1

DRAFT Traffic Flow Theory and Characteristics

AHB45: Committee on Traffic Flow Theory and Characteristics

Traffic Flow Theory and Characteristics

DRAFT Millennium Paper

List Author Here, Affiliation

Other Authors Here, Affiliations

The TRB Committee on Traffic Flow Theory and Characteristic is concerned with the development, validation, and dissemination of theoretical, experimental, and applied research on traffic flow theory and traffic flow characteristics and the determination of the relationship of traffic flow theory and traffic flow characteristics to the planning, design and operation of transportation systems. To mark the passage of the millennium, all TRB committees have mounted special efforts to capture the current state of the art and practice and their perspectives on future directions in their respective areas of focus. This paper begins with a review of the important evolutionary changes that have occurred in the field of traffic flow theory since the inception of the Committee on Traffic Flow Theory and Characteristics in 19XX. With the guideposts defined by this history, a vision of future directions the committee expects to undertake in the next decade is presented. Summary observations about the committee’s past, present and future conclude the paper.

MILESTONES

This section describes the top milestones or accomplishments in the development of the field of traffic flow theory that define the current state of the art. Historically, traffic flow theories have sought to describe in a precise mathematical way the interactions among vehicles, drivers, and the infrastructure (1). Generally these descriptions have been linked with empirical observations, but not always. As generally conceived, the infrastructure consists of the highway system and all its operational elements, including control devices, signage, and markings. These theories are an indispensable element of all traffic models and analysis tools that are being used in the design and operation of streets and highways.

The conceptualization of traffic as a science began in the 1930s with the first empirical data collection efforts and subsequent application of probability theory to the description of road traffic. This empirical thread has been woven through the developments in traffic science until today. The pioneering empirical studies conducted by Bruce D. Greenshields (2) at the Yale Bureau of Highway Traffic led to a simple model relating flow and speed, and the investigation of performance of traffic at intersections. After World War II, with the tremendous increase in the use of automobiles and the expansion of the highway system, there was also a surge in the study of traffic phenomena, with notable empirical efforts led by Leslie Edie and others involving the Lincoln Tunnel in New York. (3, 4) Along these lines, Wardrop also contributed to the scientific approach to transportation phenomena (5).

In 1959, Robert Herman of the General Motors Research Laboratories in Warren, Michigan organized the First International Symposium on the Theory of Traffic Flow (6, 7). This was the first of what has become a series of symposia on the theory of traffic flow and transportation. Tracing the topics covered in the seventeen proceedings of these symposia indicates the tremendous developments over the last 40 years in the understanding and the treatment of traffic flow processes. These proceedings also indicate exciting developments beyond the highway mode and involvement of researchers from many disciplines and many countries around the world. (7)

The field of traffic flow theory and transportation has become too diffuse to be covered by any single type of meeting, and numerous other symposia and specialty conferences about a variety of traffic-related topics are held on a regular basis. Yet, even as traffic flow theory is increasingly better understood and more easily characterized through the availability of high resolution sensor data and advanced computing capabilities, the fundamentals are just as important today as in the early days. They form the foundation for all the theories, techniques, and procedures that are being applied in the design, operation, and development of advanced transportation systems. (1)

In 1964, the Highway Research Board published the Monograph on Traffic Flow Theory as HRB Special Report 79 (8). A completely rewritten monograph was published as TRB Special Report 165 in 1975[1] (9). In 1987, the Committee on Traffic Flow Theory and Characteristics recommended that a new monograph be prepared as a joint effort of committee members and other authors. In 1999, the document was Traffic Flow Theory: A State of the Art Report (10) was published at: www.tfhrc.gov/its/tft/tft.htm. The revised monograph contains ten chapters, the subjects of which are good indicators of the major progress in the understanding of traffic flow theory over the past 50 years:

  1. Introduction
  2. Traffic Stream Characteristics
  3. Human Factors
  4. Car Following
  5. Continuum Flow Models
  6. Macroscopic Flow Models
  7. Traffic Impact Models
  8. Unsignalized Intersections
  9. Signalized Intersections
  10. Traffic Simulation

Review of several recent summaries of the development of the field of traffic flow theory (11, 12, 13) and a survey of members of the Committee on Traffic Flow Theory and Characteristics and other leading scholars in traffic flow theory revealed that certain developments in the history of traffic theory stand out above the others.

