Published: 2008-12-09
ISBN:
ISSN: 1650-3686 (print), 1650-3740 (online)
Purpose: This work proposes an innovative control chart of the Weibull percentiles using Bayesian estimators supported by bootstrap methods.
Approach: The chart offers two main advantages.
On one side; the estimation procedure is able to effectively integrate both the experimental and the technological information exploiting some specific Bayesian estimators.
On the other side; the bootstrap techniques allow to capitalize the experimental information provided by few samples.
Findings: The performance of the control chart has been investigated by means of a large Monte Carlo study.
Value of the paper: The paper presents a control chart for Weibull percentiles; where few alternative charts can be found.
Statistical Process Control; non-Normal control charts; Bayesian inference; Weibull distribution; Bootstrap methods