A review of recent trends in machine diagnosis and prognosis algorithms

Abstract

Machine diagnosis represents fault condition monitoring that may be discrete or continuous and may include preset limit i.e. false alarms, such as green (good), yellow (warning) and red (failure) light indicators to notify low lubrication or low pressure levels. Machine prognosis represents set of activities performed based on diagnostic information to maintain its intended operating condition before complete failure. Avoidance of complete failure i.e. sudden breakdowns is desired since it causes economical misfortune to manufacturing companies. There are many literatures available on diagnostic and prognostic models and tools. This paper intends to review and summarize various techniques, models, and its applications. Also, develop a methodology how some of the techniques can be applied to robotic assembly process in an automotive assembly system.