A network reduction and transformation algorithm for assessing system effectiveness indexes

Abstract

This paper presents a systematic methodology to evaluate usual system effectiveness indexes of large and complex systems using network reduction and transformation techniques. This approach, allows some or all nodes to be imperfect, is efficient and simple, and uses less computer time and memory than existing techniques. Situations where it may not be feasible to use either star-delta or delta-star transformations in its usual form, to reduce a complex network are identified. Five subnetwork configurations are suggested to facilitate simplifying the network. search and subnetwork selection scheme which is adequate for the above purpose, and an efficient array-form representation requiring minimum memory to store the information about the network are offered.