Acyclic Real-Time Traffic Signal Control Based on a Genetic Algorithm

Abstract

Abstract This paper presents an acyclic real-time traffic signal control model with transit priority based on a rolling horizon process for isolated intersections. The developed model consists of two components, including: an Improved Genetic Algorithm (IGA)-based signal optimization module and a microscopic traffic simulation module. The acyclic real-time traffic signal control model optimizes the phase sequence and the phase length with the aim to minimize the total delay of both transit vehicles and general vehicles for the next decision horizon. Numerical results show that the proposed IGA signal optimization module could provide a more efficient search for optimal solutions. The results also show that the acyclic real-time traffic signal control model outperforms the fixed-time control model. It prioritizes transit vehicles while minimizing the impact on the general vehicles.