Exploring the Nonlinear and Threshold Effects of Travel Distance on the Travel Mode Choice across Different Groups: An Empirical Study of Guiyang, China

Examining how travel distance is associated with travel mode choice is essential for understanding traveler travel patterns and the potential mechanisms of behavioral changes. Although existing studies have explored the effect of travel distance on travel mode choice, most overlook their non-linear relationship and the heterogeneity between groups. In this study, the correlation between travel distance and travel mode choice is explored by applying the random forest model based on resident travel survey data in Guiyang, China. The results show that travel distance is far more important than other determinants for understanding the mechanism of travel mode choice. Travel distance contributes to 42.28% of explanation power for predicting travel mode choice and even 63.24% for walking. Significant nonlinear associations and threshold effects are found between travel distance and travel mode choice, and such nonlinear associations vary significantly across different socioeconomic groups. Policymakers are recommended to understand the group heterogeneity of travel mode choice behavior and to make targeted interventions for different groups with different travel distances. These results can provide beneficial guidance for optimizing the spatial layout of transportation infrastructure and improving the operational efficiency of low-carbon transportation systems.