Loss estimation of interdependent infrastructures in targeted attacks

Lifelines and infrastructures are considered as basic foundations of modern society that are involved with all needed services and facilities by the society. In the modern life، the range of dependency to these infrastructures is increasing. Thus، they become an attractive target for enemies. First، based on Varden''s five rings theory، lifelines are the center of gravity in every country that in case of any destruction، defeated country will be paralyzed and will not be able to continue its vital activities. Secondly، on the basis of irregular warfare، the loss of service of the infrastructures، leading to a reduction in social welfare and the central government would be delegitimized. In addition، in order to have a better understanding of the infrastructure systems، networks dependency and their interactions in the accidents are highlighted as very important issue. The study of the interaction between water and power networks as the targeted infrastructures is the objective of this paper. In order to analysis lifelines vulnerability، water and power networks in the district of Tehran as a case study were introduced. The proposed methodology module consists of two models، graph theory and Leontief model. In this process، first، two studied networks were separately modeled based on Graph theory and then with Leontief model، the Service Flow Reduction Index was calculated for each attack scenario. These scenarios were prioritized based on the amount of damage and the remained capacity of the studied networks. Moreover، the vulnerability reduction strategies were mentioned and the effects of each one on the index were calculated. As a result، 240 scenarios in the studied district were considered. Therefore، the study was conducted for identifying the worst and the best scenario and measuring the impact of each one with the effect of dependency reduction. Finally، by considering the effect of dependency in determining the failure probability of components in the water and power networks، it can be concluded that، in some cases the dependency effect could be neglected and in some cases the effect is considerable. But in all cases، these dependencies have an ascending impact on the performance probability، and descending impact on the failure probability. According to the results، among scenarios with one targeted component، 100% vulnerability to targeted attacks، scenario 1 (explosion in water treatment A) was achieved the least serviceability index as 7. 35%. Also between scenarios with more than one targeted component، scenario 10 (explosion in both water treatment A and B and substation A) was achieved the least serviceability index as 3. 11%.