Return to search

Reliability and availability analysis of a multistate repairable system with dependent deteriorations and redundancy

Maintenance management is to design, operate, and maintain the reliability and availability of assets at a required performance level using the lowest possible cost. The standby redundancy is one of the means to achieve highly reliable system with less dependable units. As commonly used performance indicator, the reliability and availability should be precisely analysed for a repairable standby system. The reliability of a standby system is mainly studied in the framework of lifetime approach. Most existing models are developed for a two-unit standby system and a Kout- of-N system with identical units. The system units are assumed to be binary-state and the failures of system units are modelled as sudden failures. The deteriorations of units are modelled as time-dependent failure rate and are assumed to be independent. In addition, most of the existing models do not consider maintenance is carried out on the system. In reality, the deteriorations and the resultant failures in real-world systems are often interactive with each other. The systems normally experience several, often imperfect, restorations before a complete renewal. Therefore, the study of different repair policies, such as opportunistic and individual policies, in multi-unit systems need also be investigated. To address the problem, the reliability and availability models for a 2-out-of-3 cold standby system will be studied in a multi-state system reliability framework. A multistate multi-path failure mode is proposed to model the interactive deteriorations. The concept of repair matrix will be adopted to model the effect of unit level restorations on system reliability. The availabilities under individual repair policy and opportunistic repair policy will be developed.

Identiferoai:union.ndltd.org:ADTP/265796
Date January 2008
CreatorsMu, Dekui
PublisherQueensland University of Technology
Source SetsAustraliasian Digital Theses Program
Detected LanguageEnglish

Page generated in 0.002 seconds