Statistical Inference for Repairable Systems Using the Power Law Process

Abstract

The nonhomogeneous Poisson process with intensity function, u(t) = (β/θ)(t/θ)^β–1, is called the power law process and is frequently used to model the failure times of repairable systems. A FORTRAN program is presented which gives (1) point and interval estimates for β, (2) a point estimate for θ, (3) point and interval estimates for the intensity function at the time of the conclusion of testing, and (4) a goodness-of-fit test. The program applies when there is a single system and when there are several identical systems.

• Poisson Process
• Power Law Process
• Weibull Process

Additional information

Notes on contributors

Steven E. Rigdon

Dr. Rigdon is Professor and Chair of the Department of Mathematics and Statistics. He is a Senior Member of ASQ.

Xiaolin Ma

Mr. Ma is a graduate student in the Department of Mathematics and Statistics.

Kevin M. Bodden

Mr. Bodden is a graduate student in the Department of Industrial Engineering. He is a Member of ASQ.