Queue Spillback Detection and Signal Control Strategies Based on Connected Vehicle Technology in a Congested Network

Queue spillbacks create excessive delays and may result in gridlock conditions in signal controlled networks. The information from Connected Vehicles (CV) offers significant potential for developing improved signal control strategies. The paper presents the development and testing of algorithms for queue spillback detection and control in congested grid networks with CV data. Three control strategies were developed: perimeter control, spillback-based phase changing control strategy, and reduction in the system-wide cycle time with additional green times to the links with the highest vehicle accumulation. The strategies were tested through simulation in a real-life network. The findings indicate that statistically significant improvements can be obtained from the implementation of these strategies in the network throughput expressed in total distance travelled and number of vehicles served, and the network average travel speed.