交通运输系统工程与信息

交通运输系统工程与信息 ›› 2025, Vol. 25 ›› Issue (6): 229-238.DOI: 10.16097/j.cnki.1009-6744.2025.06.021

• 系统工程理论与方法 • 上一篇    下一篇

建成环境对不同群体公交出行的非线性影响研究

刘路1 ，郑浩龙1 ，李明高2 ，朱宇婷*3 ，吴珂琪4   

  1. 1. 扬州大学，土木与交通学院，江苏扬州225127；2.珠海市规划设计研究院，广东珠海519000； 3. 北京工商大学，商学院，北京100048；4.北京市交通委员会政务服务中心（北京市船舶检验所），北京100161
  • 收稿日期:2025-07-11 修回日期:2025-09-22 接受日期:2025-09-23 出版日期:2025-12-25 发布日期:2025-12-24
  • 作者简介:刘路(1991—)，女，安徽桐城人，讲师，博士。
  • 基金资助:
    江苏高校哲学社会科学研究一般项目(2025SJYB1543)；北京市教育委员会社科计划一般项目(SM202410011001)。

Nonlinear Effects of Built Environment on Bus Usage of Different Traveler Groups

LIU Lu1, ZHENG Haolong1, LI Minggao2, ZHU Yuting*3, WU Keqi4   

  1. 1. College of Civil Engineering and Transportation, Yangzhou University, Yangzhou 225127, Jiangsu, China; 2. Zhuhai Institute of Urban Planning & Design, Zhuhai 519000, Guangdong, China; 3. Business School, Beijing Technology and Business University, Beijing 100048, China; 4. Government Service Center of Beijing Municipal Transport Commission (Beijing Boats Inspection Center), Beijing 100161, China
  • Received:2025-07-11 Revised:2025-09-22 Accepted:2025-09-23 Online:2025-12-25 Published:2025-12-24
  • Supported by:
    Jiangsu University Philosophy and Social Sciences Research Project (2025SJYB1543)；R&D Program of Beijing Municipal Education Commission (SM202410011001)。

PDF

摘要： 为解析通学、通勤及老年群体高峰时段公交出行的生成机理，并在同一时空维度下，剖析建成环境对不同群体的异质性作用，本文引入随机森林模型，量化关键因素贡献度，识别阈值和边际效应，分析多因素交互作用，揭示建成环境对不同群体公交出行的非线性影响机制。研究发现：从影响因素的个体重要性看，公交线路数量对3类群体均起主导作用，但次要影响因素呈显著群体差异。其中，通勤群体高度依赖公交站密度与距市中心距离，老年群体受医疗保健与科教文化设施双重约束，通学群体主要受距市中心距离的影响。从影响因素的集体重要性看，通学与老年群体出行由土地利用因素单极主导，通勤群体由土地利用与公交站属性因素协同驱动，城市设计因素重要性相对较低且稳定。从非线性影响看，同一因素对不同群体的阈值与边际效应存在显著差异，且多因素交互作用也存在差异；此外，同一因素对通学和老年群体的影响存在明显时序分异。

关键词: 城市交通, 非线性影响, 随机森林, 公交出行, 建成环境

Abstract: This paper analyzes the generation mechanism of peak-hour public transport trips for school commuters, work commuters, and elderly travelers, and to examine the heterogeneous effects of the built environment on different groups within the same spatiotemporal dimension. A random forest model is developed to quantify the contribution of key factors, identify thresholds and marginal effects, and analyze multi-factor interaction effects. The proposed model aims to reveal the non-linear influence mechanism of the built environment on public transport trips among different groups. The findings are as follows: from the perspective of the individual importance of influencing factors, the number of public transport routes plays a dominant role for all three groups, while there are significant group differences in secondary influencing factors. The work commuters are highly dependent on the density of bus stops and distance to the city center. The elderly travelers are doubly constrained by medical and health care facilities as well as science, education, and cultural facilities. The school commuters are mainly affected by the distance to the city center. From the perspective of the collective importance of influencing factors, school commuters and elderly travelers are unilaterally dominated by land use factors, work commuters are co-driven by land use factors and bus stop attribute factors, and the importance of urban design factors is relatively low and stable. From the perspective of non-linear influence, the thresholds and marginal effects of the same factor on different groups are significantly different, and there are also differences in multi-factor interaction effects. In addition, the impact of the same factor on school commuters and elderly travelers shows obvious temporal differentiation.

Key words: urban traffic, nonlinear effect, random forest, bus travel, built environment

中图分类号: