多起讫点货物转运配送车辆调度模型及其粒子群、蚁群算法混合求解  

Study of Multi-depots Goods Transhipment Vehicle Scheduling Problem Model and Its Particle Swarm and Ant Colony Optimization Hybrid Arithmetic

作　　者：王雷震[1,2] 汪定伟[1] 王素欣[2] WANG Leizhen;WANG Dingwei;WANG Suxin(College of Information Science and Engineering,Northeastern University,Shenyang 110004,China;Northeastern University at Qinhuangdao,Qinhuangdao 066004,China)

机构地区：[1]东北大学信息科学与工程学院,辽宁沈阳110819 [2]东北大学秦皇岛分校,河北秦皇岛066004

出　　处：《信息与控制》2018年第5期564-572,共9页Information and Control

基　　金：国家自然科学基金资助项目(61273203)

摘　　要：为使多起讫点货物转运配送车辆调度结果全局最优,建立多起讫点车辆调度模型.该模型求解过程是先由粒子群算法的粒子位置向量,得到每个货物的转运点及货物转运前后运货的车辆,再把转运点加入到蚁群算法的禁忌表中,用蚁群算法优化货物转运前、转运后的车辆路径,然后粒子群算法根据优化目标对粒子进行评价筛选,重复执行以上步骤直到满足终止条件.该算法使所有车辆对所有货物的转运点及车辆路径进行优化,货物转运点的位置和数量是变化的,易于实现最优解.实例求解结果表明货物转运配送得到的车辆总路径优于货物不转运配送得到的结果.To obtain a global solution to the multi-depots goods transshipment vehicle scheduling problem （ VSP）, in this study, we established VSP models. The optimization course is as follows： First, set up a partiele position vector to obtain a gaods transshipment point and then assign goods to vehicles. Second, establish a Tabu list lot the ant colony optimization （ACO） to obtain a vehicle route. The particle swarm arithmetic then evaluates and filters the vehicle scheduling results by optimization, which continues until it meets the terminate qualification. The hybrid arithmetic optimizes the transportation point and vehicle route, and the position and nmnber of the transportation point are changeable, which makes it easy to obtain a global salution. Simulation results show that the hybrid arithmetic is effective for the multi-depots goods transshipment vehicle scheduling problem.

关 键 词：多起讫点 转运配送 车辆调度问题 粒子群算法 蚁群算法 

分 类 号：U4[交通运输工程—道路与铁道工程]