AI-Driven Predictive Maintenance: Remaining Useful Life Estimation for Turbofan Engines via Ensemble Learning, LSTM, and Uncertainty Quantification
Fuente:
Zenodo
Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Published: |
Zenodo
2026
|
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866901817683607552 |
|---|---|
| author | Mahmudul Hasan Rohan, Rohan |
| author_facet | Mahmudul Hasan Rohan, Rohan |
| contents | <p>Unplanned industrial equipment failures cause significant financial losses and<br>safety hazards across aviation, energy, and manufacturing sectors. Predictive Maintenance<br>(PdM) addresses this by estimating the Remaining Useful Life (RUL) — the number<br>of operational cycles before a machine requires intervention — enabling proactive, costoptimal scheduling. This paper presents a complete end-to-end machine learning pipeline<br>for RUL prediction on the NASA C-MAPSS turbofan engine degradation dataset [17]. We<br>compare four models — Linear Regression, Random Forest, Gradient Boosting, and a twolayer Long Short-Term Memory (LSTM) network implemented in pure NumPy under<br>rigorous Group K-Fold cross-validation to eliminate engine-level data leakage. Our feature<br>engineering pipeline transforms 21 raw sensor channels into 37 temporal features including<br>rolling statistics, a composite health index, and degradation rate. Random Forest<br>achieves the best performance: RMSE = 9.95 cycles, R2 = 0.9398, MAE = 5.77,<br>and NASA PHM asymmetric score = 1.9. We further contribute bootstrap prediction<br>intervals achieving 90.0% empirical coverage at the 95% nominal level, and a maintenance<br>scheduling module projecting $430,000 in cost savings on a 20-engine fleet. </p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_19000557 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | AI-Driven Predictive Maintenance: Remaining Useful Life Estimation for Turbofan Engines via Ensemble Learning, LSTM, and Uncertainty Quantification Mahmudul Hasan Rohan, Rohan <p>Unplanned industrial equipment failures cause significant financial losses and<br>safety hazards across aviation, energy, and manufacturing sectors. Predictive Maintenance<br>(PdM) addresses this by estimating the Remaining Useful Life (RUL) — the number<br>of operational cycles before a machine requires intervention — enabling proactive, costoptimal scheduling. This paper presents a complete end-to-end machine learning pipeline<br>for RUL prediction on the NASA C-MAPSS turbofan engine degradation dataset [17]. We<br>compare four models — Linear Regression, Random Forest, Gradient Boosting, and a twolayer Long Short-Term Memory (LSTM) network implemented in pure NumPy under<br>rigorous Group K-Fold cross-validation to eliminate engine-level data leakage. Our feature<br>engineering pipeline transforms 21 raw sensor channels into 37 temporal features including<br>rolling statistics, a composite health index, and degradation rate. Random Forest<br>achieves the best performance: RMSE = 9.95 cycles, R2 = 0.9398, MAE = 5.77,<br>and NASA PHM asymmetric score = 1.9. We further contribute bootstrap prediction<br>intervals achieving 90.0% empirical coverage at the 95% nominal level, and a maintenance<br>scheduling module projecting $430,000 in cost savings on a 20-engine fleet. </p> |
| title | AI-Driven Predictive Maintenance: Remaining Useful Life Estimation for Turbofan Engines via Ensemble Learning, LSTM, and Uncertainty Quantification |
| url | https://doi.org/10.5281/zenodo.19000557 |