Clustering Analysis of Ridership Patterns at Subway Stations: A Case in Nanjing, China

Abstract

Better understanding of urban mass transit trip mobility patterns will be helpful to increase public transit ridership and improve transit services of large cities. Therefore, a station-oriented clustering analysis on ridership patterns in subway systems based on smart card data was performed in this paper. Using the automatic fare collection (AFC) data of 89 subway stations in Nanjing, China, a similarity-based k-medoids clustering analysis approach was proposed and compared with previous studies. Then the correlation analysis between clustering results of subway stations and surrounding land uses including office and factory, residential area, scenic, university, shopping centers and entertainment venues, hospitals, and a long-distance passenger transport hub was achieved. Additionally, the station ridership on Sundays was analyzed separately to show the relationship of obvious peaks with different types of land use. The results of this research could contribute to subway station ridership forecasting and provide theoretical basis for schedule making and adjustment.