Data-driven identification of critical components in complex technical infrastructures using Bayesian additive regression trees

Complex technical infrastructures are systems-of-systems characterized by hierarchical structures, made by thousands of interconnected components performing different functions associated to various domains. Given the difficulty of deriving their functional logic using traditional risk and reliability analysis methods, we address the problem of critical component identification from an innovative perspective, which exploits the large amount of available monitored data of operation. Specifically, we develop a data-driven framework of analysis which employs Bayesian additive regression trees and validate it on a synthetic case study, which mimics the complexity of a complex technical infrastructure.