Research On Vehicle Routing Problem Based On Modified Particle Swarm Optimization Algorithm

As"the third profit source", logistics is playing a more and more important role in our life. As the same time, with the development of electronic commerce, Logistics distribution on e-commerce role in supporting the more obvious. Among of them, the logistics distribution vehicle routing optimization problem is a key part of logistics distribution optimization ,as the same time it's the indispensable content of e-commerce activities. Vehicle distribution path is optimized can improve economic efficiency and make logistics scientific. Therefore, studying the logistics distribution vehicle routing optimization problem is not only has a great theoretical significance but also useful. Particle swarm optimization is based on swarm group and made by a group of particles, it is a kind of intelligent optimization method. as the same time, it is a parallel algorithm. In problem space particle swarm has the efficiency of search efficiency through cooperative search.This article's main work includes:(1)It has a studying on vehicle routing problem and particle swarm optimization algorithm, the vehicle routing problem's model is made on the basic of studying, and using modified particle swarm optimization algorithm to solve the model.(2)It's posed modified particle swarm optimization algorithm based on the characteristics of PSO. The virtues of the PSO contains simple principle, easy realization, a few parameters and higher converging speed, but it also exist some problems, the serious problem is easily creating earliness and bad ability in local optima and made problems deviate from the optimal. considering the simulated annealing algorithm is strong with local searching ability, this paper presents a new hybrid particle swarm algorithm with the idea of simulated annealing; Because the simulated annealing algorithm's dependence on initial temperature relatively strong, so set up a relation between Initial temperature and initial particle swarm; and the roulette strategy of genetic algorithm is used in this paper.(3) Finally, using the algorithm suggested in this paper resoled the VRP and VRPTW problems, had a simulation study in the way of MATLAB, and verified high efficiency of this means.