Evaluating the influence of environmental factors and route characteristics on leisure-oriented active travel: A case study in Skåne Province

Abstract

Leisure-oriented active travel, such as walking and running, offers significant health and environmental benefits. It promotes physical activity and reduces reliance on motorized transport. Despite its importance, many studies focus primarily on urban areas, often neglecting suburban and natural regions. Additionally, these studies frequently include an excess of unfiltered features, leading to overly complex models that are difficult to apply in practice. To address these challenges, this study adopts a framework that integrates neighborhood rough set theory with methods such as Generalized Linear Models, Generalized Additive Models, Random Forest, and SHapley Additive exPlanations. By utilizing crowdsourced Komoot data and environmental datasets, the study identifies the key factors influencing the popularity of leisure-oriented active travel routes. The Hiking Index emerges as the most significant factor, followed by road surface types, surrounding environments, night light levels, and paving conditions. By accounting for both linear and nonlinear effects, the framework enhances the interpretability of the results. These findings offer practical insights for sustainable leisure-oriented active travel planning, supporting public health objectives, encouraging environmentally friendly travel, and providing valuable guidance for regional planning and policy development.