Resilience evaluation of multi-feature system based on hidden Markov model

Modern systems have become increasingly vulnerable to threats due to their growing complexity nowadays. Multi-feature systems, prevalent in the realm of complex structures, manifest their performance through a diverse array of features. In response to threats, this paper develops a resilience evaluation model for multi- feature systems based on hidden Markov models, which can describe the dynamic relationship between performance levels and external features. Quantitative resilience indicators are presented across three distinct dimensions: resistant, absorption, and recovery, whose analytical formulas are derived by generating functions and properties are proved. Meanwhile, simulation algorithms are proposed to verify the correctness of the analytic formulas. Finally, taking the system under the threat of flood disasters as an example, the resilience model proposed in this paper is applied to evaluate its resilience, and the robustness of the resilience evaluation indicators is verified.