Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
CHAPTER 1 INTRODUCTION AND RESEARCH APPROACH The 1985 Highway Capacity Manual (HCM) [1] prescribes a methodology in Chapter 9 for esti- mating the delay and level of service at signalized intersections as a function of the traffic charac- ter~stics and the signal timing plan. This methodology is widely used both nationally and inter- nationally and approaches a standard for the evaluation of the performance of signalized intersec- tions in the U.S.A The HCM is maintained by the Transportation Research Board (TRB) Committee on Highway Capacity and Quality of Service, often referred to as the "Highway Capacity Commit- tee." Working towards a major revision ofthe HCM in the year 2000, the Highway Capacity Committee has embarked on a comprehensive research program to overcome oenc~enc~es In all chapters, and to serve the needs of transportation professionals in an environment of advanced technology. The project discussed In this report addresses potential enhancements to the treatment of traffic-actuated control at signalized intersections. The specific objective is to extend the current HCM Chapter 9 analysis procedures to deal more effectively with traff~c-actuated control. PROBLEM STATEMENT The present treatment of traffic-actuated control in the HCM leaves much room for improvement. Traffic-actuated control is treated the same as pretimed control, with only two exceptions. The delay model, which was analytically derived for a pretimed controller, includes an adjustment factor to reduce the uniform delay at traffic-actuated intersections by 15 percent. 2. A primitive technique is presented in Appendix II of HCM Chapter 9 for estimating the cycle length and green times for actuated control. This technique is based on the assumption that a traffic-actuated controller will maintain 95°/0 saturation on the critical approach to each phase. Both the validity ofthis assumption and the simplistic nature of the technique have been widely criticized. The limitations ofthis treatment will be the main target ofthe research described in this report. These limitations can be overcome, but not without adding some complexity to the computational proce- dures. The HCM has traditionally dealt with "single pass" analytical models that may be described in manual worksheets. In the 1985 version, the manual worksheets were put forth as the primary method of computation. The possibility of implementation by software was not recognized in that version of the HCM Chapter 9. NCHRP Project 3-48 Final Report: Page S
RESEARCH OBJECTIVES The work camed out under NC~P Project 3-48 included the formulation of an improved methodology for predicting the operating characteristics oftraffic-actuated control, and for estimating the delay and level of service associated with a given set of operating parameters. The principal product ofthis research was an improved mode] set in a computational framework that wait facilitate testing of the model, and wait eventually become a too! for the end user. The testing and evaluation of the methodology wall rely heavily on simulation, augmented by limited field studies. PROJECT OBJECTIVES The specific objectives of this project were: I. To develop a methodology capable of assessing the capacity and level of service ~OS) for traffic-actuated intersections for typical design configurations; 2. To identifier the changes necessary in HCM Chapter 9 to implement the methodology; and, 3. To produce drabs matenal suitable for incorporation In the HCM describing how the methodology is implemented and any limitations on its use. The requirements for the methodology to be developed were established as follows: I. It must be comparable with existing HCM Chapter 9 procedures for pretimed signals. 2. It must be developed at a level suitable for both operational and planning level analysis. 3. It must be validated by a combination of field data and simulation results. 4. It should be consistent with the ISTEA act and the Federal Clean Air Amendments. 5. It should be capable of operating in the ITS environment. 6. The proposed research plan should take into account certain other research projects and initiatives sponsored by NC~P and FlIWA. The scope of the study was not intended to be all-inclusive, but was limited instead to the most typical signalized intersection designs using actuated controllers. NCHRP Project 3-48 Final Report: Page 6
MODEL DEVELOPMENT ISSUES AND SCOPE At this point, the entire HCM Chapter 9 methodology has, for all intents, reached the limits of single- pass procedures. It must be recognized at the outset that the limitations of Appendix 9-IT, as outlined above, cannot be addressed without resorting to complex iterative procedures. The development of an, ~ o - . ~ _ e ~ · ~ ~ ~ . . ~ T _· ~ `_ a useful product ·vlll dehmtely require a departure trom what the Highway capacity Committee has become accustomed to dealing with in the past. Another issue that must be addressed is the additional input data that are needed to improve the accuracy of the analysis methodology. The current pretimed models already require more data than many users would prefer, especially for planning applications. It is note however, possible to avoid the burden ofadditionaldataifan analytically defensible model for traffic-actuated operationisto be produced. This is especially true if the model is to respond to changes in the design parameters of traff~c-actuated control. One ofthe major limitations of the present HCM Chapter 9 methodology is the inability to perform effective comparisons between the pretimed and traffic-actuated control modes. This limitation is a result of the primitive treatment of actuated control and may only be overcome by improving the actuated-control model. On the other hand, much of the improvement will result from the use of more complex procedures that could apply equally well to pretimed control. In fact, if the procedures are not applied to pretimed control, then comparison of pretimed and actuated control may be even less meaningful than it is today, because the pretimed model will