An Integrated XGBoost and BiLSTM Model for Advanced Remaining Useful Life (RUL) Predictions in Aircraft Maintenance

Accurate prediction of Remaining Useful Life (RUL) in aircraft maintenance is critical to ensuring operational efficiency and safety. This study proposes an integrated XGBoost-BiLSTM model for advanced RUL predictions in aircraft components. The model leverages XGBoost for feature selection to identify the most relevant features from multivariate time-series data, reducing the dimensionality of the dataset and improving computational efficiency. BiLSTM (Bidirectional Long Short-Term Memory) is employed to capture long-term dependencies in the data, enabling more precise temporal predictions. The proposed model is evaluated using performance metrics such as accuracy, precision, recall, F1-score, and Root Mean Squared Error (RMSE) and is compared with nine existing models, including Random Forest, Decision Tree, Gradient Boosting, Support Vector Machine (SVM), CNN-LSTM, and GRU. The proposed XGBoost- BiLSTM model outperforms all the other models, achieving an accuracy of 94.76% and the lowest RMSE of 0.1524. In addition, the model demonstrates superior precision (93.45%), recall (92.56%), and F1-score (93.00%), making it highly effective for RUL prediction in aircraft maintenance. The results indicate that the hybrid approach enhances prediction accuracy and reduces errors, offering a robust solution for predictive maintenance.