Accelerated Life Test Planning With Independent Weibull Competing Risks

Author

Pascual, Francis

Author_Institution

Dept. of Math., Washington State Univ., Pullman, WA

Volume

57

Issue

3

fYear

2008

Firstpage

435

Lastpage

444

Abstract

This article studies the planning of accelerated life tests (ALT) in the presence of competing risks. The time to failure due to a specific risk is described by a Weibull model. These failure times of risks are assumed to be s-independent, and the minimum of these times is the life of the product under test. Expressions for the Fisher information matrix, often used in ALT planning, are presented. Minimum asymptotic variance of estimators and determinant-based criteria are discussed, and related to goals of testing. The proposed method is applied to the planning of ALT of Class-B insulation of motorettes.

Keywords

Weibull distribution; electric motors; failure analysis; insulation testing; life testing; machine insulation; machine testing; matrix algebra; planning; risk analysis; Class-B motorettes insulation; Fisher information matrix; accelerated life test planning; independent Weibull competing risks; minimum asymptotic variance; time to failure; $D$-optimality; $D_{s}$-optimality; Arrhenius relationship; Fisher information; asymptotic variance; maximum likelihood estimation; optimal plans; time-failure censoring;

fLanguage

English

Journal_Title

Reliability, IEEE Transactions on

Publisher

ieee

ISSN

0018-9529

Type

jour

DOI

10.1109/TR.2008.928205

Filename

4585403