Using Predicting Particle Swarm Optimization to solve the vehicle routing problem with time windows

In this paper, the predicting particle swarm optimization (predicting PSO) algorithm based on original PSO which modified by three new solution strategies has been proposed to solve the well known VRPTW problems. Predicting PSO algorithm not only can modify the variance between current status and memorial best status, but also can forward predict what the best status should be at next time status based on memory. The predicting PSO algorithm has also been tested on Solomon¿s benchmark problems. The findings indicate that the proposed algorithm outperforms other heuristic algorithms.