Abstract
This paper proposes a decision making approach based on the development of a fuzzy prognostic system to ensure the vibrations monitoring of a gas turbine based on real time information obtained from different installed sensors. In this approach the case of incomplete obtained data which may occur frequently is taken into account by using an approach of full data reconstitution form incomplete data. The proposed fuzzy prognostic system approach presented in this paper allows the analysis of the data obtained via the vibration indicators of a gas turbine system for the accurate identification of the faults to avoid the performance degradation of such systems. In order to prove the robustness of the proposed approach presented in this paper, several tested has been performed.