—Volume Relationship for the Original and

Modified Models . . . . . . . . . . . . . . . . . . . 63
5.5 Physical Network #2 . . . . . . . . . . . . . . . . . . 66
5.6 Coded Network #2 . . . . . . . . . . . . . . . . . . . 67
5.7 Khalid Bin Al-Walid Street in Jeddah . . . . . . . . . 72

5.8 Time-Space Diagram for Khalid Bin Al-Walid Street
Using the Modified Plan . . . . . . . . . . . . . . . 73

5.9 Time-Space Diagram for Khalid Bin Al-Walid Street
Using the Existing Timing Plan . . . . . . . . . . . 74

 

CHAPTER 1
INTRODUCTION TO THE STUDY

lal__lnI£QdU£IiQn

In the early 19705. as the price of oil started to increase
rapidly. oil revenue began to grow and Saudi Arabia became one of the
rich states in the Middle East. Therefore. the government initiated an
ambitious five-year development plan whose objective was to Change the
country from a pre—industrial society to a modern industrialized
country. This rapid development exerted pressure on all public
utilities and facilities including the transportation system in the
country as a whole. and in major cities such as Jeddah in particular.

Jeddah. the second major City in Saudi Arabia. after the
capital. Rijadh. lies in the West province by the Red Sea. It is
midway between Aden and Suez. at the hub of a major highway system. It
serves a dual function. It is the main commercial port for the western
part of the country and also a chief point of entry for pilgrims to
Makkah from throughout the Moslem World. It also has the biggest and
busiest airport in the country. It is a major commercial and economic
activity center. A substantial majority of all bank offices in the
Western Region are in Jeddah. There are several small and medium-sized
factories located in the industrial zone in the city. Jeddah is also

the diplomatic center of the kingdom where all the embassies and

 

consulates except the Ministry of Foreign Affairs are located. The
city has one of the major universities in the country. In the last
seven years the city has experienced a remarkable growth rate. Its
population has increased from about 600.000 in 1974 to approximately
one million persons in 1981. The metropolitan area is about 100 square
miles (35).

This expansion in area and population has influenced the
traffic performance in the City. While there is considerable traffic
between Jeddah and other Cities and rural areas. like most cities. the
road system is most congested during peak periods. Table 1.1 shows
that the number of vehicles registered increased 52 times between 1971
and 1981 to reach 690.073 vehicles (34). By 1991. the number of regis-
tered vehicles in Jeddah is anticipated to increase nearly fourfold for
the low population forecast and nearly ninefold for the high population
forecast (base year is 1974) (39L

The study of motor vehicle accidents in Jeddah between 1971 and
1981 shows that the number of accidents increased from 347 in 1971
to 2.530 in 1981. an increase of about 730 percent. In 1981. there
were slightly less than four accidents per thousand vehicles and 2.997
injuries. a rate of 1.18 injuries per accident.

The number of persons killed in traffic accidents. as shown in
Table 1.1. increased from 75 in 1971 to 323 in 1981. The fatality
rates (number of fatalities per 1.000 vehicles) in 1980 and 1981 in
Jeddah were 0.56 and 0.50. respectively. Table 1.2 shows that these

rates in the United States were 0.31 and 0.30. respectively. The table

Table 1.l.--Vehicle and accident statistics in Jeddah.

 

 

Fatalities Injured No. of

Year No. of No. of per 1.000 Number per 1.000 Acci-
Vehicles Fatalities Vehicles Injured Vehicles dents

1971 13.217 75 5.7 394 30 347
1972 25.096 65 2.6 543 22 576
1973 40.950 159 3.9 1.282 31 1.081
1974 72.269 142 2.0 1.959 27 1.531
1975 113.224 206 1.8 2.790 25 2.160
1976 185.545 287 1.5 3.340 18 2.779
1977 264.266 285 1.1 2.410 9 2.341
1978 383.108 295 0.77 3.270 9 2.607
1979 475.425 341 0.72 3.439 7 2.809
1980 602.639 342 0.56 3.387 6 2.732
1981 690.073 323 0.50 2.997 4 2.530

 

Table 1.2.--Vehic1e and accident statistics in Kuwait and the United

 

 

States.
Fatalities Injured
Year No. of No. of per 1.000 Number per 1.000
Vehicles Fatalities Vehicles Injured Vehicles
1971 158.446 233 1.5 2.718 17
1972 175.526 253 1.4 2.869 16
1973 197.777 231 1.2 2.902 15
1974 223.788 304 1.4 2.944 13
1975 272.232 367 1.3 3.168 12
1976 320.656 307 0.96 3.545 11
1977 379.101 321 0.85 3.702 10
1978 439.553 361 0.82 3.588 8
1980 164.852.000 51.077 0.31 ... .
1981 165.732.000 49.268 0.30 ..

 

 

shows that the fatality and injury rates in Jeddah are also higher
than those in Kuwait (41). a developing country close to Saudi Arabia.

In another comparison. the number of traffic fatalities per
1.000 population in the United States was 0.23 in 1980 (36). in Jeddah
it was 0.43. and in Saudi Arabia it was 0.48. Table 1.3 shows that
most of the accidents occur in the city. and Table LA shows that most
of the accidents are either run into vehicles (multiple vehicle) or run
on humans (pedestrian accidents). The high rate of these types of
accidents may be due to inefficient signal timing and a lack of coordi-
nation between signals.

There are approximately 120 traffic signals in the City of
Jeddah. and there is an average of 20 accidents per month at these
signalized intersections (49). All the signals are fixed time (no
actuated). and there are no progressive systems in the city. There are
no published studies on delay and congestion at intersections. but
experience indicates that many major intersections are congested and
oversaturated. .Appendix A shows more detailed statistics on driver and

traffic Characteristics in Jeddah.

W

The expansion of many cities in the world has made the daily
movement of people and goods an increasingly complex problem. Since
cities depend largely on their street systems for transportation
services. the traffic engineer has the responsibility of optimizing

traffic flow and safety for the benefit of the population.

Table 1.3.--Number of traffic accidents in Jeddah (1976-1981) by time
and place of accident.

 

Out of
Year Day % Evening % In City % City %

 

1976 1.494 54 1.285 46 2.375 85 404 15

1977 1.512 65 829 35 1.864 80 477 20
1978 1.740 67 867 33 2.140 82 467 18
1979 1.943 69 866 31 2.274 81 535 19
1980 1.813 66 919 34 2.411 88 321 12
1981 1.447 57 1.083 43 1.956 77 574 23

 

Table l.4.--Number of accidents by type in Jeddah (1978-1981).

 

 

Run Into Run On Run G0 Off
Year Vehicle Other Human Animal Fire Down Road Other
1978 1.169 142 849 5 2 344 5 91
1979 1.297 90 914 4 3 417 13 71
1980 1.169 133 987 .. 1 331 16 95
1981 1.077 125 843 6 17 335 8 119

 

The evaluation of comprehensive street system improvements is
complicated by the large number of alternatives available. the inter-
relationships among the design variables. and the infeasibility of
conducting large-scale experiments to test design options. In most
cases. limitations of time and cost. together with the need to avoid
undue disturbance of existing traffic movements. make extensive field
experimentation impossible. In addition. the consequences of experi-

mentation can include accidents. injuries. and even human life.

The introduction of computer-based mathematical simulation
models has enabled traffic engineers to determine the effectiveness of
proposed changes in the transportation system without actually imple-
menting and testing them. The digital computer is particularly effec-
tive in providing the medium for exercising traffic simulation models
and their interaction with external management and control measures.
Thus it provides the analyst with a very convenient laboratory for
experimentation. evaluation. and design.

These simulation models are designed to represent the behavior
of the physical system if all the variables affecting the traffic
system are identified within the model. Such variables include: road
and intersection geometrics. traffic flow and volume. speed and turning
movements. type of control. timing plans. and traffic composition.

In developed countries. such as the United States. these
variables are generally easy to measure. In fact. there are
numerous studies that have used the most sophisticated equipment
and methods to collect and gather data to be used in the models. In
addition. computerized filing systems are available to recall detailed
historical traffic data. In contrast. in developing countries such as
Saudi Arabia. the data-collection system is undeveloped and very few
traffic studies have been conducted. Most of the data collection is
done manually.

Since driver performance characteristics differ from one
society to another. depending upon their experience with modern

technology. traffic variables such as headway distributions. gap

acceptance. turning speeds. and signal phase response will also be
different. The pedestrian behavior also differs between Saudi Arabia
and the United States. as people are not accustomed to crossing the
streets at intersections only or when the signal permits. Pedestrian
conflicts are more prevalent. and greater delay in moving in the
network is expected.

Therefore. to use existing simulation models for analyzing and
improving traffic performance. at least calibration of the data is
needed. if not further modification and variation in the simulation

program.

W

This research is designed to achieve the following objectives:

1. To explore similarities and differences in traffic per-
formance on urban networks between the United States and Saudi Arabia.

2. ‘To assess the applicability of the NETSIM model to Saudi
Arabia.

3. To adapt NETSIM or another network simulation model for use
in the analysis of the traffic performance in Jeddah.

4. To collect data on a limited street network in Jeddah to
use as input data for a computer simulation of the network and to
collect additional data on a different network to test the accuracy of
the simulation model.

5. To conduct a parametric analysis on the simulation model to
determine which internal relationships (if any) need to be modified to

calibrate the model for Saudi Arabian conditions.

CHAPTER 2

LITERATURE REVIEW

The use of traffic simulation models in analyzing traffic
performance has been the subject of extensive research. in spite of the
relatively short age of these models. As the models have evolved and
become more reliable and sophisticated. their use in more complex
traffic situations has also increased. Among all traffic simulation
models. network simulation models are the most widely used. This
literature review was conducted to determine the reliability and
applicability of the network simulation models in general and the
NETSIM model in particular in analyzing traffic performance in the
United States. and developing the hypotheses to be tested in this

study.

LJJhLfimuhliQfll

Since the beginning of traffic engineering. one of the most
demanding problems has been predicting. in quantitative terms. the
effects of various traffic control strategies on real traffic. This
problem has not been easy to solve because traffic is a complex
phenomenon. difficult to characterize numerically. Mathematical models

adequately describing highly idealized and simplified conditions were

 

developed. but these early models could not portray real-world traffic
accurately.

Attention soon turned to discrete event simulation. a promising
technique that uses logic and analytical and empirical relationships to
analyze the behavior of complex traffic systems. ‘The advantages of
simulation techniques are: ‘

1. They provide the analyst with a means of addressing complex
"systems" problems made up of many interrelated parts. each of which is
subject to considerable variability.

2. They permit the engineer to focus on specific portions of
an overall problem. under conditions of at least partial "experimental
control."

3. They allow the user to experiment freely with new ideas
before committing the financial resources necessary to implement them
in the field.

4. They are generally considerably quicker. more flexible. and
less expensive than other forms of complex. analytical evaluation.

The main shortcoming of the simulation technique was the
overwhelming number of computations required to represent the many
interrelated events that take place in traffic. Therefore. traffic
could not be simulated in a practical manner until the digital computer
with its unprecedented computational speed was developed. Shortly
after the introduction of early computers in the mid-19505. traffic

simulation models. in the form of elaborate computer programs. began to

10

be created to represent single intersections. short sections of

freeways. urban arterials. and even urban networks.

Wei;

The network simulation models can be classified as either
microscopic or macroscopic in design. Macroscopic models represent the
traffic stream in some aggregate form (e4i. employing a fluid flow
analogy or a statistical representation). Daniel Gerlough (11)
developed one of the first macroscopic network simulation models in
1960. He used many approximations which made the model rather rough
and the evaluation of the effectiveness of various Changes imprecise.
James Kell (12) in the 19605 developed several specific intersection
simulations models. ‘These models dealt only with two-lane roadways and
thus had limited applicability. W. B. Cronje (22) developed a model for
the optimization of fixed-time signalized intersections in 1981 which
was applicable to undersaturated as well as oversaturated conditions.

Microscopic models describe the detailed. time-varying trajec-
tories of individual vehicles in the traffic stream. They represent
the ultimate in detailed treatment; Each vehicle is identified and its
position. speed. and acceleration are kept in memory.

Some authors (18) have identified a third class of models. the
platoon models. These models are a half-step toward detailed realism
and simulate the behavior of vehicles grouped into platoons whose
location. speed. and acceleration are tracked by the program. Platoon
speed is usually a function only of the general density of vehicles in

the platoon. thus avoiding complicated car-following calculations.

 

11

The macroscopic models offer the advantage of lower computa-
tional cost. while the microscopic models are. in general. more accu-
rate because they make fewer assumptions. However. their requirements
for computer resources retarded their development in times when these
resources were very limited. The advent of the third-generation com-
puters in the mid-19605 made possible the development of microscopic

models such as UTCS-l. which later became NETSIM (15).

2i1__I£iffi§_51mulniien_MQd§l§

Currently. there are three classes of traffic simulation
models: single road. single intersection. and network models. Among
the single road models. freeway models used to study merging. ramp
metering. the effect of traffic composition. and incident detection
phenomena are becoming comnmwu Hsu and Munjal (42) have prepared a
review of single road freeway models. and May (50) provides a compre-
hensive survey of models for freeway corridor analysis. including their
historical development and applications.

Single intersection models have generally been built for a
specific purpose and are not widely applicable. Perhaps Websterks is
the best-known example Of a single intersection model. This model was
used to study the effects of isolated traffic control signals on
intersection delay (18).

Network models are more complex. Some represent surface
streets only; others can include freeway networks. 'These models are

very useful in testing signal control strategies. traffic diversion

 

12

strategies. proposals to add or delete streets from a network. and
similar network modifications (40).

Gibson and Ross (8. 18. 40) reviewed 19 network simulation
models. ‘They reported that ten of them are obsolete and three are
limited-application signal optimization programs. Their conclusion was
that the other six traffic simulation models have been a success and
their users were generally pleased with the results obtained.

Gibson (44) provided a catalog of 104 documented computer
models for traffic operations analysis. The models were classified
according to the geometrics of the application (intersections.
arterials. networks. freeways. and corridors). Only ten of these
models were considered practical in the sense that they produce useful
results. The models are:

1. SOAP .. . .. . .. . .. . intersection optimization
2. TEXAS . . . . . . . . . detailed intersection simulation
3. PASSER II . . . . . . . . . . . . arterial Optimization
4. PASSER III . . . . . . diamond interchange optimization
5’5. SUB . . . . . . . . . . . . . . arterial bus simulation
6. TRANSYT-7F . . . . . . . . . . . . network optimization
7. SIGPO III . . . . . . . . . . . . . network optimization
8. NETSIM . . . . . . . . . . . . . . . network simulation
9. PRIFRE . . . . . . . . . . . . . . freeway optimization

10. FREOBCP . . . . . . . . . . . . . . . freeway simulation

MacGowan and Fullerton (10) have traced the evolution and

accomplishments of the Urban Traffic Control System (UTCSL The

13

initial objective of UTCS was to develop advanced operational control
programs» The project objective was later expanded to include develop-
ment and testing of control strategies using simulation techniques;
testing of the strategies in a real-life environment test facility in
Washington. DAL; and improvement of performance evaluation techniques
for measuring the efficiency of the new strategies. To test and evalu-
ate these alternative network strategies. an analytical model was
needed. FHWA sponsored the development of such a model. which was
originally designated UTCS-1 because of its relation to the UTCS proj-
ect but was renamed NETSIM.

UTCS-1/NETSIM was developed by Peat. Marwick. Mitchell. and
Company and GASL. It is based on the DYNET model and is fully
microscopic. The NETSIM model has been validated against field data
collected in Washington. 0J1. Utah. California. and New Jersey. Among
all the network simulation models. NETSIM is the most widely used. It
has been used successfully in numerous applications throughout the
country.

Hagerty and Maleck with the Michigan OCT (3) have used NETSIM
in analyzing geometric and signal system alternatives. It has also
been used to evaluate corridors at the transportation planning level
and to evaluate signal installation requests.

Labrum (4) described the experience with NETSIM studies at the
Utah DOT. They have used it extensively to evaluate traffic control
strategies for single intersections. arterials. and grid networks. as

well as to analyze pedestrian control problems. bus system plans. and

14

fuel consumption and emission rates. 'They have found that "the NETSIM
model is a very useful tool in solving a wide variety of traffic
control problems."

Hurley and Radwan (5) have used NETSIM for research in a
university environment. Most of the research described analyzes the
effects of traffic signal timing on fuel consumption and vehicle delay.
They concluded that to use this model as a research tool. improvements
in these components of the program logic and program documentation
are needed.

Nemeth and Mekemson (13. 37) compared NETSIM and SOAP in
analyzing pretimed and actuated signal controls at intersections. The
results of their studies indicated that both methods are reliable.

Although the UTCS-l model was originally developed to simulate
an urban network. its detailed treatment of intersection behavior in
addition to its great flexibility makes it an appropriate candidate for
a single intersection simulation model. Cohen (17) has modified and
validated the UTCS-15 model for use in the analysis of traffic per-
formance of single urban intersections. The modified model has been
successfully tested and compared to two other single intersection
simulation models.

Bruce Schafer (51) has used NETSIM in a comparison of
alternative traffic control strategies at a Teintersection. His
opinion was that "the NETSIM computer simulation model further expands
the traffic engineerts ability to analyze and evaluate alternatives in

a cost-effective manner."

15

Davis and Ryan (38) have compared NETSIM results with field
observations and Webster predictions for isolated intersections. They
found "no significant difference between NETSIM results and field
observations or the Webster technique for the condition simulated."

Berg and others (32) have used NETSIM to evaluate signal timing
plans for an oversaturated street network. After calibration of the
model. they reported that they were able to select the best plan.
saving a considerable amount of manpower and several months of field
observations.

Hani Mahmassani and others (52) used NETSIM in an exploratory
study of network-level relationships arising in an isolated network
with a fixed number of vehicles. The results were analyzed with
respect to the study objectives. yielding useful insights into network-
level traffic phenomena and suggesting some modifications in order to
use the NETSIM model in analyzing such problems. Their suggestions
were:

The introduction of short and long term rare events and blockages.
in addition to heavy vehicles. pedestrian interference. driveways
and parking maneuvers is likely to improve the realism of this
representation. However more fundamental modifications in the car-
following and lane-switching procedures embedded in NETSIM may be
required.

To enhance the NETSIM program. Hurley. Radwan. and Benenelli
(24) modified an existing fuel-consumption model in a form that is
suitable for insertion into the NETSIM program.

To reduce some of the difficulties associated with the NETSIM

model. such as extensive data preparation. tedious debugging. and

16

voluminous printouts. Chin and Eiger (31) developed a network
simulation interactive computer graphics program (NETSIM/ICC).
Current development in network simulation modelling is being
done by the Office of Research of FHWA.(25). They are developing a
system of traffic simulation models named TRAF. This system is
designed to represent traffic flow on any existing highway facility.
It will consist of both microscopic and macroscopic model components
for urban networks and freeways and a microscopic component only for
two-lane rural roads. NETSIM is among the components that are being
integrated into TRAF.
Regarding the application and use of NETSIM in locations other
than the United States. the investigator found only one paper. by Yagar
and Case (47). that summarizes the evaluation of NETSIM in Toronto.
The version of UTCS-l used did not have provision for changing splits.
offsets. or cycle length from one subinterval to the next. in order to
study different signal control plans between subintervals. ‘To accomp-
lish this. another subroutine similar to PRSIG (where signal codes are
primed initially) was added to the program. and changes were performed
on routine UPSIG. They concluded that:
The above modifications performed by a person who had not developed
the original UTCS-l model demonstrated that the model can be made
to perform the types of operations required of it with some
intimate knowledge of the program and its routines.

Although traffic conditions and driver characteristics in the United

States and Canada are similar. modifications in the program were

needed.

17

2.1mm:

The conclusions from the above background review are the
following:

1. Traffic simulation is an important. reliable tool for
traffic engineers and transportation planners.

2. Among all network simulation models. NETSIM is the most
widely used and among the most extensively validated models.

3. Due to the relatively recent development of NETSIM. it has
not been used widely other than in the United States. The application
and use of the model in different societies and locations may enhance

the program.

CHAPTER 3

THE NETSIM MODEL

One of the major objectives of this project is to assess the
applicability of the NETSIM model to Saudi Arabia. Since the NETSIM
model utilizes certain embedded values in simulating a network. it is
important that the effect of these values be understood. Therefore. a
brief description of the model has been summarized from the NETSIM User

Guide (FHWA. 1980>(I).

Lil—Medelfimmre

The NETSIM model was designed primarily to assist in the
development and evaluation of relatively complex network control
strategies under conditions of heavy traffic flow. It is particularly
appropriate to the analysis of dynamically controlled traffic signal
systems based on real-time surveillance of traffic movements. The
model may also be used. however. to address a variety of other simpler
problems. including the effectiveness of conventional traffic
engineering measures. bus priority systems. and a full range of
standard fixed-time and vehicle-actuated signal control strategies.

