Impacts of transportation infrastructure on the stopping behavior of heavy commercial vehicles

Restricting heavy commercial vehicles (HCV) stopping behavior is essential for traffic management and efficient logistics organization. However, insights into HCV stopping behavior conditional on infrastructure remain limited. We utilize large GPS HCV trajectory data to investigate the spatial variability of stopping duration and its associated influential factors. Results show HCV stopping duration and pre-stop trip distance exhibit marked weekday-weekend heterogeneity, with longer stops and longer preceding trips being more likely to occur on weekdays. We further propose a spatial random forest model to predict HCV stopping duration. In addition to the superiority of the proposed model, we found parking distribution and land use collectively contribute to 76.76% of the predictive power, with trip distance identified as the most influential factor, followed by distance to the nearest freight rail station and to the nearest port, respectively. Furthermore, the analyses also exhibit complex non-linear associations between influential factors and stopping duration, and significant threshold effects. These findings offer valuable insights that can serve as a nuanced guide for planning HCV parking management.