Evaluating Equipment Reliability Using Techniques Based on Kappa Square Statistics

The purpose of this paper is to describe and explain applicable techniques to evaluate the reliability of major components used in power generating stations where the quantity involved is relatively small, but the specified Mean Time Between Failures (MTBF) is significantly large, measured in terms of years. The approach must of necessity be tailored to these constraints, as opposed to techniques utilized in evaluating small electronic components and parts which are produced in large quantities. Because of this, it is possible to accumulate significantly large test times to evaluate the accept/reject criteria. To reduce test time to acceptable levels, the Kappa Square technique can be used to facilitate the design of reliability demonstration sampling plans and to optimize the probabilities of acceptance or rejection for specified criteria.