It is set in a flexible. modular format which permits its

efficient application to a wide variety of design problems. It

18

19

includes a set of "default" values for most input parameters. thereby
avoiding the need for detailed calibration in a particular area.

The model is based on a microscopic simulation of individual
vehicle trajectories as they move through a street network. Each
vehicle in the system is treated separately during the simulation. .An
array of performance characteristics is stochastically assigned to each
vehicle as it enters the network. and its behavior is governed by a set
of microscopic car-following. queue discharge. and lane-switching
rules. All vehicles are processed once every second and their time-
space trajectory recorded to a resolution of 0.1 second.

The NETSIM model is based. in part. on two earlier network
simulation models: the "DYNET" model developed by E. Lieberman and an
earlier predecessor model. "TRANS" developed by D. Gerlough and
F. Wagner. All three formulations describe a street network in terms
of a series of interconnected links and nodes. along which traffic is
processed in a series of short time-steps subject to the imposition of
varying forms of traffic control. The major differences among the
models are in their level of detail. the sophistication of their inter-
nal logic. and their capacity to respond accurately to widely varying
traffic conditions and increasingly complex control schemes.

NETSIM is the most detailed and complex of the three. The
model is divided into three major components or "modules" (see
Figure 3.1):

Module #1-—"NETSIM Pre-Processor"

Module #2--"NETSIM Simulator"
Module #3--"NETSIM Post-processor"

20

.Emomsm _oeos z_memz are--._.m «Lemma

m_m>.._<z<

FDR—.30
4<opmr~<hm

ZOZ.<JD$:m

mhmOawm

07502030

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

53805 was: mwwwwsfia «8309: V. 8.80

5 53:5 .
0a 2552 my: 5%.
i] r

 

 

 

 

 

21

The "NETSIM Pre—Processor" is designed to simplify the process
of preparing and checking data inputs. It includes a comprehensive set
of automatic "diagnostic checks" which are performed on all data
inputs. It may be operated independently. or it may be integrated
directly with the main program.

Ther"NETSIM Simulator" contains the main simulation program.

It consists of 74 separate routines. which may be linked together in a
variety of optional configurations depending on the requirements of the
user. The simulator requires as input a coded description of a street
network. together with a pre-specified control plan and a set of input
volumes. Its output includes a set of standard measures of traffic
performance. expressed as both link-specific and network-wide values.

The "NETSIM Post-Processor" consists of a set of standard data
manipulation and evaluation routines designed to operate on the outputs
of the main simulation program to compare the results of two or more
simulation runs. construct a "historical" data file summarizing their
results. and subject the resultant data set to a set of standard
statistical analyses. Figure 2 illustrates the logical flow for the
main executive routine within the "NETSIM Simulator." Figure 3 shows

the major features of the NETSIM model.

3.2—MW
The input data required to operate the model may be grouped

under two separate headings: location-specific parameters. reflecting
the particular characteristics of a given network link or intersection.

and phenomenological measures. which are assumed to remain constant

22?

 

Main Program, SIMUL

.1

(29 as: Read Title Card

J.

Clear all COMMON Storage Arrays: then
prime with embedded calibration data

1

; Read (and write) all Link Cards and
perform all Diagnostic Checks

1

Process inputs and perform additional tests

1

if any options are requested, read (and write) the
necessary input data and perform diagnostic tests

.1

Read Signal Cards and check (first subinterval, only)

1

Read all Flow Rate Cards and check data

.1.

Execute program to initialize system (first subinterval)

J.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Execute central control loop to perform traffic simulation for this subinterval.
Increment the clock in one-second intervals.

1. MOVE Routine: Process all vehicles that were queued at beginning of
time-step.

2. MOOV Routine: Process all remaining vehicles on network.

3. SVEH Routine: Emit new vehicles along entry links in accordance with
specified flow rates.

A. PRKD Routine: Emit new vehicles from intra-link source nodes as per
specified flow rates.

5. UPSIG Routine: Update all signals.
6. CLNUP Routine: Accumulate statistics and perform diagnostics.
7. Print statistical results as specified.

 

 

.1.

Subinterval complete

 

 

 

 

(-

 

Write results

 

 

 

New Subinterval Case Completed

 

4.—

C0

Figure 3.2.--Logic flow for NETSIM executive routine.

23

. MICROSCOPIC. STOCHASTIC SIMULATION OF INDIVIDUAL
VEHICLE MOVEMENTS

. SIMULATION OF FULL RANGE OF CONTROL FEATURES.
INCLUDING:

. "Stop" and "yield" signs

Turn controls

Parking controls
Fixed-time signals
Vehicle-actuated signals

Real-time traffic control and surveil-
lance systems

. MODULAR STRUCTURE INCORPORATING DETAILED
TREATMENT OF:

Car following behavior

Network geometry

Grades

Bus traffic

Queue formation

Intersection discharge

Intra-link friction and mid-block blockages

Pedestrian-vehicular conflicts

PROVISION FOR FLEXIBLE MIX OF STANDARD OUTPUT

MEASURES

Figure 3.3.--Major features of NETSIM model.

24

across the entire network to be simulated. regardless of the location
of a specific link or intersection. Specifically. the following input
data are required:

1. For each network link: number of moving lanes. length.
capacity of left-turn pocket. desired free-flow speed. mean queue
discharge rate. turning movements at downstream node. identification of
receiving links. pedestrian volume. and lane Channelization.

2. At each intersection: complete specification of signal (or
sign) control. including sequence and duration of each phase and
identification of signal facing each approach.

3. Traffic demand: specified as flow rate (VPH). percentage
of trucks admitted onto the network along input (entry) links and from
internal source nodes. and rate of extraction of vehicles at sink
nodes.

4. Duration of simulation subintervals and specification of
output options.

5. As an option. specification of bus system (routes.
stations. mean headways. and mean dwell times) and frequency and

duration of events.

LLDutpuLCbaLaetecjsflcs
The NETSIM model provides a comprehensive range of output to
describe the input data set. the status of the simulation exercise. and
results of the simulation. The input data set for each subinterval is

printed out in tabular form. The card listing of the input data is

25

also optionally printed out at the beginning of the simulation. All
input data are checked for completeness and consistency by the prepro-
cessor component of the model. When errors are found. execution is
aborted. and appropriate error messages are printed.

A comprehensive set of traffic performance measures (MOP) is
generated either as standard output at the end of each subinterval or
as intermediate output at the option of the user. Most of the MOP are
produced for each individual link and for the network as a whole. The
intermediate output option provides additional detailed information for
individual links. These data are useful in the analysis of microscopic
traffic behavior over time.

The postprocessor component of the model provides detailed
comparisons of the traffic performance measures generated during two
separate model runs. These comparisons are made for each individual
link and the network as a whole. for each time period (subinterval).
and for the entire duration of the simulation. The individual link and
network-wide measures are statistically analyzed with the paired-
comparison t-test. the Wilcoxin signed-rank test. the Mann-Whitney
u-test. and the one-way analysis of variance to determine the level of

significance of the difference.

34&__U§e£_92119n§

A number of major user options are included in the phase II
model. These options are:
1. Diagnostic checking of multiple input files with optional

execution.

26

2. Storing and revision of input data sets on tape for future
recall and update.

3. Modification of embedded input parameters. using study
format data cards. (This is one of the options used in this study.)

4. Storing of model output on tape for future statistical
module.

5. Simulation of traffic actuated signal control.

6. Simulation of a surveillance system comprising various
types of detectors.

7. Simulation of bus traffic.

8. Simulation of transient blockages within the traffic
stream.

9. A variety of standard output options. including tabulation
of origin-destination volumes.

10. Automatic system initialization.

11. Statistical analysis of model outputs.

12. Peripheral data activities.

W21

The present form of the model has certain limitations. Some of
these limitations are as follows:

1. Not all of the currently available real time control algo-
rithms can be modeled.

2. Capacity constraints include a maximum of 160 links. 99

nodes. 60 entry links. and 1.600 vehicles tracked within the system.

27

3. Extensive checking of input data validity and careful
review of output results for reasonableness are required. Erroneous
conclusions can be reached due to undetected errors in input coding or
careless construction of the link-node diagram.

4. Nonstandard traffic situations require the user to trans-
form the conditions into acceptable input. Unusual driver performance.
road geometrics. or pedestrian crossings require appropriate transfor-

mation.

Wants

Computer requirements for running the model are a function of
several factors: the size of the test network. the level of traffic
flow. the type and frequency of output reports. and the desired number
and length of simulation runs. The entire model is programmed in
FORTRAN IV. and it is fully operational on either IBM 05/360/370 or

CDC 6600 series computers.

WW
To use the NETSIM model in traffic engineering. it is not

enough to know the model structure and features. In addition. the
traffic conditions and circumstances of the site that would be analyzed
by the model should be well understood. There are several major
differences in traffic performance between the United States. where the
model was developed. and Saudi Arabia. These differences are in driver
characteristics as well as in the traffic systems. ‘They can be

summarized as follows:

28

I. .inxe£_Chana§I§£1§Ii§§

a. Due to social security. and safety factors. the law in
Saudi Arabia prohibits females from driving any vehicle.

b. The law also prohibits drinking alcohol. Therefore. few
accidents and traffic violations are caused by this factor.

c. More than one-third of the drivers don’t have a drivers
license (34). and they are not aware of most traffic regulations.

d. More than one-third of the drivers are illiterate or
uneducated (34).

9. Experience shows that violation of traffic regulations
(such as signals. stop signs. and high speed) is common and expected.

f. There are more pedestrian conflicts and delays due to
multiple pedestrian crossings in all streets.

9. Experience also shows that many drivers are accepting
smaller gaps for turning at intersections. and they have a different
response to the amber phase than drivers in the United States.

it It is common to see vehicles cluster at major intersections
instead of having a regular queue.

i. It is very comnmniin Saudi Arabia for drivers to use the
horn to accelerate traffic at intersections when the light turns green.
This habit may affect queue discharge headways and lost time of first-

queued vehicles.

29

II. W

a. At all signalized intersections. each approach has a
separate green phase. Thus. there are no conflicts with opposing
traffic. and the consideration of left-turn jumpers and acceptable gaps
in opposing traffic is negated.

b. All signals are controlled by a fixed time system.

c. Published data on specific traffic parameters such as
headways. acceptable gaps. and distributions are missing.

d. Police enforcement is weak.

e. The cycle length of many signals exceeds 100 seconds.

All of these differences may affect in one way or another the
program output. Therefore. a modification of the program may be
needed. which can be done through the embedded data cards. as will be

explained in the following chapters.

CHAPTER 4
METHODOLOGY

The methodology used in this research utilized the NETSIM model
to analyze the traffic performance in Jeddah. This required the
collection of input data. a comparison of model output with comparable
field measurements. model calibration. and validation of the modified

model in a second network.

iJ—Afiammmniflm

Simulation is essentially a simplification of a real-world
situation. The results obtained from any simulation model are only as
good as the model reflects the particular real-world characteristics of
interest to the analyst. The treatment of nonstandard traffic condi-
tions may require the analyst to represent the essential operational
Characteristics of the network by an appropriate set of quantified.
coded inputs.

Most of the embedded values used in the original NETSIM model
reflect drivers' performance and traffic conditions as observed in
Washington. ELC. These were expected to be different in Jeddah.
Therefore. to facilitate the application of the NETSIM program to the
analysis of traffic performance in Jeddah. calibration and/or modifica-
tion was considered for the following program imbedded driver-related

parameters:

30

31

1. Left-turn jumpers

2. Amber-phase response

3. Acceptable gaps at a STOP sign

4. Intersection turning speeds

5. Lane-switching acceptable lag

6. Acceptable gaps for left-turning vehicles at intersections

7. Distribution of spillback probabilities

8. Delay due to pedestrian conflict

9. Vehicle desired free-flow speed

10. Vehicle queue discharge headways
11. Lost time of first queued vehicle

These parameters are treated in the program as embedded data
inputs. They represent default conditions that may be used to repre-
sent driving behavior either in the absence of adequate data or to
reduce the preparation of data inputs. These values embedded within
the standard load module reflect driver performance as observed in
Washington. DJ; Updating of these data is provided in the program by
means of "embedded data change cards)‘

The determination of appropriate values for each of these
imbedded variables and the determination of the sensitivity of model
performance to differences in these values is not an objective of this
study. Instead. a systematic variation in the value of selected embed-
ded values was conducted. and the model output compared to field obser-
vations for selected measures of performance. When the deviation

between the model results and the field observations was within the

32

specified tolerance. the parameter was fixed. After calibration of the
model. its reliability and applicability to Saudi Arabia was tested on
another network in Jeddah. The parametric analysis of the model is

presented in the next chapter.

W

The task of selecting a suitable network site in Jeddah was not
an easy one. There were many construction and development projects in
the city. and the geometric design. signal control. and markings on the
streets in these areas were not normal. It was also difficult to
define boundaries of a chosen network due to the planning of the city
and the zigzagging of many streets.

After studying a city map and touring the city for several
days. a network that includes two major arterials and several
signalized and unsignalized intersections was selected to serve as the
study area and data-collection site. It is located in an intermediate
site between the CBD and the suburbs. To assure that site is a fair
representative of the traffic conditions of the city. the following
criteria were used in selecting the site:

1. The site should contain traffic signals. STOP signs. one-
and two-way streets. etc.

2. ‘The site should not contain major abnormal or atypical
network geometry that would cause major problems of data collection or
model formulation.

3. The area should be traversed by trucks as well as private

automobiles.

33

4. The site should include a representative range of typical
link and intersection geometrics.

-Figure 4.1 shows the physical network with the names of streets
and signals at intersections. ‘This network was coded according to the
NETSIM program. and a link-node diagram was constructed to represent
' the actual street network in the computer. Figure 442 shows the coded
network. Links may be either entry/exit type or internal to the net-
work.) Nodes are points at which vehicles enter. exit. or are con-
trolled. such as a signalized intersection. The diagram shows also the

direction of movement in the network.

Me 5 5 Fe

The NETSIM program is capable of reporting detailed data on 16
separate performance measures for each individual link and 12 separate
measures for the network as a whole. Among these performance measures.
only the following were selected as MOP for this study:

1. Travel time: average travel time per vehicle per link per
simulation interval. in veh. seconds (total duration. running and
stopped time).

2. Delay: average delay time per vehicle. by link. per
simulation interval. in veh. seconds (route delay not intersection
delay. difference between total travel time and "idealized" travel
time).

3. Speed: average traffic speed per link per simulation

interval. in mph.

311

 

 

 

*

A St. (Thu-Annoorayen Street)

 

 

3rd St. (Al-Tawbah Street)
4th St. (King Fahd Street)

 

 

 

 

 

 

 

 

* ist St. (Khalid Bin Al-Waiid Street) *

 

 

 

 

 

 

 

 

 

 

 

 

.....J
* B St. (AI-lskan Street) *
—- a
2
55
co
1:
3
T:
3
:75
'o
.._._1 5
* >1: C St. (Mohamed Suroor Ai-Saban Street) :14

 

 

—1 l

at traffic light at intersection

 

 

 

Figure 4.l.--Physical network #1.

35

 

 

°""’ 6

A St.

 

entry

 

1 st St.

exit

.-°“

entry

6 exit

.1

YA

 

 

 

 

 

BSt.
o n

 

 

00

a ,._. a
U x CI!
5 o 5
804

Q

 

Figure h.2.--Coded network #1.

 

36

4. Cycle failure: Total number of cycle failures. by link.
during simulation interval. defined as the number of times a queue
fails to clear from the discharge end of the link during a green

period.

4 4 D o

The data-collection phase was a major step in this project.
The NETSIM program requires extensive data collection to be used as a
full set of exogenous inputs. There are not enough historical traffic
data from the city of Jeddah. and it was almost impossible to obtain
all the necessary data from the files. ‘Therefore. a series of manual
field counts. signal checks. and network inventories was done. Ten
students majoring in civil engineering at King Abdulaziz University in
Jeddah assisted in the data-collection phase. Training sessions for
those students were offered and designed to provide them with an over-
all understanding of the traffic parameters needed to be collected and
the collection methods. They were provided with field sheets and stop
watches.

The data collected from the field can be grouped as follows:

1. For each link: length. number of moving lanes. length of
left- and right-turn pockets. pedestrian volume. turning movements at
the downstream mode. and lane Channelization.

2. For each intersection: type of control (STOP. YIELD. or
signal). sequence and duration of each phase. and identification of

signal facing each approach.

37

3. Flow rates (vph). and percentage of trucks admitted onto
the network along input (entry) links.

4. For comparison of MOP. the data were gathered on: travel
time. delay. speed. and cycle failure at the specified intersections
and links.

The computer output. in Appendix B. shows a summary of all the
input data specified for each link and each intersection approach in

the network.

Aa5__ELQQeflnEe_anidflflnafl£UiJHLficifixliles

The data-collection phase of this project started in the middle
Of April 1983 and lasted until the first week of June 1983. During
this period. the weather was normal and school vacations had not yet
started. and there were no abnormal conditions that might have affected
traffic volume or other traffic parameters.

Due to the different social circumstances in Jeddah. there are
three rush-hour count periods on weekdays as follows: from 7130-8a30
a.m.. from 1:30-2:30 p.m.. and from 5:30-6:30 p.m. A lS-minute count
interval was used at intersections.

The first week was devoted to selecting and coding the network.
The second week was a training period on field data collection for the
ten participants. Four participants were assigned at major intersec-
tions to collect data on traffic volume. turning movements. traffic
composition. and pedestrian movements. 'Two participants were assigned
to collect data on network geometrics. and two were assigned to collect

the required data at nonsignalized intersections. ‘Two persons with a

38

stop watch determined the cycle lengths and phasing of signals. ‘The
values of the MOP at the selected locations were determined by the
researcher and two students.

All the counts and readings were repeated at least three times.
and their averages were used in the program. Link lengths were deter-
mined from aerial photographs obtained from the municipality of Jeddah.
The same procedure was repeated for selecting. coding. and collecting

data for the second network.

CHAPTER 5

DATA ANALYSIS

The 1982 version of the NETSIM computer program was used in
this reseamdt Before using the program. its performance was checked
by running the program with input from the sample problem stored on the
program tape. The output was compared with the stored output and found

to be exactly the same except for some insignificant round-off errors.

5 ] D | B ]' I'l'l

Data reliability is essential to this experiment. since the
modifications in the program to suit Saudi Arabian conditions will be
based on these data. Data collected from the field for this project
are of two types:

1. Input data. These data include link length. number of
moving lanes. length and capacity of left-turn pockets. lane Channeli—
zation. signal phases and cycle length. pedestrian volume. turning
movements. flow rates. and percentage of trucks.

2. Output data. These data were collected for the test
networks to be used in calibrating and validating the model changes.
These measures of performance (MOP) include average travel time per
vehicle per link. average traffic speed per link. average delay time

per vehicle per link. and cycle failure by link.

39

40

The eight most heavily congested links of the 25 links in the
calibration network were selected for use in this study. For the first
two MOP. the moving-car technique was used to obtain the data. A total
of eight runs were made on each link. and the means and standard

deviations were calculated using the formulas:

n
i S =/Z (X3702
1:] °=I '

n n-l

 

>4
ii

M:

x

where X = sample mean
X1 = "i" th measurement
n = number of runs
S = standard deviation

Tables 5.1 and 542 show the individual data and the computed means and
standard deviations of the travel time and average speed of the
selected links in the calibration network.

To calculate the average delay time per vehicle. by link. the

following equation is used:

average delay ._ average travel _ "idealized" travel

time per vehicle time time

link length
desired speed

 

= average travel time

Table 5.3 shows the average delay time per vehicle for these same links.

41

Table 5.1.--Travel time (V-sec) in the calibration network.

 

 

 

.Link ........
Run #
1.2 2.1 1.9 2.11 11.2 3.4 4.3 9.8
l 23 36 56 65 59 59 51 29
2 19 31 61 61 62 54 42 33
3 17 25 65 55 67 58 44 26
4 22 39 54 62 60 51 49 30
5 21 33 59 57 65 57 55 25
6 16 40 66 66 58 52 45 33
7 15 36 6O 64 68 61 54 35
8 21 27 53 58 63 51 41 27
2 X1 154 267 473 488 502 443 381 238
X’ 19.3 33.4 59.1 61.0 62.8 55.4 47.6 29.7
S 2.96 5.42 4.78 4.00 3.69 3.89 5.40 3.66

 

Table 5.2.--Average speed (mph) in the calibration network.

 

 

 

Link
Run # -~
1.2 2.1 1.9 2.11 11.2 3.4 4.3 9.8
l 15 10 20 14 19 17 19 18
2 20 13 16 18 17 20 25 13
3 22 15 14 21 13 18 24 21
4 16 7 21 l6 18 22 20 15
5 17 12 18 20 14 19 16 22
6 22 7 13 13 21 21 23 13
7 23 9 19 15 12 14 17 12
8 17 14 22 l9 16 22 26 20
2.54 152 87 143 136 130 153 170 134
X 19.0 10.9 17.8 17.0 16.3 19.1 21.3 16.7
S 3.12 3.09 3.27 2.93 3.11 2.75 3.77 3.99

 

42

Table 5.3.--Average delay time per vehicle in the calibration network.

