Optimal allocation of change points in simple step-stress experiments under Type-II censoring
Introduction
In accelerated life testing (ALT) applications, in order to speed up the testing procedure and to reduce the time to failure, the experimental units are exposed to higher stress levels than normal. The data from such an ALT are then transformed to estimate the distribution of failures under normal operating condition. Primary references on ALT are (among others) Nelson (1990), Meeker and Escobar (1998) and Bagdonavicius and Nikulin (2002).
One popular ALT model is the step-stress model, where the stress put on the testing units is not constant and is varied at two or more ordered levels. The simple step-stress model, which plays a notable role in accelerated life testing, consists of two stages and allows for testing under two different conditions at these stages of the experiment (usually referred to as stress levels). Interest lies on inference for the expected mean lifetimes and at these stress levels, when the underlying distributions are exponential. For details pertinent to inferential and optimization aspects in the simple step-stress model under Type-II censoring for exponentially distributed lifetimes, we refer to Balakrishnan et al. (2007), Bai et al. (1989), and Miller and Nelson (1983). Wang (2009) worked under the assumptions of the exponential step-stress model, Type-I or Type-II censored, while Abdel-Hamid and Al-Hussaini (2009) considered a more general life distribution under Type-I censoring. A detailed survey of all developments on step-stress models has been provided by Balakrishnan (2009). Recently, Kateri et al. (2009) introduced a meta-analysis approach for multiple step-stress experiments. Under this more general model, independent step-stress experiments are carried out for the same population and under the same stress levels. What can vary among the samples, thus giving a flexibility from a design point of view, are their sizes, the proportion of censoring in each of these samples, and the time points at which the stress levels are changed.
The multi-sample step-stress set-up may, therefore, be viewed as a framework to analyze different experiments in the sense of a meta-analysis as well as the basis to plan experiments in a reasonable or, even better, in an optimal way. Here, we tackle this specific problem and discuss the optimal allocation of change points for the characteristic case of simple step-stress experiments.
In this paper, we focus on the special case of two independent samples and study its features, along with those of the one-sample simple step-stress model in the context of designing an ALT experiment. In terms of experimental design, it makes sense to consider a one-sample experiment with a sample of units and a total of observed failures as opposed to a two-sample experiment with sample sizes and corresponding number of failures being such that and . Given the total sample size , which is often fairly small, a choice of sub-samples may be unrealistic. However, the effects and issues pointed out here in the two-sample case remain valid and the findings may be extended to this general case as well.
One problem with step-stress models is that the MLE may not always exist. This happens when there is no observation under some particular stress level. In the one- and two-sample cases, we derive and analyze the probabilities of non-existence of the MLE and then discuss optimal allocations of the change points by minimizing the non-existence probabilities of the MLE of the parameters and .
Optimal design of step-stress tests could also be based on other criteria chosen by the experimenter. Standard criteria are Var-optimality and D-optimality, that minimize the asymptotic variance of the MLE of the mean life at a specified design stress or the determinant of the information matrix, respectively. However, since existence of the MLEs is of particular interest, we focus here on minimizing the non-existence probabilities of the MLEs. In a general step-stress set-up, due to the large number of parameters, it will not be possible to just provide tables of respective optimal plans even in one-criteria situations. A careful consideration and study of several plots as those presented in the subsequent sections will be necessary to make concrete decisions.
The structure of the paper is as follows. In Section 2, the two-sample experiment is introduced along with the MLE in the one- and the two-sample set-ups. In Section 3, the probabilities of non-existence of the MLE of and are derived for the one- and the two-sample experiments, and these probabilities are then examined, compared and interpreted. Finally, in Section 4, we address the optimal allocation of the change points and point out the importance of a good choice of the change points in the design of such a step-stress experiment.
