地铁乘客站点的选择行为分析及预测  

Analyzing and Predicting Station Choice Behavior of Subway Passengers

作　　者：王璞[1,2] 肖健和 李明伦 郭宝 WANG Pu;XIAO Jianhe;LI Minglun;GUO Bao(School of Traffic and Transportation Engineering,Central South University,Changsha 410075;Rail Data Research and Application Key Laboratory of Hunan Province,Changsha 410075)

机构地区：[1]中南大学交通运输工程学院,长沙410075 [2]轨道交通大数据湖南省重点实验室,长沙410075

出　　处：《电子科技大学学报》2022年第4期623-629,共7页Journal of University of Electronic Science and Technology of China

基　　金：国家自然科学基金(71871224);湖南省交通运输厅科技进步与创新项目(212102)。

摘　　要：采用数据互补的方法,融合智能卡数据和公交车GPS轨迹数据,获取了公交和地铁乘客在地铁新线路开通前后的乘客个体出行信息。利用居民空间行为指标——“出行质心”对地铁乘客的公交出行质心定位,并分析了地铁站点与乘客公交出行质心之间的距离对乘客地铁站点选择的影响。结果表明有86.15%的乘客选择距离他们公交出行质心更近的地铁站点,地铁站点与乘客公交出行质心之间的距离是影响乘客出行站点选择的重要因素。基于这一发现,建立Logit模型对乘客是否会选择新地铁站点进行预测,预测准确率、精确率、召回率和特异性分别为83.87%、 84.23%、83.66%和84.09%,预测结果表现良好。研究结果可用于评估计划新建的地铁站点对邻近既有站点的影响,有助于地铁营运计划的设计。Data fusion method is used to obtain the individual travel information of bus and subway passengers before and after the opening of the new subway line. The important human mobility index, center of mass, is used to locate the center of mass of passengers’ bus boarding locations, and the impact of the distance between a subway station and the center of mass on passengers’ choice of subway stations is analyzed. Results show that 86.15% of the passengers choose subway stations closer to their center of mass, and the distance is a key factor affecting passengers’ station choice behavior. Based on the findings above, we develop a Logit model to predict whether a passenger will choose to use the new subway station. The prediction accuracy, precision, recall and specificity are 83.87%, 84.23%, 83.66% and 84.09%, respectively, indicating that the model performs well.The results of this study can be used to evaluate the impact of planned new subway stations on the adjacent existing subway stations and contribute to the design of subway operation plan.

关 键 词：数据融合 出行特征 地铁新开线 智能卡数据 

分 类 号：TU984[建筑科学—城市规划与设计] U121[交通运输工程]