In this paper, we present a hybrid model combining Feedforward Neural Network (FNN) and Long Short-Term Memory (LSTM) for the prediction of the Remaining Useful Life (RUL) of turbofan engines. The proposed model leverages the strengths of FNN in capturing non-linear relationships and the temporal capabilities of LSTM to handle sequential data. Additionally, we applied a Kalman filter during the preprocessing phase to reduce noise, which significantly enhanced the model’s performance. We evaluated our model on the C-MAPSS dataset, a widely used benchmark for RUL prediction in aero-engines. Our experimental results demonstrate that the FNN-LSTM model, with the Kalman filter preprocessing, achieves superior performance, providing accurate and reliable RUL predictions, thereby contributing to more effective maintenance decision-making and improved operational efficiency.