On tuning the particle swarm optimization for solving the traffic light problem

• Autor(a) principal: Silva, Gonçalo O.
• Data de Publicação: 2022
• Outros Autores: Rocha, Ana Maria A. C., Witeck, Gabriela R., Silva, António José Linhares, Durães, Dalila, Machado, José Manuel
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://hdl.handle.net/1822/81692
• topic: Particle swarm optimization; Simulation of urban mobility; Traffic lights problem

In everyday routines, there are multiple situations of high traffic congestion, especially in large cities. Traffic light timed regulated intersections are one of the solutions used to improve traffic flow without the need for large-scale and costly infrastructure changes. A specific situation where traffic lights are used is on single-lane roads, often found on roads under maintenance, narrow roads or bridges where it is impossible to have two lanes. In this paper, a simulation-optimization strategy is tested for this scenario. A Particle Swarm Optimization algorithm is used to find the optimal solution to the traffic light timing problem in order to reduce the waiting times for crossing the lane in a simulated vehicle system. To assess vehicle waiting times, a network is implemented using the Simulation of Urban MObility software. The performance of the PSO is analyzed by testing different parameters of the algorithm in solving the optimization problem. The results of the traffic light time optimization show that the proposed methodology is able to obtain a decrease of almost 26% in the average waiting times.