交通运输系统工程与信息

交通运输系统工程与信息 ›› 2023, Vol. 23 ›› Issue (6): 296-306.DOI: 10.16097/j.cnki.1009-6744.2023.06.029

• 工程应用与案例分析 • 上一篇    下一篇

基于手机信令数据的城市区域居民出行OD预测模型

胡宝雨*，刘学   

  1. 东北林业大学，土木与交通学院，哈尔滨 150040
  • 收稿日期:2023-07-21 修回日期:2023-09-04 接受日期:2023-09-14 出版日期:2023-12-25 发布日期:2023-12-23
  • 作者简介:胡宝雨(1987- )，男，黑龙江哈尔滨人，副教授，博士。
  • 基金资助:
    教育部人文社会科学研究青年基金 (19YJCZH052)

Prediction Model for Residents Travelling OD in Urban Areas Based on Mobile Phone Signaling Data

HU Bao-yu*,LIU Xue   

  1. School of Civil Engineering and Transportation, Northeast Forestry University, Harbin 150040, China
  • Received:2023-07-21 Revised:2023-09-04 Accepted:2023-09-14 Online:2023-12-25 Published:2023-12-23
  • Supported by:
     Youth Fund Project of Humanities and Social Sciences Research of the Ministry of Education (19YJCZH052)

PDF

PDF(English version)

27

摘要： 为揭示城市区域居民出行规律和OD产生原理，基于手机信令数据探索城市区域居民出行目的地选择机制，考虑人口数量和POI数量两类数据建立一个地点机会选择(POS)模型。利用联通智慧足迹平台获取得到哈尔滨市居民手机信令出行数据，从交通小区和交通中区两个尺度，对哈尔滨市市二环、三环、四环不同范围展开实例研究。结果表明：地点机会选择模型预测结果在吸引能力和出行距离分布中基本符合实际数据规律，从交通小区和中区两个尺度上看，该模型预测精度分别能够达到 67%~72%、75%~83%，较机会优先选择模型预测精度分别提高了 13%~18%、9%~20%，较辐射模型预测精度分别提高了57%~60%、55%~60%。本文所提出模型的优越性在于模型简单且无待定参数，输入数据较容易获取，具有较高的预测精度，能够为城市交通规划提供理论参考。

关键词: 城市交通, OD预测, 地点机会模型, 手机信令数据, 居民出行规律

Abstract: To reveal the travel pattern and OD generation principle of urban area residents, the destination selection mechanism of urban area residents is explored based on mobile phone signaling data. The position opportunity selection (POS) model was developed by considering the population and the number of POI. The mobile phone signaling travel data of Harbin residents, obtained from the Unicom Smart Footprint Platform, is used to validate the model. The analysis is conducted at both the traffic cell and traffic mid-zone levels, focusing on Harbin's second, third, and fourth ring roads. The results show that the POS model predictions were generally consistent with the actual data patterns in the traffic attraction capacity and travel distance distributions. At both the traffic cells and mid-zones scales, the model achieves a prediction accuracy of 67% ~72% and 75% ~83% , respectively. These results indicate an improvement of 13%~18% and 9%~20% over the opportunity priority selection model and a superiority of 57%~60% and 55%~60% over the radiation model, respectively. The advantage of the proposed model lies in its simplicity and lack of parameters. The input data are easily obtainable, and the model offers high prediction accuracy. The findings provide a theoretical reference for urban traffic planning.

Key words: urban traffic, OD prediction, position opportunity selection model, mobile phone signaling data, resident travel pattern

中图分类号: