Research On Commuting Characteristics And Commuting Distribution Model Of Beijing Metropolitan Area Based On Hitch Data

With the rapid development of China's economy and society and the enhancement of the radiation ability of major cities,the metropolitan area has gradually become a new urbanization main form of big cities.At present,there are few studies on commuting in the metropolitan area,and they are mainly concentrated on the research of urban rail transit.In recent years,due to the development of sharing economy,with the advantages of convenience,comfort and economy,hitch has gradually become an important mode of commuting in the metropolitan area.In order to enrich the theoretical research of road commuting in the metropolitan area,this paper takes Beijing metropolitan area as an example,based on the trip data of Di Di Hitch,proposes a method to divide the metropolitan functional areas,and studies the commuting characteristics and commuting distribution model of metropolitan area on the basis of divided functional areas.The main research work of this paper includes the following four aspects:(1)This article first describes the used trip data of hitch and points of interest data,including data volume,time span and data format.Then the coordinate transformation and preprocessing are carried out for the trip data of hitch.The preprocessing process includes checking whether there are missing records,the same records and abnormal records,and eliminating the data in the above situations to obtain valid data sets.(2)Based on the basic analysis of hitch data to ensure the accuracy of the data,the grid model including the scope of Beijing metropolitan area is constructed and the valid grid is determined.Considering the latitude and longitude of the grid,the amount of OD along the hitch data and the number of POI,the K-means++ clustering algorithm with improved k-value selection method is used to cluster the valid grid.On the basis of the initial clustering results,22 typical functional areas of Beijing metropolitan area are selected according to the two conditions of regional order quantity and commuting rate.(3)Based on the divided functional area of metropolitan area,this paper studies the commuting characteristics of metropolitan area in terms of commuting intensity,space-time characteristics and commuting accessibility.Each aspect also considers the commuting characteristics between two functional areas and the commuting characteristics with a specific functional area as the starting and ending point.In the analysis process,big data processing method and spatial visualization method are used to make the conclusions more intuitive.On the basis of the above studies of commuting characteristics,this paper classifies the functional areas by using the obtained commuting characteristics indexes by using the method of hierarchical clustering,and puts forward the three-level hierarchical structure of the core area,the inner suburb area and the outer suburb area of Beijing metropolitan area.(4)The functional area is used as the traffic area for commuting,and the unconstrained gravity model is selected as the basic model.Based on the analysis of the factors affecting the commuting in the metropolitan area,an improved commuting distribution model is constructed by introducing the composite variables that affect the commuting distribution in the metropolitan area.Considering the Collinearity problem,two gravity models are calibrated by ridge regression analysis and stepwise regression analysis.The commuting distribution is analyzed by using the model,and the applicability and fitting effect of the two gravity models are compared.This paper contains 35 figures,46 tables and 71 references.