A Modeling Framework for the Inoperability Assessment of Interdependent Critical Infrastructures exposed to Spatially Distributed Natural Hazards

• A novel framework is proposed to model natural hazard-induced disruptions in multi-state interdependent CIs. • A C-vine copula model is used to simulate spatially coherent hazard scenarios across a system-of-systems spatial domain. • Site-specific marginal distributions are fitted to capture spatial heterogeneity in hazard behavior. • The C-vine copula is integrated with a multi-state DIIM to estimate hazard-induced failures and cascading effects. • Results support preventive and mitigative strategies based on criticality and dependency analysis. In this paper, a Dynamic Inoperability Input-Output Model (DIIM) is integrated with a copula-based hazard spatial model to assess the risk of inoperability of critical infrastructures exposed to spatially distributed natural hazards. Copulas are used because they can describe the natural hazard extent across a spatial domain. In particular, Canonical-vine copulas (C-vine) are here used to model the complex spatial dependencies while accounting also for the tail effects typical of extreme hazard events. The proposed modeling framework is applied to a system of systems comprising artifact gas and power networks exposed to extreme wind from Northeast of Italy meteo.