Carroll J Messer

Short-Term Freeway Traffic Volume Forecasting Using Radial Basis Function Neural Network

A radial basis function (RBF) neural network has recently been applied to time-series forecasting. The test results of an RBF neural network in forecasting short-term freeway traffic volumes are provided. Real observations of freeway traffic volumes from the San Antonio TransGuide System have been used in these experiments. For comparison of forecasting performances, Taylor series, exponential smoothing method (ESM), double exponential smoothing method, and backpropagation neural network were also designed and tested. The RBF neural network model provided the best performance and required less computational time than BPN. It seems that RBF and ESM can be a viable forecasting routine for advanced traffic management systems. There are some tradeoffs between RBF and ESM. Although the performance of ESM is inferior to RBF, the former does not need a complicated training process or historic database, and vice versa. However, even in the best performance case, 35 percent of the forecast traffic volumes showed 10 percent or more percentage errors. This means that we cannot heavily depend on the forecast traffic volumes as long as we are utilizing the models tested. Further work is needed to provide a more reliable traffic forecasting model.

Traffic Signal Optimization Program for Oversaturated Conditions: Genetic Algorithm Approach

Traffic signal optimization programs have been used widely among transportation professionals. However, none of the existing computer programs can optimize all four traffic control parameters (i.e., cycle length, green split, offset, and phase sequence) simultaneously, even for undersaturated conditions. In this paper, a genetic algorithm-based signal optimization program that can handle oversaturated signalized intersections is presented. The program consists of a genetic algorithm (GA) optimizer and a mesoscopic traffic simulator. The GA optimizer is designed to search for a near-optimal traffic signal timing plan on the basis of a fitness value obtained from the mesoscopic simulator. The proposed program is compared with the newly released TRANSYT-7F version 8.1 on the basis of CORSIM simulation program. Three different demand volume levels—low, medium, and high demand—are tested. For the low-demand and high-demand volume cases, the GA-based program produced statistically better signal timing plans than did TRANSYT-7F in terms of queue time. In the case of medium-demand volume level, the signal timing plan obtained from the GA-based program produced statistically equivalent queue time compared with TRANSYT-7F. Both programs are judged to provide superior capability for oversaturated conditions due to their queue blockage model when compared with previously available signal timing optimization software.

Enhanced genetic algorithm for signal-timing optimization of oversaturated intersections

Enhancements were provided to a previously developed genetic algorithm (GA) for traffic signal optimization for oversaturated traffic conditions. A broader range of optimization strategies was provided to include modified delay minimization with a penalty function and throughput maximization. These were added to the initial delay minimization strategy and were further extended to cover all operating conditions. The enhanced program was evaluated at different intersection spacings. The optimization strategies were evaluated and compared with their counterpart from TRANSYT-7F, version 8.1. A microscopic stochastic simulation program, CORSIM, was used as the unbiased evaluator. Hypothesis testing indicated that the GA-based program with average delay minimization produced a superior signal-timing plan compared with those produced by other GA strategies and the TRANSYT-7F program in terms of queue time. It was also found from the experiments that TRANSYT-7F tended to select longer cycle lengths than the GA program to reduce random plus oversaturation delay.

Performance of Advance Warning for End of Green System for High-Speed Signalized Intersections

A major difficulty with traffic signal operation on high-speed approaches is the dilemma faced by approaching motorists when the downstream signal turns yellow. Should the motorists stop or proceed through the intersection? Crashes that may occur at these intersections result in excessive property damage and personal injury because of the high speeds involved. The Texas Transportation Institute has developed a new system named the Advance Warning for End of Green System (AWEGS) for application at high-speed signalized intersections. Typically, dilemma zone detection strategy is based on a certain approach speed (typically the 85th percentile). AWEGS provides protection for the majority of motorists who are not covered by the dilemma zone treatment. AWEGS provides advance warning to motorists by using signs mounted on the roadside. These signs (Be Prepared To Stop When Flashing) would flash a beacon about 5 to 6 s before the onset of the yellow signal for high-speed approaches. Similar systems have been implemented in Canada and in a few U.S. states that use the trailing-green approach, which results in loss of dilemma zone protection every cycle. AWEGS, however, is almost completely independent of the traffic signal controller, and hence the signal controller would continue to provide the dilemma zone protection for which it was designed. The system was implemented at two sites in Waco and Brenham, Texas. Results of AWEGS implementation illustrated an improvement in traffic operations. AWEGS consistently enhanced the dilemma zone protection at intersections and reduced red light running by about 40%.

Modeling Impact of Ramp Metering Queues on Diamond Interchange Operations

Methodologies to model the impact of ramp metering queues on diamond interchange operations were developed in this study. The methodologies are part of an overall research effort to address the integrated operations of a diamond interchange and a ramp metering system. The methodologies were implemented in DRIVE, a computer model characterized as a mesoscopic simulation and analytical model. A meso-scopic model has the advantages of both macroscopic and microscopic models with less computing time while still considering stochastic traffic flows. DRIVE can be used to perform system analysis and evaluation of an integrated diamond interchange and ramp metering system over multiple cycles. The component concerning the diamond interchange operations is documented here, with a particular focus on the impact of potential queue spillback to the diamond interchange signals from ramp metering. The study focused on the two common diamond phasing schemes: basic three-phase and Texas Transportation Institute (TTI) four-phase. DRIVE was validated against the VISSIM microscopic traffic simulation model, and general agreement was found in the results between the two models. For oversaturated ramp conditions with queue spillback, both DRIVE and VISSIM indicated that TTI four-phase favors the frontage road approach, whereas the basic three-phase favors the arterial left-turn movement. DRIVE provides a new analysis tool for analyzing diamond interchange operations taking ramp metering into consideration, which would facilitate studies on developing operational strategies to better manage such systems.

