Simulation-Based Evaluations of Real-Time Variable Speed Limit for Freeway Recurring Traffic Congestion

Freeway sections with lane tapers are likely to experience recurring congestion due to disruption to traffic caused by merging maneuvers. These merging maneuvers tend to create shock wave effects that cause long delays and slow traffic. This paper examines the performance of variable speed limits (VSL) as a promising mitigation treatment for freeway recurring traffic congestion. To do this an interstate highway section was selected that suffers from recurring traffic congestion caused by lane drop and lane merging in Northern Virginia. The first effort endeavored to model the 5 mile section in VISSIM and calibrate it using INRIX speed data and historical traffic counts from the Virginia Department of Transportation (VDOT). Calibration efforts involved adjusting parameters embedded in VISSIM’s car following and lane changing models that led to an acceptable model performance within a selected confidence interval. A VSL algorithm developed by California PATH and programmed in AIMSUN was adopted in VISSIM to be used for determining optimum speed adjustments at three specific locations along the freeway section. Finally the VSL algorithm was applied to that section for a certain period of performance (i.e., 2 pm to 8 pm) in several VISSIM runs. These VISSIM runs used different compliance percentages. Findings show that for certain compliance rates, reductions in travel time were significant, resulting in up to 12% of network-wide total travel time savings.