This article introduces a novel Remaining Useful Life (RUL) estimation method using Machine Learning techniques, guided by domain knowledge, and applied to a dataset of aircraft engines (C-MAPSS). Predictive maintenance, or prognostics, offers the opportunity to predict the lifespan of aircraft engines, thereby reducing costs, minimizing breakdowns, and ensuring their reliability. While existing solutions in the literature primarily rely on either physical modeling or datadriven methods, they have achieved promising results, but they often face limitations, such as the interpretability of models, their reusability, and scalability. Knowledge-based methods have the potential to overcome these limitations, but they introduce their own set of challenges during implementation. In this article, we address a new hybrid method for predictive maintenance of aircraft engines. A case study combining a data-driven approach and knowledge will be presented as a proof of concept to demonstrate the feasibility of this hybrid solution and the possibilities it can offer.