David K. Hale

Simulating High-Occupancy Toll Lane Operations

Microscopic simulation is critical for evaluating the operation strategies of managed lanes. However, most existing tools are limited in their ability to simulate dynamic tolling strategies of managed lanes, particularly those with multiple segments. Three sets of modeling components are developed in this paper to demonstrate simulation of high-occupancy toll (HOT) lane operations. The first component implements three pricing strategies: responsive pricing; a closed-loop, control-based algorithm; and time-of-day pricing. The second component mimics drivers’ lane choice behaviors in the presence of tolls, and the third represents different toll structures for multisegment HOT lane facilities. An enhanced version of CORSIM, which contains these new modeling components, is validated by simulation experiments involving the 95 Express network in South Florida.

PREDICTION OF TRAFFIC-ACTUATED PHASE TIMES ON ARTERIAL STREETS

Traffic congestion in U.S. cities has grown rapidly in recent years, and numerous solutions are needed to address the problem. New research is described for producing basic improvements in the practice of traffic signal timing design, which is one of the many available weapons for fighting congestion and delay. Models in the literature for predicting traffic-actuated phase times are oversimplified. In some cases, they do not recognize the signal settings associated with today’s controllers, which affect phase times. In other cases, models from the literature do not recognize operational effects (for example, progression, queue blockage, permitted left turn, critical movement, early return to green, and stochastic effects) that affect phase times or that are applicable only to isolated signals instead of coordinated signals. Consequently, the recommended average phase times are not practical, and the resulting performance estimates and timing plan designs are unlikely to materialize in the field. The overall goal of this research is to describe and demonstrate an improved methodology for predicting actuated phase times. An improved methodology is needed to improve the overall timing plan design process. Experimental results indicate an improvement in the accuracy of actuated phase time calculation based on the improved methodology.

Estimation of Crash Modification Factors for an Adaptive Traffic-Signal Control System

AbstractAdaptive traffic-signal control (ATSC) is a traffic management strategy in which traffic-signal timings change, or adapt, based on observed traffic demand. Although ATSC can improve mobility, it also has the potential to reduce crashes because mainline stops should be reduced. This paper aims to evaluate the safety effectiveness of ATSC using the empirical Bayes method. This analysis examines 47 urban or suburban intersections where ATSC was deployed in Virginia using 235 site-years of before data and 66 site-years of after data. Installing ATSC was found to produce a crash modification factor (CMF) for total intersection crashes of 0.83 with a standard error of 0.05. This CMF was statistically significant at a 95 percent confidence level. Fatal and injury crashes did not change by a statistically significant amount, indicating that the primary safety benefit of ATSC was reduction in property damage crashes. Analyses of ATSC safety effects by crash type, by traffic volume level, and by operational...

Dynamic Hard Shoulder Running for Traffic Incident Management

Hard shoulder running (HSR), used in many large cities for reducing peak hour recurring congestion, has been proved effective. This paper looks at another perspective of using hard shoulders and proposes dynamic hard shoulder running (D-HSR) for traffic incident management. The purpose of this paper is to show the benefits and make recommendations to state departments of transportation and local agencies on how to use hard shoulders dynamically to reduce the effects of nonrecurring traffic incidents. An approach based on microscopic simulation with factorial experimental design is adopted in this study, and interesting results are obtained from the discussion and statistical analyses of simulation results: (a) D-HSR strategies are more suitable for property damage only incidents in which traffic operations centers have more flexibility in managing traffic; (b) only the part of the shoulder that is 0.5 mi upstream and downstream of the incident location needs to be opened to achieve maximum benefits for relieving a bottleneck; (c) the opened shoulder can be closed after the incident is cleared, and opening the shoulder for a longer time will not improve traffic conditions; and (d) the effectiveness of D-HSR is significant across different roadway geometry, traffic, and incident scenarios. Equations to estimate potential benefits are also available in this paper. These results are favorable particularly in practice because shoulders need to serve as refuge areas for incident vehicles and be used by emergency vehicles. It is recommended that departments of transportation and local agencies consider D-HSR for relieving congestion during incidents.

