Study On Influence Of Built Environment Around Urban Rail Transit Station On Passenger Flow

In recent years,with the development of society and economy in China,the urban area and radiation range are expanding,and the car ownership is also increasing year by year,which not only aggravates the pressure of urban road network,but also produces more pollution to the environment.Urban rail transportation has the advantages of punctuality,high speed,land saving,noise reduction,air pollution reduction and energy saving.Subway construction is a century-old project with huge cost,long construction period and wide coverage.Therefore,in the context of the rapid development of modern urban construction,the characteristics of traffic demand should be comprehensively considered to explore the relationship between rail transit passenger flow and multi-dimensional influencing factors around the station,which is of great significance to guide the scientific rationality of TOD model planning,improve the accuracy of passenger flow prediction and improve the efficiency of rail transit operation.In the existing research results,the research object is mostly single passenger flow type,less for multi passenger flow type and multi perspective type;Most of the research models are linear regression models,and few are nonlinear regression models.The modeling method mostly uses the standard linear regression model,and rarely considers the spatial difference and nonlinear relationship of variables caused by geographical location changes.In this paper,based on the collected building environment and the data of the urban railway transportation card,the relationship between the building environment around the station,the flow of passengers at the railway station,and the flow of passengers between the railway stations is discussed between the railway station and the station.There are two main aspects to the content.One is the research on the impact of the built environment around the station on the passenger flow of rail transit stations.The other is the research on the impact of the built environment around the station on the passenger flow between rail transit stations.The first part focuses on the passenger flow of rail transit stations,and selects the daily inbound and outbound volume of work day,the daily inbound and outbound volume of morning peak hours(8:00-9:00),the daily inbound and outbound volume of night peak hours(18:00-19:00),and the daily inbound and outbound volume of holidays to characterize the characteristics of the whole day passenger flow,commuter passenger flow and holiday passenger flow.Taking the OLS model of station passenger flow and environmental factors as the starting point,the key variables are selected,and the GWR model is introduced to analyze the influence of each variable in the geographically weighted regression model(GWR).The spatial distribution of each variable coefficient is visualized by GIS software.At the same time,the spatial heterogeneity of influencing factors of subway station passenger flow is discussed,and it is found that there is spatial heterogeneity in the regression model of station passenger flow for commercial and administrative office land factors.The second part focuses on the passenger flow between rail transit stations.For the 66 stations,4290 ODs are formed,and the maximum passenger flow per hour and the time period of all ODs are selected.Then,they are divided into three categories according to the time characteristics,namely,the early peak period(8:00–9:00),the late peak period(18:00–19:00)and other periods(non-early and late peak periods).At the same time,the gradient boosting decision tree(GBDT)model of rail transit OD passenger flow and built environment factors is established,and the relative importance of variable characteristics is calculated,and the variable partial dependence graph is generated to characterize the nonlinear relationship between variables.The main conclusions of the study are as follows:(1)From the point of view of the station,six influencing factors including commercial land,administrative office land,the number of bus lines,the length of main and secondary trunk roads,and whether they are transfer stations have a significant impact on the passenger flow of urban rail transit stations.From the point of view of the station,the relatively important variables are the distance of the car,three types of residential land,office land,commercial land,commercial land,primary and secondary education land,industrial land,the number of conventional bus lines,the length of the main and secondary roads,etc.(2)Conventional bus lines have a significant positive impact on passenger flow at each station in the morning and evening peaks,and commercial land has a significant positive impact on subway passenger flow in the evening peak and holidays.Through the nonlinear relationship diagram,it is found that this growth will reach the peak and then tend to be stable.(3)Travel distance has a significant impact on passenger flow in each period,and there is a nonlinear relationship between travel distance and passenger flow change,and the image is parabolic.(4)The length of main and secondary trunk roads is significantly correlated with the subway passenger flow,and the rationality and convenience of connecting public bicycles,parking lots,bus stations and subway stations should be considered.This paper describes the spatial heterogeneity of passenger flow influencing factors and describes the nonlinear relationship between variables.Based on the research perspective of urban rail transit station passenger flow and inter-station passenger flow,this paper analyzes the relationship between various factors and rail transit passenger flow in detail,which makes up for the lack of previous studies on the spatial-temporal relationship and quantitative relationship between passenger flow and variable influencing factors.