基于分布式出行信息的轨道交通出行特征辨识

为准确识别网络化运营环境下城市轨道交通乘客的出行特征,设计了可采集Wi-Fi信息的分布式交通行为识别系统,并建立出行特征识别算法.布设在各站点的检测设备可采集乘客所携带移动设备独一无二的M AC地址信息,并上传至信息中心.信息中心通过对比同一设备在各站点获取的时间戳和对应站点编号,可识别乘客的出行路径和行程时间,结合轨道交通车辆的走行时间信息可获取换乘站的换乘时间,并应用上述时间信息验证所识别出行路径信息的有效性.在西安市轨道交通系统的测试结果表明,同由客票信息获取的出行行为相比,该系统能采集所有网络形态下的乘客出行路径及行程时间,测试数据的平均采样率可达32.86%,误差为3.8%.该系统的分析结果可用于城市轨道交通系统的客票清分、站点设计等环节. 

Identification of Travel Behavior of Urban Rail Transit System Using Distributed Trip Information

A detection system and related algorithms of distributed trip behaviors are developed to identify charac-teristics of trips in urban rail transit systems using Wi-Fi information collected from passengers.The unique Media Access Control(MAC)addresses of mobile devices carried by passengers can be detected by detection devices installed in stations and be uploaded to a data center.The characteristics of route choices and travel time can be identified by comparing time stamps and related station IDs of selected mobile devices in the data center.T ransfer time can also be obtained by consid-ering operation time of the urban rail transit system,w hich can be utilized as a constraint of route identification.A case study of the rail transit system in Xi′an with actual information from the AFC data indicates that the proposed system can identify route choices and travel time for all kinds of networks with a sampling rate of 32.86% and an error rate of 3.8%. Analyzing the results of this system can be utilized for precise management,such as ticket clearing and station design.