Optimal Testing Procedures for Special Structures of Coherent Systems

Abstract

Consider the problem of minimizing the expected cost of identifying the state of a coherent system (as “functioning” or “failed”). The system is composed of components (that either fail or work), and only individual components can be tested. Efficient algorithms are presented for some special cases of coherent systems: parallel-series, series-parallel and k-out-of-n systems. Examples are given to demonstrate the algorithms.