交通运输系统工程与信息 ›› 2018, Vol. 18 ›› Issue (1): 108-114.

• 系统工程理论与方法 • 上一篇    下一篇

基于混合Logit 模型的网约车选择行为研究

唐立1,邹彤2,罗霞* 3,陈思为3   

  1. 1. 西华大学 汽车与交通学院,成都 610039;2. 上海市政工程设计研究总院(集团)有限公司 成都分公司, 成都 610000;3. 西南交通大学 交通运输与物流学院,成都 610031
  • 收稿日期:2017-08-15 修回日期:2017-10-29 出版日期:2018-02-25 发布日期:2018-02-26
  • 作者简介:唐立(1988-),女,四川遂宁人,讲师,博士.
  • 基金资助:

    四川省科技厅项目/Sichuan Provincial Science and Technology Program(17RKX0108);成都市科技局项目/ Chengdu Science and Technology Project(2015-RK00-00227-ZF).

Choice Behavior of Taxi-hailing Based on Mixed-Logit Model

TANG Li1, ZOU Tong2, LUO Xia3, CHEN Si-wei3   

  1. 1. School of Automobile and Transportation, Xihua University, Chengdu 610039, China; 2. Shanghai Municipal Engineering Design Institute (Group) Co., Ltd., Chengdu 610000, China; 3. School of Transportation and Logistics, Southwest Jiaotong University, Chengdu 610031, China
  • Received:2017-08-15 Revised:2017-10-29 Online:2018-02-25 Published:2018-02-26

摘要:

近年来,网约车发展迅速,而现有研究较少从市场角度理解用户对网约车的选择偏 好.本文构建了基于混合Logit选择模型,选取网约车依赖性、出行目的等出行相关属性,年龄等 个人社会经济属性为效用函数特征变量,假设到达时间、在车时间和费用为随机变量并服从对 数正态分布.使用D-efficient效率设计法生成问卷,在成都市开展网约车使用特征与选择偏好的 实证调查.基于观察数据标定模型,参数显著性及符号解释了城市综合交通背景下出行方式选 择的关键因素.对出租车和网约车费用进行边际效应分析,明确了价格变化对出行结构的影响.

关键词: 城市交通, 出行方式选择, 混合Logit模型, 网约车, 出行行为分析

Abstract:

In recent years, the development of taxi- hailing is rapid, but how to understand users’choice preference on taxi-hailing from the market perspective is merely seen in the current research. A mixed-Logit based choice model is built in this paper. Dependence on taxi- hailing, travel related attributes like trip purpose, individuals’social- demographic characteristics like age are chosen as the explanatory variables for the utility function. Arrival time, travel time and cost are assumed as the random variables and log- normally distributed. Generating the questionnaire using D-efficient method, an empirical survey for observing taxi-hailing use features and choice preferences was conducted in Chengdu, China. Calibrating the model based on the observed data, the significance and sign of the parameters reveal key factors of travel mode choice under the integrated urban traffic background. The marginal effect analysis on cost of conventional taxi and taxi-hailing explains the impact of price change on mode share.

Key words: urban traffic, travel mode choice, mixed-Logit model, taxi-hailing, travel behavior analysis

中图分类号: