0% found this document useful (0 votes)
360 viewsMay - Traffic Flow Fundamentals
Uploaded by
Nguyen MinhNhatCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
360 viewsMay - Traffic Flow Fundamentals
Uploaded by
Nguyen MinhNhatCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 473
TRAFFIC FLOW
FUNDAMENTALS Library of Congress Cataloging-in-Publicarion Data
May, Adolf D. (Adolf Darlington)
‘Trafic Row fundamenas/Adtt D, May
Bet
Indludes bibliographical ferences.
ISBN 0-13-920072-2
Ir Trac Now —Matheratcal models, 1. Tie
TIMG9 1990
1 de00 99-39687
or
Editorial/production supervision: Virginia L. McCarthy
Cover design: Edsal Enterprises
‘Manufacturing buyer: Denise Duggan
© 1990 by Prentice-Hall, Inc.
A Division of Simon & Schuster
Englewood Cliffs, New Jersey 07632
‘All rights reserved. No part of this book may be
reproduced, in any form or by any means,
‘without permission in writing from the publisher.
Printed in the United States of America
098765432
ISBN 0-13-926072-2
Prentice-Hall Intemational (UK) Limited, London
Prentice-Hall of Australia Pry. Limited, Sydney
Prentice-Hall Canada Inc., Toronto
Prentice-Hall Hispanoamericana, S.A., Mexico
Prentice-Hall of India Private Limited, New Delhi
Prentice-Hall of Japan, Inc., Tokyo
Simon & Schuster Asia Pte, Ltd., Singapore
Editora Prentice-Hall do Brasil, Ltda., Rio de Janeiro Contents
PREFACE xi
1 INTRODUCTION 1
Raanenl
An Analytical Approach 2
‘Textbook Organization 4
Part One: Traffic Characteristics 4
Part Two: Analytical Techniques 5
Future Challenges 6
Selected Problems 8
Selected References 9
PART ONE: TRAFFIC CHARACTERISTICS
2 MICRoscopic FLOW CHARACTERISTICS n
24
2.2
23
24
25
26
27
‘Time Headways and Time Headway Distributions 12
Classification of Time Headway Distributions 14
Random Headway State 15
Constant Headway State 19
Intermediate Headway State 23 |
Evaluating and Selecting Mathematical Distributions 36
Selected Applications 41 vi Contents
2.8 Summary 48
2.9 Selected Problems 48
2.10 Selected References 50
3 > MACROSCOPIC FLOW CHARACTERISTICS 54
3.1 Temporal Flow Pattems 55
3.2 Spatial Flow Patterns 59
3.3 Modal Flow Pattems 60
3.4 Mathematical Count Distributions 66
3.5. Estimating Hourly Flow Rates 70
3.6 Uninterrupted Traffic Applications 72
3.7 Interrupted Traflic Applications 75
3.8 Selected Problems 78
3.9 Selected References 80
4 MICROSCOPIC SPEED CHARACTERISTICS 83
4.4 Vehicular Speed Trajectories $3
4.2 Speed Characteristics under Uninterrupted Flow Conditions 92
43° Mathematical Distributions 95
4.4 Evaluation and Selection of Mathematical Distributions 101
4.5. Estimation of Population Means and Sample Sizes 108
4.6 Selected Problems 110
47 Selected References 114
5 MACROSCOPIC SPEED CHARACTERISTICS 116
5.1 Speed and Travel Time Variations 117
5.2 Importance of Mean-Variance Relationships 126
5.3 Travel Time and Delay Study Techniques 133
5.A Travel Time and Delay Study 145
5.5 Selected Problems 154
5.6 Selected References 157
6 — MICROSCOPIC DENSITY CHARACTERISTICS 160
6.1 Distance Headway Characteristics 161
6.2 Cai-Following Theories 162
6.3. Car-Following Theory Application 173,
6.4 Trafiic Stability 177
6.5 Measurements with Presence-Type Detectors 181 Contents
66
67
Selected Problems 187
Selected References 190
7 MACRoscopic DENSITY CHARACTERISTICS
TA
72
73
74
78
76
77
78
Historical Development 195
Density Measurement Techniques 197
Density Contour Maps 201
Introduction to Shock Waves 205
Estimating Total Travel Time 213
Estimating Traffic Demand 214
Selected Problems 223
Selected References 224
PART TWO: ANALYTICAL TECHNIQUES
8 DEMAND-SUPPLY ANALYSIS
a4
82
83
84
85
An Analytical Framework 227
An Example of Demand-Supply Analysis 232
Summary 244
Selected Problems 245
Selected References 246
9 CAPACITY ANALYsis.
9
92
93
94
95
98
97
98
9g
Historical Background 247
Scope and Limitations 249
Muttilane Facilities 250
Ramps and Ramp Junctions 255
Weaving Sections 260
Signalized Intersections 268
Summary 279
Selected Problems 279
Selected References 281
10 TRAFFIC STREAM MODELS
10.1
10.2
10.3
‘Stream Flow Fundamentals 284
Practical Considerations 287
Proposed Individual Models 296
vil
192
227
247
283 1
12
13
vill
