Revealing Urban Spatial Interaction Characteristics and Crowd Travel Patterns from Trajectory Data

The accelerated urbanization process has raised higher demands for urban planning and management, and a precise understanding of urban spatial interaction characteristics is fundamental to these efforts. How to uncover the frequent interaction patterns between urban spaces, especially those that exhibit stability over time, and reveal the underlying semantic information, remains a pressing challenge, however. This study proposes a time-sliced multilayer network analysis method that combines crowd travel trajectories with urban land-use data to reveal the characteristics of urban spatial interaction and the travel patterns of the population. Empirical analysis of Wuhan’s main urban area shows that residential areas, educational and research zones, and parks and green spaces are involved in most frequent interaction patterns, with residential areas playing a central role in the distribution and changes of urban traffic flow. Furthermore, the mining of frequent interaction patterns reflects the complexity and interdependence between urban functional zones, particularly the high-frequency interactions among residential areas, educational and research lands, and parks and green spaces. This study reveals frequent interaction patterns between urban functional zones from trajectory data, providing a new perspective for precisely understanding urban dynamics, which is beneficial for the formulation of urban planning and management policies.