Effect of pre-signals in a Manhattan-like urban traffic network

Abstract

Pre-signal is an additional traffic light proposed in traffic system to improve the left-turning efficiency of intersections. The efficiency of pre-signals in a networked environment will be affected by the coordination of pre-signals and the intersection traffic lights. In this paper, we study the implementation of pre-signal in a Manhattan-like urban traffic system based on a microscopic cellular automaton model. The pre-signals are located before the normal traffic lights, and have two phases (left-turning phase and through phase) for the aim of segregating left-turning and through vehicles into separate sets of lanes. We propose a reliable method to coordinate the intersection traffic light period and phases with the pre-signal period and phases. Simulation results show that the pre-signals can improve the performance of the system, for both closed and open boundaries. The macroscopic fundamental diagram of the system shows a much higher traffic flow for a wide range of densities. Moreover, the effect of pre-signal area length also needs to be considered in a networked environment. We find that the optimal length of pre-signal area should be determined by the number of vehicles passing the traffic light in a left-turning phase at the intersection. The results can help the implementation and application of pre-signal in real traffic system.