10.4 Proposed Family of Models 303
40.5 Pedestrian Stream Models 315
10.6 Selected Problems 318
10.7 Selected References 319
‘SHOCK WAVE ANALYSIS
11.1 An Historical Perspective 322
14.2 Shock Wave Equations 323
14.3. Shock Waves at Signalized Intersections 325
11.4 Shock Waves along a Highway 328
14.5. Shock Waves along a Pedestrian-Way 330
14.6 Complexities and Refinements 332
14.7 Selected Problems 334
11.8 Selected References 336
QUEUEING ANALYSIS.
12.1 Input Requirements and Classification 339
12.2 Deterministic Queueing Analysis 339
12.3. Stochastic Queueing Analysis 360
12.4 Selected Problems 371
12.5 Selected References 373
COMPUTER SIMULATION MODELS
18.1. Introduction to Simulation 377
13.2. Steps in Developing a Simulation Model 380
18.3 An Example of Developing a Simulation Model 384
13.4 Some Computer Simulation Models 396
13.5 Summary 409
13.6 Selected Problems 409
18.7, Selected References 411
APPENDIXES.
A
B
HEADWAY TABULATIONS AND PARAMETER VALUES OF FOUR
MEASURED TIME HEADWAY DISTRIBUTIONS
DERIVATION OF POISSON COUNT DISTRIBUTION
Contents
321
338
376
a5
420 rOoOnmoo
Contents,
EXPONENTIAL FUNCTION TABLES
CALCULATION PROCEDURES FOR NORMAL DISTRIBUTION
GAMMA FUNCTION TABLE AND EXAMPLE PROBLEM
CHI-SQUARE TABLE VALUES
LIST OF SYMBOLS
ENGLISH-METRIC CONVERSION FACTORS
INDEX
425
433
438
440
450
453 Preface
‘This book is designed for a one-semester first-year graduate course in traffic flow funda-
mentals, with approximately one week being spent on each chapter. Senior-level under-
graduate courses and courses with fewer credit hours can use the book with certain
‘modifications as noted. The book is also intended to be used as a ready reference by
professionals and researchers because of its in-depth comprehensive coverage of
research results and applications. People from other countries and those working with
‘other modes of transportation may find the book valuable because of its coverage of
‘basic principles and theories which cut across political and modal boundaries.
In the early 1970s the author developed a new first-year graduate course at the
University of California devoted to traffic flow fundamentals. Over the years an exten-
sive set of course notes was developed and continuously updated. In 1985 and 1987
draft chapters were tested in a classroom situation, Student comments were most help-
ful in improving the chapters and encouraging completion of the book. The author is
indebted to the students, from whom he has learned so much and who have contributed
in so many ways to this book
The author also wishes to acknowledge the contributions of four of his colleagues
who reviewed the final draft and offered many thoughtful suggestions: Professor Jan L.
Botha, University of Alaska; Professor Paul P. Jovanis, University of California at
Davis; Professor Robert Layton, Oregon State University; and Professor K. C. Sinha,
Purdue University. A word of appreciation is also extended to Mr. Doug Humphrey
and the staff of Prentice Hall for their continuous support and encouragement.
In closing, I would like to dedicate this book to my family. To my mother, who
sacrificed and encouraged my education with love and affection. To my father, who
introduced me to transportation engineering and always served as an example of a man
T would aspire to become. To my wife, who has been my partner in life, sharing the
sorrows and enjoying the happinesses. ‘They made this book possible.
xi Introduction
Knowledge of fundamental traffic flow characteristics and associated analytical tech-
niques is an essential requirement in the planning, design, and operation of transporta-
tion systems. Traffic flow characteristics include time headway, flow, time-space pro-
Jectory, speed, distance headway, and density. Traffic flow analytical techniques
include supply—demand modeling, capacity and level-of-service analysis, traffic stream
‘modeling, shock wave analysis, queueing analysis, and simulation modeling. The pur-
pose of this book is to provide an in-depth treatment of these fundamentals of trafic
flow, and one chapter is devoted 10 each of these traffic flow characteristics and associ-
ated analytical techniques.
