Discovering the relationship between travel behavior and land use: A case study of Beijing, China

The interaction between travel behavior and land use has drawn much attention for decades. Instead of satellite images, this paper aims to establish a link between human mobility and urban structure using massive taxi data and POI data based on real world. First, the 24-dimensional drop-off/pick-up temporal difference vector (DPTDV) is proposed and used to identify travel patterns. Then, the study area is classified into different clusters according to the hourly variation of the difference between drop-offs and pick-ups. Further, twelve POI features and two indicators are extracted to explore the land use characteristics corresponding to each specific travel pattern. Meanwhile, the relationship between travel activity and land use intensity is linearly fitted. The results from this case study will allow us to have a deeper understanding of urban dynamics, help alleviate traffic congestion and provide guidance for smarter urban construction.