Traffic Performance Analysis of Dynamic Merge Control Using Microsimulation

Dynamic merge control (DMC) can be used in freeway merge areas to change lane allocation dynamically at interchanges. DMC generally prioritizes the facility having higher volume and closes a lane on the lesser-volume roadway. DMC has been implemented in the Netherlands and Germany, where it was reported that the application of DMC significantly improved traffic operations. However the DMC strategy has rarely been studied and has not been implemented in the United States. This research used microsimulation studies with VISSIM to investigate the efficiency of the DMC strategy. Optimum traffic demand thresholds were sought specifically for the geometric case in which a two-lane freeway merged with a three-lane freeway and tapered into four lanes. Major-road traffic demands between 2,500 and 4,600 vehicles per hour were compared against minor-road demands between 3,000 and 4,600 vehicles per hour. The DMC strategy was applied by closing the right lane of the major road, ahead of the merging gore area. The results indicate that (a) the DMC strategy is beneficial for all the previously mentioned traffic demand combinations in regard to average vehicle delay and average vehicle speed; (b) when traffic demand on the minor road exceeds 1,900 vehicles per hour per lane, these benefits become statistically and practically significant; and (c) DMC can greatly alleviate the capacity reductions caused by lane changing in the merge area.

Revisiting the application of simultaneous perturbation stochastic approximation towards signal timing optimization

ABSTRACT Simultaneous Perturbation Stochastic Approximation (SPSA) has gained favor as an efficient optimization method for calibrating computationally intensive, “black box” traffic flow simulations. Few recent studies have investigated the efficiency of SPSA for traffic signal timing optimization. It is important for this to be investigated, because significant room for improvement exists in the area of signal optimization. Some signal timing methods and products perform optimization very quickly, but deliver mediocre solutions. Other methods and products deliver high-quality solutions, but at a very slow rate. When using commercialized desktop signal timing products, engineers are often forced to choose between speed and solution quality. Real-time adaptive control products, which must optimize timings within seconds on a cycle-by-cycle basis, have limited time to reach a high-quality solution. The existing literature indicates that SPSA provides the potential for upgrading both off-line and on-line solutions alike, by delivering high-quality solutions within seconds. This article describes an extensive set of optimization tests involving SPSA and genetic algorithms (GAs). The final results suggest that GA was slightly more efficient than SPSA. Moreover, the results suggest todays signal timing solutions could be improved significantly by incorporating GA, SPSA, and “playbooks” of preoptimized starting points. However, it may take another 5–10 years before our computers become fast enough to simultaneously optimize coordination settings (i.e., cycle length, phasing sequence, and offsets) at numerous intersections, using the most powerful heuristic methods, at speeds that are compatible with real-time adaptive solutions.

Right-Turn-on-Red Flow Profile Impactson Urban Street Capacity Analysis

The Highway Capacity Manual 2010 (HCM 2010) contains computational procedures for evaluating traffic operational efficiency of urban street segments. These procedures have been implemented within several commercial software packages and are likely used by thousands of engineers and planners across the United States. The procedures for urban street capacity analysis contain no logic for handling right turns on red (RTORs) or for handling special cases of RTORs such as shielded and free right turns. A new proposed RTOR modeling framework is described for urban streets in the HCM 2010. When significant upstream RTOR flows exist, the proposed logic is designed to generate more realistic flow profiles. Three types of experimental results are presented: they demonstrate the improved modeling accuracy of the proposed logic. First, it is shown that macroscopic flow profile shapes are now more visually sensible because they now illustrate RTOR flows moving at the appropriate times. Second, macroscopic flow profile shapes are now more consistent with microscopic vehicle trajectories. Third, a statistical analysis shows that when the proposed logic is used, HCM 2010 performance measures become more consistent with the performance measures generated by microsimulation. Finally, case study results show that when the proposed RTOR logic is not used, control delays are sometimes be inaccurate by more than 30%. Given the experimental evidence presented, it is urgent that the proposed improvements be adopted and implemented so that RTOR corridors can be accurately analyzed by the HCM 2010 procedures.