Application of fuzzy logic for selection of Machinery Health Monitoring strategies

Machinery Health Monitoring (MHM) is increasingly being adopted by industries not only as a means for asset management but also for ensuring high levels of availability with its consequent gains. However, investments on MHM strategies will be effective only to the extent that they are appropriately selected and utilized. As a part of the process for selection of MHM systems, it is necessary to ascertain the impact of MHM systems in enhancing availability at plant level and in a cost effective manner. In this paper the need to quantify features of MHM systems such as detectability and prognostic ability and a simple fuzzy logic based method for doing so are discussed. These features are then incorporated in a multi-objective maintenance optimisation model based on Markov process and genetic algorithm. The results of the optimisation serve as decision support for selection of MHM systems. A case study is presented to demonstrate the concept.