Abstract: To design delivery service efficiently,the equilibrium problem of last mile delivery service choice under customers bounded rationality was researched.Three situations related to delivery service were home-delivery service,attended pickup point service and unattended pickup point service.From the perspective of customers expected utility maximization,the piecewise functions of customer expected utility under different situations were proposed based on queuing model by considering factors included pickup distance,pickup time,charge,security,communication convenience and return convenience.The delivery service choice model was constructed with Nested Logit(NL) model to depict the customers bounded rationality as customers unable to make accurate calculations of their expected loss or waiting utility.Meanwhile,based on quantal response equilibria,the existence and uniqueness of Logit equilibrium in last mile delivery service system were proved.The correctness of the proposed model was demonstrated by numerical experiments,and the result also indicated that the last mile delivery service design was affected by the factors such as pickup distance,pickup time,charge and customer rational degree significantly.

Key words: bounded rationality, last mile delivery service, Nested Logit model, Logit equilibrium, queuing

摘要： 为设计高效的配送服务,研究了顾客有限理性条件下送货上门、有人值守自提点与无人值守自提点三种最后一公里配送服务选择均衡问题。从顾客期望效用最大化角度出发,基于排队模型,考虑取货距离、取货时间、运费、安全性、沟通便利性、退货方便性等因素,建立了三种配送服务的分段效用函数。基于Nested Logit(NL)模型,构建了有限理性下顾客配送服务选择模型,刻画了顾客无法准确计算他们的期待损失或者等待效用的有限理性。基于随机最优反应均衡,证明了最后一公里配送服务系统存在Logit均衡且均衡唯一。通过数值实验验证了模型的正确性,揭示出取货距离、取货时间、运费及顾客理性程度是最后一公里配送服务设计的重要影响因素。

关键词: 有限理性, 最后一公里配送服务, NL模型, Logit均衡, 排队