become primitive compared to the actuated model. The development of an entirely new methodology that applies universally to pretimed and actuated would clearly be desirable and probably would be feasible, however, this task is beyond the scope of the current project. One controversial issue here is whether or not pretimed and actuated control converge when the operation is billy saturated. It is a logical assumption that the length of all actuated phases will be governed by their maximum settings when saturation occurs. This would lead to the conclusion that an actuated controller will revert to pretimed operation under these conditions. On the other hand, gap settings that are appropriate for minimizing delay when volumes are below capacity may prevent some phases from reaching their specified maximums. This introduces a stochastic element that is difficult to deal with analytically. Another problem that is difficult to deal with analytically is the very low volume operation that typically occurs late at night. Under these conditions' the repetitive cyclical operation upon which the analysis is based no longer applies. The effort required to develop a model for dealing accurately with delays of a few seconds per vehicle (i.e., level of service "A") is difficult to justify. An approximation of the operating characteristics with very low volumes will generally be acceptable from a capacity and level of service point of view. NCHRP Project 3-48 Final Report: Page 7
The performance measures (delay, stops, filet consumption etc.) are very sensitive to design parameters such as unit extension times and maximum phase times. Optimization of these parameters has long been recognized as a desirable objective. The objectives of this study do not include the development of an optimization methodology or a design manual. The analysis techniques to be developed should, however, be developed in a manner that will facilitate their incorporation into an optimization and/or design procedure at some point in the fixture. Much ofthe analytical work on traffic-actuated control found ~ the literature assumes that passage detectors are used, whereas presence detectors are much more common in actual practice. Passage detectors transmit a short pulse to the controller coincident with the arrival of each vehicle. Presence detectors transmit a continuous signal to the controller as long as the vehicle remains in the detection zone. For purposes of this study, variable length presence detectors will be assumed. Passage detector operation may be approximated by using short length presence detectors. For example, a six-foot presence detector will transmit a signal of approximately one half second duration to the controller for each passing vehicle. Traffic simulation models have been developed to the point where they may be used as a productive surrogate for field data collection. The work described in this report has made extensive use of NETSIM [2] to verify some of the analytical modeling concepts proposed herein. NETSIM was chosen for this purpose because it has been used extensively by traffic engineers and transportation planners, gaining some measure of credibility in the process. The main issue regarding simulation is not whether simulation models are resell, but under what conditions they are more useful than analytical models. For example, given the proper inputs, NETSIM is capable of generating most ofthe measures of electiveness required to evaluate the level of service at an intersection or on an arterial route. Because of the more detailed and microscopic nature of the modeling process, it can be argued that simulation results are more credible than the current analytical methods now contained in the HCM. On the other hand, the analytical tools offer a much better insight into the operation of a given facility because they provide access to intermediate computations. For example, the analytical tools provide a direct estimate of the capacity of an approach to a signalized intersection, in addition to measures such as opposing queue service times for permitted led turns, progression adjustment factors, etc. RESEARCH APPROACH The research approach involved five principal tasks: 1. Assessment of current practice; 2. Development of an analytical mode} for traff~c-actuated control; 3. Implementation of the analytical mode! in a computational structure; NCHI2P Project 3-48 Final Report: Page 8
4. Evaluation of the mode} by simulation and limited field testing; and 5. Technology transfer efforts. Each of the these tasks are summarized In the following discussion. Assessment of Current Practice - The assessment of current practice involved both a user survey and a literature review. To obtain the user perspective, a survey was conducted among a group of traffic engineers who deal regularly with the design, implementation, and evaluation oftraffic-actuated signals. The objectives of the survey were to establish the range of operational configurations that should be considered in the research effort; to identify the timing plan design and analysis procedures that are commonly employed; and to determine the interest of individual agencies in providing data andIor sites for data collection. The survey responses were received and analyzed. A summary of the responses is presented in Appendix B. In general, the survey showed that the HCM is the predominant method used to evaluate the operation of traffic-actuated signals. Signal timing programs are used frequently on coordinated routes, especially for design purposes. A wide range of design parameters for detector placement and controller settings was reported. No changes to the scope of the project were suggested by the results of the survey. Review of past and current research was also an important task. There has been a substantial amount of research conducted on traff~c-actuated control related to this project. The main topics of this literature review included: Traffic-actuated control definitions, Warrants for traffic-actuated control, Benefits and operating considerations for traffic-actuated control, Effects of coordination and phase-skipping for traff~c-actuated control, Late-night, low-volume operation of coordinated actuated systems, Evaluation of traff~c-actuated control by simulation, Estimation of green times and cycle length for traffic-actuated control Delay models for traff~c-actuated control, · Signalized intersection capacity models for traffic-actuated control, and · Overview and evaluation of "Enhancement of the Value Iteration Program for Actuated Signals" (EVIPAS). The literature review is summarized in Appendix A. NCHRP Project 3-48 Final Report: Page 9
