A new model for reliability optimization of series-parallel systems with non-homogeneous components. (January 2017)

• Record Type: Journal Article
• Authors: Feizabadi, Mohammad; Jahromi, Abdolhamid Eshraghniaye
• Is Part Of: Reliability engineering & system safety. Volume 157(2017)
• Page Start: 101
• Page End: 112
• Publication Date: 2017-01
• Subjects: Reliability optimization -- Redundancy allocation problem -- Series-parallel systems -- Non-homogeneous components -- Genetic algorithm; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2016.08.023
• Languages: English
• ISSNs: 0951-8320

Abstract

Abstract: In discussions related to reliability optimization using redundancy allocation, one of the structures that has attracted the attention of many researchers, is series-parallel structure. In models previously presented for reliability optimization of series-parallel systems, there is a restricting assumption based on which all components of a subsystem must be homogeneous. This constraint limits system designers in selecting components and prevents achieving higher levels of reliability. In this paper, a new model is proposed for reliability optimization of series-parallel systems, which makes possible the use of non-homogeneous components in each subsystem. As a result of this flexibility, the process of supplying system components will be easier. To solve the proposed model, since the redundancy allocation problem (RAP) belongs to the NP-hard class of optimization problems, a genetic algorithm (GA) is developed. The computational results of the designed GA are indicative of high performance of the proposed model in increasing system reliability and decreasing costs. Highlights: In this paper, a new model is proposed for reliability optimization of series-parallel systems. In the previous models, there is a restricting assumption based on which all components of a subsystem must be homogeneous. The presented model provides a possibility for the subsystems' components to be non- homogeneous in the required conditions. The computational results demonstrate the high performance of the proposed model in improving reliability and reducing costs.