Reliability optimization for dynamic configurable system with robust redundancy strategy

The k-out-of-n system, a representative redundancy configuration, finds broad application across engineered systems. These systems are interconnected via a flexible switching mechanism that enables dynamic adjustment of redundant components in response to failures, according to task requirements. This adaptability is achieved by regulating auxiliary elements situated at component interfaces, such as switches and valves. Although previous studies have demonstrated the effectiveness of optimizing system structures using active, standby, or mixed redundancy strategies, the critical role of the switching mechanism among redundant components has often been overlooked, leading to suboptimal system configurations. To address this limitation, a robust redundancy strategy is introduced for reliability optimization in systems that can be configured dynamically. This strategy adaptively reallocates redundant components and is defined as robust because it may specify the number of operational components for each subsystem as well as the optimal switching instants. The optimization problem is represented by a nonlinear mixed-integer programming model, which is solved by a designed genetic algorithm. A case study involving the backup scheme of a fuel supply system at a marine vessel is showcased to demonstrate the effectiveness of the proposed strategy.