Macroscopic Flow Models

The survey indicated that the pioneering work by Bruce Greenshields on macroscopic flow relations (2) helped establish a scientific perspective toward traffic. Next, the recognition of the similarity between traffic flow and compressible fluid, which led to macroscopic traffic flow modeling based on application of hydrodynamics, was mentioned. In particular the hydrodynamic models developed by Lighthill & Whitham (1955) and Richards (1956), referred to collectively as LWR models, and enhanced by others, serve as a guidepost in the history of traffic theory (14, 15). Newell’s simplified kinematic wave theory was also a significant development mentioned in the survey results (16). The monograph by Prigogine and Herman (19) discussing continuum models has been the basis for more recent work including cellular automata models. The discretization of the LWR by Daganzo’s cell transmission model (18, 19) also opened a pathway of research in dynamic traffic assignment and network modeling. The development of “higher order” flow models, starting with Payne (20) and Daganzo’s important criticism (21) further serve as milestones in traffic flow history. Extensions of this work more recently in the physics research community have included development of a three-phase hypothesis, including the consideration of a synchronized flow state.

Microscopic Flow Models

It is thought that the current state of the art is focused on how to structure models of traffic flow, including developing an understanding of what traffic characteristics are important, how they relate to one another, and how they evolve over time and space in traffic networks. As a foundation for the more popular modeling approaches, the recognition of the similarity between traffic flow and moving particles led to the development of microscopic car following models, including the work by Chandler, Herman and Montroll (22) and Herman, Montroll, Potts, and Rothery (23). Also notable was that this included the recognition of the importance of human factors in traffic flow modeling, which led to microscopic traffic flow modeling based on car-following, lane-changing, and gap acceptance. Extensions to the macroscopic and microscopic perspectives have led to hybrid traffic flow models.

Data and Simulation

The third set of accomplishments in the field of traffic flow theory are related to the advancements in sensing and computing that have led to improved empirical analyses and simulation techniques. These efforts have been inspired in part by the early empirical efforts of traffic science pioneers. Edie’s work aimed at a generalized description of basic traffic stream measurements was particularly notable (24). Another example is the work by Treiterer and Myers (25) which focused on individual vehicle level trajectories measured over a short freeway segment. It is thought that the increasing availability of rich sensor data has improved our understanding of empirical spatiotemporal features of congested traffic patterns at freeway bottlenecks. Some new diagnoses, theories, and traffic management/control techniques have resulted from these empirical and theoretical results. The wide deployment of detailed traffic sensing and measurement systems has increased the focus on the variability and reliability of traffic prediction tools. The development of advanced computer simulation of traffic flow has substantially changed the field of traffic flow theory by allowing us to shift our focus to microscopic traffic characteristics and variability. The committee may be able to play a role in expanding these kinds of activities.

Overall, it is clear that major developments have been made in traffic flow theory. Much of the past work has been done outside of the auspices of this committee. While work has been reported via this committee, this perspective offers an opportunity to think about ways that the Committee on Traffic Flow Theory and Characteristics can increase its relevancy in the future. Other accomplishments mentioned in the committee survey included Webster’s pioneering work to develop signal cycle optimization formulae and his research on signalized intersections (26). Next we will turn to a discussion of the major challenges that lie ahead for the field of traffic flow theory.

FUTURE DIRECTIONS

There are many challenges for future developments in the field of traffic flow theory. The Committee on Traffic Flow Theory and Characteristics should consider ways to become a clearinghouse for understanding traffic flow phenomena based on solid empirical studies, to be a forum for debate and discussions of conflicting evidence and to serve as a repository for empirical data. Through the committee’s strategic planning process, it would be useful to continue to revisit the basic questions of why we are interested in fundamentals of traffic flow theory and why we are developing improved traffic simulation tools. Along these lines, the committee survey revealed three primary arenas where there are opportunities for major investments in research and improved systems.

