摘要
基于地铁刷卡数据,采用K-means时间序列聚类算法对北京市40个换乘站及157个中间站和首末站的进出站客流进行画像分析及客流风险识别。考虑地铁车站在轨道交通运营网络中的位置属性,分别提取高峰时段的大客流特征和换乘客流的时空分布特征,将地铁换乘站分为4类,中间站分为6类,首末站分为2类。聚类结果表明:各类型车站的进出站客流和换乘客流在高峰时段的时间不均衡性较为突出,考虑不同周边用地性质,地铁客流的空间不均衡性较为显著。通过辨识不同地铁车站类型及其对应的客流时空特性规律,提出基于特征值数据的客流风险等级评价方法,分析了典型的换乘站客流风险状态。结果表明:不同类型的换乘站客流风险时段有所区别、客流风险等级不同,其中B类型换乘车站客流风险最高。本研究能为地铁客流精细化诱导和地铁运营风险防控提供支持。
This paper uses the K-means time series clustering method to analyze 40 interchange stations and 157 intermediate stations and terminal stations in Beijing based on the data of subway incoming and outgoing swipe card passenger flow and to identify the passenger flow risk.Considering the location attributes of metro stations in the rail transit operation network,the characteristics of passenger flow during peak hours and the spatial and temporal distribution characteristics of interchange passenger flow are extracted,respectively.The stations are divided into 4 categories for interchange stations,6 categories for intermediate stations,and 2 categories for terminal stations.The clustering results show that the time imbalance of inbound and outbound passenger flow and transfer passenger flow of each type of station is more prominent in the morning and evening peaks,and the spatial imbalance of passenger flow is also more significant considering different surrounding land properties.By identifying different metro station types and the corresponding passenger flow risk states,the passenger flow risk level evaluation method based on characteristic values is proposed,and the passenger flow risk degree of interchange stations is measured.Results show that different types of interchange stations have different passenger flow risk hours and different passenger flow risk levels,among which the passenger flow risk is the highest for the type B interchange stations.This study can provide support for subway passenger flow refinement induction and subway operation risk prevention and control.
作者
刘路
郑浩龙
车宇禄
朱宇婷
光志瑞
LIU Lu;ZHENG Haolong;CHE Yulu;ZHU Yuting;GUANG Zhirui(College of Architectural Science and Engineering,Yangzhou University,Jiangsu Yangzhou 225127,China;Business School,Beijing Technology and Business University,Beijing 100048,China;Technical Innovation Research Institute of Beijing Mass Transit Railway Operation Co.,Ltd.,Beijing 100039,China)
出处
《交通工程》
2024年第7期93-99,共7页
Journal of Transportation Engineering
基金
江苏省知识管理与智能服务工程研究中心2022年度开放课题(KMIS202206)
扬州大学中国大运河研究院2023年度开放课题(DYH202305)
扬州市-扬州大学市校合作科技创新平台建设项目(YZ2020269)。
关键词
地铁车站
客流特征
画像分析
风险识别
metro station
passenger flow characteristics
portrait analysis
risk identification
