Converting Dependent Models into Independent ones, Preserving Essential Features
暂无分享,去创建一个
Abstract : Let T denote the life length of a series system of n components having respective life lengths (T sub 1), ..., (T sub n), not necessarily independent. Necessary and sufficient conditions are given for the existence of a set of independent random variables (H sub I), I a subset of (1, ..., n), such that the life length of the original series system and the occurrence of its failure pattern (set of components whose simultaneous failure coincides with that of the system) have the same joint distribution as the life length of a derived series system of components having life lengths (H sub I) and the occurence of the corresponding failure pattern of the derived system. The authors also exhibit explicitly the distributions of these independent random variables (H sub I). This extends the results of Miller (University of Missouri Technical Report, 1975) while using more elementary methods.
[1] Douglas R. Miller. A Note on Independence of Multivariate Lifetimes in Competing Risks Models , 1977 .
[2] I. Olkin,et al. A Multivariate Exponential Distribution , 1967 .
[3] Larry Lee,et al. FAILURE RATE - A UNIFIED APPROACH , 1976 .
[4] Richard E. Barlow,et al. Statistical Theory of Reliability and Life Testing: Probability Models , 1976 .
[5] Albert W. Marshall,et al. Multivariate Distributions with Exponential Minimums , 1974 .