Premium
Bayesian D‐optimal Designs for Weibull Distribution with Censoring
Author(s) -
Zohbi Ibrahim,
Wainakh Mohieldin,
Arafeh Hamzeh
Publication year - 2016
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.1964
Subject(s) - weibull distribution , censoring (clinical trials) , bayesian probability , monte carlo method , robustness (evolution) , computer science , optimal design , statistics , reliability engineering , algorithm , mathematics , engineering , biochemistry , chemistry , gene
Bayesian optimal designs have received increasing attention in recent years, especially in reliability experiments. Weibull distribution is one of the most widely used in reliability analysis, owing to its various shapes for the probability density function and its convenient representation of lifetime data, particularly when data are censored, which is quite common in most life‐testing experiments. This paper provides analytical characterizations of Bayesian D‐optimal designs for Weibull distribution, as well as proposes an algorithm for finding a Bayesian D‐optimal design that combines the point exchange algorithm with the Monte Carlo simulation method. Different censoring mechanisms are incorporated, and the robustness of designs against parameters' mis‐specification is assessed. Copyright © 2016 John Wiley & Sons, Ltd.