Aircraft maintenance is one of the most important cost items faced by the operators of air fleets and is a major contributor to the aircraft life cycle cost. An aircraft fly-by-wire flight control system has a total of primary flight control actuators ranging from 10 to 20 depending on the aircraft type, with a failure rate of 1/1000 flight-hours; therefore, a health monitoring system for primary flight control actuators, able to recognize an actuator degradation in its early stage could greatly contribute to optimize the maintenance operations, reduce the airplane downtime and prevent missions interruptions.This note presents the initial part of an ongoing research project aimed at developing a prognostic and health management system for fly-by-wire primary flight control actuators. A key feature of the project is to develop a PHM system for these actuators suitable for the flight control actuators of legacy airplanes, which are poised to operate for still a long time, and not only for those of new aircraft. The primary flight control actuators of fly-by-wire flight control systems of existing aircraft are electrohydraulic servoactuators with a typical configuration and complement of transducers, and there is no practical possibility of introducing additional sensors. For this reason, the research activity was directed towards the study of algorithms able to identify faults only by using the already available information of the servoactuators state variables.The implemented algorithms are a combination of mathematical and neural network based ones, and the identification of degradations was performed by the analysis of the response of the servoactuators to a sequence of selected stimuli provided in preflight or postflight. The servovalve current and the feedback position are processed by dedicated algorithms in order to obtain significant indicators of the servocatuator health condition. The values of the indicators obtained during the sequence of stimuli are analyzed in combination with those obtained in the past.This is performed by the neural network part of the algorithm which allows a reliable identification of presence and of type of a degradation.The results obtained from the initial part of the research activity are interesting and encouraging. Individual degradations of the servoactuator parameters have so far been addressed and the algorithms for identifying them have been developed. All that makes up the foundations of the future research activity which will be focused on analyzing the effects of simultaneous multiple degradations and to the estimation of the remaining useful life.