| Under the background of increasingly tight constraints on land use,resources and environment in megacities,advocating a compact and efficient land use pattern,focusing on the improvement of rail transit and non-motorized traffic facilities and services,has gradually been widely recognized as an important means to realize urban sustainable development.However,with the rapid development of urban rail transit in China’s megacities,the network structure is becoming more and more complex,and the passenger flow characteristics will show new characteristics with the expansion of the network.In order to timely and effectively respond to the changes of passenger flow characteristics of urban rail transit,it is necessary to analyze the passenger flow change trend at the three levels of network,line and station in the process of network expansion,and summarize the general evolution law of urban rail transit passenger flow,so as to clarify the network scale at the network level,determine the construction sequence and the capacity in different period at the line level,guide the station design and passenger flow organization at the station level.At the same time,it can also provide reference for the rail transit planning of late developing cities.In addition,the expansion of rail transit network will significantly improve the convenience and accessibility of rail transit,and then affect the scale and distribution of passenger flow.At the same time,the existing method of using land use to explain rail transit passenger flow is difficult to answer the question that land use is a slow variable and passenger flow is a fast variable.Therefore,in addition to the traditional relationship between transportation and land use,it is necessary to quantitatively analyze the impact of network characteristics on rail transit passenger flow,so as to improve the interpretability of rail transit passenger flow.According to the research idea of "putting forward problems-analyzing problems-solving problems",this paper adopts the research methods of combining qualitative and quantitative analysis,GIS spatial analysis and machine learning correlation model,starts with the analysis of passenger flow characteristics of urban rail transit,constructs the sigmoid curve model of passenger flow growth,and analyzes the evolution law and influence mechanism of passenger flow.On this basis,from the perspectives of station network characteristics and land use,this paper constructs the correlation model of boarding ridership,and quantitatively analyzes the interpretation degree of two kinds of influencing factors on boarding ridership.Firstly,taking the rail transit network passenger flow in four typical megacities of Beijing,Shanghai,Guangzhou and Shenzhen as the research objects,the growth law and factors of rail transit network passenger flow in the process of rail transit network are modeled and analyzed.The results show that the growth trajectory of the passenger volume of the network conforms to the sigmoid curve,and the sigmoid curve can be divided into three stages according to the different growth rate,namely,the initial cultivation stage of zone I,the medium-term growth stage of zone II and the later stable stage of zone III.For the passenger volume of the network,a certain network scale is needed to support its growth.Although the network scale of cities with different population size and spatial scale is different at different growth stages of passenger volume,the network length of about 100 km is needed to play the preliminary scale effect of rail transit.Then,taking Shanghai rail transit line passenger flow as the research object,the growth law and influencing factors of line passenger flow in the process of rail transit network are modeled and analyzed.The growth track of line passenger flow also conforms to sigmoid curve,and the influence of network scale is the most critical.It also gives the recommended line construction sequence of urban line,suburb line and tangential line.On this basis,from the perspectives of land use around the station and the network characteristics of the station,the correlation model of boarding ridership is established by using the method of direct demand model.The results show that the land use around the station can explain 60～70% of the boarding ridership,and the network characteristics can explain 20～30% of the boarding ridership,so as to clarify the significant impact of the network characteristics of the station on the boarding ridership.In addition,there is obvious spatial heterogeneity in the impact of land use around the station on the boarding ridership.Generally speaking,the closer to the urban central area,the better the interpretation of land use on the boarding ridership.Finally,based on the boarding ridership data of Shanghai Rail Transit in different years,the overall growth law of station boarding ridership under the condition of networking is analyzed.For the stations of new lines and mature lines,the growth law and influencing factors of boarding ridership are analyzed,and then the growth mode of boarding ridership is defined.The results show that the growth curve of station boarding ridership is generally an S-shaped curve,but it is different from the growth law of network passenger flow and line passenger flow.In addition to the three stages of cultivation period,growth period and stability period,some stations also have a balance period,showing the phenomenon of passenger flow decline,and further clarify that the network characteristics of the station have a significant impact on the growth curve of passenger flow.This paper summarizes the general evolution law and influence mechanism of rail transit passenger flow in typical megacities,which can provide technical support for urban rail transit planning and operation.In addition,it also quantitatively analyzes the impact of network characteristics on rail transit passenger flow,improves the interpretability of station boarding ridership,makes up for the shortcomings of existing research,and provides new views and solutions for rail transit passenger flow prediction. |
