Vulnerability Assessment Methods And Improvement Strategies Of Regional Multi-Modal Passenger Transport Networks

In a national strategy context of building a country with strong transportation network,the regional multi-modal passenger transport network is a lifeline engineering to provide intercity transportation services,whose significance for regional economic and social development is self-evident.The analysis of the vulnerability of regional multi-modal passenger transport network can provide a theoretical foundation for improving network resilience and reliability and reducing the impact of emergencies.Therefore,a passenger transport system consisting of highway,railway,waterway and aviation is the object of this study.The theoretical algorithms such as complex network theory,information entropy and coupled map lattice model were applied to build the network model,construct the measurement system,identify the critical nodes,clarify the evolution characteristics,and propose improvement strategies,which was the main research line.The validity of the model was verified by the simulation analysis method based on actual data of Zhejiang Province and Yunnan Province.Thus,the vulnerability analysis and improvement method applicable to multi-modal passenger transport network was proposed.The main research contents are divided into five parts as follows.(1)The multi-modal passenger transport network topology model was constructed based on complex network theory.An undirected weighted single mode passenger transport network topology model was constructed based on the Space L method,which abstracted the physical structure of each transportation mode into the network topology.Then the multi-modal interdependent passenger transport network(MIPTN)topology model was constructed from the perspective of spatial interdependence by considering the transfer relationship among transportation modes.It was applied to investigate the interactions among the network layers.The multi-modal composite passenger transport network(MCPTN)topology model was constructed by mapping the networks of each transportation mode of MIPTN to the same dimension and considering the attributes,functions,and locational relationships of nodes and edges.The multi-layer networks were abstracted into a composite network in a simplified way,which was used to analyse the overall network performance changes from a meso perspective.The association relationship between multiple transportation modes was abstractly characterized from different perspectives,which laid the foundation of differentiated network topology model for the subsequent research.(2)The vulnerability measurement system of the multi-modal passenger transport network was constructed from the perspectives of network topology,transport capacity,and transportation mode differential.A series of metrics applicable to multi-modal passenger transport network,such as station distribution balance,coupling strength,sub-network sensitivity,and synergy degree,were proposed based on the topological vulnerability metrics and combined with actual traffic demand and interactions between transport modes.The differences and commonalities of the vulnerability of comprehensive passenger transport networks in developed coastal areas of Southeast China and mountainous border areas of Southwest China were compared and analysed.The reasonableness and applicability of the metrics were verified from multiple perspectives through actual data simulation,thus providing evaluation criteria for subsequent studies.(3)The critical node identification methods of multi-modal passenger transport network were proposed.The critical nodes were identified from the perspective of network topology based on the weighted k-shell model,which integrates the diversity of transportation modes,independent transportation capacity,and connectivity.Then the critical nodes were identified from the perspective of traffic demand based on the weighted k-shell model.Both the number and influence of multi-order neighbor nodes were considered in the model.Meanwhile,the transport corridor idea was introduced to combine with the traffic demand to modify the attenuation coefficient of multi-order neighbor node contribution to differentiate the node influence.The SI model and the partial vulnerability metrics proposed in Chapter 3 were utilized to verify the validity and reliability of the model with the comprehensive passenger transport network in Zhejiang Province as an example.The analysis results show that the critical nodes identified from 2 different emphases of topology and traffic demand are reasonably valid and better than the traditional model.The results of identification can provide multi-perspective support for targeted improvement of network vulnerability.(4)A method for analyzing the vulnerability evolution process of multi-modal passenger transport network vulnerability was proposed.The coupled map lattice model was improved to be suitable for cascading failure propagation of node states between passenger networks.It was constructed by combining traffic demand and station capacity to calibrate node states and using traffic demand to modify node state coupling strength.Two major scenarios of attacking a single node and attacking multiple nodes at the same time were constructed.It combined with the partial vulnerability metrics proposed in Chapter 3 to empirically analyse the passenger transport network in Zhejiang Province as an example.The interdependent response mechanism and time-varying law of vulnerability among nodes,sub-network,and composite network of multi-modal passenger transport network under different external disturbance strength,coupling strength,disturbance scale,and disturbance objects were obtained.The analysis results show that different disturbance strength thresholds exist for each network.There is difference in the impact on the network vulnerability for different disturbance strengths acting on each subnetwork.The disturbance strength of perturbing critical nodes to occur network collapse is less than that of non-critical nodes.When the disturbance strength is large enough,the higher the coupling strength of the node states,the faster the network collapse.The method provides a new theoretical support for analysing the interactions and impacts of multi-modal passenger transport networks and provides a foundation for vulnerability improvement.(5)The vulnerability improvement strategies of multi-modal passenger transport network were proposed for both scenarios of emergency evacuation and transportation network planning.For the emergency evacuation scenario in a disaster,a node importance evacuation strategy combining node importance and cascade failure model was proposed.Its effectiveness was compared and analysed with the other five evacuation strategies.The results show that the node importance evacuation strategy is the best evacuation solution from all analysis angles.For the pre-disaster comprehensive transportation network planning scenario,the vulnerability measurement system was utilized to compare and analyse the vulnerability differences between the current and 14 th Five-Year Plan passenger transport network in Yunnan Province and Zhejiang Province.From the perspective of practical demands,the effectiveness of the planned network compared with the current network was demonstrated.The analysis results show that the vulnerability of the 14 th Five-Year Plan network has been significantly improved compared with the current network.The layout of the railway network needs to be further improved.Some regions need to enhance the diversity and substitutability of transportation modes to reduce network vulnerability.To address the problems of the planning network,further improvement strategies are proposed with a view to improving its resilience and reliability.