Cost-based importance genetic algorithm for system reliability optimization

As the counterpart to conventional reliability optimization problems (ROP), the cost minimization reliability optimization problem (CMROP) plays a critical role within the domain of ROPs. Given the escalating demands for product reliability, CMROP has emerged as a focal point in reliability optimization research. This area of study concentrates on enhancing system reliability efficiently while minimizing costs to meet specific requirements. This paper introduces a cost-importance-based genetic algorithm (CIGA) designed to enhance the efficiency of the cost-minimization reliability optimization model. The algorithm integrates a cost-based importance measure with a local search rule and a genetic algorithm. The validity of this algorithm is corroborated through its application to several typical system structures. Extensive numerical experiments are conducted to evaluate the performance of CIGA in comparison to traditional genetic algorithms (GA) and BIGA across five aspects of the solutions. The results demonstrate the superiority of the proposed algorithm in addressing cost minimization reliability optimization problems, showcasing its enhanced efficacy over the competing algorithms.