海岛无人机多模式应急配送模型与算法研究  

Model and algorithm research on multi-mode emergency distribution of drones on island

作　　者：胡志华[1] 魏月荷 田曦丹 HU Zhihua;WEI Yuehe;TIAN Xidan(Logistics Research Center,Shanghai Maritime University,Shanghai 201306,China)

机构地区：[1]上海海事大学物流研究中心,上海201306

出　　处：《大连理工大学学报》2024年第6期649-660,共12页Journal of Dalian University of Technology

基　　金：国家自然科学基金资助项目(71871136);上海市自然科学基金资助项目(23ZR1426500);教育部哲学社会科学研究重大课题攻关项目(21JZD058).

摘　　要：灾害情况下物资海岛配送是应急物流的一种特殊应用场景.无人机配送作为解决应急物流问题的一种新型配送方式受到广泛关注,并涌现出了一系列配送模式.通过考虑无人机成本、无人机飞行距离、无人机起降平台和地面配送成本,构建无人机直接配送、点对点双阶段配送、回路双阶段配送和双层物流配送网络4种配送模式.针对回路双阶段配送的多指标耦合特征,提出一种迭代启发式算法进行优化.围绕普陀山海岛场景,构建20~100个需求点和2~4个无人机起降平台的算例进行数值实验,分析并验证模型正确性和算法性能.实验结果表明,4种配送模式模型具有有效性;算法一般经过2~5次迭代,能够有效减少回路双阶段配送约19.08%的无人机飞行距离,减少双层物流配送网络约28.63%的地面配送距离.Material island distribution in disaster situation is a special application scenario for emergency logistics.Drone distribution as a new distribution method to solve the emergency logistics problem has received wide attention,and a series of distribution modes have emerged.By considering drone cost,drone flight distance,drone landing platform and ground distribution cost,four distribution modes are proposed:drone direct distribution,point-to-point bi-stage distribution,cyclic bi-stage distribution and two-level logistics distribution network.An iterative heuristic algorithm is proposed to optimize the multi-metric coupling characteristics of cyclic bi-stage distribution.Based on Putuoshan island scenario,20-100 customers and 2-4 drone landing platforms are constructed for numerical experiment to analyze and verify the model correctness and algorithm performance.The experiment results show that the models of the four distribution modes have validity,and the algorithm can effectively reduce the drone flight distance of cyclic bi-stage distribution by about 19.08%and the ground distribution distance of two-level logistics distribution network by about 28.63%after 2-5 iterations generally.

关 键 词：无人机配送 应急物流 选址-路径优化 旅行商问题 迭代启发式算法 

分 类 号：F251.2[经济管理—国民经济] G311[文化科学]