Collaboration and transportation resource sharing in multiple centers vehicle routing optimization with delivery and pickup

被引:72
作者
Wang, Yong [1 ,2 ]
Zhang, Jie [1 ]
Assogba, Kevin [3 ]
Liu, Yong [1 ]
Xu, Maozeng [1 ]
Wang, Yinhai [4 ,5 ]
机构
[1] Chongqing Jiaotong Univ, Sch Econ & Management, Chongqing 400074, Peoples R China
[2] Univ Elect Sci & Technol, Sch Management & Econ, Chengdu 610054, Sichuan, Peoples R China
[3] Oregon State Univ, Sch Civil & Construct Engn, Corvallis, OR 97331 USA
[4] Univ Washington, Dept Civil & Environm Engn, Seattle, WA 98195 USA
[5] Tongji Univ, Coll Transportat Engn, Transportat Data Sci Res Ctr, Shanghai 201804, Peoples R China
基金
中国国家自然科学基金; 中国博士后科学基金;
关键词
Cooperative game theory; Resource sharing; Multiple centers vehicle routing problem; Profit allocation schemes; Hybrid heuristic algorithm; VARIABLE NEIGHBORHOOD SEARCH; NSGA-II; COST ALLOCATION; HORIZONTAL COOPERATION; OBJECTIVE OPTIMIZATION; NETWORK OPTIMIZATION; PROFIT ALLOCATION; GENETIC ALGORITHM; TIME WINDOWS; LOGISTICS;
D O I
10.1016/j.knosys.2018.07.024
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The adoption of collaboration strategies among logistics facilities and the formation of one or multiple coalitions constitute a sustainable approach to vehicle routing network optimization. This paper introduces a collaborative multiple centers vehicle routing problem with simultaneous delivery and pickup (CMCVRPSDP) to minimize operating cost and the total number of vehicles in the network. Distribution and pickup centers are allowed to share vehicles and customers in order to increase the entire network's efficiency and maximize profit. To provide the coalition coordinators with good routing solutions, we propose a hybrid heuristic algorithm which properly combines k-means and Non-dominated Sorting Genetic Algorithm-II (NSGA-II). Based on clustering solutions, the proposed Hybrid NSGA-II (HNSGA-II) first generates a real coded population to bind our mathematical model constraints and to obtain a large number of feasible solutions which converge to optimality. Chromosomes are divided for genetic operations with partial mapped crossover and swap mutation algorithms, before their recombination to ensure the quality of our results. Comparisons with the traditional NSGA-II and the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm indicate better performances of HNSGA-II in terms of objective function values. We also apply Cost Gap Allocation method (CGA) and the strictly monotonic path selection principle to examine profit allocation schemes. Numerical analyses on part of Chongqing city's logistics network show the superiority of HNSGA-II over MOPSO and NSGA-II on the practical case study, as well that of CGA over the Minimum Costs-Remaining Savings (MCRS), Shapley and Game Quadratic Programming (GQP) methods. In addition, the proposed profit allocation approach has supported the establishment of a grand coalition instead of two sub-coalitions. CMCVRPSDP optimization reduces long-haul transportation, improves the vehicle loading rate and facilitates sustainable development. Through the rational allocation of profits, the proposed solution methodology assures the stability and fairness among coalition members. The implementation is also important to design sustainable urban transportation networks.
引用
收藏
页码:296 / 310
页数:15
相关论文
共 50 条