Research on Global-Local Optimal Information Ratio Particle Swarm Optimization for Vehicle Scheduling Problem

摘要

In order to reduce the standard particle swarm algorithm trapped in local optimal value, guarantee the convergence speed of the particle swarm optimization algorithm and improve the quality of the solution and robustness in the vehicle scheduling problem, based on the standard particle swarm optimization (PSO) algorithm, this paper proposes a new improved standard particle swarm algorithm namely global-local optimal information ratio PSO (GLIR-PSO), and the algorithm using the particle's global-local optimal information ratio weighs the particles of particle's global optimal and local optimal information and it is applied to the vehicle scheduling problem, the model of particle swarm optimization for vehicle scheduling problem is established, and compared with standard particle swarm optimization algorithm and the new particle swarm optimization algorithm with global-local best minimum. The results of simulation demonstrate that the algorithm shows a better performance in convergence speed, so it is an effective method for solving the vehicle scheduling problem.