‘Transportation system planners, designers, and operators should have a basic
Knowledge of traffic flow fundamentals. Planners assess traffic and environmental
impacts of proposed system modifications, and this can be accomplished only through &
supply-demand framework that requires understanding of flow characteristics and their
interrelations. Designers determine link sizes and configure systems, and must carefully
evaluate the trade-off between traffic flow levels and levels of service. Operators iden
tify locations and causes of existing system defects and generate operational improve-
rent plans and predict their effects. For these reasons and others, knowledge of traffic
flow fundamentals is required for all graduate transportation students at the University
of Califomia,
Although this book focuses primarily on the highway transportation system, traffic
flow characteristics and associated analytical techniques are fundamental to all transpor-
tation systems. All transportation systems have particles moving along links that
interact with each other, interact with the physical facility, and require analysis and con-
trol. In air transportation, for example, flow characteristics are studied and analysis
made of air-side and land-side subsystems. On the air side, flow control schemes have
1 2 Introduction Chap. 1
bbeen developed for the airways, simulation models developed for use in design of air-
Ports, and queueing models employed for allocating aircraft to gate positions. On the
land side, analysis has been made of baggage handling, person flow at check-in counters
and gate positions, and mode transfer at the terminal interface. In water transportation,
simulation models have been developed to study line-haul ocean operations and ferry
systems, and queueing analysis has been applied to lock operations and port facilities.
In fixed-guideway systems, time~space diagrams are in continuous use for railroad line
operations, simulation models for urban rail systems and freight yards, and queueing
analysis for transit stations. One should not overlook special transportation systems
which also require traffic flow fundamentals, such as for pedestrian-ways, escalators,
Sec. 1.4 Part Two: Analytical Techniques 5
Chapter 3 is concerned with macroscopic flow characteristics, which are expressed
as flow rates, and attention is given to temporal, spatial, and modal patterns. Empirical
evidence is provided, analytical techniques are described, and applications are presented.
Chapter 4 describes microscopic speed characteristics of individual vehicles pass-
ing a point or short segment during a specified time period. Particular attention is given
to vehicular speed trajectories over time and space as well as to statistical analysis of
individual speed measurements.
Chapter 5 is directed to macroscopic speed characteristics, which are concemed
with the speed of groups of vehicles passing a point or short segment during a specified
period. Particular attention is given to temporal, spatial, and modal variations as well as
to statistical analysis of group speed measurements. Trave! time and delay techniques
are also included,
Chapter 6 is devoted to microscopic density characteristics, which are concerned
with individual distance headways between vehicles, with particular emphasis on
minimum and average values. The chapter includes an extensive coverage of car-
following theories and automatic data collection systems.
Chapter 7 is concerned with macroscopic density characteristics, which are
expressed as the number of vehicles occupying a section of roadway. Density measure-
‘ment techniques are described and particular attention is given to density contour maps.
‘Analytical techniques using density contour maps are described.
1.4 PART TWO: ANALYTICAL TECHNIQUES
Analytical techniques associated with individual traffic characteristics are discussed in
earlier traffic characteristics chapters. The remaining portion of the book is devoted to
analytical techniques involving the total traffic flow situation. Table 1.3 provides a
framework of these analytical techniques as presented in Chapters 8 through 13.
Chapter 8 provides a transition from the treatment of individual traffic characteris-
tics to a demand-supply framework, which permits the integration of analytical tech-
niques presented later. The demand-supply framework is introduced followed by a
rather comprehensive example of traffic control and traveler response interactions.
Chapter 9 is devoted to capacity and level-of-service analysis. ‘The supply,
demand, and control features are analyzed to determine the capacity and the trade-off
between the quantity of traffic and the resulting level of service to the users. Attention
is given to multilane facilities, including ramps and weaving sections as well as to sig
nalized intersections.
