交通运输系统工程与信息

交通运输系统工程与信息 ›› 2014, Vol. 14 ›› Issue (4): 131-138.

• 系统工程理论与方法 • 上一篇    下一篇

基于有序Logistic 回归的城市人行道 服务水平研究

赵琳1，边扬1，荣建1，刘小明*1,2   

  1. 1.北京工业大学交通工程北京市重点实验室，北京10012；2.中华人民共和国交通运输部，北京100736
  • 收稿日期:2013-11-04 修回日期:2014-03-13 出版日期:2014-08-25 发布日期:2014-09-16
  • 作者简介:赵琳（1988-），女，黑龙江五常人，博士生.
  • 基金资助:

    国家自然科学基金（51108012）

Pedestrian LOS of Urban Sidewalks Based on Orderly Logistic Regression

ZHAO Lin1，BIAN Yang1，RONG Jian1，LIU Xiao-ming1,2   

  1. 1.Key Laboratory of Traffic Engineering, Beijing University of Technology, Beijing 100124, China； 2.Ministry of Transport of the People’s Republic of China, Beijing 100736, China
  • Received:2013-11-04 Revised:2014-03-13 Online:2014-08-25 Published:2014-09-16

PDF

被引次数

13

摘要：

科学的人行道服务水平评价方法可以为营造良好的步行出行环境提供有力的 理论支撑.针对人行道服务水平等级为定序变量的特点，本文提出了基于有序Logistic 回 归模型的人行道服务水平评价方法.首先，选择北京市交通枢纽地区34 条典型人行道进 行行人满意度调查，获取典型、全面的数据样本；然后，采用模糊C 均值聚类方法得到行 人满意度与人行道服务水平等级的对应关系；最后，应用逐步回归方法提取人行道服务 水平显著性影响因素并分析其影响机理，进而建立人行道服务水平等级的有序Logistic 回归模型.通过实际验证，模型精度较现有的线性回归模型有所提高.

关键词: 城市交通, 人行道服务水平, 有序Logistic 回归, 模糊C 均值聚类, 模糊隶属度

Abstract:

Scientific method measuring pedestrian level of service (LOS) on sidewalks can provide a powerful theoretical support for a comfort walking trip environment. Aiming at the characteristics that pedestrian LOS is a orderly discrete variable, this paper proposed an evaluation method for pedestrian LOS, based on the orderly Logistic regression model. First, pedestrians’satisfaction questionnaire survey was conducted on 34 representative sidewalks in Beijing transportation hub areas. Typical and comprehensive data were obtained. Then fuzzy C mean clustering was used to get the correspondence between pedestrians’satisfaction and the LOS. Finally, the significant influencing factors were extracted with step-regression method and their influencing mechanism was analyzed. And then the orderly Logistic regression model for pedestrian LOS was formulated. The field tast proved that the model has higher accuracy than the traditional linear model.

Key words: urban traffic, pedestrian level of service, orderly Logistic regression, fuzzy C mean clustering, fuzzy membership

中图分类号: