Files
Abstract
Complex multi-domain engineering systems are at the heart of modern warfare. The very nature of complexity means that the interactions between elements of such systems can lead to unforeseen consequences that are difficult to understand and predict. This is particularly true when there are varied types of disruption that can take place, such as component failure or deliberate attacks. The ability to analyse and assess how complex systems recover from disruption is critical for understanding resilience, especially as automated control design aspects are increasing. This paper proposes a triple-layer network methodology that is based on the physical, functional, and control layers of a complex system. The number of controllers and connections between controllers and functional nodes are varied for different design options, and resilience is evaluated. By identifying the control design features that have the greatest influence on resilience, the preferred design option can be chosen, ensuring that resilience meets the design objectives in the early stages with only the necessary redundancy elements. The method is suggested to be integrated into the overall process of designing high resilience monitored and controlled system architectures ultimately allowing to design for recoverability.