Combination Strategy of Dynamic Variable Speed Limit Method Based on Real-Time Crash Prediction Model for Highway

Variable speed limit (VSL) has a significant effect on reducing the risk of collisions and delays on highways. There are few variable speed limit control strategies based on the use of high-precision crash prediction models to reduce delay cost and crash cost, and quantitative research on indexes of VSL control parameters based on different traffic volume is inadequate. Based on Aimsun simulation technology, this paper selects the road section of Shanghai G15 Shenhai Expressway from Nantong Development Zone to Sutong Bridge section to build a simulated road network model for simulation experiments. Different combination strategies of real-time risk model-based VSL control strategies were explored to find the best combination strategy, including the trigger threshold, the speed limit value adjustment cycle, and the speed limit value adjustment gradient based on low, medium and high volume traffic flow. The results show that the control effect of VSL control strategy is significantly affected by real-time traffic volume. This paper firstly proposes a new kind of dynamic VSL control method based on real-time risk model. The simulation verification of the road section on May 2, 2016, 13:50:00-14:50:00 real-time traffic flow shown that the dynamic strategy can effectively reduce the crash cost by 20.30%, and reduce the delay cost by 35.80%, better than static VSL control strategy for different volume conditions. Therefore, the VSL control method is of great significance for improving the safe operation level of highways.