Statistical Distributions in Scientific Work pp 309-317 | Cite as

# Failure Time Distributions: Estimates and Asymptotic Results

## Summary

The paper deals with life distributions for coherent systems of components. Two major questions are discussed: (i) estimation of system life in terms of component lives, and (ii) asymptotic models. Both questions are related to extremes of a sequence of random variables through the path set and cut set decomposition of coherent systems, which reduce a coherent system to either a parallel or a series system. It is pointed out that the classical theory of extremes of independent and identically distributed random variables does not provide an acceptable approximation. Hence, the emphasis is on dependence or on the case when the random variables are not identically distributed. The inequalities presented when discussing question (i) are applicable not only to extreme value problems but to an arbitrary multivariate distribution when lower dimensional marginals are specified. The asymptotic models are also discussed in the light of hazard rate properties of the limiting distributions for the extremes in a particular model.

### Key Words

coherent system component life system life extreme value distribution dependent model inequalities multivariate distribution Weibull distribution hazard rate## Preview

Unable to display preview. Download preview PDF.

### References

- Ahsanullah, M. (1977). A characteristic property of the exponential distribution.
*Annals of Statistics*, 5, 580–582.MathSciNetMATHCrossRefGoogle Scholar - Barlow, R.E., Proschan, F. (1975).
*Statistical Theory of Reliability and Life Testing: Probability Models*. Holt, Rinehart and Winston, New York.Google Scholar - Esary, J.D., Marshall, A.W., Proschan, F. (1971). Determining an approximate constant failure rate for a system whose components have constant failure rates. In
*Operations Research and Reliability*, D. Grouchko, ed. Gordon and Breach, New York.Google Scholar - Galambos, J. (1975). Methods for proving Bonferroni type inequalities.
*Journal of the London Mathematical Society, Series**2*, 9, 561–564.MathSciNetMATHCrossRefGoogle Scholar - Galambos, J. (1977). Bonferroni inequalities.
*Annals of Probability*, 4, 577–581.MathSciNetCrossRefGoogle Scholar - Galambos, J. (1978).
*The Asumptotic Theory of Extreme Order Statistics*. Wiley, New York.Google Scholar - Galambos, J. (1981). Extreme value theory in applied probability.
*The Mathematical Scientist*, 6, 13–26.MathSciNetMATHGoogle Scholar - Galambos, J., Kotz, S. (1978).
*Characterizations of Probability Distributions*. Lecture Notes in Mathematics Series, Vol. 675, Springer Verlag, Heidelberg.Google Scholar - Galambos, J., Mucci, R. (1980). Inequalities for linear combinations of binomial, moments.
*Publicationes Mathematicae Debrecen*, 27, to appear.Google Scholar - Kounias, E.G. (1968). Bounds for the probability of a union, with applications.
*Annals of Mathematical Statistics*, 39, 2154–2158.MathSciNetMATHCrossRefGoogle Scholar - Kwerel, S.M. (1975). Most stringent bounds on aggregated probabilities of partially specified dependent systems.
*Journal of the American Statistical Association*, 70, 472–479.MathSciNetMATHCrossRefGoogle Scholar - Natvig, B. (1980). Improved bounds for the availability and unavailability in a fixed time interval for systems of maintained, interdependent components.
*Advances in Applied Probability*, 12, 200–221.MathSciNetMATHCrossRefGoogle Scholar - Sathe, Y.S., Pradhan, M., Shah, S.P. (1980). Inequalities for the probability of the occurrence of at least m out of n events.
*Journal of Applied Probability*, to appear.Google Scholar - Zidek, J.V., Navin, F.P.D., Lockhart, R. (1979). Statistics of extremes; an alternate method with application to bridge design codes.
*Technometrics*, 21, 185–191.MathSciNetMATHCrossRefGoogle Scholar