Abstract
Any automated process can be said to be in state 0 (operating properly) or, if a malfunction occurs, to be in state 1 (not operating properly). Periodic inspection is necessary to make sure that the process is in state 0. We derive the minimum cost inspection interval for such a process with failure probability p, based on costs of inspection, repair, and nonconforming items. We also consider the case of an unknown failure probability and suggest a sequential procedure for estimating p and, simultaneously, the optimum inspection interval. Results of a simulation study show that the proposed methods work quite well.
Additional information
Notes on contributors
Hari K. Iyer
Dr. Iyer is an Associate Professor in the Department of Statistics and a Faculty Visitor at the National Institute of Standards and Technology.
Dominic F. Vecchia
Dr. Vecchia is a Mathematical Statistician in the Statistical Engineering Division.