Multidimensional visualization of transit smartcard data using space–time plots and data cubes
- 716 Downloads
Given the wide application of automatic fare collection systems in transit systems across the globe, smartcard data with on- and/or off-boarding information has become a new source of data to understand passenger flow patterns. This paper uses Nanjing, China as a case study and examines the possibility of using the data cube technique in data mining to understand space–time travel patterns of Nanjing rail transit users. One month of smartcard data in October, 2013 was obtained from Nanjing rail transit system, with a total of over 22 million transaction records. We define the original data cube for the smartcard data based on four dimensions—Space, Date, Time, and User, design a hierarchy for each dimension, and use the total number of transactions as the quantitative measure. We develop modules using the programming language Python and share them as open-source on GitHub to enable peer production and advancement in the field. The visualizations of two-dimensional slices of the data cube show some interesting patterns such as different travel behaviors across user groups (e.g. students vs. elders), and irregular peak hours during National Holiday (October 1st–7th) compared to regular morning and afternoon peak hours during regular working weeks. Spatially, multidimensional visualizations show concentrations of various activity opportunities near metro rail stations and the changing popularities of rail stations through time accordingly. These findings support the feasibility and efficiency of the data cube technique as a mean of visual exploratory analysis for massive smart-card data, and can contribute to the evaluation and planning of public transit systems.
KeywordsSmartcard data Transit Travel behavior Exploratory data mining Space–time plot Data cube
This study was sponsored by the International Cooperation and Exchange of the National Natural Science Foundation of China (No. 51561135003) and the Key project of National Natural Science Foundation of China (No. 51338003).
- Curries, G., Mesbah, M.: Exploring transit operations performance at a network level using AVL and new GIS visualization methods. The 90th Annual Meeting of the Transportation Research Board (CD-ROM), Washington (2011)Google Scholar
- Hofmann, M., O’Mahony, M.: Transfer journey identification and analyses from electronic fare collection data. In: Proceedings of the Intelligent Transportation Systems, pp. 34–39. IEEE (2005)Google Scholar
- Lu, C.T., Boedihardjo, A.P., Shekhar, S.: Analysis of spatial data with map cubes: highway traffic data. Geographic Data Mining and Knowledge Discovery, 2nd edn, pp. 69–97 (2009)Google Scholar
- Ma, X.: Smart card data mining and inference for transit system optimization and performance improvement. Doctoral dissertation, University of Washington (2013)Google Scholar
- Ministry of Transport of the People’s Republic of China.: The Chinese transit metropolis program. http://zizhan.mot.gov.cn/zhuantizhuanlan/gonglujiaotong/gongjiaods/201310/t20131030_1505116.html (2011) (in Chinese)
- Nanjing Planning Bureau. Nanjing Transport Annual Report (2013) (in Chinese)Google Scholar
- Nanjing Statistic Bureau. Nanjing Economy and Society Developed Statistical Bulletin (2014) (in Chinese)Google Scholar
- Riedel, H.U.: Chinese metro boom shows no sign of abating. Int. Railw. J. 54(11), 46–48, 50, 52 (2014)Google Scholar
- Shekhar, S., Lu, C. T., Liu, R., Zhou, C.: CubeView: a system for traffic data visualization. In: Proceedings of IEEE 5th International Conference on Intelligent Transportation Systems, pp. 674–678 (2002)Google Scholar
- Shekhar, S., Lu, C. T., Liu, A.: High performance spatial visualization of traffic data (No. CTS 04-04,). University of Minnesota, Center for Transportation Studies (2004)Google Scholar
- Sun, L., Lee, D. H., Erath, A., Huang, X.: Using smart card data to extract passenger’s spatio-temporal density and train’s trajectory of MRT system. In: Proceedings of the ACM SIGKDD international workshop on urban computing pp. 142–148. ACM (2012)Google Scholar