Methodological Framework for Understanding Urban People Flow from a Complex Network Perspective

Cities are connected by their functions and so are the spaces within cities. With the emergence of multisource data containing abundant urban residents' mobile information and the introduction of complex network methods to urban space research, a large amount of research has focused on exploring spatial structures from spatial interaction networks. Previous studies have limitations of lacking an understanding of the urban spatial structure from a systematic perspective and have not addressed the application value of research results from the perspective of urban planning. To address this research gaps, this paper proposed an analytical framework for a multiscale urban spatial characteristics analysis that is sufficiently intuitive for understanding both the spatial structure and organization relationship from a complex network perspective based on urban resident mobility data. This framework includes the full cycle workflow from data preprocessing, the construction of people flow network, method choice, and application. Unlike other works that focus on the community identified, this analytical framework considered the spatial characteristic within the communities and the relationship between communities and then compared the research results and planned urban structure. The goal of this research was to propose an analytical framework that can be applied for the identification of urban spatial structure, division of spatial management units, assessment of the implementation effects of spatial planning, and the configuration and optimization of auxiliary public service facilities. Two weeks of mobile phone signaling data that cover the Wuhu city center area were used in the empirical analysis. The results showed that (1) there is significant spatial heterogeneity between the compactness and intensity of the people flow network, (2) the community that was identified from the people flow network is more dispersed and microscopic in space compared with the communities from master plan; and finally, (3) at the community scale, there are fewer communities in the urban north area than in other areas and the interaction between communities in the urban north area is much looser and weaker in intensity. These findings may help in understanding the organizational relationship of urban groups from a systematic perspective and provide useful information to the division of planning groups and show the effects of assessment planning implementation to the planer.