Chapter 10 is concerned with traffic stream models which provide for the funda-
‘mental relationships of macroscopic traffic stream characteristics for uninterrupted flow
situations. The relationships are for free-flow and congested-flow conditions. Extensive
sets of field measurements are described and single- and multiregime models presented
and evaluated,
Chapter 11 presents shock wave analysis, which is one of the techniques, available
for the analysis of oversaturated traffic systems. Traffic stream models serve as a 6 Introduction Chap. 1
TABLE 1.3 Framework for Trafic Analysis Techniques
‘Analytical Microscopic "| Macroscopic
‘Technique || individual Units) | (Groupe of Units)
Supply analysis |} Minimum time | Capaciy analysis
| hesdways| (Chapter 9)
(Coapter2)
ow Time—space low-speed
relationships siagrams density
(Chapter 10) | relationships
‘Chapter 10)
Shock wave Time-apace | Shock waves
alysis, diagrams (Chapter 11)
(Chapter 11)
(Queueing analysis | Discrete Continuous
‘anais analysis
(Chater 12) | (Chaper 12)
‘Simulation modeling || Disorete Contnaoue
models models
(Chapter 13) | (Chapter 13)
See also Chapter &
beginning point for shock wave analysis. The fundamentals of shock wave theory are
presented. Examples of the application of shock wave theory 10 signalized intersec-
tions, uninterrupted highways, and pedestrian-ways are illustrated.
Chapter 12 is devoted to queueing analysis, one of the techniques available for the
analysis of oversaturated traffic systems. A classification of queueing analysis tech-
niques is formulated which covers the gamut from deterministic situations to completely
Stochastic situations. Queueing theory is presented followed by a variety of applica
tions.
Chapter 13 describes computer simulation modeling techniques and their applica-
tion to traffic, systems. Both microscopic and macroscopic simulation models are
covered, The chapter includes a step-by-step procedure for developing a simulation
model, an example of actually developing a simulation model, and a description of
some models curently available.
1.5 FUTURE CHALLENGES
Like many other transportation and infrastructure systems, highway transportation
passes through periods of expansion and consolidation. In this century the 1920s and
1930s marked the rapid expansion of the highway system in both road mileage and
route quality. The construction of the interstate highway system, including the urban
freeway system, were major accomplishments of the 1950s and 1960s which provided Sec. 1.5 Future Challenges 7
for high-quality intercity and intracity travel never before obtained. These eccomplish-
ments improved the quality of travel so much that phenomenon growth in traffic
demand occurred, and soon demands approached or exceeded capacities at many loca-
tions, particularly in the urban and suburban areas.
‘The 1970s ushered in @ new era of consolidation and with it transportation system
management. As demand continued to approach or exceed capacities, attention turned
to controlling the demand while making only spot improvements on the capacity side.
During the 1980s, transportation system management has continued but has been
entianced and broadened by the anticipation of innovative vehicle, computer, and elec-
tronic technology. Two closely related challenges are envisioned for the remainder of
this century and the beginning of the twenty-first century: a more systematic and
comprehensive transportation system management activity and a gradual implementa-
tion of the most promising new technologies. Both will be directed toward making
‘maximum use of the existing transportation system Trafic flow fundamentals will play
‘an important role in meeting these two challenges.
Planners, designers, and operators of the transportation systém all have a role to
play in developing @ more systematic and comprehensive transportation system manage-
‘ment activity. ‘The skills of planners will be needed to develop and apply improved
techniques for evaluating the impacts of land use changes and in developing more pre-
cise behavior models of the effects of system changes on spatial, temporal, modal, and
total traveler responses. The ingenuity of the designer will be required to identify crti-
cal links in the system where capacity increases are urgently needed and to develop
design plans that meet the needs but with serious constraints on available right-of-way
‘and environmental impacts. The tenacity of the operator will be essential to identity
‘operational defects and to develop plans to balance system operation on a real-time
basis
Innovated vehicle, computer, and electronic technology ate on the threshold of
developments that have the potential of making maximum use of the existing ranspor-
tation system. Technologies with the greatest potential must be identified and a gradual
implementation plan developed, New vehicle technologies include in-vehicle longitudi-
nal and lateral information warning systems, radar brakes, and perhaps ultimately, fully
automatic controlled guidance systems. Computers can play an even greater role in the
future in both offline and on-line operations. Off-line computer packages are becoming
faster, more flexible, and user fiendly. The use of on-line computer systems provides
the opportunity of engaging improved control theory algorithms, such as atifical intel-
ligence, expert systems, fuzzy sets, and the like, to make maximum use ofthe highway
system under normal and unusual trafic conditions. New electronic technologies
interact strongly with vehicle and computer technologies. New detectors, communica-
tion links, and control processors are being researched intensively. New roadside and
in-vehicle information systems are being developed that may lead toward navigation
systems and route selection under dynamic traffic conditions.
‘The highway system of today carries a more significant number of vehicle-miles
of travel than ever before—greater than that for which it was designed. Demands con-
tinue to grow at faster rates than improvements are being made. The movement of per-
sons and goods has gradually deteriorated. Transportation systems management and 8 Introduction Chap. 1
new technologies offer the greatest challenge and hope for iinproving the quality of
movement. The ability to understand and apply traffic flow fundamentals is an essential
ingredient in working toward improving the transportation system. It is hoped that this
‘book will stimulate and involve the reader in the challenges that lie ahead.