 

Link Desired Average Link Length Delay in

 

 

Link Length Speed Travel Time Desired Speed X sec/veh
# in ft. in MPH in sec 1.47

1.2 540 25 19.3 14.7 4.6
2.1 520 25 33.4 14.1 19.3
1.9 1620 25 59.1 44.1 15.0
2.11 1600 25 61.0 43.5 17.5
11.2 1580 25 62.8 43.0 19.8
3.4 1520 35 55.4 29.5 25.9
4.3 1520 35 47.6 29.5 18.1
9.8 620 35 29.7 12.1 17.6

 

Cycle failure is defined as the number of times a queue of
vehicles fails to clear from the discharge end of the link during a
green phase. 'This measure was obtained by counting this failure at
signalized intersections for a period of 30 minutes. Table 5.4 shows

the total number of cycle failures for the test links.

Table 5.4.--Observed cycle failures per one-half hour in the
calibration network.

 

Total Number of

 

Link Cycle Failures
2.1 2
1.9 1
3.4 4
4.3 5
9.8 2

 

43

Wang

The initial step used to determine the required modifications
to make the NETSIM program fit Saudi traffic conditions was to simulate
the given network by the model as used in the United States. By
comparing simulated and measured values of the measures of performance.
it was anticipated that specific modifications could be determined.
Therefore. a simulation run was conducted for the calibration network.
The run simulated a 30-minute time period. Equilibrium was attained
for this stimulation period. and the average values for the MOP's were
recorded.

To compare the measured and simulated output of travel time and

average speed. the t-test was used:

= (7- u)
S/fh'

student's t-test

where t

7': mean of measurements

U = simulated value

S = standard deviation of measurements

n = number of runs
For a 90 percent confidence interval and seven degrees of freedom. the
table value of t 15.1.895. If the computed t < 1.895. the hypothesis
that there is no difference between the measured value and the
simulated value will be accepted; otherwise. this hypothesis will be

rejected.

44

Table 5J5 shows the results for the selected links. The
simulation values on three links are within acceptable limits. while
the other five are not. indicating that a modification in the program is
needed. It does not appear that a change in a single parameter will be
sufficient. since the simulation results are not uniformly high or low
for the network. Thus. there are probably differences in more than one
factor causing the differences in the measured and simulated values of

the MOP's.

W
The 1982 NETSIM model software contains a total of 11 programs.

90 subroutines. and 4 block data for fuel consumption and vehicle-
emissions computations. There is a specific function for each program
or subroutine. and they are related to each other according to their
functions in the simulation process. as shown in Figure 55L The main
executive program is UTCS-1. This program reads the link cards. sets
up the initial data storage for further processing. tests whether
cumulative and intermediate outputs are requested. activates those
subroutines that initialize the contents of the COMMON storage arrays.
and reads the remaining input data. All traffic characteristics and
relationships assumed in the model are stored in these programs and
subroutines.

The TRVL subroutine. for example. calculates the acceleration
or deceleration of a vehicle; distance traveled; new speed; whether it
will enter a left-turn pocket. join a queue. or switch lanes; come to a

halt before a signal or travel through an intersection; whether it will

45

r.

 

 

 

N o e.~_ m.~_ mwm._ a.e_ m.m_ m_o.N a.mm ..NN w.m
m o ..m. m.__ Noe.m m._N m.e_ mes.~ e.as m.~m m.:
s _ m.m~ m.w~ mem.~ ..m. w.e_ mmo.~ a.mm N.wm s.m
- - w.m_ m.m_ mwa. m.e_ _.a_ mem. w.~e M.Ne N.__
- - m.a_ m.m mma.s 0.5. o.m~ Mme.m o._e m.as __.N
_ o o.m_ N.s_ mew. m.a_ m.w_ amw. _.mm a.mm m._
N _ m.m_ m.om asa.~ m.o_ m.m emc.e a.mm a.ms ..N
- - e.s m.m ems. o.m_ w.m_ mew. m.m_ e.w_ ~._
eabcsoo emww_ seam. seam. . mmmn_u eats seam. . mmmn_u eats embm_ xc_s
- .m -su_mo -se_m 1AA mo V a ..mo: -se_m 1AA mo V b -mmaz -se_m
ec:__ma so>\oom In: own _%
o_o>u >m_oo nooam ommco>< oe_h _o>mck xcozuoz

 

 

 

 

._ovoe beeccso osu m:_m: mx:__ evade—mm mo ammuihau.m.m o_nmp

46

..ovoe co_um_:E_m xcozuoc z_mhmz mo ecsuoscum _mu_m04-i._.m ecammu

 

 

  

I min... .

 

 

finne- .:§. .895 .08 .anFu .85... .955... .33..
352.8 .368 .32.. £2! .80 65.8.. .555... .535

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

47

be stopped by a bus at a station. or by a blocked vehicle; and applies
delay due to short-term events. A user-specified speed profile and a
stochastically determined desired speed provide the bases for the

vehicle's trajectory. Figure 5.2 shows the overlay structure for the

NETSIM program.

544__Embedded_0ata
The set of embedded data used in NETSIM includes a total of 20

separate parameters. They reflect both a series of microscopic
performance characteristics and a lesser number of simple network
characteristics. These data may be revised through the use of speci-
fied card types. Values and applicability of these parameters to the
case study are as follows:

1. Left-turn jumpers: A left-turn jumper is a vehicle that is
first in queue when the signal changes to green. and executes the left-
turn maneuver (immediately) before the oncoming queues can discharge.
The program includes an embedded value of.38 as the mean probability
of a lead left-turn jumping (JMPG).

The traffic signal phasing used in Saudi Arabia prevents left-
turn jumpers by assigning a separate signal timing for each approach.
thus eliminating conflicts with on-coming traffic. To verify that this
variable was being used appropriately. a run was done using a value of
JMPG(I)=0. As expected. there was no change in the model output.
indicating that the program is not using this parameter due to the

characteristics of the described network.

’48

.cOmmooocouumOQ ucm .coum_:E_m .cOmmooocaoLa z_mhmz com ecauoscum >m_co>o-|.u.m Cc:m_u

AvOuuoaaso uozv

 

 

AEdumoum :wmzv ommmmm_
11. _

dance:
uOmmOOOumoumom

r1

 

 

AUNL
UHm<
Nhfiuc
Emma
BU<UmQ
Nh2¢m9
UHQOm
UHmsz

904m:
Zmfimo
Nh<zom
Nh<om
mmHUFQ
BU<OHm
B<Qwam
mfimmbm
mHQmDm
FmBAHm
zmmfim

:ozm
emmmm
muse
aqoa
somemo
use:
one
e<meme
ano<
Queue
mzaq

BmBZH
=U3ma

A>¢B
ZMEKA

ADZAU
UHmm:
mmmxm
=m>m
>002

 

FU<2H
930904

>mmfio

xmmfio

hhmfio

804mm Emomqh
meUD thmza
Dmmmbm ZH>KDW
ozmzmm AQ<¢AU
UHmmm HBZH

 

 

Afimumoum :fimxv ADsz

 

 

 

AsmwoOum cause oedema

 

 

.

manna: heydasfiam

 

 

mzua
me9<Z
03mm
UHm¥2A
mDmUBD

 

 

ammuoum zozzoo

 

 

wasooz nowa>uwmsm

q

mascot uOmnououmioum

L

 

 

49

2. Amber phase response: The response of the lead moving
vehicle in a lane that has no queue at the instant the signal turns
amber. to the onset of the amber signal. is expressed in terms of an

acceptable deceleration rate as follows:

where: I is the decile index of individual driver characterist cs
and d is the assigned acceptable deceleration in ft/sec .
Changes in this parameter would have a minor (and uniform)
effect on the travel time MOP. Since the differences between observed
and simulated results are not uniform. no change was made to this
variable.
3. Acceptable gaps at a STOP sign: The stored decile dis-
tribution of acceptable gaps for near side (or one-way) cross-street

traffic is:

I 1 2 3 4 5 6 7 8 9 10
g 56 50 46 42 39 37 34 3O 29 20
Where: I is the decile index of individual driver Characteris-
tics and g is the acceptable gaps in sec x 10.
To account for the time required for the entering vehicle to

find an acceptable gap in the traffic stream on the far side. the

following additional time is applied:

50

I l 2 3 4 5 6 7 8 9 10
g 12 21 26 31 35 39 42 46 49 51
where: I is the decile index of individual driver character-
istics and g is the assigned acceptable gaps in sec x 10.
The selected network at Jeddah has only two STOP signs. and they are
located on relatively uncongested links. These links were not used in
the comparison of MOP's; thus no Changes were made to this variable.
4. Turning speeds: Moving vehicles unimpeded by others must
slow as they approach an intersection if they are to negotiate a
turning maneuver. These speeds. applied deterministically. are:
Left-turn speed. ILT = 22 ft/sec
Right-turn speed. IRT = 13 ft/sec
Since the effect of changing this parameter would be a function of the
left- and right-turning volume at each intersection it would not be
uniform. Thus it is a candidate as a parameter to be changed in the
calibration of the model. The number of left-turning and right-turning
vehicles on each of the links used in the calibration was recorded to
determine if there was a relationship between these volumes and the
travel time deviations. Table 5.6 contains these data. There does not
appear to be any relationship between the travel-time differences and
the turning volumes. Thus. no changes were made to this variable.
5. Lane-switching acceptable lag: A vehicle cannot switch
lanes unless an acceptable lag is available in the target lane. This

value. deterministically applied. is IALAG = 31 tenths of a second.

51

This value has also not been modified since the effect of any change

would be uniform across all links.

Table 5.6.--Left-turning and right-turning vehicles in the calibration

 

 

network.
Link % % %
Simulated Measured Difference Left Turns Right Turns

1.2 18.6 19.3 4% 0% 15%
2.1 45.0 33.4 -25% 100% 0%
1.9 58.7 59.1 2% 14% 0%
2.11 47.3 61.0 30% 0% 0%
11.2 62.3 62.8 1% 52% 48%
3.4 58.2 55.4 - 6% I 0% 5%
4.3 52.3 47.6 -10% 11% 0%
9.8 27.1 29.7 10% 6% 3%

 

6. Acceptable gaps for left-turning vehicles: A decile
distribution of acceptable gaps in the on-coming traffic facing left-
turning vehicles is stored in the IGAP array. These values. in tenths

of a second. are:

I 1 2 3 4 5 6 7 8 9 10

g 78 66 6O 54 48 45 42 39 36 27

As explained in "Left-Turn Jumpers." this parameter is not applicable
in Saudi Arabia because there is no on-coming traffic at signalized

intersections.

52

7. Mean effective vehicle lengths:

Autos: VLNGTH (l) = 20 (feet)
Trucks: VLNGTH (2) = 37
Buses: VLNGTH (3) = 50

These values are appropriate for Saudi Arabia. and thus the default
values remained constant through the study.

8. Probability of a vehicle joining (or causing) spillback:
The probability. in percentage. of a vehicle joining a spillback

comprised of I vehicles is defined in the SPLPCT array:

I 1 2 3 4

SPLPCT 100 81 69 40

These probabilities are reasonable for Saudi conditions. and no Change
was made to this variable.

9. Delay due to pedestrian conflict: The program defined two
types of conflict: strong (or heavy) interaction at the beginning of
the green phase. and weak (or light) interaction for the remaining
duration of the green phase. The duration of vehicular delay. in
seconds. for each kind of conflict is defined by a statistical decile

distribution stored in the PDLY array:

53

where I is the decile index of individual driver characteristics
and d] is the duration of vehicular delay for weak
interaction and d2 is the duration of vehicular delay
for strong interaction in seconds.

Light pedestrian flow: PPER (l) = 0
Moderate pedestrian flow: PPER (2) = 10
Heavy pedestrian flow: PPER (3) = 25

Since the ability to choose the combination of pedestrian flow (PPEN)
permits the analyst to vary the pedestrian-delay component. no changes
were made in the (PDLY) array.

10. Vehicle desired free-flow speed: As each vehicle enters a
link. it is assigned a free-flow speed. This assignment is obtained by
applying a percentage factor to the free-flow speed specified for that

link. The embedded percentage values are:

I l 2 3 4 5 6 7 8 9 10
F 75 81 91 94 97 100 107 111 117 127
where I is the decile index of individual driver characteris-
tics and F is the assigned percentage for free-flow speed.
These values were used without any changes because the analyst can vary
the specified free-flow speed to change the travel time and average

speed on any link. This is one of the variables used in the calibra-

tion process.

54

11. Vehicle queue discharge headways: As each queued vehicle
moves up to the stop-line. it is assigned a delay until discharge.
reflecting the queue discharge headway. This headway is obtained by
multiplying the mean queue discharge headway specified for the link by

the following percentage according to the decile distribution:

K 1 2 3 4 5 6 7 8 9 10

F 170 120 120 110 100 100 90 70 70 50

where K is the decile index of individual driver characteristics
and F is the assigned percentage for discharge headway.

For the second and third vehicle in queue at the time the
signal turns green. additional delays of 0.5 seconds and 0.2 seconds.
respectively. are added to the headway.

If the vehicle is not an automobile. the value is multiplied by
a factor of'L6 to reflect the more sluggish operating characteristics
of trucks and buses.

Two runs were conducted assuming mean headway values of 2.2 and
2.0 seconds for all links. Table 5.7 shows a comparison of these two
runs with the observed values for the calibration network. For a mean
headway of 242 sec. the simulated travel time of four links and the
average speed of three links were within acceptable limits. while for a
mean headway of Zilsec. the simulated travel time and average speed of
only two links were within these limits. Therefore. a mean headway of

2.2 seconds was used for the remainder of the runs.

55

 

 

 

 

 

 

. m.m mma.m o.m~ :mN.m N.~: o.~

- m5 of 0.5 :.N
. m.m mom.m m.mN mNm.N_ ..N: ~.~
o m.m_ omm._ ..ON moo._ m.nm o.N

_ of Q: ..mm m...
o o.:_ Nmm. m.m_ mmm. ~.wm N.N
o _.m~ mwm._ w.m :mm.m :.o: o.~

N CS 3: ...mm ..N
o ~.mN omm.N m.m mmm.q m.~: ~.N
I o.m New. m.m_ _m_. ..m. o.~

- e... 0.9 Wm. N;
. m.m omo. m.m_ mum. m.w_ ~.N
oeuczou ovum. vmum_ ovum. mmm._ out: team. mmw._ out: oeum_ cow

u:E_m nae—mu -:E_m "An.mo.vu Immoz -:E_m "An.mo.vu ammo: I:E_m
>m3omo: %
ec:__wu Lo>\UOm :az dew cmoz xc_4
o_u>u >m_oo cooam ommco>< oe_h _o>mch

 

 

 

 

 

 

.m>mzomu; ucaucmooo ozone cmoe ozu mo cOm_cmoeoo <1-.m.m o_nmh

 

 

 

S6

 

 

 

. a.m. ~mm.m o.~. mmm.~ ..mm o.~
N e.a. 5.0. a.mN m.m
o N.w_ mmm.~ ..m. mmm. ~.om N.N
o o... Nom.m m.m_ om_.N m._m o.~
m ..w. m..~ e.as m.s
. ...m. mtg ...... aid 93 ~.~
o a.mw mo_.N 0.5. .o:.~ N.wm o.~
s a.mm ..m. s.mm :.m
_ m.w~ :ww.~ m.o_ mmm.N N.mm N.N
u m.m_ Nwm._ w.m_ NNN.N a.mm o.N
- w.m. m.e. m.~e N...
s ..w_ omw. N.m_ m_o. a.mm N.N
toum_ oeum_ beam. mmw._ Cot: oeum_ mmw._ out: team—
couc:0u -:E_m Tao—mu -:E_m "Am.mo.vu Immoz i:E_w uA~.mo.vu immoz I:E_m vow %
men—.md co>\oom :a: bum >Mwwmox x:_4
o_u>u >m_oo comam ommco>< OE_P _o>mch

 

 

 

 

 

 

.easc.ocoo--.a.m e.nme

57

12. Lost time of first queued vehicle: The first vehicle in
queue when the signal turns green suffers a (start-up) lost time. This
lost time can be applied deterministically by specifying its value on
the link card. or it can be extracted from a decile distribution stored
in the program. The mean of the stored values is 2.6 seconds. To test
the sensitivity of the model results to the parameter. deterministic
values of 2.4 and 2.2 sec were used in consecutive runs. Table 5.8
shows a comparison of these two runs. Using a value for lost time of
244 sec. simulated travel time of five links and average speed of four
links were within acceptable limits. while for a value of lost time of
2L2 sec. the simulated travel time of four links and average speed of
three links were within these limits. Therefore. a lost-time delay of
244 sec would appear to be the most appropriate value for first-queued
vehicles in Jeddah. This value is lower than the 245 average found in
the United States and reflects the aggressive nature of drivers in
Saudi Arabia.

These tests of variations in the embedded parameters used in
the NETSIM model calibrated for the United States accomplished two
things. First. the best estimates of mean queue headway and lost-time
delay to be used in Saudi Arabia were determined. Changing these
parameters resulted in an increase in the number of links with accept-
able travel-time deviation from three to five and the number of links
with acceptable average speeds from three to four. Second. it demon-
strated that other Changes in the program are required. as there remain

large differences between simulated and calculated delay time for

“fix ‘-

 

58

 

 

 

 

 

 

I m.m omm.m N.:N oum.__ a.ma N.~
- 0.: 0.: 0.5 ...N
u m.m mmw.m _.MN mom.m 0.5: :.~
0 0.0. Nmm. m.m. 0N0. m.mm N.N
. 0.0. 0.: ..mm 0..
o m.:_ mom. m.w_ m_:. w.mm :.N
o ..NN mo_.N o.w mam.m m.Nm N.N
N 0.0. 0.0. 0.00 ..N
0 0..N 000.. ..m .N0.. 0.00 :.N
.. m.m mom. ...m. mum. 5w. N.N
- 0... 0.0. 0.0. N..
u ...m mam. 0.9 R... 0.0. ...N
boom. ovum. .ooum_ mmw._ not: team. mmw._ not: coun—
ooucaoQ T:E_m use—mu u:E_m "Am.mo.vu immoz -:E_m "Am.mo.va immoz I:E_m Cum %
ecu—.mu ;o>\OOm Ia: OOm oe_h xcmo
o_o>u >m_oo venom ommco>< we.» _o>mch awe;

 

 

 

 

 

 

.mo_o_;o> oozonc umc.m mo >m_oo oe_uuumo_ Co com_cmaeoo <in.m.m o_nmh

59

 

 

 

 

 

 

_ 0.0— mou.~ m.m_ _m_._ ~.wN N.N
N 0.N. N.0. N.0N 0.0
o w.n_ NN_.N m.m_ :00. m.w~ :.N
o N... mum.: N.m_ a.m.m m.:m N.N
0 ..0. 0..N 0.N0 0.0
_ ..m. mmm.: m.:_ o_m.: N.mm :.N
o m.mN mmm.m o.w_ o__.N m.wm N.N
0 0.0N ..0. 0.00 0.0
_ w.m~ wmo.m m.w_ ouo.~ o.mm :.N
I N._N mum. o.m_ Nom._ ..mo N.N
- 0.0. 0.0. 0.N0 N...
I ..m. Nm_. m.m_ om:._ 0.:0 :.N
oeucon ovum. oeum_ edge. mmm._ tot: beam. mmw._ out: team. cam
i:E_m use—mu |:E_m "Am.mo.vu Immox -:E_m "Am.mo.vu Immoz I:E_m %
mes—.md LO>\UOm Id: bow mew x:_4
o_u>u >m_oo vooam ommco>< 08.0 _o>mch

 

 

 

 

 

 

.0asc.bco0--.0.0 a.0...»

60

several links. Also. these parameter changes did not improve the
comparison of simulated and counted-cycle failures.

Both Tables 5.7 and 5.8 show that the simulated average speed
is slower than the measured speed for each of the links that is outside
the acceptable limits. These links are also the most congested ones in
the network. This suggested that a modification in the speed-volume
relationship embedded in the program might be needed.

To determine the lead vehicle speed. the program uses the
following relationship:

VL = MOD (JVACC [MLJ/100.000. 100)
This equation represents. in FORTRAN IV. the speed-volume relationship
shown in Figure 55% In the stable-flow region. as volume increases.
the space-mean speed of traffic decreases until the critical density is
reached (Point C). 'Thereafter. the flow becomes unstable. and both

volume and speed decrease.

 

25 A
E
E 2011.
'8
E '5 i
U)
8. 10 .-
('0
L
3
< S )-

 

I I V I Y

J
7|

l
100 200 300 #00 500 600 700

Average Lane Volume (veh/hr)

Figure 5.3.--Speed-volume relationship for the original model.

61

Because both the mean queue headways and the lost-time—delay
parameter changes reflect more aggressive driving in Saudi Arabia. it
was hypothesized that this relationship between speed and volume might
also be different. The use of speed limits is not widely practiced.
there is little traffic enforcement. and most drivers are aggressive.
Therefore. a relationship that produces a higher speed for a given
volume was indicated. Several modifications to the relationship shown
in Figure 543 were tested; the best results were obtained when the
relationship was modified to read:

VL = MOD (JVACC [ML1/100.000. 100) *1.5
Figure 5.4 illustrates this modified relationship. Table 5.9 shows a
comparison of the simulated and measured MOP using the modified model.
All the simulated travel times and average speeds are now within
acceptable limits except the travel time on link (4.3). The simulated
number of cycle failures also matches the observed data more closely
than the previous runs. ‘The complete output of the program is
contained in Appendix B.

This change in the model has affected the performance of some
of the network links. The average speed of link (2.1). which is a
heavily traveled link. has increased from 7.9 mph to 9.8 mph. The same
thing happened to links (3.4) and 4.3). which are also heavily traveled
links and are among the longest links in the network. Link (1.9) has
the same length as (3.4) and (4.3) but is less congested. Its average
speed has decreased from 18.8 mph to 18.3 mph. There is no significant

change in the performance of the low-volume. short links (9.10).

  

 

62

 

 

N 0 0.0. 0.0. 00N.. 0.0. 0.0. 000.. 0.0N 0.0N 0.0
0 0 ..0. 0.N. 000.. 0..N 0.0N 00..N 0.00 0.00 0.0
0 . 0.0N 0.0N 00N.. ..0. 0.0. 000.. 0.00 0.00 0.0
- - 0.0. 0.0. 000.. 0.0. 0.0. N00.. 0.N0 ...0 N...
- - 0.0. 0.0. 000. 0.0. 0.0. 00N. 0..0 0..0 ...N
. 0 0.0. 0.0. 000. 0.0. 0.0. 000. ..00 0.00 0..
N 0 0.0. N.NN 000.. 0.0. 0.0 0.0.. 0.00 0.00 ..N
- - 0.0 0.0 0.0. 0.0. 0.0. 000. 0.0. 0.0. N..
0000000 0.00. 0000. 00.0. 000.. 00.: 0000. 000.. 0..0 0000.
n:E.m nso.mo -:E.m "00.00.00 immoz -:E.m n.0.mo.vu immoz -:E.m x:_0
o.:..mu zm>\UOm :0: com .0
o.o>u >m.oo cooam ommco>< OE_P .o>m.0 xcozuoz

 

 

 

 

..ocoe no.0.voe 0:» mc.m: mo: ovum—35.0 dchin.m.m o.nmh

63

(10.11). or (6.7%. The unsignalized links (1.2) and (11.2) show little
change in their performance. The average speed for link (1.2) has
increased from 19.8 mph to 19.9 mph. while the average speed for link
(11.2) has increased from 17.1 mph to 17.4 mph. Most of the signalized
links show some improvement. The average speed of link (9.8).
connecting two signalized intersections. has increased from 13.9 mph to
15.0 mph. The same thing has happened to link (7.8). No change
occurred to link (7.10). while the average speed of link (1.9) has
decreased from 18.8 mph to 18.3 mph. The travel time of all these

links is inversely proportional to the average speed.

 

 

 

modified
33
Q.
E
v
T)
o
o
Q.
(I)
Q) \
01
2 original
o
>
.<
l 1 l 4 L l l
I I T I I f

 

100 200 300 400 500 600 700

Average Lane Volume (veh/hr)

Figure 5.4.--Speed-volume relationship for the original
and modified models.

64

The comparison of the network statistics using the original and
modified models shows significant improvement. The average speed
increased from 15.32 mph to 16.39 mph. STOPS/vehicle decreased from
1.47 to 1.45. Average delay/vehicle decreased from 54.70 sec to 47.52
sec. Travel time/veh-mile decreased from 3.92 min/v-mile to 3.66
min/v-milea Stopped delay as a percentage of total delay decreased

from 67.1 to 60.6.

Wei
The analysis of data collected in Jeddah shows that the NETSIM

model can be applied successfully in simulating traffic performance in
Saudi Arabia with the following traffic and program Characteristics:

1. Queue discharge rate: A mean time-gap (headway) between
vehicles discharging from a standing queue of 2.2 sec.

2. Lost time of first queued vehicle. A lost-time or queue
start-up delay of 2.4 sec.

3. A modified speed-volume relationship. The formula used in
the TRVL subroutine to determine lead-vehicle speed should be multi-

plied by 1.5.

5.6—ModeLMalidaLLQn
Since each network of streets has its own unique set of
characteristics (volumes. street lengths. signal settings. eth. the
fact that the NETSIM model could be modified to reproduce the MOP from

one network is not conclusive evidence that these modifications are

65

suitable for other networks in Jeddah. The more critical validation
test is whether the model can reproduce MOP from another network.

To validate the modified model. data for the input parameters
and the MOP for a second network in Jeddah were collected. The network
was similar in size to the first one. with 24 internal links and 16
nodes. Figures 55 and 5.6 show the physical and coded representations
of the network. The eight most congested links were selected to obtain
the measures of performance. Tables 5.10. 5.11. 5.12. and 5.13 contain
the mean travel time. average speed. average delay time. and cycle

failure for each link.

Table 5.10.--Travel time (sec) in the validation network.

 

Link
14.15 15.14 15.16 16.15 16.17 17.16 18.19 19.18

 

Run #

 

—l
U)
U)
m
N
\O
\l
U)
01
J}
.h
0
01
_J
\l
(D
.5
—l
-—l
N
O
N
U)
N
U1

3.99 3.87 3.93 3.54 3.54 4.27 3.99 3.89

 

 

66

 

 

96

D St. (A-Tathaamun AI-Arabi Street)

 

T—

 

5th St. (Makarona Street)

 

 

 

 

 

 

E St. (Al-Rabetah Al-lslameyyah Street)

 

9(— 7th St. (Prince Maied Street) *

 

6th St. (Al-Oroubah Street)

 

 

 

 

 

 

 

31¢ F St. (Palestine Street)

 

a: traffic light at intersection

 

 

 

Figure 5.5.--Physica1 Network #2.

 

67

 

 

 

 

 

 

 

 

 

 

 

 

 

5th St. A
\‘F ‘

 

 

E St. )- “ entry
a 20 1 5H
m fl 9X“
«ii
5
0

6th St. ,

 

 

 

 

 

 

 

Figure 5.6.--Coded Network #2.

 

68

_“E'-‘-;8-z'

Table 5.11.--Average speed (mph) in the validation network.

\ .

 

 

 

Link
Run #

14.15 15.14 15.16 16.15 16.17 17.16 18.19 19.18
1 7 20 9 6 20 14 25 12
3 14 14 14 14 27 11 22 16
3 18 15 11 16 26 7 16 14
4 9 12 15 12 29 13 23 11
5 6 11 12 6 21 17 27 15
6 13 17 8 11 19 8 19 17
7 17 18 13 15 26 15 18 10
8 15 12 16 9 30 7 24 9
Xi 99 119 98 89 198 92 174 104
X 12.4 14.9 12.3 11.1 24.8 11.5 21.7 13.0
S 4.53 3.22 2.82 3.87 4.20 3.85 3.77 2.93

 

Table 5.12.--Average delay time per vehicle in the validation network.

 

 

 

Link Desired Average Link Length Delay in
Link Length Speed Travel Time Desired Speed X sec/veh
# in ft. in MPH in sec 1.47
14.15 800 40 42.3 13.6 28.7
15.14 800 40 37.1 13.6 23.5
15.16 780 40 45.5 13.3 32.2
16.15 800 40 50.8 13.6 37.2
16.17 800 40 22.3 13.6 8.7
17.16 760 40 51.4 12.9 38.5
18.19 820 35 25.3 15.9 9.4
19.18 800 35 40.6 15.5 25.1

 

69

Table 5.13.--Cycle failure of second network.

 

Total Number of
Link Cycle Failures

 

14.15
15.14
15.16
16.15 1
16.17
17.16
18.19
19.18

U'INJSW—IOOQ

 

A run using the modified model was conducted for this valida-
tion network. Table 5.L4 shows the simulated values and the observed
values for the MOP. The t-test for all links indicates that none of
the simulated values are significantly different from the observed
values. 'This validation test confirms that the model can accurately
simulate Saudi Arabian traffic flow. The complete output of the pro-
gram is in Appendix C.

Comparing the simulated values of the original and modified
models for this validation network. the following variations in links
performance can be observed. The performance of the high-volume links
has been slightly improved. The average speed of link (14.15) has
increased from 13.5 mph to 14.3 mph. The average speed for link
(17.16) has increased from 8.5 mph to 9.8 mph. and the average speed
for link (19.18) has increased from 14.2, mph to 15.4 mph. This network
is located in a recently developed area where all links have almost the

same length; therefore. the model does not show any variation in links

7O

 

 

 

0 0 ..0N 0.NN . 00... 0.0. N.0. 000.. 0.00 0.00 0..0.
N 0 0.0 0.0 0N... 0..N N.0N 000. 0.0N 0.0N 0..0.
0 0 0.00 N.00 00N.. 0... 0.0 0N... 0..0 ..00 0..0.
0 0 . 0.0 0.0 0.0.. 0.0N 0.0N .0N.. 0.NN 0.0N 0..0.
.. 0. N.00 0.00 000. .... 0.0. .00.. 0.00 0.N0 0..0.
0 0 N.N0 0.00 00... 0.N. N... 000.. 0.00 0.00 0..0.
0 . 0.0N 0.0N 000. 0.0. 0.0. 0N0.. ..00 0.00 0..0.
0 0 0.0N 0.0N 000. 0.N. 0.0. 0.0.. 0.N0 0.00 0..0.
0.0. -.00 ...N... ".00..: -.0. 0.0. "00...... -.0. ...N...
..0.“. ..0.;

 

 

 

 

.xcozuo: vcoomm any .0 do: team.:E.m mgkll.:..m o.nmh

 

7]

performance due to variations in Tinks Tength. There is no significant
change in the performance of unsignaTized Tinks. whiTe the simuTated
cycTe faiTure of signaTized Tinks has generaTTy improved. The cycTe
faiTure of Tink (T3.12) has decreased from 7 to 3. the cycTe faiTure of
Tink (T7.T6) has decreased from T4 to 6. whiTe the cycTe faiTure of
Tink (T9.18) has increased from 0 to 3.

The comparison of network statistics using the originaT and
modified modeTs shows some improvements. Theraverage speed has
increased from T5.T3 to 15.74 mph. The stops/vehicTe has decreased
from T.32 to 1.30. The average deTay per vehicTe has decreased from
52.65 sec to 50.37 sec. and traveT time/veh-miTe has decreased from

3.94 to 3.74 min/v-miTe.

5.1 AQQJanIan 9f the Mode]

One of the important features of the NETSIM modeT is its
capabiTity to be used to anaTyze and evaTuate traffic signaT timing
pTans and strategies. A modification in signaT timing parameters. such
as offsets and the duration and sequence of the signaT phases. can
resuTt in fewer stops. Tess deTay. reduced traveT time. Tess fueT
consumption. and reduced accidents.

To demonstrate the use of the modified NETSIM modeT in Saudi
Arabia. the modeT was used to simuTate the effects of a modification in
the existing signaT timings of a major street in the caTibration net-
work in Jeddah. Figure 5.7 shows a representation of KhaTid Bin AT-

WaTid street. which is a one-way street with three traffic Tights. The

 

72

objective of the proposed timing
pTan is to coordinate the three
signaTs to produce a better pro-
gressive system. This change
shoqu reduce deTay and traveT
time and increase the average
speed on the street. The offset
at each intersection was set so
that the first vehicTe of the
pTatoon wiTT receive the green
indication just as it reaches the
intersection. SignaT (T) was con-
sidered a base signaT for the sys-
tem. The offset of the other

signaTs was determined using the

foTTowing formuTa:

offset

 

[—

 

 

* 1“ St. (Khalid Bin AI-Wllld Sttoot) *

 

F___

 

 

 

 

 

: tame light It Interaction

Figure 5.7.-—KhaTid Bin AT-
WaTid Street
in Jeddah.

 

== Distance between signaTs. in ft

Desired speed in ft/sec

 

 

- _ 1 620
ff '- ’ = o
0 set for Signal 9 T8.3 x 1.47 60 0 sec
. 2 240
ff f 1 = ' = .
0 set or sugna 8 18.3 x 1.47 83 1 sec

The desired speed for the system was considered

is the average speed on Tink (T.9L

to be 18.3 mph. which

The highest cycTe Tength in the

TIME (seconds)

73

system is 84 seconds at signaT T. Figures 5.8 and 5.9 show the time-
space diagrams for the seTected street for the modified and existing
timing pTans. There is an increase in the desired speed for the system
from 15.5 mph to 18.3 mph. and an increase in the bandwidth (minimum

green time) from 30 sec to 35 sec.

 

 

 

 

 

 

 

 

 

 

160
”'0 18.3 mph
120 . /
3 sec
TOO ¢ 35
. C“
T 80 ““6 ‘
:5 <63“
0')
8 60 '1F’
-— u
m 3 60 Sig
341' [+0 1 04-) H-
50 I620 33’. 53%
u.
2 20 8‘8
0
N _ , _
(I) (9) (8)
1620 zzho

SPACE (feet)

Figure 5.8.--Time—space diagram for KhaTid Bin AT-WaTid Street
using the modified pTan.

 

 

7h

 

 

 

 

'60 15.5 mph
.
Th0 /////;7
120 0 9e."
17. .3
'2 100 .352“ fl?"
0 $\
0 v-
0) 0
01 $6
" 8O
1...: .c
z 4.: f
I: g 60 0
Q) 0 cu
-—u a) m
0):.) I40 0 8 "’ 00
F5: . 8’) U") m 0‘
$500 N p '-
~ 2° 0 T5
c
o0 1

 

 

 

 

 

 

(T) (9) (8)
T620 zzho

SPACE (feet)

Figure 5.9.—-Time-space diagram for KhaTid Bin AT-WaTid Street
using the existing timing pTan.

TabTeELTS shows a comparison between the simuTated vaTues of
the measures of performance of KhaTid Bin AT-WaTid Street using the two
timing pTans. The tabTe shows an improvement in aTT measures of per-
formance due to the improvement in seTecting the offset time between
the signaTs.

This experiment demonstrates the potentiaT benefits of using

the modified NETSIM modeT in Saudi Arabia.

75

 

 

 

0.0 - ..0. ..0. 0.0+ 0.0. 0.0. 0.0- ..0N 0..N 0.0
0.0N- m... 0.0. ...+ ..m. 0.0. m..- 0.00 0.00 0..
00.. 00.. 00.. 00.. 00.. 00..
omflm00 00.0: 00.0.0 mmflm00 00.0.: 00.0.. omfl000 ...0.0: 00.0.:
. 00...0oz m0.3.x. . 0o...0oz m0.3.x. . 0m...0oz 00.00.x0
00..
.cm>\oom In: oom
oE.h >m_oo nooam omm.o>< we.» _o>m.k

 

 

 

.mcm_a mc.E.u ucoLo.m.v ozu mc.m: mo: .0 mo:_m> woum.:s.m coozuoa com..maeoo <nn.m..m o_nmp

 

 

CHAPTER 6

CONCLUSIONS AND RECOMMENDATIONS

W5

ATthough the NETSIM modeT was originaTTy designed to simuTate
traffic performance in the United States. this study has demonstrated
that it can be appTied in different traffic conditions. The modeT is
not difficuTt to understand; it can be used properTy with TittTe modeT-
ing experience. The modeT requires more extensive data than some other
modeTs. The payoff. however. is that the output is more representative
of reaT conditions.

The modeT is reTativeTy expensive to operate. However. it is
more economicaT than coTTecting actuaT fier data to determine the
change in system performance resuTting from modifications in traffic
controT. The avaiTabiTity and ease of use wiTT continue to expand the
appTication and acceptance of NETSIM.

The modeT caTibration and vaTidation was conducted by comparing
the resuTts of the modeT simuTation with a sampTe of fier data. Since
the sampTing error is unknown. the true accuracy of the modified modeT
is aTso unknown. The t-test used in the anaTysis assumes that both the
fier-coTTected sampTe and the simuTated resuTts are from normaT dis-
tributions. An extension of this study to incTude a Targer sampTe

shoqu be conducted to verify this assumption.

76

 

77

Based on the data coTTected and anaTyzed in this study. the
foTTowing concTusions were reached:

T. Traffic performance in Saudi Arabia can be simuTated and
anaTyzed using a modified NETSIM modeT.

2. A modification in the speed-voTume reTationship embedded in
the modeT is needed to suit the Saudi traffic conditions. The formuTa
used in the TRVL subroutine to determine Tead vehicTe speed must be
muTtipTied by'T.5 to accurateTy simuTate Saudi driver acceTeration
characteristics.

3. A mean time-gap (headway) between vehicTes discharging from
a standing queue of’242 sec and Tost time for queue start-up deTay of

2.4 sec are most suitabTe for traffic in Jeddah.

QJ—Becommendaflons

WhiTe this study demonstrated that the NETSIM modeT can be
modified to simuTate traffic on the street network in Jeddah. the
Timited appTication to the two networks does not answer aTT the ques—
tions. AdditionaT studies and anaTysis shoqu be performed. incTuding:

T. The modeT shoqu be tested and vaTidated in different
cities in Saudi Arabia such as Riyadh and Dammam.

2. Measures of performance of aTT Tinks and/or nodes of both
the caTibration and vaTidation networks shoqu be obtained to estabTish
the vaTidity of the simuTation over a broad range of Tink characteris-
tics.

3. The use of automated techniques in data coTTection shoqu

be deveToped to reduce the effort required for data coTTection. These

 

78

techniques coqu incTude: voTume counts from detectors. photoTogging
for both traffic operationaT characteristics and the physicaT charac-
teristics of the network. aeriaT photography for a muTtipTicity of
data. street-TeveT photography for parking characteristics and network
data. and a portabTe event recorder for gathering information about
vehicTe speeds. journey times. and fueT consumed.

4. To provide a more streamTined version of the modeT. the
options of simuTation of bus traffic and transient bTockages within the
traffic stream were suppressed in this study. However. it is recom-
mended that simuTation of these options shoqu be tested in Saudi
Arabia because of the increased use of pubTic bus transit and the
frequent occurrence of temporary bTockages due to accidents. emergency
vehicTes. or unusuaT congestion.

5. A deveToped data-information system is needed to coTTect
extensive data on major traffic parameters in Saudi Arabia and to
estabTish statisticaT distributions of these parameters to update the
embedded input data of the modeT.

6. Traffic performance in Saudi Arabia shoqu be simuTated
using other modeTs (such as TRANSYT or TEXAS) and the resuTts compared
with the modified NETSIM.

7. The NETSIM User Guide needs improvement. It woqu be more
heTpfuT to the user if it incTuded the program Togic. the major assump-
tions. and the traffic modeTing used in the program.

8. Among the recommended appTications of the modeT in Jeddah

and other cities in Saudi Arabia are the foTTowing:

 

79

a. 'Traffic-fTow improvement on major streets. ‘TraveT time.
deTay and average speed and other traffic parameters can be
improved by evaTuating different timing pTans for the seTected
street or network. This improvement wiTT reduce congestion and
increase the street capacity.

tn ControT methods and operationaT timing at singTe isoTated
intersections to seTect from either traffic-actuated voTume-density
signaT controT. two- and four-way stop signs. or fixed-time signaT
controT.

c. Effects of adding bus Tanes to the CBD network or other
sites.

d. Effects of parking Tot or shopping center deveTopment
on the surrounding area in the city.

9. Effects of geometric changes such as additions of Teft-
turn bays or right-turn pockets.

f. InstaTTation of future signaT systems and the effects
on surrounding syStems.

g. SimuTating the effects of the predicted future changes in
traffic voTume on the existing network. This woqu heTp the city
in seTecting and prioritizing street-improvement projects.

9. Introductory courses on modeTing and the NETSIM program are

recommended for the traffic engineers who are working in the traffic

departments in Saudi Arabia.

