摘要
中欧班列运输需求高速增长,导致旺季时间偏长,运能供不应求。部分时间敏感型货物无法长时间等待,进而转向了其他运输方式。为满足时间敏感型货物的运输需求,本文考虑货物时间价值并提出多等级定价方案,以班列运营企业收益最大化为目标建立了考虑货主选择行为的动态舱位控制模型。由于多维状态空间模型难以获得精确解,采用值函数近似方法近似求解动态规划模型,利用最小二乘近似策略迭代算法对线性近似、分段线性近似两种形式的近似值函数进行参数标定。应用学习得到的参数进行仿真实验,对比两种值函数近似方法与基准运价下先到先得、多级定价下先到先得方案的收益性能和舱位预订情况。结果表明,在需求旺季供不应求的场景下,本文提出的基于多级定价的舱位控制策略能把有限的运能分配给最需要的客户,同时可以有效提升铁路企业的运输收益,其中分段线性近似方法对收益提升的效果更佳。
The rapid growth in transport demand for China Railway Express has led to a long peak season and shortage of transport capacity.Some time-sensitive goods cannot wait long and are diverted to other modes of transport.To meet the transport demand for time-sensitive goods,in this study,a multi-level pricing scheme was proposed considering the time value of goods,and a dynamic slot inventory control model considering the choice behavior of goods owners was established to maximize the revenue of train operation enterprises.Because of the multi-dimensional state space,it is difficult to obtain an exact solution for the model.Therefore,the dynamic programming model was approximately solved using the value function approximation method,and the least squares approximate policy iteration algorithm was adopted to calibrate the parameters of the linear and piecewise linear approximation value functions.Applying the learned parameters,simulation experiments were performed to compare the revenue performance and slot booking status of the two value function approximation methods with those of“first-come first-served scheme under benchmark price”and“first-come first-served scheme under multi-level pricing”.The results show that under the scenario of short supply in the peak demand season,the slot inventory control strategy based on multi-level pricing proposed in this paper can allocate limited capacity to customers most in need,while simultaneously effectively improving the transportation revenue of railway enterprises,among which the piecewise linear approximation method has a better effect on revenue improvement.
作者
徐如君
张小强
XU Rujun;ZHANG Xiaoqiang(School of Transportation and Logistics,Southwest Jiaotong University,Chengdu 611756,China;National Engineering Laboratory of Integrated Transportation Big Data Application Technology,Chengdu 611756,China;National United Engineering Laboratory of Integrated and Intelligent Transportation,Chengdu 611756,China)
出处
《交通运输工程与信息学报》
2024年第2期77-89,共13页
Journal of Transportation Engineering and Information
基金
国家自然科学基金资助项目(U2368212)。
关键词
铁路运输
多级定价
动态舱位控制
货物时间价值
选择行为
railway transportation
multi-level pricing
dynamic slot inventory control
time value of goods
choice behavior