1.6 SELECTED PROBLEMS
1. Define the six traffic ow characteristics identified in this chaptet.
2, Describe the six trafic flow analytical techniques identified in this chapter.
53. Expand the examples in the text on how planners, designers, and operators use traffic flow
fundamentals,
4.
5
Develop an expanded Table 1.1
Describe the elements contained in Figure 1.1. Consider extensions such as environmental
Conditions, feedback loops, and other elements that you feel should be included.
6. Why do interrupted oversaturated flow situations require more complex analytical tech-
niques?
7. What are some of the factors to be considered in selecting between microscopic and
macroscopic analyses?
Study the organization of the book. Are there othe organizational schemes that you would
consider?
9. Select one chapter ofthe hook that interests you especially and waite a half-page summary
deserbing its contents.
10, During the original work onthe book a separate chapter on time-space diagrams was con-
sidered, to illstate another analyial technique. It was finally decided to include this
‘atrial in other chpters. What other chapters contain materials on time-space diagrams,
and do yo fel tht the coverage is adequate?
11, Define tansporation system management and provie ahistorical perspective.
12, Write a new Section 15 and éiseuss what you think the fare challenges wil be.
1B, The year is 2000, What new technologies would you expect to see on your new autorno-
vite?
14, Wha impact do you feet the following new vehicle technologies wi have on capacity?
(a) Longitudinal an lateral infomation warning systems
(b) Radar brakes
(©) Fully automatic convlied guidance systems
18, Conduct brief literature search of one ofthe following control theoy agoritims
(a) Arica intelgence
(b) Expert systems
(Fuzzy ws
16 Conduct bie literature seerch of in-vehicle information systems.
17. Do you fel that in-vebice information systems will rove tobe costeffetve? Why ?
18, In relation tothe materials covered in tis chapter, what professional activites do you see
for yous inthe fist quarter ofthe twenty-first cent? Ctiap. 1 Selected References 9
1.7 SELECTED REFERENCES
1. W.D. Ashton, The Theory of Read Traffic Flow, John Wiley & Sons, Inc, New York, 1966,
178 pages.
2, Paul C. Box and J. C. Oppenlander, Manual of Tratfic Engineering Studies, 4th Edition,
Institute of Transportation Engineers, Washington, D.C., 1976, 233 pages.
3. Per Bruun, Port Engineering, 3rd Edition, Gulf Publishing Company, Book Division, Hous-
ton, Tex., 1981, 787 pages.
4, Don Drew, Traffic Flow Theory and Control, McGraw-Hill Book Company, New York,
1968, 467 pages.
5, John J. Fruin, Pedestrian Planning and Design, New York Metropolitan Association of
Urban Designers and Environmental Planners, lnc, New York, 1971, 206 pages.
6, Denos Gazi, Trafic Science, John Wiley & Sons, Inc, New York, 1974, 293 pages.
7. Daniel L. Gerlough and Matthew J. Huber, Trafic Flow Theory—A Monograph, Transporia-
tion Research Board, Special Report 165, TRB, Washington, D.C., 1975, 222 pages.
8, Danie! L. Geriough and Matthew J. Huber, Statistics with Applications to Highway Traffic
‘Analyses, Eno Foundation, Saugatuck, Conn, 1978, 179 pages.
9, Frank A. Haight, Mathematical Theories of Traffic Flow, Academic Press, Inc., New York,
1963, 242 pages.
10, William W. Hay, Railroad Engineering, 2nd Edition, John Wiley & Sons, Inc., New York,
1982, 758 pages.
11, Wolfgang $, Homburger and James H. Kell, Fundamentals of Tragfic Engineering, 12th Eali-
tion, University of California Press, Berkeley, Calif, 1988.
12, Robert Horonjeff and Francis X. McKelvey, Planning and Design of Airports, 3rd Edition,
‘McGraw-Hill Book Company, New York, 1983, 616 pages.
13, J.J. Leeming, Statistical Methods for Engineers, Blackie & Son Lid., Glasgow, 1963, 146
ages
14, Wilhelm Leutzbach, Introduction 10 the Theory of Traffic Flow, Springer-Verlag, Beslin,
1988, 204 pages.
