Multiobjective optimization of reliability–redundancy allocation problems for serial parallel-series systems based on importance measure

Abstract

For complex equipment, designers are challenged to reduce the expense while satisfying high requirements of system reliability. This article focuses on the multiobjective reliability–redundancy allocation problem for serial parallel-series systems to balance the conflicts between system reliability and design cost. The multiobjective reliability–redundancy allocation problem model for serial parallel-series systems is established with constraints on system reliability and design cost. An importance measure–based and harmony search–based multiobjective particle swarm optimization algorithm is proposed to solve the multiobjective model effectively based on the importance measure–based harmony search and multiobjective particle swarm optimization algorithm. The performance of the importance measure–based and harmony search-based multiobjective particle swarm optimization algorithm is verified by comparison with the nondominated sorting genetic algorithm and importance measure–based multiobjective particle swarm optimization algorithm. In Experiment 1, the performance of the importance measure–based and harmony search-based multiobjective particle swarm optimization algorithm is better than that of the nondominated sorting genetic algorithm and importance measure–based multiobjective particle swarm optimization, and the importance measure–based and harmony search-based multiobjective particle swarm optimization algorithm also can get the Pareto front with better uniformity. Compared to the nondominated sorting genetic algorithm, four cases with different constraints of system reliability and design cost are considered in Experiment 2, and the importance measure–based and harmony search–based multiobjective particle swarm optimization algorithm applies to the systems with the lower system reliability constraints and the higher design cost constraints.

Maintenance optimization of reconfigurable systems based on multi-objective Birnbaum importance

“…These multi-objective problems are regarded as non-deterministic polynomial-time (NP)-hard problems, which should be solved by some evolutionary algorithms. 27,28 For reconfigurable systems, electrical distribution network reconfiguration with constraints of the system reliability and the power loss was optimized by an enhanced GA. 3 A robust optimization algorithm based on hybridization of a shuffled frog leaping algorithm and particle swarm optimization was developed to obtain the optimal distribution feeder reconfiguration, which could satisfy both operation and security expectations. 29…”

Minimal path set importance in complex systems

“…In the recent literature, numerous importance measures have been introduced for binary systems, and they are applicable in several phases of different systems design such as telecommunication, transportation, energy (including petroleum and nuclear), health, risk analysis, etc. An extensive review of importance measures can be found in Borgonovo, Kuo and Zhu, Amrutkar and Kamalja, and Zhao et al [1][2][3][4][5] In probabilistic safety assessment, Borgonovo 1 and Vaurio 6 considered the roles of several importance measures. While previous research has mainly concentrated on component importance measures, [7][8][9][10][11][12][13] the importance measure of minimal path sets, 14 as required in the context of complex systems, has received very little attention.…”

Stress-strength interference-based importance for series systems considering common cause failure

“…Natvig [23] raised the dynamic and stationary importance measures in repairable and nonreparable multistate coherent systems. Zhao et al [45] introduced the redundancy importance measure into the multi-objective optimization of reliability-redundancy allocation problems for serial parallel-series systems. Si et al [29] proposed the concept of integrated importance measure, which concerned the probability distributions and transition intensities of the component states simultaneously.…”