文章编号： 1672-5328（2019）03-0019-11

钮心毅，谢琛
(同济大学建筑与城市规划学院，高密度人居环境生态与节能教育部重点实验室，上海200092)

摘要： 使用手机信令数据识别职住地时，采用不同规则和参数取值可能导致结果差异。这种差异对 职住地识别结果可靠性的影响程度值得研究。根据手机信令数据特性，提出职住地识别的时间连续 性、空间位置分辨率、数据时间序列3 个关键时空因素。采用4 种时间规则、3 种空间聚合距离、3 种数据时间序列进行组合，对同一城市同一批手机信令数据进行职住地测算。比较不同时空因素下 职住地识别结果的差异，采用识别率、平均直线通勤距离、共同识别用户一致性等指标验证结果的 可靠性。研究表明，不同的时空因素对职住地识别结果产生显著影响。最后，探讨在实际应用中时 空因素取值需注意的问题，并提出相关建议。

关键词： 交通规划；手机信令数据；居住地识别；工作地识别；时空因素

中图分类号： U491.1+2

文献标识码：A

Identifying Residence and Workplace Locations with Cellular Signaling Data: Spatial- Temporal Factors and Impacts

Niu Xinyi, Xie Chen
(College of Architecture and Urban Planning, Tongji University, Key Laboratory of Ecology and Energy- Saving Study of Dense Habitat, Ministry of Education, Shanghai 200092, China)

Abstract： Using different rules and parameters in residence/workplace identification based on cellular signaling data may lead to different results. Therefore, it is necessary to study the impact of the difference on the reliability of residence/workplace identification. According to the characteristics of cellular signaling data, this paper proposes three key spatial-temporal factors in residence/workplace identification, namely time continuity, spatial position resolution and data time series. With the combination of four types of time rules, three types of spatially aggregated distances, and three types of data time series, the residence and workplace are analyzed based on the same batch of cellular signaling data from the same city. The results under different spatial-temporal factors are compared to discover the differences, and the reliability of the findings is verified with indicators such as identification rate, average linear commuting distance, and user consistency. The results show that different spatial-temporal factors have significant impacts on the identification results of residence and workplace location. Finally, the paper provides suggestions on several noticeable problems in the value-taking of spatial-temporal factors.

Keywords： transportation planning; cellular signaling data; residence identification; workplace identification; spatial-temporal factors