The nonlinear effects of the subjective and objective built environment on active travel behavior: A case study of Shanghai, China

Active travel refers to modes of transportation powered by human effort without external assistance, such as walking and cycling. This form of travel not only enhances the health of urban residents but also helps alleviate urban congestion. While the relationship between the built environment and active travel has been widely studied, the nonlinear relationship between the subjective and objective aspects of the built environment and active travel remains unclear. In this study, we employ eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanation (SHAP) models to investigate these nonlinear and threshold effects and identify the determinants of active travel choices, using Shanghai as a case study. The results show that the objective built environment influences active travel choices significantly more than subjective perceptions. Among the objective factors, road network density, population density, and distance to city center are the top contributors. In contrast, in terms of the subjective built environment, only perceptions of traffic violations and social security make notable contributions. Furthermore, road network density, population density, and distance to city center exhibit significant threshold effects on active travel choices. Additionally, by analyzing cases with a mismatch between subjective environmental perceptions and active travel choices, we find that commuting distance and annual household income are key determinants of active travel, with road network density, population density, and distance to city center also playing crucial roles. These findings offer valuable policy implications for developing sustainable, transportation-oriented built environment plans in Shanghai.