Machine Criticality Level Assignment with Fuzzy Inference System for RCM

Abstract

The general goal of the manufacturing companies is reducing the production costs which includes the maintenance cost that comprises the main percent of overall production cost in different industries. Maintenance strategies may differ from each other according to labor force requirement, applicability and financial costs. Especially in the companies which have various machines and equipment, achieving the optimum maintenance activity for each machine or equipment is very difficult because of the cost of the maintenance strategy or inadequate labor force. Recently, Reliability-centered maintenance (RCM) is basically found to be the most efficient strategy in comparison with the existing supervision of maintenance strategies. Among the basic stages of RCM process, identification of critical component, which has a considerable influence on system reliability, is very crucial. However, machine criticality level assignment is not an easy problem since the desired real data may include uncertainty and subjectivity. Fuzzy set theory is an important solution when there is uncertain, linguistic or subjective data to handle. In this study, we applied a Fuzzy Inference System (FIS) to determine the machine criticality levels for maintenance activities. Machines and equipment are classified into criticality levels according to specified input variables such as MTBF (Mean Time Between Failure), machine occupancy status and monthly breakdown cost. These criticality levels will provide companies an important data to assign optimum maintenance strategies to the corresponding machines and equipment.