A Hybrid Particle Swarm Optimization and Tabu Search algorithm for adaptive traffic signal timing optimization

In this paper a hybrid Particle Swarm Optimization and Tabu Search Algorithm for adaptive traffic signal timing optimization is proposed. We present a novel algorithm that uses the information of the particle best neighbor in updating velocity and position at the ith iteration; particle and her best neighbor exchange their best local position with a certain frequency. Then, the historical best solutions of both particles will be stocked in the Tabu list to avoid trapping into local optimum and premature convergence. In our contribution first, we give a new way of moving for each particle depending on her best historical position and whether it is included in the Tabu list. Second, we prove the effectiveness of the proposed algorithm for solving the real time traffic at isolated intersections. In this case study, we aim at optimizing and regulating the real time traffic in Moroccan cities with equipments that have an adaptive programming. The system shows good results and provides cyclic signal operation based on a real time control approach by minimizing total delay at intersections.