基于NSGA-II-RD的电商物流配送中心选址–路径–调度联合优化研究

Research on Joint Optimization of E-Commerce Logistics Distribution Center Location-Route-Dispatch Based on NSGA-II-RD

摘要: 针对电商平台需要规划向消费者配送商品的具体路径和库存调度量的情境，进行了全过程联合优化，提出了低碳视角下电商物流配送中心选址–路径–调度联合策略研究。首先，考虑到配送过程中时间存在不确定性，引用前景理论刻画消费者对预期配送时间的心理期望。同时，考虑到促销期间订单量激增导致库存有限，因此用订单满足率衡量消费者的满意度水平，与前一指标相结合计算配送效率。其次，在计算配送成本时，除了基本的运输成本，还创新性地引入了缺货成本系数这一变量，结合订单满足率，衡量缺货损失成本，从而建立起配送效率最大、配送成本最小的多目标优化模型。最后，使用改进设计的NSGA-II-RD (Non-dominated sorting genetic algorithms for route and dispatch)算法进行求解，将第一部分选址中的部分最终备选点作为本部分的配送中心，使用上海市相关数据进行案例分析，验证了该模型和算法的可行性。

Abstract: In response to the situation where e-commerce platforms need to plan specific delivery routes and inventory dispatch quantities to consumers, a joint optimization of the entire process was carried out, and a low-carbon perspective was proposed to study the joint strategy of e-commerce logistics distribution center location, route, and dispatch. Firstly, considering the uncertainty of delivery time, prospect theory is used to depict consumers’ psychological expectations towards expected delivery time. Meanwhile, considering the limited inventory due to surging orders during promotional periods, the order fulfillment rate is used to measure consumer satisfaction level, and combined with the previous indicator to calculate delivery efficiency. Secondly, when calculating delivery costs, in addition to basic transportation costs, an innovative variable called the stockout cost coefficient was introduced. Combined with the order fulfillment rate, the stockout loss cost was measured to establish a multi-objective optimization model that maximizes delivery efficiency and minimizes delivery costs. Finally, the improved NSGA-II-RD algorithm was used to solve the problem. Some of the final candidate points in the first part of location were selected as distribution centers. Case analysis was conducted using relevant data from Shanghai to verify the feasibility of the model and algorithm.