 

APPENDICES

80

 

APPENDIX A

STATISTICS ON DRIVER AND TRAFFIC CHARACTERISTICS

IN JEDDAH

8]

 

82

 

 

000 000..0 .00.0 000.00 00..0 000.00 000.0 ... 000.00 000.00 .00.
000 000.00 000.0 000.00 00..0 0.0.00. 000.0 ... 000..0 000.00 000.
0.0 000.00 000.0 000.00 000.0 0.0.00 000 00..0 00..00 000.00 000.
.00 000.00 000.0 000.00 000.0 000.0.. 000 000.0 000.00 000.00 000.
000 000.00 000.0 000.00 000.0 .00.00 000 000.0 000.00 .00.00 000.
000 000.0. 000.. 00..0. 000.0 .00.00 000 000.0 000.00 000.00 000.
000 000.0. 000 000.0. 000.. 000.00 000 .00.0 000.0. 00...0 000.
00. 0..... .00 000.0. 000.. 0.0..0 000.. 000.. 000.0. 000.0. 000.
00 000.0 000 000.0 .00 000.0. 0.0 000 000.. 000.. 000.
00. 000.0 000 000.0 000 000... 000 000 00..0 ..0.0 000.
00. 000.0 .00 000.0 00. 000.0 00. 000 000.. 000 .00.
mmmcmo_n_ mmmcoomg m®mc00_n_ mmum—m mwum—n— mmum—m mmum—m mmum—m
.00... 0...... 0.00 00. 0.0. 0...... 0.00. 00.0..
.0000 .0 .02 .0 .02 .0000 .0 .02 .0 .02 .0 .02 .0 .02

 

...00.-.00.. 000000 0. 000000.. 000 00.0.00>--..< 0.000

 

 

83

 

 

000 000.. 0.0.0 000.. 000.0 000.. .00.0 00. 000 00... 000.. 000 .00.
000 0.0.. .00.0 000.. 000.0 000.. 000.. 00 .00 000 000.. 00. 000.
000 000.0 000.. 000.. 000.0 000.0 0.0.. 00 000 000 000.0 00. 000.
000 0.0.0 000.. 000.. 00..0 0.0.. 000.. .0 000 ..0.. 000.0 .0. 000.
00%...“ 0.0000. 00.00.... -WMMM .00..... ..00. 00...... mm... 0mm... .00“. 00...0... 0...... 0...

mooa w 00: mooa

 

.A_wm_uwmm_v cmvnow :. mucov_oom :. nm>_o>c_ mgo>_0n mo mo_um_0ouomgmzo oEOmnu.N< o_nm0

 

84

Table A3.--Number and causes of accidents in Jeddah (T978-T98T).

 

 

Circu- Traffic High Alcohol
Year STOP Tation Outreach Violation Speed of Drugs Other
T978 43 T97 285 66 2,000 T9 322
T979 #2 309 296 TOO 2,702 IT 289
T980 68 40“ 274 TTh 2,560 T7 369
T98] 32 TT3 9T 284 2,375 T3 hOT

 

Table Ah.--Number and type of vehicTes involved in accidents in Jeddah

(T978-T98T).

 

 

Year Sedan Jeep Bus Pick-up Truck Tank-Truck Other
T978 T,765 78 T26 862 3T3 63 679
T979 T,906 73 T30 807 296 99 808
T980 1,507 l03 Th5 519 28T 79 730

1981 2,154 93 155 469 4h] 76 602

 

 

85

 

 

0. . 000 0.. 0 000.0 0 000.0 .00.0 000.000 .00.
0. . 000 00. 0 000.0 0 000.0 000.0 000.000 000.
0. . .00 00. 0 000.0 0 000.0 000.0 000.000 000.
.. . 000 00. 0 000.0 0 000.0 000.0 00..000 000.
0. . 000 00. 0 0.0.0 0 .00.0 000.. 000.000 000.
0. 0 000 00. 0. 000.0 0. 000.0 000.. 000.00. 000.
0. 0 000 00. 00 000.0 0. 00..0 000 000.0.. 000.
0 0 00. 00. 00 000.. .0 .00.. 000 000.00 000.
0. 0 00. 0.. .0 000.. 00 .00.. 0.0 000.00 000.
.. 0 00 00 00 000 00 000 00. 000.00 000.
00 0 00 0.. 00 000 00 000 00. 0.0.0. .00.
.0.00< 00.0.00> 00.0 .000 00. 00.0.00> 00.0.00> 00000 000000.000
00. 00. 000. 00. -..000. 00. 000. .0. 00.000. 000. 00. -.00< x000. 00.0.00> .000
..mumu ..mumm 00 .oz vogsmc_ no0:m:_ 00 oz mucon.oo< 00 .02 00 .oz

 

.A_wm_n—mm_v :mcvow c. mucov_oum 00 80>“ cam LonE:ZI|.m< o_nmk

 

APPENDIX B

PROGRAM OUTPUT OF CALIBRATION NETWORK

86

87

nuuh mu «upcxuzmu «mats: 1092a: xcu chum
a.mqaq mosqu.1«ooma .huuzpw o<1<u uzux
cmm~ 20a aq¢4< ou;q1

rm\cm\u,c L
unugq.> up
n mm<u uc hxuhw

zo~h<4;1~m

 

SI”UL£TION OF TRAFFIC

THC NFTSIH PODrL

 

TRAFFIC

CF

SIHULATION

HAHED ALBAR JAN 1980

05/?F/q3

O

7581
PFD LINE CP‘N

SAUCI ARABIA

JEDDAHo

STREET.

KING FAHAD

SEED FOR FANDOH NUHPER GENERATOR IS

IDEKTIF

2 3 0 B TYPF G L

l

LOST DFN

[AME SPAN

LINK

88

HNMQ mun-can» GHC'JV)CUI\LF&0-GH¢0IO «1.005431% 90-.
p.m.-«Hun «co-«4 HNC u-JNWNNNNNMM

(\NNNNNNNnhnr‘JvNWham“NNNNNNNNNNNN
O‘HHH”HF‘HHHFHHHHHHH—thfififif‘finHHF‘
GOOOLJDC OC'C CCOGC CGGDtbc CLO: COC'OO
CCOOGGLOCC CCOL CCLODOCCCDOL LUC cc
OOOODODODODOOOOOOOODCGDOOGOGOOO

OCOOODOOOOODDO OOCCOCOOD OCODOGDO

OOOOCOOOOOC-OCO OGCODOOCO OGOGCOCO

GHHCDCMDOHH—HHP}”Ftrfl'fl'onncnv-H

DDOOQCODOCOOOODOOOOODODCDOOOODO

'4ch c cmo-ccgorwtc-moc Q rec—uncut (\rzuu‘ a
H H H H no c

G: CL C;
FoamOo-«DDIDNIHIOOONODIDNOQuDDIDHOG‘éCOhO
Cit-0 H H
duh a; do

00‘ 9NHDHGC-5020VJOCOHDUHSOOF OGNOWCUD
“CH Ft

6 g“
DOC)OOOOODDOODOOOODOODODOODDODO D

UJOG‘OOQTIOWOOaflufldibo—c\DOOOI’JODOMDNFO
H O comm ”C D D mm
H H

IDDWODOOIDONNfiOOIOONlflU‘OOmOHOOO‘FNNMO
Q C HO U‘CU‘U- 0 ONqu ma; 0‘: fica‘la‘N
H

ootNDcNoucioccmomocowe¢oo~cw¢oc
comma In H c c r. Anew-n u N

NNNNNNNNNNNNNNNNNNNNNNNNNNNN
NNNNNNNNGNNNNNNNNNNNNNNNCJNNN

>>>>>>
WIDODO OOOIDIDODOUIOOK “KKK K
nNNNNNNNNNNNNnNNPhbhhh
zzzzzz
wuuuuu

DOODOODOOGOOOOODOO 0000990000600

35

OQDDOOOGNDNOOOOOQO OOOOOQOOOODOO
H H

OODODOOODDOOOOODOO DODDOODDOOOOO

CNN‘LfiDQNNNOGQO ‘QLfiONCUQLVmcfiw‘fNND

mmwmmwmmmr-hnmwm«Nahunwmmcncmnnn
H HHv-‘H H

NNHNNNNQ C OCNNNNNNNNMNNIONNMQ O'NNN

AAAAAAAAAAAAAA AAAAAnanaaa-AAAAAA
WHO‘HNSNQ VOID cucmmhonwwbomo~nnmhmm
H H HH

0... OOOOOOOOO. 00..000...
«NHNnNancmchwmh
H d

UUUUUUUUUUUVUUVUU---UUUUUUUUU- U

 

89

f'l

mzcnahmunun 02

new u>om< u
can 0» 0mm H
0mm c» 00” 0

acxxmLUL. usnao> beau
wmmzouuhuu >~um2mc ¢<thwuuua

>uqmmbguz up Leap b:nz~ a mark waaauaa are >u~au> unquan

. an .com.

am on u a» 0H» 0 am o¢ . an own .0 AN onu . an 0~ v am ..u u
hzmxu>c& ammonaoza L>aJ usz4 quJOJJGu MI» team mzaDhlhou

. gu cc 0 gr orca- as 0w v as 0m w an .mcm.

aw om - Au .h . Ac 0m w .0 or 0 an .Now. an .o . aw .H a

azuwooao 22~a qumcaao xZHJ oszoaao xzuq wszonno

kzwzu>oz zmaptpuua 1pm: mxz~4
xaczpu¢ uc wuaap<uu aquHQOALao»

A-
”F:

V-

AA
NC

 

90

‘ I1.“

m m a Ana . c
..u ..w .Hu .0; .m .
oLxuqcnnuq LuhqupLza ua.uqu whee u .qzw.m humane
Asapzou zuhw ”moz: m~ w 0002
n c m .nwm . we
a a N .aam 0 mm
c r n .00. 0 mm
a r n . .ann . mu
0 r 0 Anna 0 an
s r H .aos . be
- . ..u .00 0. ..m ..m . ... .0 .
unlucoaaaQ cupggpc 2a ozhguu mucou daze u » nuke
n c m age. 0 an
n a m .afin . Nw
a r n AQON . m—
. 0 n 32. 0
s v 0 amp 0 w
0 r w. Ammo . cm
. 0 0.0 .u 0.: .Hn 0 ..a r .
VLluqcocmu pm».p»L2m uzwuqu w.cc U :« chm human:
n 0 u Anon . .m
a a n .awu . an
L N n “runny w
H n n Aanm . em
h m o aawm . «w
0 m 0 Anon . cm
. opn .0 ..m .~ ..v .mam.
wuxucoanaq Laban—L2— wz~u<u wuc coo quuum humuuo
u a ~ Aug 0 c
.. . ..0 .0. . ..m .0 ..r .u .
Luzpqcmguq LL»090L7~ ungqm wmccu a<zc~w bwmuuu
acahzou zohm nucz: mu m was?
0 n .aum . cm
H n ”cum 0 me
n o Ammo . oo
o H .ao . 0
. ..n .m ... .H0m.
wuzu<oaAuq cu»<0~;2. 02.040 c..oo cu .q‘awv p.muuc

IchuLA< mat» «on hbuuum 2m ac»: wur<u~uz~ 0
ch<t d<zmmw Uakquh

 

 

Anon”. 0w
zo~hqmao
no m
ne« 0“
me n
men 0H
am e
arm 0:
zanpqmao
no m
an" m
n0 n
uoa 00
aw t
nmm me
zeapqaso
a: n
a." 0"
A: n
as" NH
am 0
a¢m an
zcqumao
Anna". om
zcuhcaso
Au: . n
.nrc . pm
An: a m
.05.. 0e
zen—«man

n
A>th

HRHQVWD

4>hz~

HNHQUH‘:

u>w2n

ammo mu:

a>p2~

4>hz~

H01"? 6

J>hzn

m
was:

D mu'umnunn

LA!

Cl 000000

D
Z

kl

D mmrmnfln

ml

92

“no:

 

9]

U) C
lua-

quUaocanc

MLIUQCQAQQ

.
mutuaocaaq

.
Ipqoaaau

V)
L:

auhcuch~

umkquhczu

m:
H

.05
unknown

cam.
H QZHU

.
o... .

N

w

o... .0. .

uz awau whoou aczcnm

CONNHD

a:
uz~u<m who:

Cob. (Mo-to

o.o
uzhpau mucc

oNNo-ONN

..m

o
u

dompzou zo.m muczs m. H. 0002

.
aazon

..c.
m

0.x

azHU¢m mkoou quuwm

th—‘CO'N

Oak

Car

at mLaou aq¢oum

OHCJNN MC. “1“.

MD h: NF?-

9.

0H0 CJQCJ(\

.ao . 0
.umuuo
.amu 0 n.
.00. . on
.000 . 00
.000 0 o.
humane
.00 . c
.0.0 0 00
.00. 0 co
.aw. . 0
..muuo
.mo. 0 N.
.amm . mp
.000 . u.
.amm . on
.amn . hm
.am. . m.
hummuo
.an. . h
.amm . ¢c
.anm 0 me
.asm . mm
.awm . mm
.nmu . 0.
humuuo

ance“. om
zonh<uno

UV----
N
O

zo~>«¢:o

.am
.am~
.am
.am.
.nw
.awo -

zanpdmno

~508va
nanny!)

4>hz~

HNMO

J>sz

HNWQ

4>h2~

HNnou‘hD

d>hz~

HNMQLOQ

d>th

«A
uooz

0 GOOD
Flo-004d

Lu'
0
2

D (“000‘

O
2

uJ u)
D NNFFNF D mmmcocnco

U

 

 

 

 

92

w m— wszJ ruhzm uc mutzaz

«a mu mumrzz Unoz zztuxcx

.20.0 02. 002. 20000-200. .004 02
02000. 20000 .20.. 020 .00. 2..: 000
20000 002. 20.00-0200. 02

2000. 002000.0 20000 :..a 000
0....2000 200. .00.0 2..a 000 00 00.0
30000 .000 20000 2..2 000

20000 .10.: 2.000 2..: 000

0.0

20000

000:. 00 0.0..

mmccu Adzwnw uh 20whdhuzumfihz~

U: \L F m 0‘

OHCVI'.Q

 

93

mpg
wwn

mucflwxhh bu
mocuulhb h:

“01.» c» mud Hutu» xcau mczaumm cw «Om Am oh a zap; zc ou4~4>u¢& m<z zu¢¢44~am
nuznh c» "on Hutu» zcmu mozouum «N man .m 0h . 12H; 26 ouJH4>U¢Q m4: 29(zdduaw

¢p<o 0<u~hwnh¢pm zuxhco oz< lambdaatnm muzuttcu

WCZOUum amc «upuc cu—uaatou acmmua zo~h<N~4¢~hH2n

was ruz<a3uuc 2n wuzQIU kuz ozmoshwz wzh ounaauuo muau~xw> cum 0w4u>u hw<m czuuao

an «a: >uz<nzpuc Zn w¢zczu huz ovmozhwz uzp ownasuuc mm40n1u> cmw oUJu>u hmdu OZMKDD
m ‘ a—N n
«m mnn N
But uwm u
unauzu >uzcaauuc banana

ou2~«bh¢ tanzmmqmnou

ma 5—2: ch zuhhhua ma aqua <b¢a rzohuufidmh uau—Ih:
cmmwupcuu um JJ~3 mconmmnzw 02¢ zenhnznmzou Juan

ozod on b
3C: min» xUOJU

_0>amh2»0m:w uznuac mupazuz c >zm>w ¢¢unu¢ 44m: hanhDo u>~h143tnu
mchuuv com no w4¢>muhz~ h< mauaan Jan: hDGIhZHma 0mozouum com no oeuzun a mom
.d>¢kh2nl£=w no c2»zz~6um ¢u~u¢ mazouuw cow muuzuxzou hDAhDO arduout¢u~z~

czcuum th¢>¢sz~ canzzduw
oLLZCUum cmmnn dqaaMFZH wank zonhgaaznm

 

9.4-

mo. "02.. 0. 00. "02.. 2000 0020000 0. 000 .0 .. 0 02.0 20 000.02000 00: 000000.00
0 0.0 0 0 0 0 0 0 0 0 0.0 0 0 0 0 0 00 .. 0 00 m .0 ..00.
0 0.0 0 0 0 0 o 0 0 0 0.0 0 0 0 0 0 .0 00 0 .0 0 .0 .000.
0 0.0 0 0 c 0 0 0 0 0 0.0 0 0 0 0 0 0. n. 00 00 0 .. .000.
0 0.0 0 0 0 0 c 0 0 0 0.0 0 0 0 . . g 0 000 .0 00¢ .. .0 .000.
0 0.0 0 0 0 0 0 0 0 0 0.0 0 .. 0. 0 0 0. 0.0 0 00. an .0 .000.
0 0.0 0 0 0 0 c 0 . 0 0 0.0 0 0 m c N 0 00w 00 ..m n. .. ..00.
00. a... 0 0 0 0 a 0 0 00 0.0 0 0 0 0 . m0. 0 0 n0. 0 ... .0. .
0. 0.0 0 0 0 .0 0 0 0 u. ..0 0 0 0 . . 0 .0 0 00 ~ .0. .0 0
.0. 0.0. 0 0 0 0 0 0 s0. 00 0.0. 0 0 ~ 0 . 0. .0~ .. ~00 00 .0 .0 .
0. ..0. 0 0 0 0 a 0 0 0n . 0.0 0 0 0 0 0 .0 0 0 cm 0 .0. .. .
r. 0.. 0 0 0 0 0 0 0 00 0.0 0 0 0 . 0 0 00 n 00 n .. .0 0
0o 0.. 0 0 0 0 0 0 0 00 ...N 0 0 N 0 0 0 mo 00 00. 0 .0 .. .
00. 0.0 0 0 0 0 o 0 0 00 0.00 0 0 0 c. . 00 0 00. 00. n0 .0 ... .
0 0..N 0 0 0 0 0 0 0 0 ..0 0 0 0 0 0 .. mm 0 m0 0 ... .0 .
0 0... 0 0 0 . 0 0 0 .0 ... 0 0 0 0 0 0 00 .0 me 0 .0 .. 0
.. ..0 0 0 0 0 0 0 0 0. .... 0 0 0 . . 0 00. 0 00. 0 .. .0 .
o. 0.0. 0 0 0 0 o 0 0 a 0.0. 0 0 0 0 0 no 0 .0 m0. 0. .0 .0 .
0 ..0. 0 0 c 0 0 0 0 0 0.. 0 0 0 0 0 00 0. 0 00 0 .0 .0 0
o 0... 0 0 c 0 c 0 0 0. 0.0. 0 0 0 0 0 0 0 0 .. 0 .0 ... .
.0 ..0. 0 0 0 0 0 0 0 m. 0.0 0 0 0 0 . 00 0 0 00 n ... .. 0
0.. 0.0. 0 0 0 0 0 0 0 mm 0.0. . 0 0 0 0 0 000 00 000 00 .0 .0 .
r. ..00 0 0 0 c 0 0 0 .0 0.. 0 . . . . 0. 00. 0 000 0. .0 .0 .
cm. 0.00 0 0 0 0 c 0 0 0. n... 0 0 0 0 0 0 0.0 00 000 .0 .m .0 0
«N. 0.0. 0 0 0 0 0 0 . 00 ...n 0 . 0 0 c 00 ~00 0 000 mm .0 .n 0
no ..0. 0 0 0 0 0 0 0 00 0.00 0 0 0 m 0 .0 0 00 No 0. .0 ... 0
.0 ..00 0 0 0 0 0 0 0 00 0.0 0 0 0 0 0 0 e. 0 00 0 ... .0 .
0. 0.0. 0 0 0 . 0 0 0 00 0.0 0 0 0 0 0 0 0 cm 00 0 0 .n .
0. 0.0 0 0 0 0 c 0 0 00 .0..n 0 0 0 0 . 00 0 am 00 .. .0 .N 0.
.n. ..0. 0 0 0 0 c 0 0 .0 ..0. 0 0 0 0 . 0 ..~ 00 .00 00 .0 .. 0
0. 0.0 0 0 0 . 0 0 0 .. ...N 0 0 0 N 0 0 0 00 .0 e .. .0 0
u. 0.0. 0 0 0 c 0 0 0 0. ... 0 0 0 0 0 .. cm 0 ~0 m .0 .. 0
- . . - - . a . .
..0.... ..00; 2000002§500 .0050... :30; 000.000 .00 ,.

