For modern engineering industry, Prognostic hasbecome a key feature in maintenance strategies since it enables toenhance system availability and safety while reducing operationalcosts and avoiding unscheduled maintenance. Prognostic canbe seen as the prediction of the system’s remaining useful lifewith the purpose of minimizing catastrophic failure events. Suchtask could be performed on the basis of an accurate physicalrepresentation of the system behavior and/or by using availablehistorical data that have been collected.In this paper, a novel prognostic approach is proposed, basedon data-driven category techniques. This approach uses mainlyhistorical data, regardless of the underlying physical process,and it can be divided into two steps. First, an original signalprocessing technique is used to develop life prediction models.In the second step, the system’s current health state is predictedand the RUL is estimated based on a proposed formula. Thisapproach is validated by using four different data sets generatedfrom the NASA’s turbofan engine simulator (C-MAPSS) and theobtained results are compared with relevant existing approachestested using the same collected data. The main outputs of ourstudy attest that the proposed approach is robust, applicableand effective even in the presence of various fault modes andoperating conditions.