Search

This article applies general engineering rules for describing the reliability of devices working under variable stresses. The approach is based on imposing completeness and physicality. Completeness refers to the model's capability for studying as many stated conditions as possible, and physicality refers to the model's capability for incorporating explanatory variables specified and related each other by the physical laws. The proposed reliability model has as many explanatory variables as necessary but only three unknown parameters, and hence, it allows the engineer to collect reliability data from different tests campaigns, and to extrapolate reliability results towards other operational and design points. [ABSTRACT FROM AUTHOR]

In this article, we consider the change-point hazard rate model which arises quite commonly in mechanical or biological systems, which experience a high hazard rate early in their lifetime due to infant mortality and then a constant or steady hazard rate after the threshold time. We first derive the corresponding mean residual life function (MRLF) and observe that the MRLF is initially increasing and then constant. Here, we derive a test statistic for exponentiality against Increasing Initially then Constant Mean Residual Life (ICMRL). We also derive the asymptotic distribution of the test statistic and compare the power of the test with other existing tests such as likelihood ratio, Weibull, and Log gamma tests considered in the literature. The test performs quite well as compared to other alternatives studied. [ABSTRACT FROM AUTHOR]

The Weibull distribution is composited with Pareto model to obtain a flexible, reliable long-tailed parametric distribution for modeling unimodal failure rate data. The hazard function of the composite family accommodates decreasing and unimodal failure rates, which are separated by the boundary line of the space of shape parameter, gamma, when it equals to a known constant. The least square and maximum likelihood parameter estimation techniques are discussed. The advantages of using the proposed family are demonstrated and compared by illustrating well-known examples: guinea pigs survival time data, head and neck cancer data, and nasopharynx cancer survival data. [ABSTRACT FROM AUTHOR]

A two-parameter family of lifetime distribution which is derived from a model for static fatigue is presented. This derivation follows from considerations of the relationship between static fatigue crack extension and the failure time of a certain specimen. The cumulative distribution function (cdf) of this new family is quite similar to the cdf of the half-normal distribution, and therefore this density is referred to as the generalized half-normal distribution (GHN). Furthermore, this GHN family is a special case of the three-parameter generalized gamma distribution. Even though the GHN distribution is a two-parameter distribution, the hazard rate function can form variety of shapes such as monotonically increasing, monotonically decreasing, and bathtub shapes. Some properties of this family are given, and examples are cited to compare with other commonly used failure time distributions such as Weibull, gamma, lognormal, and Birnbaum-Saunders. [ABSTRACT FROM AUTHOR]

The usual likelihood used to make inferences about parameters of lifetime distributions when lives may be censored consists of the product of probability densities for completed lives and reliabilities for the censored lives. This works well in many situations. We consider two special cases of a model which assumes a Poisson birth process for the items: in one case the usual method works well, but in the other case it does not, and better estimates are found. [ABSTRACT FROM AUTHOR]

In finite mixture models, maximum likelihood estimators have good properties, such as efficiency, consistency, and asymptotic normality under some uniform integrability assumptions on the mixture and its derivatives up to the third order. These conditions are frequently not easy to check because complex computations on bounding a lot of derivatives are involved. We give results implying these conditions for a new class of families of distributions, W-type families, which make it easier to check the conditions in many cases. Many useful and known families of distributions such as Weibull, Generalized Gamma, Log-gamma, inverse Log-gamma, inverse Gaussian, and all of the exponential families are W-type families. Hence, these results have broad applications. [ABSTRACT FROM AUTHOR]

We consider a two-way classification model with interaction and assume that the errors have a location-scale nonnormal distribution. From an application of the modified likelihood estimation, we obtain efficient and robust estimators of the parameters. We define F statistics for testing main effects and interaction. We analyze the Box-Cox data and show that the method developed in this paper gives accurate results besides being easy theoretically and computationally. [ABSTRACT FROM AUTHOR]

In a linear regression model of the type y = θX + e, it is often assumed that the random error e is normally distributed. In numerous situations, e.g., when y measures life times or reaction times, e typically has a skew distribution. We consider two important families of skew distributions, (a) Weibull with support IR: (0, ∞) on the real line, and (b) generalised logistic with support IR: (-∞, ∞). Since the maximum likelihood estimators are intractable in these situations, we derive modified likelihood estimators which have explicit algebraic forms and are, therefore, easy to compute. We show that these estimators are remarkably efficient, and robust. We develop hypothesis testing procedures and give a real life example. Symmetric families of distributions, both long and short tailed, will be considered in a future paper. [ABSTRACT FROM AUTHOR]