Optimal burn-in procedure for mixed populations based on the device degradation process history

• Published in: European journal of operational research Vol. 251; no. 3; pp. 988 - 998
• Main Authors: Cha, Ji Hwan, Pulcini, Gianpaolo
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 16.06.2016; Elsevier Sequoia S.A
• Subjects: Burn-in; Cost optimization; Degradation; Degradation process; Demography; Failure; Infant mortality; Mathematical models; Mixed population; Non-homogeneous gamma process; Numerical analysis; Optimization; Parameter optimization; Populations; Research methodology; Simulation; Studies; Burn-in; Degradation process; Cost optimization; Mixed population; Non-homogeneous gamma process
• ISSN: 0377-2217; 1872-6860
• DOI: 10.1016/j.ejor.2015.12.019

•This paper proposes a new degradation history-based burn-in procedure.•The proposed procedure utilizes degradation process history of the item observed during burn-in.•This paper provides a new ‘general insight/perspective’ on the burn-in models.•This work opens new possibilities for further developments of similar approaches. Burn-in is a method of ‘elimination’ of initial failures (infant mortality). In the conventional burn-in procedures, to burn-in an item means to subject it to a fixed time period of simulated use prior to actual operation. Then, the items which failed during burn-in are just scrapped and only those which survived the burn-in procedure are considered to be of satisfactory quality. Thus, when the items are subject to degradation phenomena, those whose degradation levels at the end of burn-in exceed a given failure threshold level are eliminated. In this paper, we consider a new burn-in procedure for items subject to degradation phenomena and belonging to mixed populations composed of a weak and a strong subpopulation. The new procedure is based on the ‘whole history’ of the degradation process of an item periodically observed during the burn-in and utilizes the information contained in the observed degradation process to assess whether the item belongs to the strong or weak subpopulation. The problem of determining the optimal burn-in parameters is considered and the properties of the optimal parameters are derived. A numerical example is also provided to illustrate the theoretical results obtained in this paper.