Scholarlink

This chapter focuses on the Shuangliu District, a typical urban-rural fringe in Chengdu, to investigate children’s travel characteristics. It employs the Gradient Boosting Decision Tree (GBDT) model with SHAP values to analyze the impact of various indicators on children’s travel distances. The results indicate that: (1) Travel destinations exert the strongest influence on travel distance, followed by five built environment variables, while the socioeconomic attributes of children’s families have minimal impact. (2) The built environment exhibits significant nonlinear effects on travel distances, particularly population density. When population density exceeds 8800 people/km2, traffic congestion caused by over-concentration leads to increased travel distances for children. (3) The built environment not only independently influences travel distances through nonlinear mechanisms but also generates interactive effects among different indicators. These findings provide decision-making support for creating child-friendly travel environments.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The Nonlinear Influence of Neighborhood-Built Environment on Children’s Travel Distance

  • 章节

  • 发表日期：

    2026年

查看全文
  • Yibin Ao,
  • Yi Long,
  • Homa Bahmani

摘要

This chapter focuses on the Shuangliu District, a typical urban-rural fringe in Chengdu, to investigate children’s travel characteristics. It employs the Gradient Boosting Decision Tree (GBDT) model with SHAP values to analyze the impact of various indicators on children’s travel distances. The results indicate that: (1) Travel destinations exert the strongest influence on travel distance, followed by five built environment variables, while the socioeconomic attributes of children’s families have minimal impact. (2) The built environment exhibits significant nonlinear effects on travel distances, particularly population density. When population density exceeds 8800 people/km2, traffic congestion caused by over-concentration leads to increased travel distances for children. (3) The built environment not only independently influences travel distances through nonlinear mechanisms but also generates interactive effects among different indicators. These findings provide decision-making support for creating child-friendly travel environments.