Peer-reviewedCC BY-NCReliability: Theory & Applications
Two-level Factorial Life Testing with Type-II Censored Data
- 1Beer Sheva, Israel
Abstract
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.
Related
- Goodness-of-fit testing for the Pareto type-I distribution based on a mean residual life characterization
- Practical Guidelines for Goodness-of-Fit Testing for Right-Censored Data
- Testing for Unobserved Heterogeneity in Censored Duration Models: EM Approach
- Multi-Absorbing Phase-Type Distributions for Right-Censored Competing Risks Data
- Relative Variability Estimation for the Power Lindley Distribution with Progressive Type-I Interval Censored Data
BibTeX
@article{gertsbakh2026twolevel,
title = {{Two-level Factorial Life Testing with Type-II Censored Data}},
author = {Ilya Gertsbakh},
journal = {Reliability: Theory \& Applications},
year = {2026},
month = sep,
doi = {10.66721/7wk19v40},
url = {https://doi.org/10.66721/7wk19v40},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Two-level Factorial Life Testing with Type-II Censored Data
AU - Ilya Gertsbakh
PY - 2026
DA - 2026/09/01
JO - Reliability: Theory & Applications
DO - 10.66721/7wk19v40
UR - https://doi.org/10.66721/7wk19v40
AB - 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.
SN - 1932-2321
ER -