Enhancing the resilience of combat system-of-systems through a two-phase recovery framework

A combat system-of-systems (CSoS) comprises various types of functional combat entities that interact to meet corresponding task requirements in the present and future, such as neutralizing enemy forces and building advanced military capabilities. Enhancing the resilience of CSoS is crucial for optimizing network structure, improving survivability and fulfilling task requirements. Accordingly, we propose an integrated framework called CSoS resilience enhancement (CSoSRE) to enhance the resilience of CSoS from a recovery perspective. Specifically, we present a two-phase resilience optimization model and provide the resilience optimization objective for CSoS. This model accounts for task baseline, recovery cost, time constraints, and both emergency and comprehensive recovery characteristics. Moreover, we extend it from the deterministic case to the stochastic case to describe the uncertainty in the recovery process. Based on this, a resilience-oriented recovery optimization method based on deep reinforcement learning (RRODRL) is proposed to determine a set of entities requiring restoration and their recovery sequence. This method introduces a phase-adaptive discount factor, effectively distinguishing and adapting to different recovery phases. Finally, extensive experiments are conducted to test the feasibility, effectiveness and superiority of the proposed CSoSRE. The results offer valuable insights to inform the operation and design of a more resilient CSoS.