基于破损信息不确定的共享单车新车投放-旧车调拨-坏车回收联合调度优化研究  

Joint Scheduling Optimization of New,Used and Broken Bicycles in Bicycle-Sharing System Under Uncertainty of Broken Bicycles

作　　者：肖建华[1,2] 贾楠 潘钰雅 杜经国 XIAO Jian-hua;JIA Nan;PAN Yu-ya;DU Jing-guo(Research Center of Logistics,Nankai University,Tianjin 300071,China;Collaborative Innovation Centre for China Economy,Tianjin 300071,China)

机构地区：[1]南开大学现代物流研究中心,天津300071 [2]中国特色社会主义经济建设协同创新中心,天津300071

出　　处：《数学的实践与认识》2021年第12期89-101,共13页Mathematics in Practice and Theory

基　　金：国家自然科学基金(61772290,62072258)。

摘　　要：针对共享单车的供需不平衡、破损信息不确定等特征,融合鲁棒优化方法,以系统调度时间最短为目标,构建了基于破损信息不确定的新车投放-旧车调拨-坏车回收共享单车联合调度优化模型.针对模型的NP难特性,利用多目标分治策略和变步长扰动策略,提出一种改进变邻域搜索算法求解该问题.最后,通过维也纳标准测试集和SABB案例,验证模型和算法的有效性及高效性.In view of the imbalance between supply and demand of bicycle-sharing system and the uncertainty of broken bicycles,the paper integrates the robust optimization method and constructs a joint scheduling optimization model of new bicycle release,used bicycle rebalance and broken bicycle recycling with the goal of minimizing the system allocation time.Aiming at the NP-hard problem,the paper proposes an improved variable neighborhood search algorithm based on the multi-objective divide-and-conquer strategy and the variable step-size perturbation strategy.Finally,the validity of the model and the efficiency of the algorithm are verified by Vienna standard test set and SABB case.

关 键 词：信息不确定 共享单车 联合调度 鲁棒优化 变邻域搜索 

分 类 号：U491.225[交通运输工程—交通运输规划与管理]