并行烟花算法求解多货场电子商务物流运输调度问题  

Parallel Fireworks Algorithm for Multi-depot E-commerce Vehicle Routing Problem

作　　者：陈厚仁[1] 蔡延光[1] 黄戈文[1,2] 蔡颢[3] CHEN Hou-ren;CAI Yan-guang;HUANG Ge-wen;CAI Hao(School of Automation,Guangdong University of Technology,Guangzhou 510006,China;Information and Network Center,Jiaying University,Meizhou 514015,China;Department of Health Science and Technology,Aalborg University,Aalborg 9220,Denmark)

机构地区：[1]广东工业大学自动化学院,广州510006 [2]嘉应学院信息网络中心,广东梅州514015 [3]奥尔堡大学健康科学与工程系,丹麦奥尔堡9920

出　　处：《嘉应学院学报》2020年第3期21-27,共7页Journal of Jiaying University

基　　金：国家自然科学基金(61074147);广东省自然科学基金(S2011010005059);广东省教育部产学研结合项目(2012B091000171,2011B090400460);广东省科技计划项目(2012B050600028,2014B010118004,2016A050502060);广州市花都区科技计划项目(HD14ZD001);广州市科技计划项目(201604016055)

摘　　要：根据电子商务物流运输调度过程中的特征,考虑了不同区域司机配送成本、多货场、时间窗等约束条件,建立了一种多货场电子商务物流运输调度模型,并提出一种并行烟花算法进行求解.所提出的算法采用货场车辆数与最大位置法结合的实数编码方式,并提出一种并行交流策略加强了算法的全局搜索能力以及提高运算速度.对比实验表明,相对于改进前的烟花算法,并行烟花算法具有较好的求解精度和求解时间.According to the characteristics of transportation scheduling process in e-commerce logistics,a multi-depot e-commerce vehicle routing problem is established in consideration of the constraints of variable costs for drivers in different regions,multi-depot and time windows.The parallel fireworks algorithm is proposed to solve the problem.The proposed algorithm uses the real number coding method which adopt the maximum position method including the vehicles number of depots.The parallel communication strategy is proposed to enhance the global search ability of the algorithm and reduce running time in solving problems.The comparison experiment shows that the parallel fireworks algorithm outperform the original fireworks algorithm in solution accuracy and running time.

关 键 词：烟花算法 多货场 车辆路径问题 并行交流策略 最大位置法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]