15, M. H. Lighthll and G. B, Whitham, On Kinematic Waves: A Theory of Traffic Flow on
Long Crowded Roads, Proceedings of the Royal Society, Series A, Vol. 229, 1957, pages
317-345,
16, Louis J. Pignataro, Traffie Engineering—Theory and Practice, Prentice-Hall, Tne., Engle-
‘wood Cliffs, NI, 1973, 502 pages.
17. Vukan R. Vuchic, Urban Public Transportation: Systems and Technology, Prentice-Hall,
Inc., Englewood Cliffs, NIL, 1981, 673 pages.
18, F. V. Webster and B, M. Cobbe, Traffic Signals, Her Majesty's Stationery Office, London,
1966, 111 pages.
19. Marin Wohl and Brian V. Martin, Traffic System Anabysis for Engineers and Planners,
McGraw-Hill Book Company, New York, 1967, 558 pages.
20. American Association of State Highway and Transportation Officials, A Policy on Geometric
Design of Highways and Streets, AASHTO, Washington, D.C., 1984, 1087 pages.
21. Organization for Economic Cooperation and Development, Route Guidance and In-Car Com:
‘munications Systems, OECD, Pars, February 1988, 104 pages. i 10
2,
28,
Introduction Chap. 1
‘Transportation Research Board, Highway Capacity Manual, Specisl Report 209, TRB, Wash-
ington, D.C., 1985, 474 pages.
Institute of Transporation Engineerings, Manual on Uniform Trajfic Control Devices, Wash-
ington, D.C., 1989, 562 pages.
Great Britain Road Research Laboratory, Research on Road Traffic, Her Majesty's Stationery
Office, London, 1965, 505 pages.
Organization for Economic Cooperation and Development, Dynamic Traffic Management in
Urban and Suburban Road Systems, OECD, Pasis, April 1987, 104 pages,
(Organization for Economic Cooperation and Development, Traffic Capacity of Major Routes,
OECD, Paris, July 1983, 120 pages.
‘Transportation Research Board, Traffic System Management, Special Report 172, TRB,
Washington, D.C, 1977, 163 pages.
Institute of Transportation Engineers, Transportation and Traffic Engineering Handbook, 2nd
Eaition, Prentice-Hall, In., Englewood Cifis, NJ, 1982, 883 pages. 2
Microscopic Flow
Characteristics*
‘The time headway between vehicles is an important flow characteristic that affects the
safety, level of service, driver behavior, and capacity of a transportation system. A
‘minimum time headway must always be present to provide safety in the event that the
lead vehicle suddenly decelerates. The percentage of time that the following vehicle
‘must follow the vehicle ahead is one indication of level or quality of service. The dis-
tribution of time headways determines the requirement and the opportunity for passing,
merging, and crossing. The capacity of the system is governed primarily by the
minimum time headway and the time headway distribution under capacity-flow condi-
tions, For these reasons it is important that the designer and operational manager of
transportation systems have a theoretical knowledge of time headways and their distri-
‘butions and the ability to apply this knowledge to real-life traffic problems. The pur-
pose of this chapter is to provide a theoretical base with practical applications of this
important microscopic flow characteristic.
Although primary attention will be given to highway traffic systems, time head-
ways and time headway distributions are also important in other modes of transporta-
tion, such as in pedestrian, rail, water, and air transportation. The theoretical principles
and practical applications contained in this chapter can, in large measure, be utilized in
these other modes, particularly where the human beings control the movement, Tran-
sportation modes in which the individual vehicles are controlled or scheduled automati-
cally provide special cases for study.
‘This chapter begins by defining and describing time headways and time headway
distributions. Next, a classification scheme is presented which provides systematic
‘IF Limi time is avaiable, the more advanced materials included in Section 2.5 can by scanned ot
mines
" 2 Microscopic Flow Characteristics Chap. 2
structure for all of the most relevant theoretical distributions. Then the mathematical
formulations and characteristics of each such theoretical distribution are given. At this
point real-world time headway data are introduced and techniques are presented for
Evaluating and selecting most appropriate mathematical distributions to represent the
real-world time headway distributions. In the final section examples are provided for
applying the principles covered in this chapter to a variety of highway traffic problems.
