Fault diagnosis has become an issue of primary importance in modern process automation as it provides the prerequisites for the task of fault detection. The ability to detect the faults is essential to improve reliability and security of a complex control system. When a physical parameter change due to failure has occurred in a system, the failure effect will hardly be visible in the output performance. Since the failure, effect is reflected as a change in the predictor model. In this paper we describe a completed feasibility study demonstrating the merit of employing hybrid parameter-estimation and fuzzy logic for fault diagnosis. In this scheme, the residual generation is obtained from input-output data process, and identification technique based on ARX model, and the residual evaluation is based on fuzzy logic adaptive threshold method. The proposed fault detection and isolation tool has been tested on a magnetic levitation vehicle system.