Two-level Factorial Life Testing with Type-II Censored Data
DOI:
https://doi.org/10.66721/7wk19v40Keywords:
Orthogonal design, type - II censoring, location-scale family, order statistics, life testingAbstract
We consider two-level factorial designs with the response being the observed item’s lifetime. For each factor combination, we put on test n items and stop testing when exactly r items have failed. Our principal assumptions are that there exists a monotone transformation of the random response which belongs to a location-scale family, and that only the location parameter linearly depends on the factors involved. We develop a simple time-saving testing scheme which permits an efficient computational procedure for estimating the factor effects.
References
George E. Box and N.R. Draper. Model Building and Response Surfaces, Wiley & Sons, 1987.
R.G. Bullington et al. Improvement of an industrial thermostat using designed experiments. J.Qual. Tech., 25, 1993, 262-270.
I. Gertsbakh. Statistical Reliability Theory. Marcel Dekker, 1989.
M.Hamada. Using statistically designed experiments to improve reliability and achieve robust reliability.IEEE Trans. Reliab., 1995, 44, 206-215.
M. Hamada and C.F. Wu. Analysis of censored data from highly fractioned experiments. Technometrics, 33, 1991, 25-38.
M.Hamada and S.K. Tse. On estimability problem in industrial experiments with censored data. Statistica Sinica, 2, 1992, 381-391.
R.A.Johnson and D.W. Wichern. Applied Multivariate Statistical Analysis, Prentice Hall, 1982
W. Nelson. Accelerated Life Testing. Wiley, 1990.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