Section snippets
Preliminaries
We restrict ourselves here to introducing the notation in the two-sample case, since the corresponding one for the multi-sample case becomes apparent. Let two independent samples of sizes and be placed on a life test, subjected to an initial stress level of . During the experimental period, the stress level is increased to at pre-fixed change point for the sample, for . Moreover, under Type-II censoring, the experiment is terminated when a pre-fixed number of
Probabilities of non-existence of the MLE
The goal of a simple step-stress experiment is to estimate the mean lifetimes and of the units under stress levels and , respectively. In the one-sample case, both MLE exist only if failures are observed at each stress level. For the two-sample experiment, if for both samples (), then the test is terminated without entering the second stress level and consequently the MLE does not exist for the mean lifetime under the second stress level . Otherwise, in at least one
Optimal allocation of change points
The probabilities of non-existence of and for the one- and two-sample simple step-stress experiments have been derived and discussed in the preceding section. In planning an experiment of either type, we would like to minimize the non-existence probabilities. Hence, we would like to control the total probability of non-existence of and and keep it to a minimum. In other words, the worst case of not having any estimate is intended to be minimum in probability.
The total
Discussion
In this paper, we have studied the non-existence probabilities of the MLE and and compared them for the one-sample and the two-sample simple step-stress experiments. Furthermore, we have shown that these probabilities of non-existence of the MLE can be minimized by appropriately allocating the points at which the stress levels are changed. It has been illustrated via a simulation study that this allocation of the change points leads to estimates of acceptable precision in terms of both
Acknowledgements
The authors are grateful to the associate editor and the referees for their helpful comments and suggestions, which led to a clearer presentation.
References (11)
- et al.
Estimation in step-stress accelerated life tests for the exponentiated exponential distribution with type-I censoring
Computational Statistics & Data Analysis
(2009) - et al.
A meta-analysis approach for step-stress experiments
Journal of Statistical Planning and Inference
(2009) Testing for the validity of the assumptions in the exponential step-stress accelerated life-testing model
Computational Statistics & Data Analysis
(2009)- et al.
Accelerated Life Models: Modeling and Statistical Analysis
(2002) - et al.
Optimum simple step-stress accelerated life test with censoring
IEEE Transactions on Reliability
(1989)
Cited by (13)
A new statistical inference method for multi-stress accelerated life testing based on random variable transformation
2021, Applied Mathematical ModellingCitation Excerpt :As a solution, the censoring scheme is usually used in ALT to save time and expense. Generally, there are two types of censoring schemes reported in the literature [7–10]: Type-I censoring and Type-II censoring, where the former terminates the test once the pre-fixed time is reached, and the latter terminates the test once the pre-specified number of faulty units is reached. Recently, progressively censoring, a natural extension of the above censoring schemes, is becoming popular due to its flexibility.
The step-stress tampered failure rate model under interval monitoring
2015, Statistical MethodologyMeta-analysis of general step-stress experiments under repeated Type-II censoring
2015, Applied Mathematical ModellingCitation Excerpt :The baseline distribution function F, which is assumed to be absolutely continuous, can be arbitrarily chosen. Step-stress models with multiple samples under Type-I and Type-II censoring allowing for meta-analysis of single step-stress experiments have recently been introduced and studied in Kateri et al. [35,12,36] focussing on MLEs of exponential parameters and respective properties. The multi-sample set-up consisting of simple step-stress tests with pre-specified numbers of observations under each stress level, and underlying exponential distributions is briefly mentioned in Kateri et al. [37].
Optimal acceptance sampling plans for log-locationscale lifetime models using average risks
2012, Computational Statistics and Data AnalysisCitation Excerpt :This type of censoring has been widely treated in the literature. Papers by Bhattacharyya (1985), Schneider (1989), Ebrahimi and Soofi (1990), Leslie and van Eeden (1993), Xu and Yang (1995), Wu et al. (2001), Xiong et al. (2003), Jun et al. (2006), Ismail (2009), Fernández (2010a,b,c), Kateri et al. (2011), Khan et al. (2011), and Lemonte and Ferrari (2011) are just a sample. The remainder of this paper is organized as follows.
A literature review on planning and analysis of multi-stress accelerated life test for reliability assessment
2021, Recent Patents on EngineeringMeta-analysis of a step-stress experiment under Weibull distribution
2021, Journal of Statistical Computation and Simulation
- 1
Visiting Professor at King Saud University (Riyadh, Saudi Arabia) and National Central University (Taiwan).