Dynamic PCU Based Adaptive Signal Control for Heterogeneous Urban Traffic Using Microsimulation

Abstract

This research develops an adaptive traffic signal control scheme to address heterogeneous traffic conditions prevalent in developing nations, where poor lane discipline and frequent lane changes limit the effectiveness of traditional adaptive signal control schemes based on static Passenger Car Unit (PCU) values. Dynamic PCU values were estimated by considering the effective area occupied by vehicles and their intersection traversal times, enabling the signal controller to adapt to varying traffic conditions and vehicle compositions. The adaptive control algorithm was implemented in the Simulation of Urban MObility (SUMO) environment at Angoda signalized intersection in Colombo, with the simulation model calibrated using field data collected from two analogous intersections. Performance was evaluated using throughput, delay, queue lengths, and carbon dioxide emissions for each approach. Confidence interval analysis across simulation replications indicated that observed performance differences between static and dynamic PCU control were generally consistent. Dynamic PCU values were influenced by road geometry, with higher values observed on multilane approaches. The proposed method improved most performance measures, particularly reducing queue lengths and carbon dioxide emissions at single lane approaches, although throughput improvements were limited. These findings demonstrate the potential of dynamic PCU based adaptive signal control for heterogeneous urban signalized intersections.