Spatial allocation of heavy commercial vehicles parking areas through geo-fencing

Inadequate parking planning for heavy commercial vehicles (HCV) exacerbates urban road congestion. As an effective means of parking management, geofencing that identifies the virtual boundary for geographic areas is essential to ensure these vehicles do not impede traffic and urban spaces. However, geofenced areas must be rationally designed to prevent mismatches between parking areas and real parking needs. This paper presents a data-driven approach that integrates the Spatial-temporal Density-Based Spatial Clustering of Applications with Noise (ST-DBSCAN) methods and a Gaussian mixture model for identifying and predicting potential parking areas for HCVs. Leveraging the HCV trajectory data and land use data in Shanghai, China, we characterize the spatial distribution of parking demand and create a probabilistic model to predict active HCV traffic patterns and the spatial confidence regions under varying land use conditions. The results show that clusters of HCV parking demand tend to congregate near ports, comprehensive transportation hubs, logistics centers, and commercial hubs. These clusters correspond to five distinct parking demand patterns (i.e., day-long HCV stops, morning peak time HCV stops, daytime HCV stops, afternoon peak time HCV stops, and nighttime HCV stops), each reflecting specific spatiotemporal characteristics. The geofenced spatial domain was found to be very sensitive to the timing of parking, emphasizing the importance of using advanced geofencing technologies. The methodological framework introduced in this study holds significant value for policymakers and HCV operators as it aids in determining parking at strategic levels, offering valuable insights and tools to enhance the effectiveness of parking management.