Abstract
Freeway networks are vulnerable to natural disasters and man-made disruptions. The closure of one or more toll stations of the network often causes a sharp decrease in freeway performance. Therefore, measuring the probability and consequences of vulnerability to identify critical parts in the network is crucial for road emergency management. Most existing techniques only measure the consequences of node closure and rarely consider the probability of node closure owing to the lack of an extensive historical database; moreover, they ignore highways outside the study area, which can lead to errors in topological analysis and traffic distribution. Furthermore, the negative effects produced by the operation of freeway tunnels in vulnerability assessment have been neglected. In this study, a framework for freeway vulnerability assessment that considers both the probability and consequences of vulnerability is proposed, based on the perspective of network cascade failure analysis. The cascade failure analysis is conducted using an improved coupled map lattice model, developed by considering the negative effects of tunnels and optimizing the rules of local traffic redistribution. The perturbation threshold and propagation time step of network cascade failure are captured to reflect the probabilities and consequences of vulnerability. A nodal vulnerability index is established based on risk assessment, and a hierarchical clustering method is used to identify the vulnerability classification of critical nodes. The freeway network of Fuzhou in China is utilized to demonstrate the effectiveness of the proposed approach. Specifically, the toll stations in the study area are classified into five clusters of vulnerability: extremely high, high, medium, low, and extremely low. Approximately 31% of the toll stations were classified as the high or extremely high cluster, and three extremely vulnerable freeway sections requiring different precautions were identified. The proposed network vulnerability analysis method provides a new perspective to examine the vulnerability of freeway networks.
Highlights
A freeway network is the backbone of the intercity transportation network and meets the enormous transportation demand
Some events caused by external natural disasters may cause extensive damage to the freeway network, whereas others that originate within the transportation system, such as car accidents and short-term heavy flow, may lead to local road closure and even trigger cascade failure of the network
Hierarchical clustering is a data-driven method, which can Vulnerability assessment on freeway network from a perspective based on network cascade failure overcome subjective bias and has been widely applied in hierarchical rank studies [52, 53]
Summary
A freeway network is the backbone of the intercity transportation network and meets the enormous transportation demand. To address the problems discussed above, this study attempts to provide an alternative method to obtain the probability and consequences of vulnerability by using cascade failure analysis and proposes a vulnerability assessment framework based on risk analysis. The topological coupling coefficient considers the negative impact of freeway tunnels on segment operations, and the flow redistribution algorithm is integrated into the flow coupling coefficient This improved model is employed to perform a cascade failure analysis of the freeway network to capture both the probability and consequences of vulnerability. Virtual peripheral nodes are introduced to describe scenarios where there is a path extending outside the study area, eliminating border effects This framework eliminates the need for an extensive historical database and facilitates a more comprehensive and convenient vulnerability assessment of the freeway network. The nodal vulnerability is assessed based on the risk analysis and the critical nodes of the road network are identified using a hierarchical cluster algorithm
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
