主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2026, Vol. 34 ›› Issue (10): 80-93.doi: 10.16381/j.cnki.issn1003-207x.2023.2045

• • 上一篇    下一篇

带时间窗的多中心开闭混合式电动车配送路径优化问题研究

苟梦圆1,2, 王勇1,2(), 罗思妤1, 许茂增1   

  1. 1.重庆交通大学经济与管理学院,重庆 400074
    2.绿色物流智能技术重庆市重点实验室,重庆 400074
  • 收稿日期:2023-12-05 修回日期:2024-04-07 出版日期:2026-10-25 发布日期:2026-10-09
  • 通讯作者: 王勇 E-mail:yongwx@cqjtu.edu.cn
  • 基金资助:
    国家自然科学基金项目(72371044);国家自然科学基金项目(71871035);重庆市教委科学技术重大项目(KJZD-M202300704);重庆市自然科学基金面上项目(CSTB2025NSCQ-GPX0848);重庆市高等教育教学改革研究重大项目(251030);重庆市教委人文社科研究项目(26SKJD096)

Open-Closed Mixed Electric Vehicle Routing Optimization of Multi-center Distribution with Time Windows

Mengyuan Gou1,2, Yong Wang1,2(), Siyu Luo1, Maozeng Xu1   

  1. 1.School of Economics and Management,Chongqing Jiaotong University,Chongqing 400074,China
    2.Chongqing Key Laboratory of Green Logistics Intelligent Technology,Chongqing 400074,China
  • Received:2023-12-05 Revised:2024-04-07 Online:2026-10-25 Published:2026-10-09
  • Contact: Yong Wang E-mail:yongwx@cqjtu.edu.cn

摘要:

针对多中心电动车辆路径优化研究在资源共享和开闭混合式车辆路径设计相结合方面存在的不足,研究了带时间窗的多中心开闭混合式电动车配送路径优化问题。首先,构建了包含机械功率、放电效率和能量转化效率等因素的电动车电量消耗最小化和运营成本最小化的双目标优化模型。然后,提出了基于贪婪算法的改进NSGA-Ⅱ混合算法求解模型,该混合算法设计了基于时空距离的贪婪算法生成初始种群,并结合问题特征进行了种群异位交叉与变异操作,增强了种群多样性,以及设计了充电站插入策略和资源共享策略,进而提高了算法收敛速度。其次,将所提算法与多目标粒子群算法、多目标差分进化算法和多目标遗传算法进行对比分析,并应用帕累托解数量、超体积与平均理想距离算法性能衡量指标进一步比较分析,综合验证了所提算法的有效性。最后,结合实例对提出的模型和算法进行验证,并对不同开闭混合模式和不同合作模式下的多中心电动配送车辆路径优化方案进行了对比分析。结果表明,多中心开闭混合式电动车配送路径优化方法能够有效降低电动车配送网络的运营成本,减少电动车的总电量消耗和使用数量,并有利于提高多中心电动车配送网络的资源配置效率,进而为城市电动车物流配送网络优化调度以及新能源物流网络构建提供方法支持和理论支撑。

关键词: 多中心电动车配送网络, 开闭混合式路径, 基于贪婪算法的改进NSGA-II混合算法, 资源共享策略, 合作模式

Abstract:

With the substantial increasement of energy consumption in the logistics transportation industry and the scale and cluster development of urban logistics distribution networks, promoting the green transformation of the logistics and transportation industry and deepening the resource integration of urban logistics distribution networks have become the key development directions. The design of open-closed mixed vehicle routes and the application of high-efficiency, low-carbon and low-cost electric logistics vehicles in the urban multi-center distribution network are conducive to the integration of transportation resources in the logistics network, and promote the low-carbon development of the logistics industry. However, due to the widespread phenomenon of unreasonable configuration and unbalanced use for facilities in the urban multi-center logistics distribution network, electric logistics distribution vehicles face problems such as difficult charging and slow distribution time. Thus, in order to overcome the insufficiency of the multi-center electric vehicle routing optimization in combination with resource sharing and open-closed mixed routing design, an open-closed mixed electric vehicle routing problem of multi-center distribution with time windows is proposed. First, a nonlinear function is applied to calculate the mechanical power of electric vehicles by considering the factors such as air resistance, rolling resistance and gravity, and then the energy loss and driving time are combined to obtain the energy consumption calculation method of electric vehicles. Based on the calculation method and resource sharing constraints of charge stations and electric vehicles, the multi-center electric vehicle routing optimization model is constructed to minimize the total energy consumption and operating cost. Second, an improved NSGA-II hybrid algorithm based on the greedy algorithm is designed to solve the model. The hybrid algorithm divides the service period according to the customer time window characteristics, and designs the greedy algorithm based on spatial and temporal distances to generate the initial solutions, which improves the convergence speed of the hybrid algorithm. The ectopic crossover and mutation operations are proposed to realize the iterative updating of the solutions, and the charging station insertion and resource sharing strategies are designed to achieve the reasonable planning of vehicle routes. Additionally, the proposed hybrid algorithm is compared with multi-objective particle swarm optimization, multi-objective differential evolution algorithm, and multi-objective genetic algorithm. The number of Pareto solutions, hyper-volume and mean ideal distance are used as the measurement indicators to further compare and analyze, so as to comprehensively verify the effectiveness of the proposed hybrid algorithm. Third, combined with a multi-center electric vehicle distribution network in Chongqing city, China, the electric vehicle routing optimization scheme is studied, and the optimization results under different open-closed mixed routing designs and different cooperation modes are compared and analyzed. The research results show that compared with the open route design and the closed route design of electric vehicles, the application of the open-closed mixed route design of electric vehicles can shorten the travel distance, reduce the energy consumption and the number of used charging stations. At the same time, the resource integration of multi-center electric vehicle distribution network can effectively save total operating cost, reduce the number of electric vehicles and energy consumption. Therefore, this study can improve the efficiency of resource allocation in the multi-center electric vehicle distribution network, and then can provide theoretical support for the scheduling optimization of the urban electric vehicle distribution network and the construction of new energy city logistics networks.

Key words: multi-center distribution network of electric vehicle, open-closed mixed route, improved NSGA-II algorithm based on the greedy algorithm, resource sharing strategy, cooperation mode

中图分类号: