In this article, we consider the problem of predicting survival times of units from the exponential distribution which are censored under a simple step-stress testing experiment. In this article, Type-II censoring are considered for the form of censoring. Two kinds of predictors—the maximum likelihood predictors (MLP) and the conditional median predictors (CMPs)—are derived. Some numerical examples are presented to illustrate the prediction methods developed here. Using simulation studies, prediction intervals are generated for these examples. We then compare the MLP and the CMP with respect to mean squared prediction error (MSPE) and the prediction interval. Finally, we used a real data to apply the prediction methods developed in the article.
NelsonW. Applied life data analysis. New York, NY: John Wiley & Sons; 1982.
2.
BalakrishnanNCohenAC. Order statistics and inference: Estimation methods. San Diego, CA: Academic Press; 1991.
3.
MillerRNelsonWB. Optimum simple step-stress plans for accelerated life testing. IEEE Trans Reliab. 1983; 32:59–65.
4.
BaiDSKimMSLeeSH. Optimum simple step-stress accelerated life test with censoring. IEEE Trans Reliab. 1989; 38:528–532.
5.
DeGrootMHGoelPK. Bayesian estimation and optimal design in partially accelerated life testing. Nav Res Logist Q. 1979; 26:223–235.
6.
KateriMBalakrishnanN. Inference for a simple step-stress model with Type-II censoring and Weibull distributed lifetimes. IEEE Trans Reliab. 2008; 57:616–626.
7.
XiongC. Inference on a simple step-stress model with Type-II censored exponential data. IEEE Trans Reliab. 1998; 47:142–146.
8.
BalakrishnanNHanD. Exact inference for a simple step-stress model with competing risks for failure from exponential distribution under Type-II censoring. J Stat Plan Inference. 2008; 138:4172–4136.
9.
BalakrishnanNKunduDNgHKTKannanN. Point and interval estimation for a simple step-stress model with Type-II censoring. J Qual Technol. 2007; 9:35–47.
10.
KaminskyKSNelsonPI. Prediction of order statistics. In: BalakrishnanNRaoCR, editors. Handbook of statistics, 17–Order statistics: Applications. Amsterdam: North-Holland; 1998. pp. 431–450.
11.
XiongCMillikenGA. Prediction for exponential lifetimes based on step-stress testing. Commun Stat-Simul Comput. 2002; 31:539–556.
12.
LeeJElmoreRKennedyCGrayMJonesW. Lifetime prediction for degradation of solar mirrors under step-stress accelerated testing. 2011 Workshop on Accelerated Stress Testing and Reliability, San Fransisco, USA; 2011.
13.
KaminskyKSRhodinLS. Maximum likelihood prediction. Annal Inst Stat Math. 1985; 37:707–717.
14.
BasakIBasakPBalakrishnanN. On some predictors of times to failure of censored items in progressively censored samples. Comput Stat Data Anal. 2006; 50:1313–1337.
15.
BasakI. Prediction of times to failure of censored items for a simple step-stress model with regular and progressive Type I censoring from the exponential distribution. Commun-Stat Theory Methods. 2014; 43: 2322–2341.
16.
RaqabMZ. Modied maximum likelihood predictors of future order statistics from normal samples. Comput Stat Data Anal. 1997; 25:91–106.
17.
BasakPBalakrishnanN. Maximum likelihood prediction of future record statistics. In: Series on Quality, Reliability and Engineering Statistics: Vol. 7 Mathematical & Statistical Methods in Reliability. 2003. pp. 159–175. New Jersey, USA.
18.
RaqabMZ. Approximate maximum likelihood predictors of future failure times of shifted exponential distribution under multiple Type-II censoring. Stat Methods Appl. 2004; 13:43–54.
19.
YangC-HTongL-I. Predicting Type-II censored data from factorial experiment using modifed maximum likelihood predictor. Int J Adv Manuf Technol. 2006; 30:887–896.
20.
RaqabMZNagarajaHN. On some predictors of future order statistics. Metron. 1995; 53(1): 185–204.
21.
GannounASaraccoJYuK. Nonparametric prediction by conditional median and quantiles. J Stat Plan Inference. 2003; 117:207–223.
22.
BasakIBalakrishnanN. Predictors of failure times of censored items in progressively censored samples from normal distribution. Sankhya Indian J Stat. 2009; 71-B, part 2:222–249.
23.
RaqabMZAhmadiJArabliBA. Comparisons among some predictors of exponential distributions using pitman closeness. Comput Stat. 2013; 28:2349–2356.
24.
WangRFeiH. Uniqueness of the maximum likelihood estimate of the Weibull distribution tampered failure rate model. Commun Stat-Theory Methods. 2003; 32:2321–2338.
25.
MaoSWangL. Accelerated life test (in Chinese). Vol.137. Ha Noi, Vietnam: Publishing House of Science. 1997. pp. 66–69.
26.
BalakrishnanN. A synthesis of exact inferential results for exponential step-stress models and associated optimal accelerated life-tests. Metrika. 2009; 69:351–396.
27.
DevoreJ. Probability and statistics for engineering and the sciences. Boston, USA: Cengage Learning; 2016.
28.
KannanNKunduDBalakrishnanN. Survival models for step-stress experiments with lagged effects. (Special volume dedicated to W. Meeker.) In: NikulinMLimniosNBalakrishnanN editors. Advances in degradation modeling. Boston, USA: Birkhäuser, 2010. pp. 355–369.
