A Vision for Probabilistic Analysis of Re- configurable Resilient Engineered and Infrastructure Systems

Resilience is a challenging concept that is given many definitions and perspectives in the engineering infrastructure, systems engineering, and economic literatures. An emerging understanding of resilient systems is as a management principle or framework allowing for reconfiguration or adaptation in the face of threats or shocks. This is in contrast to the prevailing approach to engineered system resilience: systems may focus on resistance to threats, and speedy recovery if vulnerabilities are breached. In this paper, we develop a probabilistic approach loosely related to the Functional Dependency Network model developed by Garvey and Pinto in which reconfiguration in the face of threats is actively investigated in conjunction with component fragility and interdependency. The framework consists of five objects. First, the analyst must specify the system of systems in terms of a partially directed graph. Second, the analyst must specify the links between the nodes in a link matrix. Third, the analyst must describe the fragility of links between the nodes in terms of a three dimensional link transition matrix in which link states are characterized by threat layers (i.e., each layer is a link matrix in which the probability of that link persisting to the next time step) and a connection possibility frontier (i.e., the set of nodes or node types to which the disconnected node may be reconnected). Finally, each node or system component requires a fragility curve for the threats to which the system is subjected. Keywords—complexity, resilience, functional dependency network analysis, Bayesian networks, technology of humility