Analysis of passenger flow characteristics and their relationship with surrounding urban functional landscape pattern
Yaolin Liu,Mingyang Tang,Zihao Wu,Zhide Tu,Zihao An,Nannan Wang,Ying Li
DOI: https://doi.org/10.1111/tgis.12665
IF: 2.568
2020-01-01
Transactions in GIS
Abstract:The passenger flow characteristics of rail transit stations and their relationship with surrounding urban space play an important role in urban planning and transit-oriented development policies. However, passenger flow data are high-dimensional and complex, and corresponding characteristics are difficult to extract. Moreover, the integration of natural, social and economic factors in urban space may satisfactorily explain passenger flow characteristics. In this study, we clustered stations with passenger flow by using a dynamic time warping (DTW) distance-based affinity propagation (AP) algorithm. We introduced the urban functional landscape to reveal the urban space surrounding the station by landscape metrics. Then, we explored the relationship between passenger flow characteristics and urban functional landscape pattern. The DTW distance-based AP algorithm method identified passenger flow characteristics under total, inbound and outbound conditions, with cluster numbers of 3, 5 and 5, respectively. With the distance from the station, the proportion of natural landscape increased (from 26.11 to 36.62%), whereas that of socioeconomic landscape decreased (from 59.87 to 51.57%). The multivariate regression analysis indicated that types of urban functional landscape had significant influence on the passenger flow characteristics in different accessible areas of the station. These findings highlighted the effect of urban functional landscape pattern on passenger flow.