Evaluation Of Freeway Link Vulnerability Based On Improved Node Deleting Method

Freeway system plays an important role in the regional economic and social development as a modern transport corridor. However, performance of freeway system degrades due to various events, for instance road closure or traffic congestion happens after adverse weather, which reflects system vulnerability. Identifing and analyzing vulnerable sections of freeway system ensures normal operation of system and improves the efficiency of emergency evacuation.First, the definition, related concepts and characteristics of freeway system vulnerabilty are introduced. Due to concealment and excitation conditions diversity are the main features of vulnerability, mining system vulnerability excitation condition is the basis of vulnerability research. Vulnerability influencing factors are divided into macro and micro levels according to whether the factor impacts links simultaneously and directly. And the hierarchy model of freeway link vulnerability factors is established based on DEMATEL/ ISM integrated model, which transforming the complex relationships into an intuitive and hierarchical relationship. The research result shows that atrocious weather and geological disaster are deep factors and network topology is inducing factor for vulnerability. Hence, the changes of network topology structure and the failure characteristics of the road section need be explored.Secondly, links failure characteristics under adverse weather are analyzed, and find there are two types of links failure. One is forced to close for deeply influenced by adverse weather. The other is congestion failure due to demand overcomse supply, especially adjacent links of closed link share diverted traffic volume and traffic demands increase, while the capacities are also reduced owing to the bad weather and traffic supplies decrease. However, the traditional node deletion method only considers deleting failure nodes from network, without considering the adjacent section capacity reduction, which does not consist with the failure characteristics of freeway system under the bad weather.Finally, an improved node deleting method considering cascading failure is proposed, it is more applicable to freeway system vulnerablity evaluation under adverse weather. The improced vulnerability assessment model focuses on solving three problems: collecting basic data, selecting appropriate vulnerability assessment index and algorithm designing. The basic datas needed for vulnerability assessment include network topology, link length, number of lanes, speed and OD matrix. And structural vulnerability, link failure probability, failure consequence severity are selected as assessment indexes. The volume is obtained by equilibrium assignment model, the capacity reduction factor depends on the event level and link speed. It is proved that the improved node deletion method is more suitable for freeway vulnerability assessment under adverse weather than the traditional method. Heilongjiang Provincial expressway network is taken for example to show the effectiveness of the new method, and corresponding improvement measures and recommendations are proposed.