Clustering of Passenger Flow and Land-Use of Beijing Urban Rail Transit Stations Based on Multi-Source Data

Abstract

With the deepening of transit-oriented development (TOD), the construction of high-density, multi-functional urban spatial structures around public transportation stations has become a significant focus in urban development planning. This study is aimed to explore the spatial and temporal cluster patterns and the relationship between taxi passenger flow and land use types around urban rail transit stations. Taking Beijing as an example, this study extracts the time series of pick-up and drop-off points from the taxi GPS track dataset and employs the affinity propagation (AP) method to conduct the spatial and temporal cluster analysis on these taxi pick-up and drop-off points. Then, based on the urban POI dataset, this study classifies these stations into four functional types by adopting the K-means method. Finally, the partial least squares (PLS) method has been used to explore the relationship between the taxi passenger flow and the land use types under different time intervals. The result indicates that: first, there is a regional aggregation effect for the taxi passenger source in the rail transit station area, and the cluster areas are mainly located in Chaoyang and Haidian districts, as well as along the circular metro line. Besides, there is a significant rush hour in the morning and evening for this passenger flow. Second, the commercial, residential, and comprehensive transit stations are mainly located in the central urban districts of Beijing with balanced spatial distribution, while the industrial transit stations are dispersed to the outer suburbs of the city, which conform to Alonso's spatial distribution of land. Third, there is a significant positive correlation between taxi passenger sources in the rail transit station area and land use types containing transportation service, public facility and daily life service. This study analyses the population mobility and land type cluster form around the urban rail transit from the micro-perspective, which verifies and supplements Alonso's transportation location theory. Therefore, this study contributes to the rational planning of urban spatial resource allocation and the construction of the urban rail transit network.

Keywords

• clustering algorithm
• land-use
• passenger flow
• rail transit station