PROGRESS IN GEOGRAPHY ›› 2016, Vol. 35 ›› Issue (1): 126-134.doi: 10.18306/dlkxjz.2016.01.014

• Orginal Article • Previous Articles    

Classification of subway stations in Beijing based on passenger flow characteristics

Qin YIN1(), Bin MENG2(), Liying ZHANG3   

  1. 1. College of Environment and Planning, Capital Normal University, Beijing 100048, China
    2. College of Arts and Science of Beijing Union University, Beijing 100191, China
    3. College of Geophysics and Information Engineering, China University of Petroleum, Beijing 102249, China
  • Online:2016-01-31 Published:2016-01-31
  • Supported by:
    National Natural Science Foundation of China, No.41171136;Beijing Philosophy and Social Science Foundation Grant, No.14CSA002;The Importation and Development of High-Caliber Talents Project of Beijing Municipal Institutions, No.IDHT20130322


In cities there is a high concentration of social and economic activities at subway stations. Different types of subway stations have different functions, as reflected in their regional characteristics, traffic function, and land use. Meaningful station classification helps to understand urban functional partition and evaluate rail transportation infrastructure development. Based on the usage data of subway passes, this study classified the subway stations with time series clustering. The result shows that (1) subway station passenger flows have clear temporal and spatial differences. It reflects the temporal and spatial differences of urban functional partitions; (2) this study uses a time series clustering method. By considering passenger flow characteristics, subway stations can be divided into residential-oriented stations, employment-oriented stations, spatial mismatched stations, mixed mainly residential-oriented stations, mixed mainly employment-oriented stations, mixed type stations, commerce- and attraction-oriented stations, and other stations; (3) Using traffic data at subway stations is an effective way to compare spatial behavior and physical space.

Key words: passenger flow characteristics, time series clustering, subway stations, Beijing