Meaningful System Reliability Requirements.

The major purpose of this paper is to present some reliability measures which will reflect user needs and be operationally verifiable. It will both demonstrate the incompleteness in using one measure and a risk associated with requiring multiple measures. My objective is to help the reader tailor the solution of this dilemma to individual circumstances. The impact of the downtime distribution shape on system requirements is also examined. The Lognormal distribution illustrates one useful system downtime model, and the implication of this empirical result is briefly explored.