Research On Critical Node Identification Of Urban Public Transportation Network

With the continuous development of the social economy and urban construction,urban public transport has become increasingly important in improving the operation level of transport services and the travel experience of urban residents.However,in the daily operation of urban public transport,some metro stations or bus stops may cause extensive traffic congestion after extreme weather,resulting in a serious decline in the structural connectivity of the transport system.At the same time,there is an uneven development of the urban public transport system,with some metro stations and bus stops carrying most of the passengers of the urban public transport system.In order to improve the resistance of the transport network to disruption and to avoid station failures that significantly reduce travel efficiency,it is necessary to master critical urban public transport stations.Therefore,based on the theory of network science,this paper carries out two main research contents centering on the topology structures of the transportation network and the temporal and spatial characteristics of passenger travel.First,it only considers the influence of the characteristics of passenger flow of urban rail transit and carries out the identification of critical nodes of the urban rail transit network.Then,considering the interaction between the passenger flow of urban rail transits and the passenger flow of urban buses,the critical nodes of the bus-metro double-layered network are identified.(1)To address the problem that the critical station identification method does not fully integrate the time-varying characteristics of passenger flow,a critical node identification method for urban rail transit networks is proposed,which improves the efficiency of critical station identification and reveals the dynamic change characteristics of station importance by combining network topology information and time-varying characteristics of passenger flow.Firstly,the node load network of urban rail transport is constructed based on an L-space to capture the topology of the transport network and to characterize the time-varying passenger flow.Then,a topology-passenger flow centrality method is proposed to obtain local and global information on the network topology by fusing various centrality indicators and using the entropy weighting method to determine the weights of different indicators.At the same time,topology-centrality also quantifies the important impact caused by station passenger flows.The experiments based on Shanghai urban rail transit data show that topology-centrality can effectively assess the importance of stations and uncover critical stations in urban rail transit while revealing the dynamic change of station importance.(2)Aiming at the problems that the critical station identification method pays less attention to the coupling relationship of different transportation systems,lacks effective use of operating lines,and is difficult to quickly obtain global information of the network structure,a critical node identification method for bus-metro double-layered networks is proposed,which reveals the critical stations of urban public transport by fusing line information and community structure characteristics.Firstly,the bus-metro double-layered network is constructed based on L-space and geographic proximity,so as to portray the interaction between urban bus and urban rail transit and record passenger flow information and line information through node attributes and edge attributes respectively.Then,a line-community centrality method is proposed to identify critical nodes of the bus-metro double-layered network based on the operating lines and community structure.The method is based on line information to characterize the tightness of connections between stations and accordingly,quantifies the degree of influence of different neighbors on the importance of nodes.At the same time,a bipartite diagram is used to portray the mutually reinforcing relationship between stations and lines.In addition,line-community centrality uses community structure and passenger flow information to quickly capture the important impact of stations that are far apart.The experiments based on Shanghai urban public transport data show that line-community centrality can effectively explore the critical stations in the double-layered network.