需求导向的高铁列车差异化定价与票额分配综合优化  

Collaborative optimization of differential pricing and seat allocation for multiple high-speed trains considering passenger demand

作　　者：苏焕银 徐光明[2] 曾琼芳 彭舒婷 贾旭东 SU Huanyin;XU Guangming;ZENG Qiongfang;PENG Shuting;JIA Xudong(School of Railway Tracks and Transportation,Wuyi University,Jiangmen 529020,China;School of Traffic and Transportation Engineering,Central South University,Changsha 410075,China;School of Tourism Management,Hunan University of Technology and Business,Changsha 410205,China)

机构地区：[1]五邑大学轨道交通学院,广东江门529020 [2]中南大学交通运输工程学院,湖南长沙410075 [3]湖南工商大学旅游管理学院,湖南长沙410205

出　　处：《铁道科学与工程学报》2022年第11期3138-3147,共10页Journal of Railway Science and Engineering

基　　金：五邑大学高层次人才科研启动项目(2017RC51);国家自然科学基金资助项目(72171236,71901093)。

摘　　要：结合我国高速铁路旅客出行需求的特征,研究高速铁路多列车差异化定价和票额分配综合优化方法。在分析旅客出发时段偏好特征的基础上,综合考虑旅客的出发时间偏差、旅途时间和票价支出构建旅客出行成本函数,基于多项Logit模型模拟旅客的出行选择行为。以客票收入最大化为目标建立高速铁路多列车差异化定价和票额分配的协同综合优化模型,将解的生成分为2个阶段。第1阶段,采用3种方法在票价约束范围内随机搜索构造票价方案,分别为变尺度柯西分布随机生成方法、组合随机生成方法和自适应随机生成方法。第2阶段,在保持优化目标不变的前提下,将综合优化模型转化为票额分配优化模型,采用Cplex优化器求解给定票价方案下的最优票额分配方案。基于解的两阶段生成方法,设计改进的直接搜索模拟退火算法对优化模型进行求解,选取武广高速铁路多趟列车进行实验分析。研究结果表明:1)列车客票收入具有明显的提高,受弹性需求系数的影响较大,提高了9%~27%,旅客出行需求满足率均达到97%以上。2)在算法降温200次时,优化目标均已经实现了97%以上,算法收敛效率较高。3)不同出发时段的列车票价具有差异性,列车在各个运行区段上的票额分配方案与旅客出行需求高度匹配,说明了优化方法的需求导向性。Differential pricing and seat allocation of multiple high-speed trains were jointly optimized considering with travel demand characteristics of high-speed railways in China. The preference of the passenger demand regarding departure times were analyzed. The departure time deviation, travel time and ticket fare were considered simultaneously in the travel cost function. The multinomial Logit model was employed to simulating passenger choice behaviors. A collaborative optimization model was constructed with the aim of maximizing the ticket revenue and the decision variables for pricing train legs and allocating seats for multiple high-speed trains.New solutions were constructed in two stages. In the first stage, three methods of generating new solutions were used to search new solutions randomly within the price constraints, which were respectively the method based on the Cauchy distribution with a variable scale parameter, the method by combining randomly and the method based on the standard normal distribution. In the second stage, the collaborative optimization model was translated into a seat allocation optimization model with the same optimized objective. Based on the above two stages, a modified direct search simulated annealing algorithm was designed to solve the optimization model. The experimental analysis containing dozens of trains is performed on Wuhan-Shenzhen high-speed railway. The results are drawn as follows.(1) The ticket revenue of the optimized solution increases obviously by 9%~27%,influenced greatly by the elastic demand coefficient, and more than 97% of the passenger demand is satisfied.(2) Over 97% of the optimization objective is obtained while the temperature of the algorithm falls 200 times, and the algorithm converges efficiently.(3) Differentiation is reflected for trains with different departure periods and the seat allocation of each train leg is highly matched with the passenger demand, illustrating that passenger demand are fully considered in the optimization method.

关 键 词：铁路运输 高速铁路 差异化定价 票额分配 模拟退火算法 

分 类 号：U293.22[交通运输工程—交通运输规划与管理]