Abstract: In view of the transaction default problems in the current urban logistics distribution process, such as customers cancel orders in the midway and return goods without reason, this paper presents a measurement method of customer credit degrees. First, customer credit values are calculated through the customers' historical transaction default data, and then the customer credit degrees are addressed. The vehicle routing optimization model includes transportation cost, lease cost, and penalty cost of the time window is established. Second, a genetic (GA)-tabu search (TS) hybrid algorithm is devised to address the model, and the elite retention strategy is used for loop iteration optimization in the process of the hybrid algorithm. Finally, the effectiveness and feasibility of the model and algorithm is verified through the example data of a certain takeout logistics distribution network in Chongqing. The experimental results show the logistics distribution scheduling schemes under different service strategies, and the comparison of customer delivery service adjustment based on the customer credit and the sensitivity analysis is carried out. The study shows that the reasonable division of the customer credit rating can effectively reduce the total cost of logistics distribution and improve the customer service level.

Key words: vehicle routing problem, GA-TS hybrid algorithm, customer credit degree, logistics distribution, elite retention strategy

摘要： 针对当前城市物流配送过程中普遍存在的客户中途取消订单、无故退换货等交易违约问题,引入客户信用度的测度方法。根据客户历史交易违约数据计算客户信用值,并转化求解客户信用度,构建了包含车辆配送成本、租赁成本以及违反时间窗惩罚成本的配送路径优化模型。设计了一种遗传(GA)-禁忌搜索(TS)混合算法进行模型求解,在算法过程中应用精英保留策略进行循环迭代寻优。结合重庆某外卖物流配送网络的实例数据,验证了模型和算法的有效性和可行性。实验结果给出了不同服务策略下的物流配送调度方案,并进行了基于客户信用度的客户配送服务序列调整比较和敏感度分析。研究表明客户信用等级的合理划分可以有效降低物流配送成本和提高客户服务水平。

关键词: 车辆路径问题, GA-TS混合算法, 客户信用度, 物流配送, 精英保留策略