基于粒子群算法的满载需求可拆分车辆路径规划

Split vehicle route planning with full load demand based on particle swarm optimization

为了更加合理地规划车辆配送路径,尽可能使用最少的车辆数和最短路径长度来完成整个客户点的配送任务,提出一种基于粒子群算法的满载需求可拆分车辆路径(F-SDVRP)规划策略,在配送过程中通过确保任何一辆满载的配送车辆从配送点出发后均以"最优"的配送路径进行配送来达到配送的总路径"最优"要求,并通过粒子群算法不断优化整个客户点的配送顺序.仿真结果表明,在求解相关客户点配送问题时,所提出的车辆规划策略得到的结果优于对比文献中的求解方法,在配送车辆数相同的情况下,最大的路径长度减少率达到8.21%.此外,各算例的仿真结果表明,所提出的策略的寻优结果稳定,粒子群算法可以解决满载需求可拆分车辆路径规划问题.

In order to plan the vehicle distribution path more reasonably, we use the minimum number of vehicles and the shortest path length as possible to complete the distribution task of the entire customer point. The paper proposes a particle swarm algorithm based on a full load demand split vehicle path(F-SDVRP) planning strategy, in the distribution process, by ensuring that any fully-loaded distribution vehicle has "optimal"distribution path for delivery after starting from the distribution point to achieve the "optimal"requirement of the total distribution path, and the particle swarm optimization algorithm is used to continuously optimize the distribution order of the entire customer point. The simulation results show that the vehicle planning strategy proposed is better than the solution method in the participating comparative literature when solving the related customer point distribution problem. With the same number of delivery vehicles, the maximum path length reduction rate reaches 8.21 %. In addition, the simulation results of the examples also show that the optimization results of the strategy are stable, and the particle swarm algorithm can solve the problem of split-vehicle path planning with full load requirements.