2.1 TIME HEADWAYS AND TIME HEADWAY DISTRIBUTIONS
Distance
‘A microscopic view of traffic flow is shown in Figure 2.1 as several individual vehicles
traverse a length of roadway in single fle for a certain period of time. The arrival time
of each vehicle at the observation point is noted as f, f2» 15, andt.. The elapsed time
between the artival of pairs of vehicles is defined as the time headway, and thus the
time headways can be shown as
Mra =t2
‘Observe that the time headway (f) actually consists of two time intervals: the accu-
ancy time for the physical vehicle to pass the observation point and the time gap
between the rear of the lead vehicle and the front of the following vehicle. In theory,
‘an individual time headway does not necessarily have to be the elapsed time from the
passage of the leading edges of two consecutive vehicles but only the elapsed time
between the passage of identical points on two consecutive vehicles. However, in prac-
tice the leading edges are used whether the measurements are taken automatically by
detectors or manually by observers, Occupancy time is discussed in Chapter 6 where
(dps stata, ete. en
Figure 21 Microscopic View of
Time “Tratfic Flow Sec. 2.1 Time Headways and Time Headway Distributions 18
microscopic density characteristics are covered. The time gap parameter is used infre-
quently in highway waffic analysis.
An observer could continue to record individual time headways at a specific loca-
tion for periods of time representing different flow situations. Then the individual time
headways for each flow situation could be sorted into time headway intervals and plot-
ted as shown in Figure 2.2.* ‘The four time headway distributions are for four traffic
flow levels: 10-14, 15-19, 20-24, and 25-29 vehicles per minute. The vertical scale
{is minute flow rate and the four measured time headway distributions are positioned on
this scale at their average minute flow rate values. The horizontal scale is time
Hexcy oc
2 2 4 ‘ ‘ 10
Ney
tobi 158 298 sm sm
= ew
raf
T=s0
138
2 sk 4
=
g
z
i
Zoe a
i omy
as 4
7 L
Figure 22 Measured Time Headway Distributions (Prom Reference $)
"The time headway distibutions should normally be shown a ba graphs, but fr illustration purposes
‘only are shown in Figure 2.2 a¢ continuous distributions. 4 Microscopic Flow Characteristics Chap. 2
headway in seconds, and each of the four distributions is ploted on this scale based on
O.s-seosnd time headway intervals. The height of the shaded areas represents the pro-
portion of observed headways in each 0.5-second headway interval. Finally, there ae
ev sets of contour lines superimposed on top of the distributions, which represent the
ean time headway and the cumulative percentage of headways (1, 15, 33, 50, 67, 85.
tind 99%), Headway tabulations and parameter values for the four measured time head
\way distributions are summarized in Appendix A.
‘This extensive data set of some 14,570 individually measured time headways
presented in Figure 2.2 and Appendix A provides a basis for the reader to obtain
fnsights and identify pattems of measured time headway distributions [5]. Some of the
pertinent observations are listed below.
« Individual time headways are rarely less than 0.5 second (on the order of 1 to 2
percent).
«Individual time headways are rarely over 10 seconds unless the minute flow rate
is below 15 vehicles per minute.
© The time headway mode is always less than the median, which is always less than
the mean. However, they tend to converge as the minute flow rate increases
toward capacity.
© ‘The mean time headway tracks the 67 cumulative percentile curve for the entire
minute flow rate range.
«The ratio of the standard deviation to the mean time headway approaches 1 under
ow flow conditions but decreases continuously as the minute flow rate increases.
“The subject of time headways and related mathematical distributions has been stu-
died at least since the 1930s and the reader is encouraged to review some of the early
eferences 10 get a historical perspective (10, 15, 16, 35, 40, 42, 44, 45]. Many data
kets of individually measured time headways are available in the literature and the
Teader is encouraged to study them to validate the observations identified above (5, 6,
31, 32, 34, 36, 38, 39, 58, 62, 68].
2.2 CLASSIFICATION OF TIME HEADWAY DISTRIBUTIONS
‘The shape of the time headway distribution varied considerably as the waffic flow rate
increased. This was observed in Figure 2.2 and was due to the increasing interactions
between vehicles in the traffic stream. For example, under very low flow conditions,
there is very litle interaction between the vehicles and the time headways appear t0 be
Somewhat random. As the traffic flow level increases, there are increasing interactions
between vehicles. In simple terms, some headways appear to be random, while other
headways are of vehicles, that are following one another. As the trafic flow level
approaches capacity, almost all vehicles are interacting and are in a car-following pro-
cece. In this process all time headways are approximately constant. A classification Sec.2.3 Random Headway State 15
scheme is proposed for time headway distributions consisting of a random distribution
state for low flow levels, an intermediate distribution state for moderate flow levels, and
4 constant distribution state for high flow levels. The following three paragraphs
describe this three-level classification scheme.
Under very low flow conditions, all the vehicles may be thought of as traveling
independent of one another. In other words, any point in time is as likely to have a
vehicle arriving as any other point in time. The only modification in real life would be
due to a minimum time headway always being required from a safety point of view.
