Solution methods for reliability optimization of a series system with component choices

Reliability has been considered as an important design measure in various industrial systems. We discuss a reliability optimization problem with component choices (ROP-CC) subject to a budget constraint. This problem has been known as an NP-hard problem in the reliability design fields. Several researchers have been working to find the optimal solution through different heuristic methods. In this paper, we describe our development of simulated annealing and tabu search algorithms and a reoptimization procedure of the two algorithms for solving the problem. Experimental results for some examples are shown to evaluate the performance of these methods. We compare the results with the solutions of a previous study which used ant system and the global optimal solution of each example obtained through an optimization package, CPLEX 9.1. The computational results indicate that the developed algorithms outperform the previous results.