The NASA Aircraft Engine dataset has been extensively analyzed to predict the remaining useful life (RUL) using various machine learning and deep learning models.
Among the models tested, the Multilayer Perceptron (MLP), XGBoost, Long Short-Term Memory (LSTM), and a hybrid version of LSTM combined with Convolutional Neural Networks (CNN)
have shown particularly convincing results. These models have demonstrated superior performance in terms of accuracy and robustness in predicting RUL, making them standout
choices for this application
| # | Model | Test RMSE | Test R2 Score |
|---|---|---|---|
| 1 | Linear Regression | 43.18 | 0.46 |
| 2 | Random Forest | 6.68 | 0.42 |
| 3 | XG Boost | 17.35 | 0.65 |
| 4 | Multilayer Perceptron | 4.51 | 0.52 |
| 5 | LSTM | 15.93 | 0.75 |
| 6 | CNN + LSTM | 13.34 | 0.86 |