Strategy selection for multi-objective redundancy allocation problem in a k-out-of-n system considering the mean time to failure

This paper presents a redundancy allocation problem (RAP) to maximize the mean time to failure (MTTF) and minimize the costs for a k-out-of-n system with several subsystems. In this model, each subsystem requires at least K flawless components to be active. The components can be configured in parallel, standby or single modes. However, several types of components can be selected in each subsystem. The importance of this study is due to selecting the conditions in which the system is at its best considering the trade-off between total cost and reliability. The main contribution of the article is to propose a more realistic multi-objective model in which the MTTF of the system, weight and volume limitations, and the mixed redundancy strategy (including the cold standby RAP) are considered to achieve the optimal strategy for the configuration of components in a series system with several subsystems. Beside these contributions, four multi-objective metaheuristic algorithms are applied to solve the proposed model. The algorithms are compared by C (L, B) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) methods. According to the results, the non-dominated sorting genetic algorithm (NSGA-II) is selected as the superior algorithm. Finally, by comparing the selected final solutions of the algorithms in different weights of the TOPSIS method, it was found that the parallel strategy has the least solutions in the final result among all algorithms. Besides, the best solutions with the highest MTTF and the lowest cost are related to more use of the cold-standby strategy. As a result, configuring components with more use of the cold-standby strategy and combined with the single-component strategy can provide better conditions for the system.