Key points are not available for this paper at this time.
Transport data are important for understanding human mobility and urban interactions within a city. As China’s transportation infrastructure continues to grow, more research is needed to analyse the spatial patterns of travel flows and to understand how these patterns change over time. With the development of online car-hailing and ride sharing services, floating car data have become a new resource to facilitate the analysis of human mobility patterns and the interactions of urban mobility within a city. The detection of urban communities based on urban networks is a helpful way to represent urban interactions. However, understanding community changes using online car-hailing data remains an underexplored topic. To this end, this study applies a community detection method to explore community changes over time based on the newly available floating car data (DiDi Chuxing (‘DiDi’)) in Chengdu, China. We applied undirected graphs to examine the spatial distribution of DiDi usage and the spatial patterns of travel distance. In addition, we explored the spatial-temporal variations of the communities at the taxi zone level using Blondel’s iterative algorithm, a modularity optimization approach. Results suggest that: 1) taxi zones on the south and west sides of Chengdu have more average daily trips compared to those in other areas; 2) residential taxi zones in the northeast area have a long median travel distance, indicating people living in those areas travel longer distances; and 3) the detected community structures change at different times. These findings provide valuable information for urban planning and location-based services in Chengdu.
Yang et al. (Mon,) studied this question.