Simulation

There are many challenges for improved simulation tools that should be addressed. The committee has recognized this and has formed a joint Simulation Subcommittee with several other committees to address this issue. In general, more reliable, powerful, and realistic tools that include improved realism of driving behavior models—particularly in congested and oversaturated traffic conditions, for freeways and arterials. Accurate assessment of differences between traffic in various countries is also needed. Links between improved and simulation models and improved predictions of emissions, effects of intelligent transportation systems (particularly advanced traffic management systems) on traffic flow, and traffic safety and crash probabilities are needed. Further, survey respondents call for better traffic simulation tools that encompass the full range of modes (including bicycling and walking), that account for the interaction of these modes. These tools should also incorporate more realistic characteristics that introduce heterogeneity in traffic flow.

Moving to a real-time network control environment is also a major challenge that was noted in the survey. Work should be done to advance the online deployment of traffic simulation for the benefit of traffic management and route guidance. This should include large scale adaptive traffic control in urban networks, in-vehicle applications with user-specific routing strategies. Along with these elements would be the need for better short-term prediction of traffic conditions, and efficient dynamic network loading models. Finally, it was suggested that consideration be given to the development of nanoscopic traffic flow models by incorporating engine performance, vehicle dynamics, intelligent drivers, and vehicle movement on a true two dimensional surface. Driving forces come from the need for more fidelity in safety, emissions, automatic cruise control, potential automated highway vehicle systems, etc.

Data

The next decade will involved new challenges for developing improved data collection, management, archiving, dissemination, and analysis techniques. In the spirit of those pioneers of empirical analyses, a challenge for the committee is to obtain the right data in sufficient quantity, and to manage it so as to learn useful things. Included in this challenge are issues of the accuracy and reliability of automatic data collection systems, the locations and types of traffic detectors, etc. Included in this arena will be the incorporation of vehicle-based probe data, including vehicle diagnostic data, infrastructure system data (e.g. traffic signal phase information) for optimized prognosis, routing and traffic control. Developing improved data acquisition tools will also be needed for origin-destination matrix estimation and online travel time estimation.

Improved data infrastructure and analysis techniques will also lead to improved understanding of basic traffic phenomena. If possible, this understanding should be accomplished in an open fashion, with researchers sharing data, analysis techniques and working cooperatively across international boundaries. This improved understanding could include such issues as synchronized traffic, capacity drops, hysteresis, platoon dispersion, lane changing, and multiple user classes. The committee may embark upon a program to improve traffic model validation by using high resolution field data. There is also interest in quantifying uncertainty (and reliability) regarding traffic flow phenomena, determining how much of the uncertainty is irreducible, and figuring out how to respond to uncertainty. In order to meet these challenges, the committee may want to discuss the formation of a joint Data Subcommittee, perhaps with other TRB committees that have similar interests.

Driver Behavior

In the future we need to improve our understanding of the behavioral basis for traffic flow phenomena. In particular, we need this improved understanding of tactical behavior and its consequences for traffic flow on a larger scale.

Other

Other challenges and opportunities in the next decade may include a desire to combine traffic flow models with transportation planning applications. Experiments, empirical analysis, and control strategies aimed at avoiding capacity drops (breakdowns) on freeways could also be conducted. Further work on improving traffic signal coordination in congested urban networks is also needed.

SUMMARY

Put summary here.

REFERENCES

1. Lieu, H. Traffic Flow Theory. Public Roads, Volume 62, No. 4, January/February 1999.

2. Greenshields, B.D. “A study of traffic capacity,” Proceedings of the Highway Research Board, Vol. 14, p. 468, 1935.

3. Edie, L.C., and Foote, R.S., “Traffic flow in tunnels,” High. Res. Board Proc., 37, pp. 334-344, 1958.

4. Edie, L.C. and R.S. Foote, “Experiments on Single Lane Flow in Tunnels,” Theory of Traffic Flow, Proceedings of Symposium on the Theory of Traffic Flow, ed R. Herman, pp. 175-192. Elsevier, New York, 1961.