9 cc 5 0.0:» ha wu-m-<hw xz~4

 

95

0 0.0 .. 0 0 0 o 0 0 0 0.0 0 0 0 0 0 00 c. 0 .0 0 .0 ..00
0 0.0 c 0 c 0 e 0 0 0 0.0 0 0 0 0 . 00 00 0 .0 0 .0 .000
. ..0 c c 0 0 c 0 0 0 0.0 0 0 0 . . 00 .0 00 0.. 0 .. ..00
0 0.0 . 0 0 0 0 0 u c 0.0 0 0 . 0 0 g 0 0.0 00 000 0 .0 .000
0 0.0 c 0 0 0 0 0 o 0 0.0 0 0. 0. 0 0 0. 0.0 0 .00 o. .0 .000
0 0.0 0 0 c 0 0 0 0 0 0.0 0 0 c a 0 0 000 00 ... 0 .. ..00
as. 0... c 0 c 0 a 0 0 .0 ..0 0 0 0 0 . ... 0 0 00. 0 ... .0.
o. ..0 0 0 . c 0 0 0 0. ..0 0 0 0 0 . 0 00 0 .0 . .0. .0
c0. ..0. 0 0 0 0 c 0 0 00 0... 0 0 0 0 0 0. 000 00 .0. . .0 .0

0 . .0. 0 0 0 c a 0 0 .0 0.0 0 0 0 0 0 0. 0 0 0. . .0. ..
.. ... c c 0 0 0 0 0 .. 0.0 0 0 0 0 0 0 0. 0 0. 0 .. .0
0.. ... 0 0 c 0 0 0 0 00 0.00 0 0 0 0 0 0 00. 00 .0. . .0 ..
000 0.0 0 0 c 0 . 0 0 00 0..0 0 0 0 c. . 00. 0 00. 0.0 0. .0 ...
0 0..0 0 0 0 0 0 0 0 c 0.0 0 0 0 0 0 0. .0 0 00 0 ... .0
0. .0. 0 0 0 . 0 0 0 0. 0.. 0 0 0 0 0 0 .0 .0 00 0 .0 ..
0a 0.0 0 0 0 0 0 0 0 o. 0... 0 0 0 0 . 0. 00. 0 00. 0 .. .0
.0 c.o 0 0 0 0 c 0 0 00 0.0. 0 0 0 . 0 0. 0 0. .0. .. .0 .0

0 0.0. c 0 0 0 c 0 0 0 0.. 0 c 0 0 0 00 00 0 .. 0 .0 .0
0. 0... 0 0 0 0 a 0 0 0. ..0. 0 0 0 0 0 .. 0 c 0. 0 .. ...
... 0.0. 0 0 0 0 a 0 0 0. . 0.0 0 0 0 0 0 00 0 0 00 0 ... ..
0.. .... 0 0 0 0 c 0 0 .0 ..0. 0 . 0 0 0 0 «00 .0 0.0 0. .0 .0
.. ..00 0 0 0 c c 0 0 00 0.. .0 0 . 0 . 00 000 0 .00 0 .0 ..
.0. 0..0 0 0 0 0 0 0 0 0. 0... . 0 0 0 0 0 .00 0. 000 00 .0 .0
.00 0.0. 0 0 0 0 0 0 . 00 0.00 0 0 0 .. .. 00 000 0 0.0 .0 .. .0
00. 0.0. 0 0 0 c . 0 0 00 ..00 0 0 0 . 0 00 0 0. 00. 0. 0 ...
.0 0.00 0 0 0 0 0 0 0 0. 0.. 0 0 0 0 0 0 00 0 .0 e ... .0
0. 0.0. c 0 0 . 0 0 0 0. 0.0 0 0 0 0 0 0 . 0. 0. 0 .0 .0
00 0.. 0 0 0 0 c 0 0 00 0.00 0 0 0 0 0 .0 0 0. .0. .. .0 .0
a0. 0.0. 0 0 0 0 0 0 0 0. 0.0. 0 0 0 . 0 0 000 00 000 00 .0 ..
00 0.0 0 0 0 . 0 0 0 00 0.00 0 0 0 0 0 0 0 .. 0. 0 .. .0
c. 0.0. 0 0 0 0 c 0 0 0. 0.. 0 .0 0 0 0 0. o. 0 00 q .0 ..

. , . - .

00 00; 22000.2.Effi .00.00; :35. 0000.00 0.0

om me p 0.2.. ha mu~rm~><hm x2...

96

u a o.o a c u a c o o o 9.: o a o o u om ed a Ne w .m .5.
L c c.o a c c a c as“ c c 9.: o o a H c m. an a mm a .c .¢.
L c c.c c u c c u a c a- e.a o a a m w mu op 5n mm" c .5 ...
. a a.» c o c o o o m c c.n o m a o a a. map as me» n .u .n.
g a g.o c o c u a 9 ~_ o ..o.c o ea .” DH mu m~ mna a «dc om .n .~.
F a 5.: . o c a c c o o a c.a o a o m c o «a. no“ "am Nu a” .d.
u can a.- a o o o a c c we ~.o o o a o a "an o 9 wk" m an" .o
. om ¢.w c o o o c c c «s ..m o o o o o 0 pm a mm a .9" .o
. ecu «.ma u o o c c c c pm m.rn a a N o a h" on. on mw¢ a~ . .m .m
w ms «.ma 9 a c c c a a h“ c.m a o o a 0 mm o o no u .DA .5
fl n~ w.» a o c c c c a .5 n.o a a o o o o mm a mm H .h .o
m mm~ n.n o c a a a c o «x o.o~ o c a o o a «fig a. cum a .c .sh
. man n.u a c c c . c c om c.~m o o o ca u nan a «ma smm on .¢ .«
c w ~.H~ ; c g p o o a n «.N c o o o o om am 0 cm n .ug .o
. cu c.u~ a c c p o a c cm ~.~ o o a o a c wn a. on a .w .p
n on. ¢.u c o c c a o c or m.- o a o o o ca op“ o mm. m p .m
. no w.w c o c c c c c a: 5.0“ o c a m a pm. o mm es. s .n .¢
» o c.c~ c o o o c c a c m.H o o o a a on mm o “o a .o .n
a ca a.~« o o a c c o 9 hp m.- o o a a o w" o s n~ o .c .~
m a." m.m~ o c o a c o a m¢ m.c o o o o a nu" o a me" e .~« ..
a on“ a..« c o c c c 9 HF r¢ . p.mfi m c o o g o ans cs map cu .c .m
0 mm «.mm c a o c c o 9 pm a.» a w a o a un amc o nmc x m .o
r snm ¢.n~ c p c o c o m «m o.n~ a o 9 ¢ . a spa n: "up n. .n ..
a now ~.s~ ; a o a c o n am m.a~ a p o m m cc smo o ow: H. .o .n
m om~ ¢.w~ a o c a c a c on n.o« o o a H a a» a do and H“ .w .u
u a. ¢.n~ c c o o a o c ”N ~.m a o o o c a can a mo“ ¢ .Hfi .m
a pm c.nu c c o w a a c «m n.a~ o a o o a a m ~o mm c .m .n
n «ma ~.¢ o a o c e c a ma ~.nc o a o m a as a am “ma OH .n .m
g was ~.a. a c c c c a a m¢ r.m~ o a n m m a we: as am. pm .m .4
L cw w.u a c c ~ c o o as a.mm o o o o o a o no pm H .H .M
H o. p.o~ c a c c c a a pa o.m a c a a o ma mm a no“ 0 m ..
_¢¢u sop“ ouunw zchb¢~H4mzz‘1u pz>u a4» .a.>ga .zu> m c n m a .p: :mz» hum; was .uuo x2“.
c.m .oz .u>¢ pzuaaau erg nepm \».4uc uz.4 >m zhezu; mamas pzusu>ox zmah xu>

co me p L1H» pa mu-m-<pm nan;

 

97

«cannula» Oh nbwnuhtnh to”: wczouum a «on 9% 99. a 9.9;; 20 cua~<>u¢u mt: ¥u¢944nmm
.mownuuxhh Ch honnuwtnr lock WDZUUbw .99 «on. we 99. a 9.2"; 20 OUJH<>u¢a w<= BUCGJJHaw

9 9.9 9 9 9 9 c 9 9 9 9.9 9 9 - 9 9 9 99 9~ 9 999 9. 99 .999.
9 9.9 9 9 9 9 9 9 9 9 9.9 9 9 9 9 9 99 99 9 99 .9 99 .9999
9 9.9 9 9 9 9 9 9 9 9 9.9 9 9 9 9 9 99 99 no 999. ~ 99 .9999
9 9.9 9 9 9 9 9 9 a 9 9.9 9 9 9 9 9 . 9 999 m9 9~9 9 99 .9999
9 9.9 9 9 9 9 9 9 99 9 999 9 99 99 9 9 99 999 9 .99 9c 99 .999.
9 9.9 9 9 9 9 9 9 9 9 9.9 9 9 9 9 9 9 99¢ 9n9 999 o 99 .9999
999 9.99 9 9 9 9 . 9 9 99 9.99 9 9 9 9 9 999 9 9 999 99 . 999 .99 9
99 9.9 9 9 9 9 9 9 9 99 9.9 9 9 9 9 9 9 999 9 999 9 999 .9 9
999 9.99 9 9 9 9 9 9 9 .9 9.99 9 9 9 9 9 99 999 99 999 .99 99 .9 9
9a 9.99 9 9 9 9 q 9 9 99 9.9 9 9 9 9 9 999 9 9 999 n 999..9 9
99 9.9 9 9 9 9 9 9 9 .9 9.9 9 9 9 9 9 9 99 99 99 9 99 .9 9
999 9.9 9 9 9 9 9 9 9 99 9.99 9 9 9 c 9 9 wow 99 9¢~ 99 99 .9 9
999 9.9 9 9 9 9 9 9 9 99 9.99 9 9 9 99 9 999 9 ~99 999 «n 99 .99 9
9 9.99 9 9 9 9 9 9 9 9 9.9 9 9 9 9 9 99 99 9 99 9 999 .9 9
99 9.99 9 9 9 9 9 9. 9 9. 9.9 9 9 9 9 9 9 9c :99. 999 9 99 .9 9
999 9.9 9 9 9 9 9 9 9 .9 9.99 9 9 9 9 9 99 999 9 9- 9 99 .9 9
99 9.99 9 9 9 9 a 9 9 99 9.99 9 9 9 n 9 999 9 99 99w 9 99 .9. 9
9 9.99 9 9 9 9 9 9 9 9 9.9 9 9 9 9 9 9. 99 9 999 9 99 .9 9
99 9.9 9 9 9 9 o 9 9 c9 «.99 9 9 9 9 9 9N 9 99 9» 9 9c .99
999 9.99 9 9 9 9 a 9 9 99 9.9 9 9 9 9 9 ~99 9 9 9~9 9 999 .o H
999 9.99 9 9 9 9 9 9 99 99 9.99 9 9 9 9 9 9 999 99 ,999 99 9c .9; 9
999 9.99 9 9 9 9 9 9 9 99 9.9 9 9 9 9 9 99 9999 9 9999 99 99 .9
999 ..99 9 9 9 9 9 9 9 99 9.99 9 9 9 9 9 9 999 999 999 99 .99 .o
999 9.99 9 9 9 9 9 9 9 99 9.99 9 9 9 w 9. 99 .999 9 9999 99 9. .9 H
099 9.99 9 9 9 9 . 9 9 99 9.99 9 9 9 9 9 99 9 999 999 . «9 .9 .99¢
99 9.99 9 9 9 9 9 9 9 99 9.9 9 9 9 9. 9 9 999 9 999 w 999 .~ 9
99 9.99 9 9 9 9 9 9 -9 9. 9.9 9 9 9 9 9 9 99 999 999 9 99 .9..9
9.9 9.9 9 9 9 9 c 9 .. 9 99- 9.9. 9 9 9 99 9 99 9 999 999 99 99 .9
99- 9.99 9 9 9 9 9 9 9 99 9.99 9 9 9 9 9 9 999 99 999 99 99 .9 9
99 9.0 9 9 9 9 9 9 9 99 9.99 9 9 9 9 9 9 9 999 999 9 99 .9
9. 9.99 9 9 9 9 9 9 9 99 . 9.9 9 9 9 9 9 99 999 9 999 c 99 .9
.999 99999 99999999922999 9299 999 99.999 .999 9 9 9 9 9 .9 99:9 9999 999 .999 9299
.92 .99. 9299999 999 9999 999999 9299 99 999299 99999 92 9999: 2999 :99

c on 9. Lin» h< WUHPWMP(hW KZNJ A. {$33332

 

98

u4—10>\2Hx moor

9;“ o>.cw "maun;u>\>a.ga ¢>q
he." umdunzw>\v;cru
; mu cu.— a. cow
. on up. «cu :.m
c ya Cc. hon" coma
c u «a. mom a.c—
\ a mm. r. H.&
c mm «Po Foo go!
on soon now“ Not
- r «a. oofi m.—n
. q ~a. co” .m—
: hm G». won aoh
. a ct. ro¢ coca
c e we. o.~ m.¢n
c n rm. 1. $.0—
. h n or ror nor—
hu 1n um. noon Noe"
; a" «a. x.¢~ moan
. an wn. woum a.mw
r rm up. oocw a.>p
~ we. seen cohu
c m 5:. .o» born
c c mm. #0“ now"
e um ch. F.n~ mom
a mu um. «ova noun
u p who can moo
g c wno cow coon
.au wwm .z» \ .uuo puma“
.>u u>q oaowm .c>a .e>¢
VQZOULm c

cc
ch
mm
c

an
cu
mm
mm
—m
ww
mm
om
mm
no
to
55
m—

>aauc
acww

rqmnwnov

no>¢u><auc aq~or kc uu<wzuumua < m< rqauo QMQAOpm

nua~tllu>xu1nh du>a¢p

ozu> c.c>m ">924a3uoc cam:
mam: n“ohm*. maumpuwduwzu>

cochc
a.mwc
m..m~
eocw
couW
nonhc
p.mnw
horn
roen
comp“
hocwu
nova
u.nc—
h.“
hope"
«owm
gouc
Uocm
«.0m
coco
¢ono
uo¢nm
roac
momwm
hour

manxxuuw
LAFXOIu
\u;

omupDZ—a cm

«w achqzuzuu ammxnz r0024“ «on cuwm

u4—2n>\z~x wmou nuaaxlxu>x><4mo

monmam

«uphmnhqhm xmonhuz

~.cn
o.c
phww
w.p
p.m
w.w~
o.H¢
w.m
m.~
o.w~
h.pu
won
m.n~
coca
wow”
w.>
w.-
m.»~
w.h~
w.o~
«ecu
comm
nor“
womm
con

I-IU

"WU

PDW
\

docs»
a.mow
momma
Nome“
momcc

nom¢ow

mochh
cohmn
oomnw
mococ
women
momma
toocn
vonhN
honmm
woman
chmd
hooom
comoN
Honow
«omnw
moomo
momma
mgmon
Hogan

come
m.nn
c.>~
o.nn
“.0"
m.nn
momm
n.u~
moan
coum
oohn
n.~w
monn
mom“
w.nn
a.mm
m.n¢
norm
H.no
c.aw
w.mN
pooh
cooc
n.cn
m.mn

uum

qu> \

uzwp-

mgmpwnpchm ¥z~4
Mn win» Cup<43t~m Gumm<4u

»

hootw
«gnu
"on":
N.00
moan
women
moccs
coon
noon
oomud
mouow
moan
monw
coma
newch
women
Nowtm

nowwmu

wommw
woman
nook

conn
modem
«econ
comm

‘hfli
ht!
OH!

hfl->

Nnomauazazu ouuaw ou><
unuszHtIUJUHIu>

ON.
on.
cc.
who
nto
bu.
we.
00.
«no
at.

nwo
cm.

«ct
mo.
as.
mm.
a».
n».
do.
”w.
m».
an.
«a.

h\x

02H: oowcon u>¢4uo J<p°h
Hutu» much 44h0b\uz~30t
cooONGNuwuanxlwau~zm>

wNmo

m.cs~
c.o~
o.«n~
o.-
o.-
p.mun
«.ocm
n.m.
o.o
o.oo
o.nm
~.c
a.o~
o.~n
..omc
~.~a~
c.~a~
m.n~p
a.mm
~.nn
«.nn
~.-~
m.-~
~.nm
o.n~

.o a 0 mm wish

chh<_:;~w no cZHZZ—uum uuz~w muHhm~h<hw u>~h<aaxau

~.~»
p.9u
n.no~
«.nm
m.c
c.n~
c.oou
p...
~.n~
c.~c
o.pa«
«.mm
o.n~
n.cw
m.o~n
n.-n
«$3
o.coc
~.~H~
o.nc~
o.oc
«.mp
o.omw
n.9c
a.mc

hzwmuxa

Nan
Nnn
ham
Fwd
mun
won
the
n¢n
Own
Own
man
own
nc
Fwd
meta
hwmn
mans
bond
cmN
Nmn
mad
omN
Who
«hm
«mu

«.Nm .«n .e.
n.» 2: ..o
o.no~.o .o
o.a~ .9fi..~
wwu .p ..r
«.o 8 :.
¢.~o .o .u
a.»~ .«n .o
....p. 3 .n
a..« .~ .o
a.mn .m .u
.13 G 1......
~.c .c .«.
n.- .«u .o
p.oo~.c .n
«.co«.n .o
o.o~n.n .r
~.on..c .n
~.~a .u .4
—.nn .«a .N
~.o~ .u .n
n.~n .n .«
o.oo~.o .a
o.o~ .« .~
92 .u ..a

mud": xz~
uzu>

 

99

ooc

coc
coc
soc
coc
coc
coc
cog
coc
coc
soc
coc
coc
coc
coc
coc
coc
coc
coc
coc

ccoc

coc
coc
coc
coc
coo
coc
coc
coc
coc
coc
coc
coc
coc
co:
co:
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc

cNom

mom
mow
Nob
«on
Ooh
coc"
mom
to:
mom
hon
won
Non
con
won
coc
cob
not
«on
con
com
com
now
ton
now
com

mam
cco:

coc
Ea
coc
coc
coc
coc
coc
coc
coc
coc
coc
Coo
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc

nu_h1\mxdoc. wzo~mm~xL L4U~Iu>

r mar» oxuzm»

ccoc

coc
coc
Coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
Co:
coc
coc
coc
coc
coc
co:
co:

a

Cu

«noam

oonmn
Noa¢w
moom
mocm,
poem"
coomm
moan“
roam
comm
coca“
wouh
monc
monh
momw
mo—s
coco
coon
so»:
mom.
pone
hon»
homo”
«oco
womo
Mono
n

a war» oaks: Ubnmoaxou

ccoc

coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc

P

ococ

coc
coc
coc
coc
coc
coc
coo
coc
coc
coc
coc
coo
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc

m

u:

smon

m.»
¢.m
nod
Non
mow
mom“
co—H
w.u.
Non
now
~o¢
m.m
goo
~.n
o.c
non
o.m
«on
wow
wow
hon
wow
wow
mom
won
a

u u ua>h muonzu>

oo.o move cm.a~
wm_hmwp¢»m uoaaoxzozpbz

coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coq
coc
coc
coc
coc
coc
coc
coc
coc
coc

n

con «om
mom mo¢
moo com
mom coma
non o.m
mom con
non coc
“on com”
non coca
con now
moo soc
com now"
moc coc
mo. pom
mom com
moo «oau
sow down
Now noun
coo won"
con somu
not coca
com com
wow Nona
coc nos
coc moca
m H
oooaox

ZOHhAIDMZOU Juan

ccoc

coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc
coc

n

wac—mm—zu no CZQ leurazsmzou Junk no mu34¢> u>-¢datau

...mm n~.mp~
a. n.o
_. ».
n.~ «.au
.. Q.“
a. o.
n. m.~
a." m.-
u. n.”
o. o.
o. ~.~
o. o.n
w. n.~
o. m.
m. o.«
o.. o.a~
n.n ~.on
».o a.mw
n.o a.mn
my” o.m
o. N..
o. m."
o. m.m
u.. ~.m~
o. ~.N
w. ~.H

N a
szJJqu

an“
ace

an

«on

as
aw
aw

and

am
ah
am
am
at

and

at
an

.n

a.
«N

nun

aw
an
Am
AH
AN
tuarh

oca
om
om
ob
oc
oh
own
.0
oh
ow
oo
on
onu
.0
on
.0
o.
on
on“
on
on
on
on
ow
on
MJUH

12";

 

APPENDIX C

PROGRAM OUTPUT OF VALIDATION NETWORK

100

 

101

«mmh
nm\cw\mc u

Mu mercuuzuw mmxtbz zooz<m mom ouMm

<nm<m¢ “03(mo zqccmwo bummm D~5<x muzczn

<DGU5 zH xzozhm7 ozauum no u~uL<~F
N umnu uo hthm

no 20~b<439mw

SIRULSTION 0F TRtFFIC

THE VETSIH HODFL

SIWULATION OF TQIFFIC OF SECOND NEIUORK IN JEDOA

35/26/83

I“!

9 SAUDI ARABIA

[I Q JEDDAH

PRINCE PAJIO sn-

.4

758
PC0

L‘SI DEN

S’ED FOR RINDOH "UNRER GENERLTOP IS

IDENTIFIC‘TI

LANE CHAN
2 3 9 5 TYPE B L

1

OVEN
RT

POCK ‘EAN WURNING N
SP.” L R u-F H LEFT THRU

LANE

LINK

102

HNtfltu MUN U0 udN'fltu‘uohma o_ oo-«xun trunh .mmuo-o
ddenHHH—u-tNNNNNNNNNNnI-I

(‘JNNNIVNN“HUN(VNNNdenlNthdtVNCdU‘C‘M‘JN NNNN

flH—O'4HHHO4HH HH-‘HHHHHHHHH Ho-Ov-‘HHo-Ooddv-O

OCIOCU‘UL‘ c-nurar OCODOC ODDHODC‘DDOOOIDO
00000696 canccuoaouDUOcaocouncacuoo
00006006009000 (3DODC-0Uo3063L3ODODQ'.)
UOODDODODOOODOOOOOUC)OO°OOOOCIOGO

among-Jonas.) ~Onoooaua 9.360 oomnonooc
connccnnnnocnnocccnnnnoc

00¢ cccccdc cccccccctconrOotccd ctcd
nut-u lru'uxlfx"=H-'(‘v’9'-Jf‘.'f.lf‘ cw H' « :wt ow?.uo-u\|¢\m-m(~u‘u

OUOOfJCDDOC‘oCJDCJOOQODOC; CJDOC‘V‘OOOOhOC)

O‘DU'IUJDO CL-OFHPMUDP‘O umuuchwuna tunes.

