Predictive maintenance: remaining useful life of jet engines

The brief

An airline wants to service engines just before they need it, not too early (wasted money) or too late (safety risk). Using run-to-failure sensor data, predict each engine's remaining useful life and turn it into a maintenance schedule. Good looks like: low RUL error and a scoring that punishes late predictions, a maintenance plan, and an honest discussion of uncertainty.

What you will build

1. Explore sensor trends as engines degrade
2. Build RUL labels and handle operating conditions
3. Train classical and sequence (LSTM or 1-D CNN) models
4. Evaluate with RMSE and an asymmetric late-prediction score
5. Add prediction intervals
6. Turn predictions into a maintenance schedule and dashboard

Free, open data to build it with

• NASA Turbofan Jet Engine (C-MAPSS) ↗ Kaggle (NASA data)

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.