Life tests for highly reliable products often take a long time, even when using accelerated life testing with censoring. When the production process is monitored with lifetime as the key quality characteristic, the time spent on life testing can result in significant delays for practitioners in making decisions after sampling. However, shortening the test duration, which leads to excessive right-censored observations, inevitably reduces the test power for anomaly detection. In this talk, we will investigate the optimal design of censoring time in life tests when monitoring lifetime data with likelihood-based control charts. Both finite-sample analytical and large-sample asymptotic results are examined for type-I censored exponential lifetimes. We further generalize the results to other common lifetime distributions, including Weibull, lognormal, and gamma distributions. Our investigation uncovers the twofold impact of censoring time on the actual performance of control charts under various scenarios and provides valuable references for practitioners in setting more reasonable censoring times in life testing on real-world production lines.

Dr. Xun Xiao is currently a Lecturer in Statistics at the Dept. of Mathematics and Statistics, University of Otago, New Zealand. He received B.Sc. in Statistics from the University of Science and Technology of China in 2011 and Ph.D. degree from the Dept. of Systems Engineering and Engineering Management at City University of Hong Kong in 2016.  His current research focuses on industrial statistics and point process modelling.