交通运输系统工程与信息

交通运输系统工程与信息 ›› 2016, Vol. 16 ›› Issue (5): 233-240.

• 案例分析 • 上一篇    

基于售检票数据挖掘的轨道交通乘客居住区辨识

姚向明*，赵鹏，韩宝明，邹庆茹   

  1. 北京交通大学交通运输学院，北京100044
  • 收稿日期:2016-02-23 修回日期:2016-04-15 出版日期:2016-10-25 发布日期:2016-10-25
  • 作者简介:姚向明（1987-），男，湖北宜昌人，博士后.
  • 基金资助:

    国家自然科学基金/National Natural Science Foundation of China(51478036); 中央高校基本科研业务费专项资金资助/ Fundamental Research Funds for the Central Universities(T16JB00260)；中国博士后科学基金/China Postdoctoral Science Foundation(2016M591062).

Home District Identification for Urban Rail Transit Travelers by Mining Automatic Fare Collection Data

YAO Xiang-ming，ZHAO Peng，HAN Bao-ming，ZHOU Qing-ru   

  1. School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China
  • Received:2016-02-23 Revised:2016-04-15 Online:2016-10-25 Published:2016-10-25

PDF

被引次数

4

摘要：

以自动售检票数据潜在包含的时空关联关系为基础，构建城市轨道交通乘客居住区辨识方法.假设轨道交通大部分乘客出行满足：一日内首次出行起始站与末次出行终点站相同，首次出行起始站与前日末次出行终点站相同，连续时期内首次出行起始站与末次出行终点站总是紧邻“家”的位置，以此为基础构建居住区辨识中心点法.以北京市轨道交通为对象进行实证分析，通过连续一周自动售检票数据挖掘能对88.7%的公交卡(不包括单程票、员工卡)所对应乘客的居住区进行辨识，验证了本文方法的准确性与有效性.本文研究提高了售检票数据应用价值，为乘客出行行为及需求特征分析提供了方法支持.

关键词: 城市交通, 居住区辨识, 数据挖掘, 城市轨道交通, 自动售检票数据

Abstract:

A methodology to identify home location for urban rail transit travelers is proposed in this paper, based on the spatial- temporal relationship within automatic fare collection (AFC) data. It is assumed that: most of travelers end their last trip at the start of their first trip of the day, and most travelers start their first trip at the end of their last trip of the day before, and travelers always start the first trip or end their last trip near their home location, based on these three assumptions a center point estimation algorithm is constructed. Finally, a practical rail transit network from Beijing in China is used to verify the effectiveness and accuracy of the proposed method, and result shows that 88.7 percent of passengers’home district can be identified from one week’s AFC records. This study improve the value of AFC data and provide a new method for analyzing the travel behavior and traffic demand characteristics.

Key words: urban traffic, home district identification, data mining, urban rail transit, automatic fare collection data

中图分类号: