交通运输系统工程与信息

交通运输系统工程与信息 ›› 2017, Vol. 17 ›› Issue (3): 67-73.

• 智能交通系统与信息技术 • 上一篇    下一篇

基于个体出行数据的公共交通出行链提取方法

翁剑成*1，王昌1 ，王月玥2，陈智宏3，彭曙4   

  1. 1. 北京工业大学交通工程北京市重点实验室，北京100124；2. 北京市轨道交通指挥中心，北京100101； 3. 交通运输部路网监测与应急处置中心，北京100088；4.井冈山大学机电工程学院，江西吉安343009
  • 收稿日期:2016-09-09 修回日期:2016-12-13 出版日期:2017-06-25 发布日期:2017-06-26
  • 作者简介:翁剑成(1981-)，男，浙江人，副教授，博士.
  • 基金资助:

    国家自然科学基金/National Natural Science Foundation of China(51108013)；交通运输部建设科技项目/ Ministry of Transport of the People's Republic of China(2015318221020)；北京市“科技新星”计划项目/“Beijing Nova”Program by Beijing Municipal Science and Technology Commission(Z171100001117100).

Extraction Method of Public Transit Trip Chains Based on the Individual Riders’Data

WENG Jian-cheng 1,WANG Chang 1,WANG Yue-yue 2, CHEN Zhi-hong 3, PENG Shu 4   

  1. 1. Beijing Key Laboratory of Traffic Engineering, Beijing University of Technology, Beijing 100124, China; 2. Beijing Metro Network Control Center, Beijing 100101, China；3. Highway Monitoring & Response Center, Ministry of Transport of P.R.C., Beijing 100088, China; 4. College of Mechanical & Electrical Engineering, Jinggangshan University, Ji’an 343009, Jiangxi, China
  • Received:2016-09-09 Revised:2016-12-13 Online:2017-06-25 Published:2017-06-26

PDF

摘要：

公共交通个体出行信息的提取对掌握公共交通出行的时空特征，改善居民通勤出行效率具有重要意义.研究从公交刷卡数据、公交定位数据、轨道AFC数据等海量公共交通多源数据的关联匹配与处理方法入手，提出了公共交通出行链信息提取中，换乘关系判断、通勤行为判别及出行起讫点匹配的方法与规则，标定了出行链匹配阈值参数，建立了基于个体出行数据的公共交通通勤出行链提取模型.提取模型的准确度验证表明：出行链结构提取及通勤出行判别的成功率均达到100%，出行阶段起讫点匹配成功率为 87.5%，准确性为97.1%，满足了公共交通出行特征提取的需求.该方法为公共交通通勤出行判别及基于个体的微观通勤出行时空特征的深入分析奠定了基础.

关键词: 城市交通, 出行链, 公交多源数据, 通勤出行, 个体出行感知, 出行特征

Abstract:

The extraction of public transit passengers’travel information has great significance to master the time-space characteristics of public transit travel and to improve the efficiency of residents' commuting. Through matching and processing the multi-source public transport data which derived from bus smart card data, bus location data and subway AFC system data, this paper mainly studies the methods and rules of the transfer relationship judgment, commuting travel identification and trip starting point matching, which are essential steps for the extraction of public transportation trip chain information. The thresholds for trip chain matching and connecting is also calibrated, and the public transit commuting chain extraction model is established based on the individual riders’travel data. The results of the extract model validation show that the success rate of trip chain structure extraction is, and commuting travel identification reach 100%, and the success rate of origin stop and destination stop of passengers’trip is 87.5% and the accuracy is up to 97.1% The study provide essential foundation for the public transport commuter travel identification and the public transit trip chains time-space features analysis based on the individual travelers’ridership data.

Key words: urban traffic, trip chain, bus multi-source data, commuter travel, individual travel perception, travel characteristics

中图分类号: