Assessing temporal–spatial characteristics of urban travel behaviors from multiday smart-card data

The rail transit has difficulties in meeting daily travel needs of passengers owing to a large population and accelerating urbanization. Analyzing urban travel behaviors with big data helps the design in infrastructures and the optimized personnel allocation. Furthermore, travel behaviors are characterized by dynamic at different time and locations, displaying the rule of urban traffic operation. This paper utilizes smart card data in two cities with different geographical features to analyze the temporal–spatial characteristics of urban travel behaviors. More specifically, by creating travel networks based on the pick-up and drop-off stations and the passenger population among these stations, an interesting observation is that the community structure of travel networks owns a metabolic trend and a stable feature simultaneously. The finding shows that the traffic system can be managed in several parts. Moreover, similar mobility patterns exist in some stations, which can be organized and controlled in the same way. Finally, travel behaviors are related to the urban layout and structure, so the distribution of urban areas can be understood better. Experiments provide enlightening insights for policy makers to comprehend the urban travel behaviors, thus improving the rail transit service plans and scheduling strategies.