中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2021, Vol. 40 ›› Issue (01): 29-35.DOI: 10.3969/j.issn.1674-0696.2021.01.06

• 交通+大数据人工智能 • 上一篇    下一篇

基于二元Logit模型的驾驶员收费车道选择研究

杨少伟1,2,梁汉冕1,潘兵宏1,2   

  1. (1. 长安大学 公路学院,陕西 西安 710064; 2. 长安大学 特殊地区公路工程教育部重点实验室,陕西 西安 710064)
  • 收稿日期:2018-12-17 修回日期:2019-09-10 出版日期:2021-01-11 发布日期:2021-01-11
  • 作者简介:杨少伟(1956—),男,陕西延安人,教授,博士,主要从事交通运输、路线总体工程方面的研究。E-mail:g106@chd.edu.com 通信作者:梁汉冕(1994—),男,广西钦州人,硕士研究生,主要从事交通运输、路线总体工程方面的研究。E-mail:gsoshou@qq.com
  • 基金资助:
    中交科技基金项目〔(2018)05〕

Drivers Toll Lane Selection Based on Binary Logit Model

YANG Shaowei1,2, LIANG Hanmian1, PAN Binghong1,2   

  1. (1. Highway College, Changan University, Xian 710064, Shaanxi, China; 2. Key Laboratory for Special Area Highway Engineering of Ministry of Education, Changan University, Xian 710064, Shaanxi, China)
  • Received:2018-12-17 Revised:2019-09-10 Online:2021-01-11 Published:2021-01-11
  • Supported by:
     

摘要: 通过分析驾驶员对收费站收费车道选择行为的规律特征,讨论了驾驶员收费车道选择行为的复杂性,确定模型的效用函数及概率计算公式。选取收费车道的排队长度、变换车道数、大型车数量、收费车道位置作为模型的特征变量,建立了驾驶员对收费车道选择行为的二元Logit模型。基于兰海高速钦州西收费站的实际观测数据,对各特征变量进行模型参数估计,采用Hosmer-Lemeshow检验分析模型的拟合优度。研究结果表明:驾驶员收费车道选择行为受多种因素综合作用影响,模型拟合精度较高;模型的建立为收费站交通组织、安全评价、收费车道布置提供了技术支持。

 

关键词: 交通工程, 收费车道选择行为, 二元Logit模型, 收费站, 特征变量

Abstract: By analyzing the regular characteristics of drivers behavior of toll lane selection, the complexity of drivers behavior of toll lane selection was discussed, and the utility function and probability calculation formula of the model were determined. Taking the queue length of the toll lane, the number of lanes changing, the number of large vehicles and the location of the toll lane as the characteristic variables of the model, a binary Logit model for drivers behavior of toll lane selection was established. Based on the actual observation data of the Qinzhou West toll station of Lanhai Expressway, the model parameters were estimated for each characteristic variable, and the goodness of fit of the model was analyzed by Hosmer-Lemeshow test. The research results show that the drivers behavior of toll lane selection is comprehensively affected by many factors, and the model fitting accuracy is high; the establishment of the proposed model provides technical support for traffic organization, safety evaluation and toll lane layout of toll station.

Key words: traffic engineering, behavior of toll lane selection, binary Logit model, toll station, characteristic variable

中图分类号: