3. Main hybrid methods

A large number of hybrid RUL diagnostic and prognostic methods have been successfully tested in various industrial sectors, particularly on rotating machines as part of predictive preventive maintenance. They use a combination of two or even three of the methods developed in the previous section: experience- and knowledge-based, data-based and physics-based. Given the number of possible combinations, a number of classification proposals have been put forward, none of which is internationally recognized in the scientific community. The one that seems to be the most widely accepted is that proposed by Linxia Liao