基于地铁刷卡数据的城市通勤与就业中心吸引范围研究

地理时空大数据为通勤行为和城市空间结构研究带来了新的机遇。基于一周地铁刷卡数据，采用出行链（trip-chain）数据模型对用户一天的出行进行描述，建立通勤识别规则，识别出上海市域的职住空间和通勤格局。上海市域平均通勤时间为35 分钟，通勤主流向是向心流，说明上海市域依旧表现出强单中心结构。市域就业单中心与居住郊区化并存，中心城区多核心就业已经形成。在此基础上进一步分析中心城区13 个就业（次）中心的吸引范围、模式以及在整个市域的影响力，讨论多中心发展对职住平衡与通勤消减的意义。结果表明：多中心发展有利于城市通勤的合理分配，虽然次中心的平均通勤时间不一定减少，但是就业（次）中心已经表现出对邻近地区的通勤吸引和对主中心的通勤分流。

ANALYSIS OF COMMUTING BEHAVIOR AND EMPLOYMENT CENTER USING SUBWAY SMART CARD DATA

With the development of information communication technologies, it becomes easier to access to the spatio-temporal data such as smart card data, GPS-enabled taxi data and social media data, which gives fresh impetus to research on commuting behavior and urban spatial structure. What's more, compared with the traditional data like questionnaires, this kind data has advantage of good continuity, wide coverage and dynamic update. At the same time, it is the impact of the urban space on commuting behavior that is the important topic of urban geography, urban planning and urban traffic study. And most people choose to travel by subway because of its convenience in many large cities worldwide. Therefore, this research aims to analyze commuting behavior and employment center based on subway smart card data.The subway smart card data used in this study was collected by the fare collection system for one week in April, 2015 in Shanghai. In this paper, we designedthe Trip-Chain data model corresponding to the passengers' travels in one day. According to the rules of identifying commuting built, wegraphically illustrated the spatial relation between residence and employment in Shanghai. Also the commuting pattern is depicted clearly.Meanwhilethe analysis outcomedemonstrates that the average commuting time in Shanghai is 35 minutes and the mainstream of commuting is centripetal. More interesting, the average commuting time increases concentrically from the inner city to the fringe area. In addition, the place of employment gathers in the core area while the place of residence in the suburbs. On the other hand, we identified the passenger-attracting scope of the 13 employmentcenters in the downtownseparately and they could be divided into three types: distance attenuation mode, enclave attracting mode and composite mode. Based on the number of the employment center's passenger-attracting, we summary all the passenger-attracting scopes in one map. And the identification results indicate the polycentric structure can adjust and decentralize the employment spatially, which improves the efficiency of the commuting integrally.