Identification and optimization of traffic bottleneck with signal timing

Abstract

Abstract: In urban transportation network, traffic congestion is likely to occur at traffic bottlenecks. The signal timing at intersections together with static properties of left-turn and straight-through lanes of roads are two significant factors causing traffic bottlenecks. A discrete-time model of traffic bottleneck is hence developed to analyze these two factors, and a bottleneck indicator is introduced to estimate the comprehensive bottleneck degree of individual road in regional transportation networks universally, the identification approaches are presented to identify traffic bottlenecks, bottleneck-free roads, and bottleneck-prone roads. Based on above work, the optimization method applies ant colony algorithm with effective green time as decision variables to find out an optimal coordinated signal timing plan for a regional network. In addition, a real experimental transportation network is chosen to verify the validation of bottleneck identification. The bottleneck identification approaches can explain the features of occurrence and dissipation of traffic congestion in a certain extent, and the bottleneck optimization method provides a new way to coordinate signal timing at intersections to mitigate traffic congestion.