Three computational intelligence methods for system reliability

Abstract

Nowadays, the competitiveness in the industrial world became more and more harsh, which requires that the system must be as reliable as possible. In many optimization problems, hard fitness functions are considered. Those functions can- not be solved by the traditional mathematical programming methods. An alternative solution to the conventional approaches. The meta-heuristic optimization tech- niques are used, due to their ability to obtain global or near-global optimum solu- tions. In the present paper, we address the system reliability-redundancy allocation optimization problem, using three well known algorithms namely, the Cuckoo Search (CS), Particle Swarm Optimization (PSO), and the Fractal Search (SFS). The constraints defined here in the problem are handled with the help of the penalty function method. The results of the numerical case study are compared for high- lighting the superiority of an algorithm over another.