Predictive maintenance of Jet engine using Machine learning

Predictive maintenance plays a crucial role in the aviation industry by preventing equipment failures and enhancing jet engine reliability. Traditional maintenance methods, such as reactive and preventive approaches, often lead to inefficiencies and higher costs. This paper presents a machine learning-based predictive maintenance framework that integrates XGBoost, Long Short-Term Memory (LSTM), and Support Vector Regression (SVR) to estimate the Remaining Useful Life (RUL) of jet engines and detect anomalies. By utilizing real-time sensor data, the system improves maintenance planning, reduces unplanned downtime, and enhances engine health monitoring. SHAP (Shapley Additive Explanations) is employed to increase model transparency and interpretability. Experimental findings show that the ensemble model delivers superior predictive accuracy compared to individual algorithms.