基于手机信令数据的居民活动空间及影响因素分析  

Analyzing Residents'Activity Space and Its Influence Factors Based on Mobile Phone Signaling Data

作　　者：段征宇[1] 赵浩然 DUAN Zhengyu;ZHAO Haoran(Key Laboratory of Road Traffic Engineering of the Ministry of Education,Tongji University,Shanghai 201804,China)

机构地区：[1]同济大学道路与交通工程教育部重点实验室,上海201804

出　　处：《交通与运输》2023年第2期10-14,共5页Traffic & Transportation

基　　金：教育部人文社会科学研究项目(18YJAZH018):宏微观数据融合的城市居民生活活动与出行行为研究。

摘　　要：采用手机信令数据,结合人口普查数据和兴趣点(POI)数据,以上海市外环以内的中心城区居民为对象,研究居民活动空间特征及其影响因素。从活动频率、活动半径、基家活动范围3个方面构建活动空间描述指标。以街道或镇为空间分析单元,通过聚类得到居民典型活动空间模式及其空间分布。采用单因素方差分析,从建成环境、社会人口属性中识别显著影响活动空间等因素,并建立随机森林模型量化各影响因素的重要性。研究发现,上海市中心城区居民的活动空间模式可分为3类;建成环境因素对活动空间的影响大于社会人口因素;活动空间的影响因素,按重要性依次为设施密度、可达性、住房选择、婚姻状况、儿童比例和区位。Using mobile phone signaling data,combined with census data and point of interest(POI)data,this paper takes the residents of the central area within the Outer Ring Road of Shanghai as the research object to study the residents’activity space characteristics and corresponding influence factors.From three aspects,namely the activity frequency,the activity radius,and the scope of activities based on home,indicators are established to describe the activity space of resident.Taking sub-district or town as the spatial analysis unit,the typical activity space patterns of residents and their spatial distribution are obtained,through the cluster method.Using one-way analysis of variance(ANOVA),factors that significantly affect the activity space are identified from the built environment and socio-demographic attributes,and a random forest model is built to quantify the importance of these factors.It is found that the patterns of residents'activity space in the central area of Shanghai can be divided into three types;the impacts of the built environment factors on the activity space are greater than those of the socio-demographic factors;the factors that affect the activity space in order of importance are facility density,accessibility,housing choice,marital status,proportion of children and location.

关 键 词：居民活动空间 手机信令数据 建成环境 随机森林模型 单因素方差分析 

分 类 号：U491[交通运输工程—交通运输规划与管理]