(Ma «“4 HrmhmHHwWNWHH «nuwndpooacuuwa
w 0 u- 0- co m c

O’DfiQJONF‘DdJOtfiOO‘o-GQNQVDMONOUIOO‘IDWC‘FQDN
o-q “Ono-0H d'fiHHfl‘fidfiHHo-o H o-o HHHNHHH
mm mm 0 mm

U2"; (IQ-“t mt‘w-«flwnhocnmccn Duhnnmnmonnhm
o-o o-IoCo-chIO-tfl H Ho-c fifiHo-oodv-OH
a. a. a. m (I.

ma coc-accoocoucooc ocuoccdaocc cocoa

cmsammcwoucr~r-mmnsc:c-c‘pcumunmcucwonnr.
Hm MN mm H mm H NEH-Mb 0‘ NN H
o-t H H

mccwocmcocwmuanocutcccncnnccnom
an Cowman wfififiob month: a: w 0 520014133th
H d

coracccoar- ”(DID ODONCJCDO woulwchnc FFUH
w N m Clo-1N: HnN

NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
GINNCJNCJNNQVN‘UNNNNNNlVNNNNNNOH‘JNNIVNMJ

>>>>>>>
OOWWODmmooccmmmucummmmnnaamuzmm
nnnnnnNNOOOONNQ{CONNnnNNhhhhhhh
77772,,
uuwuwuw

ccocccccccccccuocccucccnoccucccc

Cl OOOOOOD‘JNNOOUONOOO OOOOUODénDDN
Hui F. vi H

CIOOOOOOODOOOOctot IOOOOOC3000ODOOOO
NNONHF‘VDOCHJID'So'JsDOGmQO ONOOCCND ”VON”
OMQQFQFQQQFOCFQFFQFQW3F5F5$WFFF

thNNtJNh’NC ¢¢ QNCJ. Q0 CoVTINfiJNNhIC¢V£VONC

AQAAAAAAAAAAAAAAAAAAAAAAAAAAQAA
”NONQVDOWMDOwnicmhwt'JNIJFU‘GDC-NC Catfish”
HdddMV-ONHHHH‘NHHfiHHNHHW‘VHHHdHHv-flfl

......‘OOOOOOOOOOO......OOOOOOO
NHN$n¢nDC nmwnOOFNQhOOO‘O'OQOOv-ONHC
Clio-4aid—l-ddhOo-‘o-‘o-O00.000duddde‘HdflOnl'Ofidfl—IH

P'.
l

7 i
:

M

I

   

TOPOLOGICAL FEATURES OF NETDORK

.ENT

EFT-TURN MOVE“

LINKS VITH L

OPPOSING LINK OPPOSIVG LIN< OPPOSING LINK OPPOSING

LINK

“AAA
canw
"Ho-0H

O O O O
044mb
(..HHO4

U...

“A“-
QIONWD
HMHH

O O O O
onset.“
“(80H

tum
vevv

“ARA
IDOCJIH
HNNH

O O o o-
“3011'".-
HMO-“V

U...

A AAA
oncom
HNNH

O ’0.
QWOW-i
MHvOu-O

vwvvv

AAAAA
OIUUO'D
Nv-OCJo-Om

. O O O O
F'I'HO"
v-OHHHF‘

v --vv

noon-AA
cmu¢m
(‘JHNF‘H

o o o o o
FHDNO¢
HAM—0H

00

eve-vb

“AAA“
comes»
HHHHH

o o o o- o
OVINIHU':
FOO-4940'!!!
G'QID

wvvvv

“RAG“

O June's]:
MHHNH

O O 0 0 O
nmhon
o-oo-o—ooao

----“

LEFT-TURNS FROM THE FOLLOUING LINKS HAVE UNOPFOSED MOVEMENT

( 199 15) I

139 12)

(

103

ESSARY

NEC

RI‘Y AND PREPARE TYPE 8 INPUT CARD I‘

PLEASE V

NSITV CATEGORIES

VOLUME (REDS/HR)

PEDESTRIAN 9

CODE

100 TC 255

25C

'3

I“
L]

I‘Pa

ABOVE
NO PEDESTRIANS

(‘4

3

TRAFFIC SIGNAL DtTt
0 INDICATES RTCR IV EFFECT FOR THIS APPROACH

ES

ED APPROACH

CAT
2)-

RBI
1

"0‘

001050191“?-

Cl.

(

ODES FACING
12"

.
g."-

L

4'

ON

H
m

(898, 12"

OF=SET

DURATION

INTVL

NODE

NNF‘DCJC‘J

HL°(‘J OJ" P

AAAAAA
0.0-CLEO.“
c-O "€003
Nn‘FT‘H

VUVV'-

~DH¢ 0‘ Nu:
nnanu-o

AAAAGG
(LEO-0.0.0.
CT' Qfltuun
H 0'0 l")

“U---”

TN'NIH'TOC
H N 0')

HNHCGMD

NNNNNN
woo-Ground

UNDER SIGN CONTROL

NODE 13 IS

APPROACHES

DICATED
13).

IN
209
S

IGNAL
“C

S

I 129 13).
I

=sct

OF

DURATION
89

INTVL

NOOE.

(1OOP)

13

V).

M“
It
UH

10h

<1
C) no“) a a NNNN

an
mu:

AP
(

SIGNAL CODES FACING INDICATED
(3109 14).

14):

('4
CI
to

CF‘SET

DURATION

INTVL

NODE

( 15: 1A).

JNCVNOOQO‘U‘

3.9.00 U NNNOJ

CKJCJN“ «IO U-

aflflflflflflfl
aaaaaama
doduoquw
nmmwcm n

vwvvévvv

N1 HO‘NNCJQ
NNC'U‘MDP- N

RRAAAGAR
QQQQO.Q&0.
ccc-cc-muu»
H N H 0')

vwvvvwvv

OMQHOCNC
H '4 F. N

HNlficunUh‘D

CC {C Ctt C
HflHHHfio-OH

d
oocwwmmwmm

cc:
0.03

AP
(

( 169 15).

15).

SIGNAL CODES FACING INDICATED
(311'

S).

I 1‘9 1

:SET

DURATION OF

INTVL

NOOE

NNNN‘L'HO' 0'

“4:110 O-(‘J‘U‘VOI

(‘Jt‘UCJCvD-UO U-

Aananaaa

“filo-Qua“-
NW. 0'" 5':ch

”'9'---”

o‘JrJno-‘Q ogo’) ‘0
NnnUolanP-H

AAAAAAAA
QLKQQQQQ
ncccoman
'0 N H N

UVUUVVUV

C'MD'ONQN‘
0'. H 'H N

V'anruuf‘ 3'

omnmmuwmw»
HHHHHv-‘Hfi

APPROACHES

SOOES FACING INDICATED

SIGNAL

DFESET

DURATION

INTVL

NOOE

( 170 16).

(8130 16).

16).

(512

16):

m
v-c

U

CCU‘O‘NNNGJ

(VNG‘C‘JO’CLU‘O

ca: w tr NNN N

NUICJNQ'U U 0'

aaanaana

anamanao.
mwncmman
NFTW‘DFW

UVUVVVU-

hhdwuoon
Nmmmhh

Ahflflflflflfl
ILQBQQQQQ
N‘FQHC’CC‘
N (‘4 d (V

UUUVVUVU

OCWC'OCCIV
N N d N

#101” VWWFQ

0 4100 0000
dflHflo‘Ov-‘FOH

 

‘33 1

I I

105

H m H m Ham v n
oHuN omH v oHcN «H - oocm omH oncm onn v
mqu calad ouhcuHozH 02HU<u muocu Jazwnm hwmnuc
Jompzou ach xmcz: wH an uaoz
H m H Hac. . c
w o. HmH omH H onH o,~ H oHcH oNH H
mqu<Oxua¢ om k4 duHozH aZHU<u mMDOU$4<2on humnuc
Hoahzou onm muez: mH oH moo:
m u n .amm . m»
m H m Hawp . mm
w w n .asw . :n
w m H Hume . on
a m p Hanc . mm
H u p Ham . a
. oHaH .FH . onH .«H o.mH .cHa.
muxuqomamd awh¢UHoz H ozHUdk MMOOL Jaszm hMMuLG
m .H H ..aa . A
v HNH ocN a loH omH . «Hsu ooH .
mqu<O¢aaq Gum ;UH .H w2H04u munch AdzuHm humuuo

Homhzou szm mch: m— hm uooz

.aaoH. am
onquao

Hamcuc cm
20~H4¢Do

AU

00

vvvvvv

n
c
c
m
c
mm
:c
HacaH. cm

onH<¢cc

4>th

H>th

Hum 43 .nw

4>th

4>h2~

('J
OJ

wooz

Ca (aroma: mu
0 HHHHHH

U
2

Is
7‘.

uaoz

 

 

 

106

r mm mxzuq >«hzw kn ”mats:

_)
‘J

x mm uumtaz moo: rDtqut

pqux oz¢ aux» zumuu-2¢:» hum; oz
m:o¢¢. zuucu hzmHa o2. puuc zLHa cue
nomad aux» zuuszW7a=A oz

. acum. 4¢2ou4Ho zuuau :LH: aux
ouhthmma zmap Azqu :AH; cum mo oo»m
no¢¢4 hum; zuumo zLH: cum

=c~¢« AzuHa zummolzbwa cu.

cum

zumau

“Emzq «a DJEH>

mmcou quumm no zonhdhumaxukzm

In \D T‘ '0 a‘

OHNMG

107

NHCN
hwm
HNmH

3....
UPC

Hmm
nmh
hHm

d) 0‘

C:
H

:0ch

mxuamh ohua Hulxxm>c whoa 304m

muHhmthhm yrmd rahzw

u Hd>oub7HIm3w

.oH ooHa.
.mH .nHm.
.mH .NHm.
.mH .HHm.
HoH .OHm.
..H .oum.
.NH .mom.
sz4

 

108

«has aqumhmeQFm mw1h<u 07¢ onhddame muzmxsou

wczouwm «mm aubud cuhudnzou oo~xu¢ onhdwqunb~zH

ammumznh »4 on mda >u2<nauuo 2H mu242u hmz armcahuz HI» ouunauuo mu4061u> cmH ow4u>u hmfla cszao
«mmumauh #4 H w<3 ruz<aauuo 2H muz<zu but oxxczhwz wt» canDuuo muJuH1w> mm“ om49>u hm<n czHanc

on owH n

H mmH N

nut NmH H
wm7«zu >uz<aauuc conua

cu2H<hh< zahmmndnscw

nw kHz: O» thme3 mm 44H: «baa >mo»uw1¢¢b u40~zu>
Dummwucaa mm Jan: monmexu 02¢ ronhatnwzou Juan
oxoa cm b
no; maH» xuoau
4a>auth|n3m czHarn.muthHz w >¢w>m «(mama 44H: Panhan u>nhqgctau
mc70umm ch no MJ4>mthH ha m<umuq 44H: ~500thaa omc?ouum c 0 ac oonua a mom
A¢>amthomam no wszzHuum «Uhmd mozouum nco wuuzuzzou baccac uthoutzusz
0299mm Hu4¢>mmk7H mszz<um
oworuumm ccqu J<>mwh9~ UtHb onhaJDxum

0‘)

A ‘J

(\I

u!

109
I; .1

’4

OK \I’ P'-
ZA- H on

 

a.“ H H H H H N N a H.a NH .N H a H - NAN mm nNn m» HNH .oHN.
H.H u H H H o H N H H.o a H H NH HH UH am we ea mN .mH .HHN.
H.H c H H H H H N a °.o H H m N H HH NmN mm mun HN .oH oNHmrv
H.H c H o H H a o N 9.9. o I a c H o .H HH NH a. a HmH .HHo.
H.“ c a m a n o H N . N.o o H a n H HN Ho on Na o HoH .oHN.
H.H H a o H o o m r 9.0 H e m H N H NHH HN nNH NH ..H .uom.
0.0 o a N a a a a H 0.0 H H H N N . NoH an mnH o HNH .mbo.
a.mH H a H a o o H N0 o.N H a H H o Hm H a HH H .mH .ON .
o.NN H H H H o D o HN N.N H H H H H H mH a HH 0 HoN .mH .
H.¢H H H o r o n O NH m.ON a ; u o H HHH _ mm moH NH HNH .oH .
H.nN H H H H H o a on N.» c . H H H ,H Hm o a» m HmH .mH H.
o.NH H a a H o a a He m.m a u o o H H N o A o HHH .oN .
N.mH H H H L o N o H1 H.m a H H a H H oH a aN H HoN .NH .
N.NN N ( . N o o H N m.m H . I u H Hnn H NHn mH HNH .NH .
m.mH H H u H o o H N, ”.mH H N o L H IN ANN u nmN N HNH .NH x
N.HH H H H E H o H Na o.on H H H m g .n NAN NH mHn an .wH .NH .
NIHN H o H . H o c N a.» c u N H H HH wNN N oNN cH HHH .cH H
a.mH H H o c o o H gr .N.nH a H r N H HH H N NH o HmH .ON .
n.NN o H H a o. o o N o.N c H r u H HH NN . o. o HHN .mH .
H.oH a a H H u a m HH n.°¢ o o H H H a No HN nHH m .mH .mH .
n.HH . H H H r a 5 Nb N.NH H H . o 3 HH HHH mN OHH HN .oH .mH .
m.HH H H t H H c H m» n.mm H H N N n “ om Hm nHH nH HoH .mH .
”.mH H H u E o N N N. m.oN H m H H p ”H oNH ”H noH N HmH .oH .
H.NH H H a H o H o NH H.oH a . H a H H a o oN N .HH .oN .
N.NH I H H H o H c .H N.m a ‘ H N H H N N w c .HN .nH .
m.oN H u H . H o a HH 9.: c H a E H NH 9 a» c .nH ..H .
N.HH H H H E u o N HMD m.Hn H , H g a .H co HN N» m HoH .nH .
N.HH n H H o H a H H» m..N H u H N N N uh H mm H .NH .mH .
N.NN n H - H H a H H4 H.m U - N E . HNH o NmH . HaH .NH .
H.HH H H H H o o H Nb N.NN n p a w H "N - mm H» n .NH .nH .
N.NN o H H H o o a HH n.m n s a H . H mm o mm N HHH .NH .
omuam onHHNHHszqxu Hz>u aHu HnH>Ho .;N> m o H N H .H« aux» HENH mHo .uuo szH
.u>q Hzmauau u»u aon \NHHND NZHH >H xHuzNH Hausa Ezuzuao: zxnh :u:

oe N.. qu» b< mUththpw 172..

HO

'H
C

0 0|
(J

c
N

{V

qfin

 

H.H N H N H N N NH H 0.9 HH «H H .
N.H N r H H H N o N N.N N N N H
H.H H N N H N H N r H.N N N N H
N.H H H N H H o H L m.H H e N H
N.c - H L H N H m H .H.N ; L H H
N.H . N - N H N m ‘ N.N H H H H
N.H , N N N H N H H H.N N N H c
N.HH u N H H o H N m. N.H c N H H
N.HN H H N H H H H HN H.H = , N N
N..H H H H H c H H HH H.HN . L N H
N.NN N N H H o o N NN H.N N H H N
H.HH r N H H o H N HH N.H N N o N
a.mH H N H H N o N HH H.N H H N N
H.HN . N . H c N H N H.m N . J H
N.NH H N H H H N N HH N.NH c H H H
N.NH N N N o H H H H” ..oH H N N H
N.NN r H H H N N o H H.N H N H H
...H H N H N a N H H, N.HH N . N H
N.NN N H N H N H N H H.H H r N H
N.NH N N H N H w eH H.0H N m H N
H.HH H . . H H H N N- N.NN u H H N
m.HH : H N H N N H N. H..H H : N H
m.mH H N H o N H N N» H.HN H H H N
H.NH c H H H e o H H4 N.m H c H H
H.H . H N H H H H «H a.» N H N .N
NOON r . n o c o c u" do. w e a n
H.H , H , N c H H. .H; H.NH N . H m
H.¢H H N N H H N N H» m.NN a , N a
N.NN . N N H N e H NH N.m N H H H
N.mH H N H H N H H NH H.NN H H N m
N.HN H H H N H o N HH H.m H H N N
oNNHH onHHNHHszgzu H2>N NHu HnHHHo .:N> N c H N
.u>. Hzmaazu u>u aon \»«HNN NZHH Hm :chNH

‘N no h wruk h< wunhw~h<hw

L‘

' (\

H

fig,

‘ \M (‘4 f a (D

II

o
I

m
D
IJH
D
G

22»;

5

NH

1.4 I7: .'\
n

\H
P)

p.

(‘l " N "‘5 .fl '0‘ '1‘ \h
n H N H H N n

0‘.

a»

|

n

.HN park HHNH
quxuaax zmah

mwn
mm
hnn

mnc
hen
mom
ahn

cm

mm"

mm

mun
hm
as
mm
on
cw
Nm

CI

aduom

N O .D C)
U)

H

mm
me
mH

“100’

DC)

ab
9

one
Fwd
one
am

mad
Asa
Nwm
mm

HN

mm“
can

ow
wmc
nhn
Hoe
son
Hm
mm
HmH
«on
nm«
em»
mn

mad
no
«NH
pmH
mo
oh

HHo
1N)

w

a.
N F)

Qfitho‘

nhQOOq’p—‘N
NW

N

mmcooahmoo—aum
H

.000

How .cum
HoH onum
HoH .NH»
Hm" .Hum
HoH .OHH
.oH owom
H~H .mom
am“ ocN
How .a«
«an own
Hon .mu
aha .cw
Ham .hu
ahH .mH
.mH or"
.0H ohu
Hsu own
Hm“ .ow
Ho~ on“
HmH .oH
HoH .mH
«cu .m«
HnH .cH
and ooN
HHN .nH
.nn ooa
HcH .nH
HNH .mn
HmH .NH.
H~H ona
Ana .«H
12H;

Cl

1]]

N N N o o o NN N c.n NN Nn N N , ch NNN «mm o. NNN .cNo.
N N , o N o NN N N.N N N o NN NN NN NNN co NNN Nn NwN .NNN.
N N N N o o N r N.N .N, N n o N “N cN¢ mm pNn w NmN .NNN.
- N N N a a N N NNN N N a N N N ~N aN on ma N NmN .NNo.
: N : o N N NN N N.o N N , w c "N Wm N. ocN m NcN .ONa.
. N N o N 5 CN L N.N c u N N . N ”NN mN mNN w NcN .woo.
N c N. o o a N N o.o : N N N N qu Nm NNN c NNN .oou.
N N a o c o N N. o.N g N r N h we . a N. N. N¢N .ow .
N N r N N N a NN N.n N . N N N N «N o N N New .mN N
N N o o c a N NN N.NN N N N a N NNN . NN NcN NN NNN .oN .
r a c a a N N a: N.» c N N N N .N NNN a NNN N NNN .NN N
N N N o c N a a: ..NN N o N N N a a a m N NNN .o~ N
N N N o N N a «N N.N N N N N N NN NN N Nn N NNN .NN .
N N N a N o N N . w.m ( N N N N Ncn N mNm NN NNN NNN .
N N N N N N a m\ N.NN L N a N N m UN. N ch wN NoN .NN .
c N N a N o n No N.oN N N N N N . N. no. Np ocn NN NNN .NN v
N N ‘ a a o o N N.o r N N c N Nn Nu. a a». NN .NN .uN N
. N N N c N N NF N.NN a . N N N .N N NN ac N NmN .oN N
N o J o N o o N N.n N N N U N NN N. m mw N Now .mN .
N N N o o N p Nb N.NN N N N N N NN NmN .mN me n NmN .aN N
N N N o a N c «N N.NN N N N N n WN npN c. «NN nN NoN mN N
N N N N o o N on N.NN N . u a c -N nNN a. moN N NoN .mN .
N N a o o N N NM N.NN r N N . N .N NNN NN an nN NmN ocN .
N N N o c u .N aN N.m a N c o a ‘c , c m» N NNN .NN N
N N N a L a .«w NN . N.N N . c N t N N m NN N New .NN N
n o n a N c c m N.c c N N u ( mmN o mwN N .nN ocN N
N N o o o a . N7 0.NN N N N N N NN «N .N NNN . ON NcN .NN N
m N o o o N N n» N.NN N N a N a Nn NNN N cmN 0N N~N .mN N
o o o u u o N NN N.m u a c J N . um N ncN a NwN .NN y
N a .N N o N N N. N.NN . N N N N ,. mm «NN N NNN .nN N
N a N N N o N NN N.m NJ a N N a NN mm o mm o NNN .NN N
l ) 0 I O
.5433ng sawgmffiy szma .NWNNNXNNN NNN

c we 5 »:~» h¢ mumbm-<hw KZHJ

112

 

