利用OD数据的人群活动分析

Crowd Activity Analysis using OD Data

OD数据包含了城市居民活动模式,对于分析城市居民的活动、把握居民出行模式和城市空间结构具有重要意义。因此,本文提出了一种区域流动计算方法,运用区域流动倾向指数和区域主体流向来挖掘居民活动的模式。首先,根据时空粒度对区域的OD流进行统计,把OD数据映射到相应的区域内;然后,计算每个区域在不同时间段的流动倾向指数(包括流入倾向指数和流出倾向指数),并计算区域内OD流的主要来源和目的地;最后,根据流动倾向指数进行空间模式、时间模式分析和聚类分析。实验以2017年6月海口市滴滴OD数据为例进行分析,结果表明该方法能根据流动倾向指数挖掘出绝对流量小但来源或目的地集中的区域模式,并体现了海口市的城市结构特点。

OD data contains the activity patterns of urban residents, and the activity analysis of urban residents is great significance to grasp the travel patterns of urban residents and urban spatial structure. Therefore, this paper proposes a regional flow calculation method, which uses the regional flow tendency index and the regional main flow direction to excavate the patterns of residents’ activities. Firstly, we count the OD flow in the region according to the spatiotemporal granularity, mapping the OD data to the corresponding region. Then, we calculate the flow propensity index(including inflow propensity index and outflow propensity index) and the main source and destination of OD flow of each region in different time periods. Finally, spatial pattern analysis, temporal pattern analysis and cluster analysis were conducted according to the flow tendency index. The experimental results show that the method can mine the regional pattern with small absolute flow but concentrated source or destination according to the flow tendency index, and reflect the urban structure characteristics of Haikou.