An Improved Particle Swarm Optimization for the Multi-Depot Vehicle Routing Problem

Abstract

This paper proposes a formulation of the multi-depot vehicle routing problem (MDVRP) that is solved by the particle swarm optimization (PSO) algorithm. PSO is one of the evolutionary computation technique, motivated by the group organism behavior such as bird flocking or fish schooling. Compared with other search methods, such as genetic algorithm, ant colony optimization and simulated annealing algorithm, PSO has many advantages like only primitive mathematical operators, high precision and fast convergence. However, it may premature and trap into the local optima sometimes. In order to overcome the drawback, this paper introduces a modified PSO algorithm with mutation operator and improved inertia weight. The simulation results shown that this modified method could not only avoid premature automatically according to the convergence level but also get a better optimal solution than the basic one.