Mining Evolution Patterns Of Urban Metro Station Functions Based On Smart Card Data

In order to solve a series of social problems,such as traffic congestion,environmental pollution,energy consumption and so on,all countries in the world are vigorously developing metro systems.The rapid development of metro system makes it a skeleton of the city,and metro stations can be used as probes to study urban problems.This paper defines the activities of passengers who reach the surrounding areas of metro stations through metro as station functions such as living,working,catering and leisure.The changes of metro station functions,such as the enhancement of working function or the weakening of living function,are the evolutions of station functions.In the actual operation of metro systems,there are various problems,such as uncoordinated land use with the surrounding areas and inconsistent with the planning expectations.In order to explore and analyze the possible problems and understand the development situations around metro stations and even the whole city,it is urgent to study the evolution patterns of the station functions.Most of the existing researches are based on the current situation or short-term data to identify the metro station or urban area functions,with few researches of longterm evolutions of station functions.The combination of smart card data and POI(point of interest)data,land use,travel survey,statistical bulletin and so on is generally used to study identifications of metro station or urban area functions.As it is difficult to obtain consistent data and carry out data fusion,the combination of multi-source data is not suitable for rapid and low-cost detection of the evolutions of metro station functions.In the current background of big data research,there are two research methods from different perspectives,method with prior feature and no-prior feature,respectively.It is very meaningful to explore the two methods based on the same problem.Based on smart card data,this paper proposed a method of mining the evolutions of metro station functions from the perspectives of prior feature and no-prior feature,respectively and carried on the empirical research in Shenzhen,comparing and analyzing the research results,using the Baidu map and the remote sensing images to carry on the result verification,discussing the advantages and disadvantages and the applicable scopes of the two methods.The results show that both methods can effectively excavate the evolution patterns of metro station functions,and the results are basically consistent for typical stations with single function type and stations with obvious trend of function evolution.But there are dissimilarities for stations with mixed function and stations with less obvious trend or complex rules of evolution.Using Baidu map and remote sensing images to verify the results,it is known that the two methods are relatively reasonable and effective for the empirical research results of Shenzhen.For future research,the two methods proposed in this paper can be used to verify each other,and the Baidu map,remote sensing image,land use,street view can be used to further modify for the parts with inconsistent results.The main contributions of this paper are: based on smart card data from the perspectives of prior feature and no-prior feature,this paper proposed two methods to mining evolution patterns of metro station functions,implementing comparative analysis of the advantages and disadvantages of the two methods and the scopes of application,which is a deep thinking under the big data research paradigm.This paper studied the long-term evolution patterns of metro station functions based on smart card data in Shenzhen from 2014 to 2018,which is a new exploration and attempt to study the evolution patterns of metro station functions.Through the empirical study of the metro stations in Shenzhen,we identified that the function of DengLiang changed from living-oriented to working-oriented,and LiYuMen changed from working-oriented to living-oriented.It was found that the separation of working and living in Shenzhen was enhanced,and the spatial core positions of station functions were shifted.The cores of living function were shifting to the surrounding areas of Shenzhen,such as Baoan District,Longhua District and Longgang District.The cores of working function were shifting from Luohu District(developed traditional industry)to Futian District(developed financial industry)and Nanshan District(developing high-tech industry).In addition,the weekday lunch places shifted from the traditional catering industry gathering areas to areas near the workplaces.The results are helpful to understand the urban development of Shenzhen in recent years,and providing reference for planning evaluation,decision-making of the government,selection of commercial locations,investment of real estate and so on.