Fuzzy System Reliability Analysis Using Triangular Fuzzy Numbers Based on Statistical Data

In this article, we use the fuzzy concept to consider the reliability of serial system and the reliability of parallel system. Since the population reliability R(subscript j) of the subsystem P(subscript j) (j=1, 2, …, n) is unknown, if we use the point estimate (average)R(subscript j) to estimate R(subscript j) from the statistical data in the past, we don't know the probability of the error (average)R(subscript j)-R(subscript j). Moreover, the reliability of the system may fluctuate around the point estimate (average)R(subscript j) during a time interval. It follows that to use the point estimate (average)R(subscript j) to estimate the population reliability R(subscript j) is not suitable for the real cases. Therefore, it is more desirable to use the statistical confidence interval. Moreover, the probability of the error (average)R(subscript j)-R(subscript j) can also be solved. In this paper, we use the statistical confidence interval instead of the point estimate. We transfer the statistical confidence interval into the triangular fuzzy number. Through these triangular fuzzy numbers, we consider the fuzzy reliability system. We fuzzify the reliability of both the serial and parallel systems. Through defuzzifying the fuzzy reliability using the signed distance method; we get a fuzzy estimate of reliability in the two systems.

並列關鍵字

fuzzy reliability ； statistical data ； signed distance ； i-v fuzzy number ； triangular fuzzy number

國際替代計量

全文下載

主題瀏覽