Prognostics and health management: Utilizing the life cycle knowledge to reduce life cycle cost

Abstract

Prognostics is the process of predicting the future reliability of a product by assessing the extent of deviation or degradation of a product from its expected normal operating conditions. Health monitoring is a process of measuring and recording the extent of deviation and degradation from a normal operating condition. Traditional reliability engineering has attempted to quantify reliability as a failure probability and physics of failure based reliability assessment determines a range of possible times to failure. However, each product experiences its own unique life cycle that impacts its rate of degradation. With sensors as the eyes and ears to determine the actual environment, on board or off site computational platforms, and algorithms to process the information give us the ability to assess the reliability of each product and make operations, maintenance, and support decisions in an individualized manner. This talk with introduce the science and mathematics of prognostics and provide examples of use of the tools and techniques as applied in various applications such as batteries, fans, power electronics, and computers. It will also provide information on the development process of the IEEE standard on prognostics for electronics.