The performance measure of a data driven prognostic system: Application to an aircraft engine

Recently, Prognostics and Health Management (PHM) solutions are increasingly implemented in order to complete maintenance activities. PHM predicts the future behavior of a system as well as its remaining useful life (RUL). One of the main approaches of the prognostic is data-driven approach who offer an advantage of being able to learn models based on empirical data and uses artificial intelligence methods. Present paper offers an implementation of PHM solution. We were interested by the estimation of the RUL of the aircraft engine by using historical data. We have implemented two technics: Artificial Neural Network and Neuro-Fuzzy System. To compare between these methods, we have studied the performance of the prognostic system according to the accuracy, precision, MSE (Mean Squared Error) and the training time. The best method was concluded finally.