A Framework for the Network-Based Assessment of System Dynamic Resilience

In recent years, network-based resilience assessment has aroused attention because of its strong link to the stability and dependability of complex systems. Previous network-based studies have contributed to the definition and quantification of system resilience, but an integral and consistent framework is still lacking for the procedure of resilience analysis for general complex systems, and system responses and strains induced by multiple rounds of disruptions have not been well studied. In this manuscript, dynamic resilience is defined as a system's ability to resist loss of resilience and to adapt to successive resilience processes. We employ a four-factor measurement system, instead of a single-factor measurement, for the resilience analysis. A comprehensive framework for resilience assessment is proposed for dynamic resilience modeling in general complex systems to address various concerns in complex systems. A case study demonstrates the application of the proposed framework by simulating disruption intensity and recovery volume on a model communication system. We find that the assessment of dynamic resilience produces distinct results for different resilience aspects, while optimizations can help us identify solutions when all resilience factors are stabilized in the long-term dynamic resilience process. The dependability of the simulation results is verified using noise techniques in signal processing.