Modelling car dependence: Pathways to sustainable urban mobility

This study investigates car dependence for short trips (travel time ≤15 min) using an integrated Partial Least Squares-Structural Equation Modelling (PLS-SEM) and Artificial Neural Network (ANN) approach applied to Dutch travel diary data (ODiN 2022). The observed outcome of the present study demonstrates that short trips are associated with lower car dependence (car ownership and frequency of car use), particularly for non-work destinations (e.g., educational institutions, hospitals, supermarkets). However, employed individuals exhibit higher car dependence, likely due to workplace commuting demands. Car ownership emerges as the critical mediator, shaping travel patterns for age, employment, household size, income, education, and residential location. While PLS-SEM identifies employment status as the most influential factor affecting frequency of car use, ANN prioritizes car ownership, suggesting unmeasured contextual factors (e.g., urban form, travel habits) influence these relationships in ways linear models cannot fully capture. The study makes three key contributions: (1) it quantifies car-reduction potential for short trips through advanced hybrid modelling, (2) establishes car ownership as a pivotal leverage point for sustainable mobility policies, and (3) identifies the importance of contextual factors in shaping short-trip travel behaviour. The research findings emphasise the need to locate both essential amenities and workplaces in shorter distance to residences to achieve sustainable urban mobility goals in direction of removing car dependence.