".9 r H - H o n mH H H.H HH .m H n N cnn mmH wno no HoH .cHoH
. n.“ H n m o m o nH H 0.9 n . H on .H r~ onH Hp n¢H on HHH .nHoH
. 0.9 H n H o o o N H H.o ¢ m m w n H~ so. mHH poo mH HoH .NHHH
. o.o a u a H H H H H 0.6 a H H H H . N Hm wn mp N HmH .HHoH
o.u . H c n H .H - H.w . H H N N H mm a. ohH . .oH .oHuH
. o.n H n H o o n nH , ..a H o m o a HH nHu m» Hmu HH HoH .aooH
. H.H r H n u o n H H o.H m m H e n HmH no nhu n HNH .oooH
an a.mH H o o n c o a pm n.» o m m H 0 mm c H mm o HoH .ou H
.H ».H~ H H H n o a n mp n.n a H o H o H Nn H mm o Ho~ .m« H
-QN n..H . H . H c a H HH c.~u c o o e a moH H moH ecu OH HoH .m« H
an n.n~ . u H a H n H mm c.m H a o o a Hn HcH o omH o HmH .o« H
H m.uH H n n o a o a ¢¢ w.HH a h a H 5 HH H a OH H HHH .o~ H
-H n.wH . u 3 n H n H H- H.m H L H H H HH HN H »n n How .FH H
mm H.H~ - H . - c H a m m.m c H o n c ‘ mnm a mwo m HhH .wH H
-mu n.HH ” a r n o H u (r a.mH H m m n H Ho “Hm a mom nn HHH .pH.H
-mm H..H . u H n o a m sq “.mn n m H m o .m awn om ocm H~ HHH .pH H
Ho u.w~ H H H n a o o H ~.w n ( n . n m amm o mum ~ HFH .HH.H
H 5.4H H a H H n n cH o.nH a H o H H «u n oH me n HmH .Hu H
H.- a a H c a a H H H.n H a o n o mw me n mp H Ho~ .mH.H
“pH ~.nH H o H o n o m or H.oq a u u H m .H shH on .ouu m HmH .mH H
amm H.HH H o - a o o o .H m..m a m o u H Hm (Nu on wow o HoH .mH H
HHH H.~H H H H n o o H HM H.ou H H n H H .H mnH mm -~ m HoH .mH H
noH a.mH : u H o o a w om ~.Hw o a o w a H~ mnm n~ com o HnH .o« H
“a H.mH a H H o e o H N” m.» H o H H n no H o m¢ a HnH .cu H
uH H.wH ( H H o H a u a” m.m H . H H n H c H mH a Ho~ .nH H
mm H.o~ . o H n n H H u 0.; r h H H H HmH o an H HnH ..H H
-HH H.c . a H H o H m m, o.em H . a m H .n no Hn noH » HoH .nH.H
-nH ~.eH - H H H o H Hp n.nm L . a o H mm muH a HmH o H~H .oH H
La a.mn H u H H H a o nH m.w H H m m n oon a paw o HmH .~H.H
«HH H.HH r a o H u n m or m.- H . u c m m c maH omH o H~H .nH H
mm H.n~ c a n n o a a HH m.¢ o H a a c HH mm o «HH H HnH .uH H
QHHm auHaw ;cHHH~HHm2e<zH HHHH «Hm HQHHHQ .;m> m c n m H .Hm aux» HuuH mHo .Huo szH
.c: .u>n unamau xHHH ach \H.Huo uzaH Hm :chuH mamao quxu>oz 2:2» zua

H mm s. 5:» ha mu~hmnhdhm v.2”.—

. '. ‘u r‘

‘

(W “I 0‘) 1o 0\

H3
" U '0 f) P) g) a

1.!

t) C)

I I

I"

k!

mmwmwawm mm mo»«a;2mw mum’s: tooz¢¢ «Du owmm
oorwurqdmc 4<hch no uw<hzuu¢wa 4 md rdauo eunuchw

ua~zo>\z~t cm.” nugnxlxw>xutuh Ju>duh wAth>\2~1 buom.nuJH1IIw>\>¢4mo ozfil nomumN ">44uo Jake»
umw nnoom Hugo—Iu>\><4uo u>q qu).boo¢~.n>uz«a:uuo zcu: cmomanaxatH ouuaw .o>¢ one. umxub aux» J<hoh\62u>ox
«non uwau~2w>xwa0hm anon mwobwwH va~¢h0u40~zu> «cacao nmu>92~x0maunxu> uaommauumwantlwaquu>

muHFMHhuhm xacnbuz

N do.“ w." hohu 0. Norm «Ha noncm nown wooc who cow“ hon» wm coma and How
H mm. we comm Nu Oonw non Noah” monN n.w« am. now ao0d at Now How omu
mu Na. coo Wood uh «00cm moNN honmw comm. monam we. wommn nowuu coc hocm “mu cod
m ~0- ~on NonN nN comm «HF Nonmu monN some Hm. hoNn 0.00 mtw «own and How
a no.“ no Nora“ mn comm coca oomom nonm now um. mow ham mm com Aha oo~
m econ a. hood c~ momm nah ~o~0~ m.m~ Mohw Hp. ooh Coma um mom How ohm
h am. noon cohN N coon new coana moan moan” up. cone ocean mnm mowedahu one
an mm. been now" as mound coma «ooNN momn name: ac. nonmw coohu mum noomuaaa ohu
hm hm. mopm mofl ha noohw Neat momwn Honm o.n~w 0N. conwm nooow «no orcnuamu oh“
p nu. com nomN « oonc new moon" room m.oaw poo Noam wocon new «ohwuahu own
N 90. Non Donn «h mono mot" Ho&mm uohn mohn “we not" N.NN cw mom and ooN
m no.0 mod noNN a moan . «on comma ¢o¢~ cow: Hm. mom . moot can «ohm .oN own
a 5 am. mom noun nh coumw noWn ~oomn momm «oumw m~o comma hock onn mom: and omu
h hr. Nada Nod“ up homwm honn 0.NN” most Morn» am. aoanm mshm kw: oon Amu 9nd
a ho. no» one" no wooma ,bwmw mohmw comm. mcnuw Hm. comm“ N.NN mNn N.00 nod oma
h moo coo won" me comhu mowN howmw none boonN "no woman mono out Homo and Hen
N noon no" mohu «N mon n.m Nomou moon nown me. now“ nohw sh moan and oo~
u nooa no nohu ow meow a.» mohmm momN wood ms. won com mm. «am “am ona
c nu. how 0..N h oomN aom mowed 0.NN Noow rm. oatd memo mam a.mn and .0"
a a mh. new 0.0 cm «omcw Nown ootmn moum comma. .no co&u« Noam mum comm “cu one
w who how n.na nh momma Nomm momo~ moat ~.om~ mm. momma Mnofi wow mote «Nd Hod
s N. wow comm NH sown mom hone" moNN n.0mu ah. .oon0 Honmu Noe nook and .Na
0 hr. ccm cone oh women «gnu momww boa. mom.“ moo mon n.mm tau Nonn ANA onn
n wmo now comm on comm woe room” honw noow um. moNH .coum nwu «omw and own
pun In: ><4uo u4~1\uwc umm u4~r\uMm uum 2mtl> autl> z~z0>
an paw Im>x .uuo owuaw norm uqnzlzu> rm» \ wquxozm> o1w> \ Him» »\I urn» utnh am» mmqmz szJ
; u>¢ mmCFm om>< ou>¢ hum \utuhlo utnplo \uznhu» wt~bth Aahuh ><4uo 0,0: xua oxw>

muHhmHh<hm xz~4
wczouum u omUHDZH: on ma win» cwhqdaxnm numaqdw .m n n ma uxuh h2mmuma
onhadatmm no wznzzmoum uuzum mu~hm~pabm u>mthDIDU

f‘ (o o) u)
aunt-roman

r:

(.1

1H4

(5 ('c

')

 

mam u n um>h oxuamh n N mar» oOb3< wh~montou n H mm)» uduuxw>

um.n -... ac.a mm.mm Ho.c Ha.” m;.¢ om.a H¢.«.oo.m 04.0 oH.ou mm.oHH
mHHHmHHHHm uoH:nx¢o:Huz

u.~ -.n H.o H.oc H.o H.H m.m 0.0 H.s ~.¢H 0.0 H. m. HaH .oa
H.. H.” ”.0 m.Ho u.o H.o n.“ n.o «.. N.NH H.n H. e. Ho~ .mu
H.. ... 2.0 m.n¢ H.o ... ~.m n.n 3.H c.HH ”.3 H. n.m HoH .o«
».H ... h. ~.n¢ H.n c.u m.n o.n m.m n.HH m.o m. ~.n HoH .mm
... o.” H.o m.~e u. H.a H.n °.n ¢.m m.HH H.c a. N. HsH .o~
H.. H.H a.o a.wc H.n 9.0 u.m u.H H.n m.HH mum H. o. How .pH
v.0 mom can boo: mom a.” top . woo non h.m poo . moN nonn «ha .0“
H.m H.“ c.o u.em n.o o.“ m.m H.H m.m ~.HH ..c H.~ H.m HmH .5H
..m ... H.o n.moH H.o o.“ o.m 0.0 ¢.m m.m H.H m.H a.mH HoH .nu
v.m a.” H.o p.m. n.o a.“ o.n 0.9 o.n H.0H o.H H.n o.HH .nH .mH
m.n H.” H.o o.~m ..o o.; H.m o.n H.o o.HH m.m H. s. HmH .ou
... a.» o.o ¢.¢» u.o H.H m.n u.o c.m o.nH o.o ~. ~.H Haw .nH.
..w a.“ 9.0 ~.xgH n.n H.H H.m n.o H.m s.m 0.; o.H p.o HnH .oa
H.m a.» 0.9 H.co u.o a.m e.m a.o o.n h.~ v.5 H.H «.5 HoH .mu
H.” ... o.c H.Hm m.n c.u m.. 0.0 H.. o.m H.o m. H.m HcH,.nH
com moc uoo nomm moo can “on u.o con c.m goo New Nah and Hon.“
... ... o.o ..me “.0 m.. N." o.o n.m m.HH H.o m. m. HnH .ou
N.. 5.” o.o n.m¢ o.o a.” H.» o.H m.m n.~H H.“ H. m. Ham .nH
m.¢ u.” o.o «.9o 0.» H.. m.“ 9.0 m.. ~.HH 9.“ o. w.~ HnH ..H .
... J.“ o.c m.nm H.H H.n H.m o.“ m.. ¢.H H.” m. o.n HoH .nH.“
H.m ..H o.o m.ah u.o .H.m H.a H.a H.u H.oH o.o m. H.¢ H~H .m« H
M.H ... 0.0 ¢.c¢ H.o H.o H.n o.n 0.. s.HH m.o ~.H m.m . HmH .NH H
..H H.” o.o m.om a. H.” m.n n.H H.m ¢.HH H.o n. ~.~ HmH .nH H
w.m H.“ H.a 0.N. n.n 0.. u.m . H.o m.c m.HH H.u . m. m.H HnH .uH H
H n N H n u H m m H m m H nua»H uHqun
ou . H: .o.a.: monH<H .
HmHHzxmz<muH monmmeu MHquu> zonhazamzou Juan szH

mzonmmmxu ur 02¢ onhaznmzou Juau no munada u>HH<JDSDU

BIBLIOGRAPHY

HS

BIBLIOGRAPHY

"Traffic Network Ana1ysis With NETSIM" Imp1ementation Package. A
User Guide. FHWA. January 1980.

Lieberman. E. "Enhanced NETSIM Program." TRB Specia1 Report
#194. 1981.

Hagerty. 8.. and Ma1eck. T. "NETSIM: A User's Perspective." TRB
Specia1 Report #194. 1981.

Labrum. w. "App1ication of NETSIM Computer Simu1ation Mode1 to
Traffic Contro1 Prob1ems." TRB Specia1 Report #194. 1981.

Hur1ey. 3.. and Radwan. A. "Traffic F1ow Simu1ation: User
Experience in Research." TRB Specia1 Report #194. 1981.

Labrum. W.. and others. "Ana1yzing Intersection Performance
with NETSIM." Pub1ic Roads. June 1978. pp. 24-29.

Fer1is. R.. and others. "Experimenta1 App1ications of the UTCS—1
Network SimuTation Mode1." Highway Research Record #669. HRB.
1978.

Ross. P.. and Gibson. D. "Review of Road Traffic Network Simu-
1ation Mode1s." TRB #664. 1977.

Rode1at. G.. and Ti11er. G. "A New System for Traffic Simu1a—
tion." Pub1ic Roads. V01. 45. No. 1. June 1981. pp. 19-26.

MacGowan. John. and Fu11erton. Iris. "Deve1opment and Testing of
Advanced Contro1 Strategies in the Urban Traffic Contro1
System." Pub1ic Roads. V01. 43. No. 2. Sept. 1979; N0. 3.
Dec. 1979; No. 4. March 1980.

Ger1ough. D.. and Wagner. F. "Simu1ation of the Network." Con-
ference on Traffic Survei11ance. SimuTation and Contro1.
Washington. D.C.. Sept. 1964. pp. 152-65. U.S. Dept. of
Commerce. Bureau of Pub1ic Roads.

Ke11. James. "Simu1ation of the Intersection." Conference on
Traffic Survei11ance. 1964. pp. 127-43.

116

 

13.

14.

15.

16.

17.

~ 18.

19.

20.

21.

22.

23.

24.

25.

26.

117

Zo1tan. Nemeth. and Mekemson. James. "NETSIM and SOAP--A Test
of Compatibi1ity." ITE Journa1. V01. 52. No. 10. Oct. 1982.
pp. 38-41.

Ross. Pau1. "The Future of Traffic Simuiation." Pub1ic Roads.
V01. 45’ N0. 2) septa 198], pp. 75-790

Rade1at. Guido. "Simu1ation Deve1opments in Progress." TRB
Specia1 Report #194. 1981.

Cooper. D. L.. and Wa1inchus. R. J. "Measures of Effectiveness
for Urban Traffic Contro1 Systems." Traffic Systems Office.
TRW Systems Group. Houston. Texas.

Cohen. S. L. "App1ication of Network Simu1ati0n Mode1s to the
Ana1ysis of Urban Intersection Performance." FHWA-RD-74-25.
Fina1 Report. September 1973.

Gibson. David. and Ross. Pau1. "Simu1ation of Traffic in Street
Networks." Pub1ic Roads. V01. 41. N0. 2. Sept. 1977. pp. 80-90.

Kube1. Les1ie. and others. "What Network SimuTation (NETSIM) Can
Do for the Traffic Engineer." Pub1ic Roads. V01. 41. No. 4.
March 1978. pp. 162-68.

Lieberman. E.. and others. "Logica1 Design and Demonstration of
UTCS-1 Network SimuTation Mode1." Highway Research Record
#409. Highway Research Board. January 1972.

Box. Pau1. "Manua1 of Traffic Engineering Studies." ITE.
Fourth Edition. 1976.

Cronje. W. B. "Modei for Optimization of Iso1ated Signaiized
Traffic Intersections." Paper presented at the 62nd Annua1
Meeting of the TRB. Jan. 1983.

Hasse11. John. "Operating Efficiency of Traffic Signa1s."
AASHTO. V01. 58. No. 3. Ju1y 1979.

Hur1ey. Jamie. and others. "Sensitivity of Fue1 Consumption and
De1ay Va1ues From Traffic Simu1ationJ' TRR #795. Transporta-
tion Research Board. 1981.

Lieberman. E.. and Andrews. 8. "TRAFLO: A New T001 to EvaTuate
Transportation System Management Strategies." TRR #722.
Transportation Research Board. 1980.

Institute of Transportation Engineers. "Transportation and
Traffic Engineering Handbook." 2nd Edition. Prentice-Ha11.
1982.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

118

Pignataro. "Traffic Engineering Theory and Practice."
Prentice-Ha11. Inc.. 1973.

Hamburger. W.. and Ke11. J. "FundamentaTs of Traffic Engineer—
ing." 9th Edition. Institute of Transportation Studies.
University of Ca1if0rnia. Berke1ey. 1977.

Sa1ter. R. "Highway Traffic Ana1ysis and Design." Addison-
WesTey Pub1ishing Co.. Inc.. 1974.

Woh1. M.. and Martin. B. "Traffic System Ana1ysis for Engi-
neers and P1anners." McGraw-Hi11 Book Co.. 1967.

Chin. Shih-Miao. and Eiger. Amir. "Network SimuTation Inter-
active Computer Graphics Program." Rensse1aer Po1ytechnic
Institute. Troy. New York. Paper presented at the 61$t TRB
Annua1 Meeting. Washington. D.C.. Jan. 1982.

Berg. W.. and Do. Cheu1-Ung. "An Eva1uation of Network Traffic
Performance Measures Using Computer Simu1ation Mode1s."
Paper presented at the TRB Annua1 Meeting. Jan. 1981.

"Network Fiow Simu1ation for Urban Traffic Contro1 System--
Phase II." V01. 5. App1ications Manua1 for UTCS-1 Network
SimuTation Mode1. FHWA Fina1 Report. March 1974.

"Traffic Statistics. 1970-1981." Ministry of Interior. Pub1ic
Security. Genera1 Dept. of Traffic. Saudi Arabia.

"Summary of Saudi Arabian Third Five-Year Deve1opment P1an.
1980-1985." Pub1ished by TIHAHA. Jeddah. Saudi Arabia.
1981.

"Fata1 Accident Reporting System 1980." U.S. Dept. of Trans-
portation. NHTSA. Washington. D.C.

Nemeth. Z.. and McKemson. J. "Comparison of SOAP and NETSIM:
Pretimed and Activated SignaT Contro1s." Paper presented
at the TRB Annua1 Meeting. Washington. D.C.. Jan. 1983.

Davis and Ryan. "Comparison of NETSIM Resuits with Fie1d
Observations and Webster Predictions for IsoTated Inter-
sections." TRB Specia1 Report #194.

Master P1an Report. Jeddah. Regionai and Town P1anning
Department. Ministry of Interior. Municipa1 Affairs.
Kingdom of Saudi Arabia. 1973.

\, 4o.

4].

-\'42.

43.

\,44,

45.

46.

47.

48.

49.

\/50.

51.

52.

119

Gibson. D.. and Ross. P. "Simu1ation of Traffic in Street Net-
works." Transportation Engineering. V01. 47. No. 12. Dec.
1977.

Jadaan. K.. and Sa1ter. R. "Traffic Accidents in Kuwait."
Traffic Engineering and ControT. V01. 23. No. 4. Apri1 1982.

Hsu. U.. and MunjaT. P. "Freeway DigitaT Simu1ation ModeTS."
TRR #509. TRB 1974. pp. 29-41.

"The App1ication of Traffic Simu1ation Mode1s." TRB Specia1
Report #194. Conference Summary. Findings. and Recommenda-
tions. 1981.

Gibson. Danie1. "Avai1ab1e Computer Mode1$ for Traffic Opera-
tions Ana1ysis." TRB #194. 1981.

Berg. W.. et a1. "Case Study Eva1uation of A1ternative Signa1
Timing P1ans for an Oversaturated Street Network." ITE
Journa1. ApriT 1982.

Chin. Shih-Miao. "Interactive Computer Graphics User Interface
for Traffic Simu1ation Mode15." TRB. Specia1 Report #194.
1981.

Yagar. Sam. and Case. E. "Summary Eva1uation of UTCS-1/NETSIM
in Toronto." TRB. Specia1 Report #194. 1981.

Brugggeman. Jeffrey. "Network F1ow Simu1ation for UTCS.
Technica1 Report." FHWA. June 1971.

Ministry of Interior. Pub1ic Security. Genera1 Department of
Traffic. Jeddah Office. An officia1 Tetter.

May. Ad01f. "ModeTS for Freeway Corridor Ana1ysis." TRB
Specia1 Report #194. 1981.

Schafer. Bruce. "Comparison of A1ternative Traffic Contro1
Strategies at a T-Intersection." TRB Specia1 Report #194.
1981.

Mahmassani. Hani. and others. "Investigation of Network-Leve1
Traffic F1ow Re1ationships: Some Simu1ation Resu1ts." Paper
presented at the 63rd annua1 meeting of the TRB. Washington.
D.C.. Jan. 1984.