Except for the minimum headway specification, these headways could be considered as
random time headways and the time headway distribution as a random time headway
distribution. This situation will be classified as the random headway state and can be
‘considered as one of the two boundary conditions,
The other boundary condition is that when the traffic flow level is near capacity.
This situation is discussed next, and the intermediate case is-discussed in the following
paragraph. Under heavy-flow conditions, almost all vehicles are interacting, and if an
observer stood at @ point on the roadway, the time headways would be almost constant.
The only difference from real life would be that while the driver would attempt to main-
tain a constant headway, driver error would cause some variation regarding this constant
headway. Except for this variation, these headways could be considered as constant
time headways, and the time headway distribution as a constant time headway distribu-
tion. This situation will, be classified as the constant headway state and can be con-
sidered as the other boundary condition.
‘The intermediate headway state lies between these two boundary conditions. That
is, some vehicles are traveling independent of one another, while other vehicles are
interacting. It will be shown in the next section that this is the most difficult to analyze,
yet itis this situation that is most often encountered in the real world, Several different
analytical approaches are employed for the intermediate headway state. One approach.
is to select a mathematical distribution that contains parameters that will modify the
‘mathematical distribution in a continuous spectrum from a random distribution to a con-
stant distribution. A variety of other approaches have been proposed and will be
presented. This situation will be classified as the intermediate headway state and can be
considered the state between the two boundary conditions
2.3 RANDOM HEADWAY STATE
The negative exponential distribution is the mathematical distribution that represents the
distribution of random intervals such as time headways. For time headways to be truly
random, two conditions must be met. First, any point in time is as likely to have a vebi-
cle arriving as is any other point in time. Second, the arrival of one vehicle at a point
in time does not affect the arrival time of any other vehicle,
The negative exponential distribution can be derived from the Poisson count dis-
tribution, Note that the negative exponential distribution is an interval distribution;
‘That is, itis the distribution of the number of individual time headways in various time 16 Microscopie Flow Characteristics Chap. 2
headway intervals, This can be seen graphically in Figure 2.2. The Poisson count dis-
tnibution, on the other hand, is a count distribution; That is, it is the distribution of the
number of time periods which contain different flow levels.
“The Poisson count distribution derived in Appendix B will be utilized in Chapter
43, The equation for the Poisson count distribution is
mie™
a 2)
Po)
probability of exacting x vehicles arriving in a time interval ¢
average number of vehicles arriving in a time interval ¢
"x = number of vehicles arriving in a time interval being investigated
‘a constant, Napierian base of logarithms (¢ = 2.71828...)
selected time interval
Consider the special case where x=0. That is, consider the case when no vehicles
arrive in a time interval. Then equation (2.2) can be rewritten
PO
23)
‘What is the significance of P(0) in terms of individual time headways? If no vehicles
arrive in dioe interval (), the individual time headway must be equal to or greater than
1. Therefore,
PO=Ph2D) e4)
Pen es)
Recall that m is defined as the average number of vehicles arriving in time interval . If
the hourly flow rate is specified as V and ris expressed in'seconds, then
n= (gh) ao
and equation (2.5) becomes
Pia ene Qn
‘The mean time headway in seconds, 7, can be determined from the hourly flow rate, A
by
360,
v
Equation (2.7) can be modified by equation (2.8), and an alternative formulation
ccan result:
28)
Paenae™ 29)
In this form, given the mean time headway, f, the analyst can select various ‘time inter-
vals f, such as 0, 0.5, 1.0, 2.0,...40% and P(lt 2 0, 0.5, 1.0, 2.0,...» 2) can be caleu-
Tated. Note that when t = 0, P(h 2 1) = 1.0 which is obviously correct. As the value of t
increases, P (ht 2 1) decreases until finally when ¢ —> ee, P(h 21) 0. Sec.2.3 Random Headway State 7
Although analysts may be interested in probability distributions, such as P(h >),
it is often desirable to calculate the probability for specific time headway intervals, such
as the probability of a time headway between r and r + Ar seconds, This can be accom-
plished by calculating P( 21) and P(h>1+Ar) and submacting in the following
manner:
PO Sh 1)
values are calculated and are shown in Table 2.1. Exponential function tables are
included in Appendix C to aid the analyst in solving equations (2.9) and (2.11). The
next step is to calculate the probabilities for each time headway interval by utilizing
equation (2.10), and these results are also shown in Table 2.1. The final step is to con-
vert the calculated probabilities of each time headway interval into the frequency of
headways for each time headway group. This is accomplished using the equation
NIP SA St +A0)] 2.12)
Fsh
