数据驱动下基于旅客需求的高铁差异化定价优化研究
A Study on the Optimization of Data-driven High-speed Railway Differentiated Pricing Based on Passengers'Demand
陈成 1彭定2
作者信息
- 1. 安徽黄梅戏艺术职业学院综艺系,安徽安庆 246052
- 2. 芜湖职业技术学院汽车与航空学院,安徽芜湖 241006
- 折叠
摘要
为提高旅客服务质量,提升铁路运输企业收益,多条高铁线路实行了市场化票价机制.基于此,通过集成的机器学习(EML)算法深入分析旅客需求数据,预测旅客对价格和旅行时间的敏感度和需求变化,并将这些变化应用于动态定价策略中,形成一种集合了三种算法的ESA(EM L-SA-ALNS)算法.京沪高铁的实际数据验证了模型的效果,结果显示模型能有效平衡旅客需求与铁路运输企业的收益.
Abstract
In order to improve the passenger service and increase the revenue of railway transport enterprises,a number of high-speed railway lines have implemented a market-based fares mechanism.Based on this,the ensemble machine learning(EML)algorithm is used to make deep analysis on the data on passengers'demand to accurately predict passengers'sensi-tivity to fares and travel time and the changes in their demand,and these changes are applied to the high-speed railway dy-namic pricing strategy to form an ESA(EML-SA-ALNS)algorithm that aggregates three algorithms.The effect of the mod-el is verified by the actual data from the Beijing—Shanghai high-speed railway,and the results show that this model can ef-fectively balance passengers'demand and the revenue of railway transportation enterprises,providing innovative methods and ideas for the dynamic optimization of high-speed railway fares strategy.
关键词
高速铁路/旅客运输/差异化定价/旅客需求分析/机器学习/ALNS算法Key words
high-speed railway/passenger transportation/differentiated pricing/analysis on passengers'demand/machine learning/ALNS algorithm引用本文复制引用
基金项目
2020年安徽省高等学校自然科学重点项目(KJ2020A1027)
出版年
2024
NETL
NSTL
万方数据