An Efficient Search Economics Based Algorithm for Urban Traffic Light Scheduling

Instead of building more traffic infrastructures, developing a better solution for traffic light scheduling is probably the fastest and most money-saving way for improving the traffic condition or solving the traffic problem. This paper presents a novel metaheuristic algorithm, called search economics for traffic light scheduling (SE-TLS), for urban traffic light scheduling with two goals; namely, maximizing the number of vehicles reaching the destination and minimizing the trip time they take. One of the characteristics of SE-TLS is that it is not easy to get stuck in a local optimum, so it can continue to find better solutions during the convergence process. The traffic simulator, named Simulation of Urban Mobility (SUMO), was used to verify the performance of SE-TLS by applying it to the traffic scenarios of three cities. The experimental results show that SE-TLS outperforms all the other algorithms evaluated in this paper in all cases, thus implying that it provides a better traffic light scheduling for effectively improving the traffic congestion problem.