5. Wardrop, J.G., “Some theoretical aspects of road traffic research,” Proc. Inst. Civ. Eng. Part 2, 325-378, 1952.

6.

7. Theory of Traffic Flow, Proceedings of Symposium on the Theory of Traffic Flow, ed R. Herman, pp. 175-192. Elsevier, New York, 1961.

8. Monograph on Traffic Flow Theory, Highway Research Board Special Report 79, 1964.

9. Monograph on Traffic Flow Theory, Transportation Research Board Special Report 165, 1975.

10. Traffic Flow Theory: A State of the Art Report, Transportation Research Board, www.tfhrc.gov/its/tft/tft.htm, 1999.

11. Hoogendoorn, S. and Bovy, P. State-of-the-art of Vehicular Traffic Flow Modelling., Proceedings of the I MECH E Part I Journal of Systems & Control Engineering, Volume 215, Number 4, 19 August 2001, pp. 283-303.

12. Gazis, D. “The Origins of Traffic Theory,” Operations Research, Volume 50, Issue 1, Jan. – Feb. 2002, pp. 69-77.

13. Newell, G.F. “Memoirs on Highway Traffic Flow Theory in the 1950s,” Operations Research, Volume 50, Issue 1, Jan. – Feb. 2002, pp. 173-178

14. Lighthill, M. J. and Whitham, G. B.. “On kinematic waves: a theory of traffic flow on long crowded roads,” Proceedings of the Royal Society, Series A, Vol. 229, pp 317-345, 1955.

15. Richards, P. I. “Shock waves on the highway.” Operations Research 4: pp. 42-51, 1956.

16. Newell, G.F. “A simplified theory of kinematic waves in highway traffic, I. General theory, II. Queuing at freeway bottlenecks, III. Multidestination flows,” Transportation Research 27B, 281-313, 1993

17. Prigogine, I. and Herman, R.. Kinetic Theory of Vehicular Traffic. Elsevier, New York, 1971.

18. Daganzo, C.F. “The cell transmission model: A dynamic representation of highway traffic consistent with the hydrodynamic theory”, Transportation Research B, 28(4), 269-287, 1994.

19. Daganzo, C.F. “The cell transmission model. Part II: Network traffic”, Transportation Research B, 29 (2), 79-93, 1995.

20. Payne, H.J., “Models of freeway traffic and control.” Simulation Councils Proc. Series: Mathematical Model of Public Systems, 1(1):51-60, 1971

21. Daganzo, C.F. “Requiem for high-order fluid approximations of traffic flow,” Transportation Research B, 29 (4), 277-287, 1995

22. Chandler, R.E., Herman, R., and Montroll, E.W. “Traffic Dynamics: Studies in Car Following,” Operations Research, Vol. 6, No. 2, pp. 165-184, 1958.

23. Herman, R., Montroll, E.W., Potts, R.B., and Rothery, R.W. “Traffic Dynamics: Analysis of Stability in Car Following,” Operations Research, Vol. 7, No. 1, pp. 86-106, 1959.

24. Edie, L.C, “Discussion on Traffic Stream Measurements and Definitions,” Proc. 2nd Int’l Symp. on the Theory of Traffic Flow. (J. Almond, editor), 139-154, OECD, Paris, France, 1963.

25. Treiterer and Myers, “Hysteresis phenomenon in traffic flow,” Proceedings of the 6th International Symposium on Transportation and Traffic Theory, ed. D.J. Buckley, pp. 13-38, Elsevier, New York, 1974.

26. Webster, F. V. Traffic Signal Settings. Road Research Technical Paper, No. 39, Her Majesty's Stationary Office, London, U.K., 1958.

DRAFT May 2006

[1] See http://web.pdx.edu/~bertini/TFT/Traffic_Flow_Theory_Monograph_1975.pdf