Homogeneous redundancy optimization in multi-state series-parallel systems: A heuristic approach

Abstract

This paper presents a heuristic for a series-parallel system, exhibiting a multi-state behavior, minimizing the cost in order to provide a desired level of reliability. System reliability is defined as the ability to satisfy consumer demands which is presented as a piecewise cumulative load curve. The components are binary and chosen from a list of products available on the market, and are characterized in terms of their feeding capacity, reliability and cost. The solution approach makes use of a homogeneous collection of components to provide redundancy in a subsystem. The algorithm is applied to power systems found in the literature for various levels of reliability requirement. The heuristic offers a straightforward analysis and is efficient both in terms of solution quality and computational time, as compared to existing genetic algorithms and heuristics. Thus, the developed heuristic is attractive, and it can be easily and efficiently applied to numerous real-life systems.

Keywords:

• Multi-state
• series-parallel system
• redundancy optimization
• heuristic approach

Acknowledgement

The authors wish to thank the reviewers for their valuable suggestions to improve this paper.

Notes

*The solution obtained by RAMC as published in [Ramirez-Marquez and Coit (2004)]: x ₁₂ = 2, x ₂₅ = 6, x ₃₂ = 3, x ₄₇ = 3, x ₅₂ = 1, Cost: 16.52, Reliability: 0.980 7596; is incorrect. It is as given above. See Agarwal et al. (2006).

*As mentioned by Lisnianski et al. (1996), the running time of the GA on a DEC station 5000/240 did not exceed 30 seconds for a single optimization problem.