|
|
马上注册，结交更多好友，享用更多功能，让你轻松玩转社区。
您需要 登录 才可以下载或查看，没有账号？注册
x
Preface
“ Everything should be made as simple as possible—but not simpler”
Albert Einstein
Traffic Theory, like all other sciences, aims at understanding and improving a physical phenomenon. The phenomenon addressed by Traffic Theory is, of course, automobile traffic, and the problems associated with it such as traffic congestion. But what causes congestion?
Some time in the 1970s, Doxiades coined the term "oikomenopolis" (and "oikistics") to describe the world as man';s living space. In Doxiades'; terms, persons are associated with a living space around them, which describes the range that they can cover through personal presence. In the days of old, when the movement of people was limited to walking, an individual oikomenopolis did not intersect many others. The automobile changed all that. The term "range of good" was also coined to describe the maximal distance a person can and is willing to go in order to do something useful or buy something.
Traffic congestion is caused by the intersection of a multitude of such "ranges of good" of many people exercising their range utilisation at the same time. Urban structures containing desirable structures contribute to this intersection of "ranges of good".
In a biblical mood, I opened a 1970 paper entitled "Traffic Control --From Hand Signals to Computers" with the sentence: "In the beginning there was the Ford". The Ford was indeed the beginning of man';s modern age of transportation, and the enlargement of individual oikomenopolis, and this beginning took place shortly after the beginning of the 20th century. For a few decades after that, people exploited the capabilities of the new medium by trial and error, (and I am suppressing here the urge to add "mostly error").
The Ford, and its follow up companion cars, increased the mobility of people and gave birth to urban sprawl and its concomitant traffic congestion. Traffic signals, borrowed from railroad train management technology, were brought into the scene to improve safety and combat congestion, with mixed results due to inadequate understanding of the inner workings of automobile traffic. And then Traffic Theory was born.
Around the middle of the 20th century, some notable scientists laid the foundations of understanding of automobile traffic. You can find the names of some of these scientists in an article I wrote for the 50th anniversary issue of Operations Research, entitled "The Origins of Traffic Theory". Much work followed after those pioneering contributions, some of it productive and some readily classified as "solutions in search of a problem". Traffic Theory followed the tradition of other physical sciences, with many scientists entering the field in order to apply what they knew to model traffic. Some of them did not make the necessary effort to justify their modeling effort in terms of realistic models of human behaviour, often ignoring the admonition of Albert Einstein shown at the beginning of this preamble. Comments on such misguided effort will be made later in this book. Suffice it to say here that mere good fit with data is not sufficient to justify a theory, which must also be based of a rational consideration of physical properties of the things that the theory purports to describe. This book concentrates on theoretical work tested over the years, which is based on such foundations of proper pursuit of science.
Traffic theory, after modelling of traffic movement, moved into modelling of interactions of traffic movements aimed at improving them through appropriate control of traffic movement through devices such as traffic signals or other means of regulating the rate of flow of traffic past a control point. Thus, traffic control theory for congested or uncongested systems was developed, once more with mixed success in producing the desired improvement of traffic flow.
If we were to evaluate the overall effect of progress in traffic theory on traffic management, we would find "mixed success" as the universal measure of this effect. This is so in spite of considerable
effort made in recent years to utilise advanced computer technology in order to improve the management of transportation systems, under the label of "Intelligent Transportation Systems (ITS)". A short explanation of why success is mixed may be the following:
a) Some good traffic theoretical models are inadequately used.
b) Some other good theoretical models are missing, and inadequate attention is paid toward their development.
c) Some of the suggested solutions to the traffic problem are aptly described by the edict of Harold Mencken : “For every problem there is a solution that is simple, neat, and wrong” .
In this book, I present key models of traffic flow and associated traffic phenomena such as conflicts in traffic, traffic generation and assignment, and traffic control. The book builds on my earlier work, (*Traffic Science, edited by Denos C. Gazis, Wiley Interscience, New York, 1974), and the work of other transportation scholars. I have focused on identifying the appropriate use of the various models developed over the years for the improvement of traffic systems. I discuss how proper use of some of these models may accelerate the successful deployment of ITS, and also enumerate opportunities for development of additional models needed for continued improvement of ITS. The book comprises four chapters, namely:
1.Traffic Flow Theories: A description of the movement of traffic and the interaction of its component vehicles with each other.
2.Delay Problems at Isolated Intersections: A description of queuing and delays at isolated intersections.
3.Traffic Control: A discussion of algorithms for optimum control of single intersections and systems of intersections, including networks.
4.Traffic Assignment: A discussion of traffic generation, distribution and assignment in a transportation network.
I would like to express my gratitude to Walter Helly, Donald McNeil and George Weiss, and the late Leslie Edie, whose contributions to the aforementioned book “Traffic Science” provided the basis for the corresponding chapters of this book.
The book is intended as a textbook for a college Transportation Science curriculum, and as a reference book for researchers in Transportation Science. I have concentrated in presenting the fundamental concepts and methods in the various areas of traffic theory, but not necessarily covering all the contributions, which do not alter substantially these concepts and/or methods. Some of these contributions are listed in the References section of each chapter as “additional references”.
Denos Gazis
CHAPTER 1. TRAFFIC FLOW THEORIES
Macroscopic Traffic Flow Theory – Kinematic
Waves in Traffic
Shock waves
Flow versus concentration relationships
The Boltzman-like Model of Traffic
The two-fluid model of traffic 19
Microscopic Traffic Flow Theory – Car-Following Models
Local stability
Asymptotic stability
Acceleration noise
Non-linear car-following models
Some Questionable Paradoxes
The Smeed paradox
The Braess paradox
Additional Flow Theory Work, and Future Challenges
CHAPTER 2. QUEUEING AND DELAYS AT ISOLATED INTERSECTIONS
Traffic Characteristics
The Gap Acceptance Function
The Delay to a Single Car
Delays to Pedestrians
Queueing and Delays of Vehicles
The Delay at a Traffic Signal
Evaluation of delays
Optimization of traffic signal settings
The vehicle-actuated traffic signal
CHAPTER 3. TRAFFIC CONTROL
Objectives of Traffic Control
Single, Isolated Intersection
Synchronization Schemes for Arterial Traffic
A mixed integer-linear programming approach
The effect of queues on progression
The TRANSYT method
The Combination method
The SIGOP program
Traffic Responsive Operation of Traffic Lights
Single intersection
Systems of intersections
Additional contributions
The SCOOT program
The OPAC, PRODYN, CRONOS, and COP programs
The Onset of Oversaturation
Oversaturated Systems
A single oversaturated intersection
Complex oversaturated systems
Oversaturated Store-and-Forward Networks
Freeway Control
Systems Affected by Geometric Details
CHAPTER 4. TRAFFIC GENERATION, DISTRIBUTION,AND ASSIGNMENT
Network Representation of a Transportation System
Trip Generation and Distribution
Trip generation
Trip distribution
Time-Independent Traffic Assignment
Deterministic traffic assignment models
Discrete choice, stochastic, models
Global Network Optimization Models
Dynamic Traffic Assignment
Traffic Assignment in Congested Systems
Delivery of traveller information services
APPENDIX
Application of Kalman Filtering for Density Estimation in Traffic Networks
Indexnfei
[ 本帖最后由 水寿松 于 2009-4-28 12:58 编辑 ] |
|
IP卡
狗仔卡
发表于 2006-5-26 22:00:26
提升卡
置顶卡
沉默卡
喧嚣卡
变色卡
显身卡
楼主