Program for Optimizing Diamond Interchanges in Oversaturated Conditions

Traditionally, optimization of diamond interchange timings has been done with PASSER III for standard and special diamond phasing sequences. PASSER III is limited because it is designed for undersaturated conditions. It applies vertical stacking of queues and is not capable of modeling queue spillback conditions in its current form. This deficiency is addressed by the arterial signal coordination software (ASCS), whose capabilities in timing diamond interchanges in under-saturated and oversaturated conditions are presented here. ASCS consists of three modules: (a) input module, a user interface through which volume and geometry inputs can be provided to the program; (b) optimization module, a genetic algorithm-based optimization routine that can optimize signal timings; and (c) analysis module, which consists of a bandwidth analysis routine and a delay analysis routine (DAR). The DAR is a mesoscopic simulator that applies a second-by-second analysis of flows for modeling flows accurately. DAR applies horizontal stacking of queues and shock wave analysis to estimate the performance of traffic operations. Validation of ASCS for oversaturated arterial links against PASSER III and CORSIM was conducted. The results indicate that delay and throughput estimation in ASCS are realistic. The genetic algorithm-based optimization routine in ASCS was applied to estimate diamond interchange timings for three scenarios. Where queue spillback occurred, ASCS clearly outperformed PASSER III. ASCS produced near-optimal results for all scenarios studied.

Extension and Application of Prosser-Dunne Model to Traffic Operation Analysis of Oversaturated, Closely Spaced Signalized Intersections

Traffic congestion continues to have a serious impact on the productivity of the nation’s transportation system. Signalized arterials and service interchanges in urban areas serve as critical surface street facilities. These facilities often experience serious congestion problems because of high traffic demands that exceed capacity and also because of their common, closely spaced intersections and ramp terminals. Moreover, service interchanges are usually part of a signalized cross arterial and also may have short spacings to adjacent signalized intersections. Major deficiencies exist in the analysis technology and software tools available for analyzing potential traffic congestion problems at signalized arterials and service interchanges having closely spaced ramp terminals. This paper presents extensions of work originally published by Prosser and Dunne in Australia for analyzing the operational impacts of queue spillback on the capacity and delay of closely spaced signalized intersections. Coding of the described algorithm into FORTRAN was conducted, followed by experimental testing of the model using a calibrated version of the microscopic computer simulation program TRAF-NETSIM. Satisfactory comparisons were obtained between the initial version of the newly developed Prosser-Dunne Extended (PDX) model and NETSIM, but more testing and enhancement of the PDX model are recommended before implementation in existing operational software packages is considered.

Phase Capacity Characteristics for Signalized Interchange and Intersection Approaches

Described in this paper are the development, calibration, and application of models that collectively can be used to predict the saturation flow rate and start-up lost time of through movements at signalized interchange ramp terminals and other closely spaced intersections. These models were calibrated with data collected at 12 interchanges. It is concluded that saturation flow rate decreases as the distance to the downstream queue decreases. This queue is formed by the signal at a downstream intersection. Saturation flow rate increases with traffic pressure, as quantified by traffic volume per cycle per lane. It is recommended that an ideal saturation flow rate of 2,000 passenger-car units per hour of green per lane be used for signalized ramp terminals and other high-volume intersections in urban areas. The data collected for this research indicate that start-up lost time increases with saturation flow rate.

Simulation Studies of Traffic Operations at Oversaturated, Closely Spaced Signalized Intersections

Traffic congestion continues to have a serious impact on the productivity of the nation’s transportation system. Signalized service interchanges in urban areas serve as a critical connection between freeway and arterial surface street systems. These facilities often experience serious congestion problems because of high traffic demands exceeding capacity and because of their common, closely spaced ramp terminals. Moreover, service interchanges are usually a part of a signalized cross arterial and may also have short spacings to adjacent signalized intersections. Major deficiencies exist in the analysis technology and software tools available for analyzing traffic congestion problems at oversaturated signalized interchanges and closely spaced arterial systems. Results are provided of an investigation into the nature of oversaturated systems and also undersaturated systems that may become congested (oversaturated) because of poor signal timing and deficient spacing between the signalized intersections. Although the initial research focus was on signalized service interchanges based on calibrated NETSIM simulations, the research results are also applicable to closely spaced signalized urban arterials. The research shows that traffic congestion is complex, but it can be characterized and modeled. An upper bound on signal control delay is provided for oversaturated arterial operations.

Integration of Diamond Interchange and Ramp Metering Operations

The aim of this research was to develop operational strategies for integrating the operations between diamond interchange traffic signals and ramp metering signals. Integrated operational strategies were developed on the basis of the two commonly used diamond interchange phasing schemes: the basic three-phase scheme and the Texas Transportation Institute four-phase scheme. The key elements of the integration system and its operations include a proposed enhanced detection system and an operational algorithm. Through implementation of special signal timings at the diamond interchange, the traffic flows feeding the ramp meters can be effectively controlled and thus minimize ramp queues and the resulting need to flush the queues, which would possibly lead to freeway breakdown. The operational strategies were evaluated with VISSIM microscopic simulation under three general traffic demand scenarios–-low, medium, and high–-as characterized by the volume-to-capacity ratios at the metered ramps. The results of the evaluation indicated that the integrated operations were most effective under a medium traffic demand scenario in preventing or delaying the onset of ramp metering queue flush, thereby reducing freeway breakdown and system delays.