摘要
常旅客价值分析与分类是铁路运输部门客户关系管理中的重要内容,其有助于掌握旅客信息特点,满足旅客的多方面需求,提供差异化服务。根据改进的RFM价值模型计算常旅客的价值,采用K均值聚类对旅客进行分类,通过常旅客价值达到降维的目的,并设计了一种算法以实现自动确定聚类数目以及选取初始聚类中心,从而能够高效快速地进行客户分类。通过问卷调查收集福厦高速铁路的常旅客数据进行实证分析,结果表明算法自动选取的聚类数目是合理的,初始聚类中心可以提高聚类效率和准确性。最后根据分类结果,有针对性地提出相应的服务提升建议和优化策略,可为铁路部门在运营服务方面提供参考。
Analysis of frequent rail-traveler value and customer classification are significant components of the operation management and customer relationship management in the railway transportation department. They are conducive to grasping the characteristics of frequent rail-traveler information, meeting the diverse needs of different travelers, and providing differentiated services. This paper calculated frequent rail-traveler value via the modified recency, frequency,and monetary value(RFM) model and classified travelers via K-means clustering. In addition, the study designed an algorithm to automatically determine the best clustering number and select the initial clustering center through dimension reduction based on frequent rail-traveler value, which can efficiently classify rail-travelers. Then, the availability of the algorithm was verified through the data of frequent rail-travelers on the Fuzhou-Xiamen passenger dedicated line collected by questionnaire. The results show that the best clustering number selected by the algorithm is reasonable,and the initial clustering center can improve the clustering efficiency and accuracy. Lastly, some specific corresponding suggestions for upgrading service according to the classification results were proposed to provide a scientific reference for the Railway Department on operational services.
作者
郭星
许旺土
任冲
GUO Xing;XU Wangtu;REN Chong(School of Architecture and Civil Engineering,Xiamen University,Xiamen 361005,Fujian,China;Transportation and Urban Planning and Design Institute,China Railway Eryuan Engineering Group Co.,Ltd.,Chengdu 610031,Sichuan,China)
出处
《铁道运输与经济》
北大核心
2022年第6期48-55,共8页
Railway Transport and Economy
基金
中铁二院工程集团有限责任公司科技开发计划(KSNQ202053)。
