Introduction to Prognostics and Health Management

As technologies advance, modern engineering systems generally have very high reliability
requirements and are anticipated to operate continuously under complex, and often severe,
conditions. To reduce the potential downtime and maintenance costs, it is desirable that the machines
can understand their health states and take real-time decisions on their own, in order to reduce the
likelihood of having failures. Prognostics and Health Management (PHM) is a technology that provides
a path towards such a vision. In this lecture, we introduce the main theoretical foundations, as well as
some successful applications of PHM from various industries (e.g. aviation, machinery), including data
collection and signal processing, degradation state estimation and online reliability assessment, and
remaining useful life prediction.