Application of social network theory to prioritizing Oil & Gas industries protection in a networked critical infrastructure system

As a typical process industry, the Oil & Gas industries play a key role within a networked critical infrastructure system in terms of their interconnection and interdependency. While the tight coupling of infrastructures increases the efficiency of infrastructure operations, interdependency between infrastructures may cause cascading failure of infrastructures. The interdependency between critical infrastructures gives rise to an infrastructure network. In this paper, we apply social network analysis, an analytical tool used by social scientists, to study human interactions and to analyze characteristics of the critical infrastructure network. We identify Oil & Gas, Information & Communication Technologies (ICT), and Electricity as three infrastructures that are most relied upon by other infrastructures, thus these may cause the greatest cascading failure of the infrastructures. Among the three, we further determine that Oil & Gas and Electricity are the more vulnerable infrastructures. As a result, priority toward critical infrastructure protection should be given to the Oil & Gas and Electricity infrastructures since they are most relied upon but at the same time depend more on other infrastructures.     

Modeling flood induced interdependencies among hydroelectricity generating infrastructures

This paper presents a new kind of integrated modeling method for simulating the vulnerability of a critical infrastructure for a hazard and the subsequent interdependencies among the interconnected infrastructures. The developed method has been applied to a case study of a network of hydroelectricity generating infrastructures, e.g., water storage concrete gravity dam, penstock, power plant and transformer substation. The modeling approach is based on the fragility curves development with Monte Carlo simulation based structural–hydraulic modeling, flood frequency analysis, stochastic Petri net (SPN) modeling, and Markov Chain analysis. A certain flood level probability can be predicted from flood frequency analysis, and the most probable damage condition for this hazard can be simulated from the developed fragility curves of the dam. Consequently, the resulting interactions among the adjacent infrastructures can be quantified with SPN analysis; corresponding Markov Chain analysis simulates the long term probability matrix of infrastructure failures. The obtained results are quite convincing to prove the novel contribution of this research to the field of infrastructure interdependency analysis which might serve as a decision making tool for flood related emergency response and management.