A Novel Reliability Analysis Method for Fuzzy Multi-State Systems Considering Correlation

Fuzzy multi-state systems (FMSSs) exist widely in practical engineering. It is usually difficult to evaluate the reliability of FMSSs because the reliability data is usually fuzzy due to the inaccuracy or imperfection of information, and there is often correlation between the main components and others which constitute the FMSSs. Although many research works with respect to the independent failure of components have been carried out, the master-slave relationship between the main components and others of the FMSSs is often ignored, thus unrealistic results are often obtained with this treatment. Based on fuzzy universal generating function (FUGF), an effective reliability analysis method of FMSSs considering the correlation and fuzziness is proposed in this paper. In the novel method, the fuzziness of reliability data and the master-slave relationship between the main and other components are taken into consideration, and the performance levels and corresponding probabilities of the non-main components are considered as conditional probability distributions. A case study with respect to the reliability analysis of hydraulic system is presented to illustrate the application of the proposed method.