Multidimensional resilience decision-making for complex and substructured systems

Complex systems, such as infrastructure networks, industrial plants and jet engines, are of paramount importance to modern societies. However, these systems are subject to a variety of different threats. Novel research focuses not only on monitoring and improving the robustness and reliability of systems, but also on their recoverability from adverse events. The concept of resilience encompasses precisely these aspects. However, efficient resilience analysis for the modern systems of our societies is becoming more and more challenging. Due to their increasing complexity, system components frequently exhibit significant complexity of their own, requiring them to be modeled as systems, i.e., subsystems. Therefore, efficient resilience analysis approaches are needed to address this emerging challenge.

This work presents an efficient resilience decision-making procedure for complex and substructured systems. A novel methodology is derived by bringing together two methods from the fields of reliability analysis and modern resilience assessment. A resilience decision-making framework and the concept of survival signature are extended and merged, providing an efficient approach for quantifying the resilience of complex, large and substructured systems subject to monetary restrictions. The new approach combines both of the advantageous characteristics of its two original components: A direct comparison between various resilience-enhancing options from a multidimensional search space, leading to an optimal trade-off with respect to the system resilience and a significant reduction of the computational effort due to the separation property of the survival signature, once a subsystem structure has been computed, any possible characterization of the probabilistic part can be validated with no need to recompute the structure.

The developed methods are applied to the functional model of a multistage high-speed axial compressor and two substructured systems of increasing complexity, providing accurate results and demonstrating efficiency and general applicability.