nebeleben/aircraft-predictive-maintenance

✈️ Aircraft Engine Predictive Maintenance

Predict the Remaining Useful Life (RUL) of aircraft engines using machine learning.

About

This application uses a LightGBM model trained on aircraft engine sensor data to predict how many more operational cycles an engine can safely run before requiring maintenance.

How to Use

1. 1.Enter the Engine ID and Current Cycle Number
2. 2.Enter all 21 Sensor Readings
3. 3.Click Predict RUL
4. 4.View the predicted remaining useful life in cycles

Model Performance

• —Algorithm: LightGBM
• —MAE: 13.65 cycles
• —R²: 0.777
• —Training Data: 100 engines, 20,631 cycles

ML Zoomcamp 2025 Project

Capstone project for Machine Learning Zoomcamp by DataTalks.Club.

GitHub: nebellleben/capstone-predictive-maintenance