ADVANCED METHODS FOR THE RISK, VULNERABILITY AND RESILIENCE ASSESSMENT OF SAFETY-CRITICAL ENGINEERING COMPONENTS, SYSTEMS AND INFRASTRUCTURES, IN THE PRESENCE OF UNCERTAINTIES

Safety-critical industrial installations (e.g., nuclear plants) and infrastructures (e.g., power transmission networks) are complex systems composed by a multitude and variety of heterogeneous ‘elements’, which are highly interconnected and mutually dependent. In addition, such systems are affected by large uncertainties in the characterization of the failure and recovery behavior of their components, interconnections and interactions. Such characteristics raise concerns with respect to the system risk, vulnerability and resilience properties, which have to be accurately and precisely assessed for decision making purposes.In general, this entails the following main steps: (1) representation of the system to capture its main features; (2) construction of a mathematical model of the system; (3) simulation of the behavior of the system under various uncertain conditions to evaluate the relevant risk, vulnerability and resilience metrics by propagating the uncertainties through the mathematical model; (4) decision making to (optimally) determine the set of protective actions to effectively reduce (resp., increase) the system risk and vulnerability (resp., resilience).New methods to address these issues have been developed in this dissertation. Specifically, the research works have been carried out along two main axes: (1) the study of approaches for uncertainty modeling and quantification; (2) the development of advanced computational methods for the efficient system modeling, simulation and analysis in the presence of uncertainties