Development of an Analytical Model for TrafOlc-Actuated Control Both the Unctions requirements and data requirements are of interest at this point. Of requirements will be addressed separately. Functional Requirements These two sets The mode! developed is capable of providing reasonable estimates of the operating characteristics (timing and performance measures) of traffic-actuated controllers under the normal range of practical design configurations. It must be sensitive to common variations in design parameters. Examples of design parameters include: I. Traffic-actuated controller settings (initial interval, allowable gap, maximum green time) 2. Conventional actuated vs. volume-density control strategies _. ~Detector configuration (length and setback) 4. Pedestrian timing (Walk, and Flashing Don't Walk, POW) 5. Left-turn treatment (permitted, protected, perrrutted and protected, not opposed) 6. Left-turn phase position (leading or lagging) The functional capabilities of the proposed mode! are directly related to the level of detail ofthe input data. It is therefore essential to consider the items of input data that are not currently required by the HCM Chapter 9 methodology. Data Requirements The information that is already required by the HCM Chapter 9 procedure will naturally be used to the fullest extent to avoid the need for new data. Most of the additional data items relate to the operation of the controller itself The evaluation mode! will be based on the standard eight-phase dual-r~ng control concept that is more or less universally applied in the U.S.A. From a capacity and level of service point of view, less complex phasing concepts (including simple two-phase operation) may generally be represented adequately as a subset of the dual-ring scheme. For purposes of this report, a standard assignment of movements to phases will be adopted as pre- sented in Figure I. This will greatly simplify the development and illustration of all modeling procedures without affecting the generality of the capacity and level of service results. NCHRP Project 3-48 Final Report: Page 10
5 EBL ~ 2 1 ~ EBT 6 WBT Left side of barrier E - W Movements Ring 1 Ring 2 Barrier N 7 <~BL Right side of barrier (N-S Movements) Figure I. Dual-ring concurrent phasing scheme with assigned movements A set of worksheets was developed to facilitate the compilation of the necessary data and to descnbe the computational procedures. The input requirements and procedures were descnbed in the fonn of worksheets whenever possible. The worksheet format, together with narrative explanations and examples, offers a very clear and concise way to document this information. This format is also consistent with the current HCM. No claim is made, however, as to the practicality of carrying out usefill computations manually using the worksheets. Many of the computational procedures involved in the proposed methodology are highly iterative. While the individual worksheets are, for the most part, conceptually simple, it would be necessary to repeat the procedures on some of the worksheets many times as the values converge to the final results. This would be a very pained! manual process. Model Imulementation The implementation ofthe mode! involved the development of both worksheets and software. The worksheets will play a very important part in overcoming the "black box" image of complex models such as this. The computations themselves will be carried out using software. The worksheets are intended to provide software developers with a clear set of instructions for understanding and implementing the procedures. More importantly, they provide a standardized structure for presenting the results of intermediate computations in a common form that is compatible with the prescribed techniques. The main feature that unifies all HCM Chapter 9 related software products today is that they all produce their output in the form of the standard HCM worksheets. The methodology developed under this project should be adaptable to that strategy. VCHRP Project 3-48 Final Report: Page 11
Two specific software products were developed. The first implements the mode! to determine the average cycle length and phase times for isolated traffic-actuated control. The second exercises this mode} iteratively in a coordinated system environment to converge to a specified cycle length for progression. The software products are descnbed in Appendix G. Mode! Evaluation The models were evaluated using a combination of simulation studies and field studies. The results, which are reported In the next chapter, were very encouraging. These results demonstrate the clear superiorly ofthe proposed analytical mode} relative to the existing methodology. They also indicated a good agreement with simulation data. The evaluation provides strong support for a recommenda- tion to modify the current procedures. Technology Transfer Efforts The software developed as a part of this project will be very useful as an end-user product. It will promote immediate and widespread use of these research results. In addition, a draPc revision to Appendix 1[ of HCM Chapter 9 was prepared for consideration by the Highway Capacity Committee. This material is In a form suitable as a direct replacement for the existing appendix. It expands on the general discussion of traffic control equipment Dom the perspective of the analyst, and summarizes the analytical mode! described in this report. Several examples are presented to provide a clear understanding of the methodology. NCHRP Project 3-48 Final Report: Page 12