Industrial Prognostics via Ensemble Machine Learning: An Uncertainty Aware Framework for RUL Estimation on NASA FD004 Telemetry

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2026 | Volume : 16 | 02 | Page :
    By

    Taufique khan Asadulla Khan,

  • Sonali Nimbhorkar,

  1. Student, Department of Computer Applications, GH Raisoni University, Amravati, Maharashtra, India
  2. Assistant Professor, Department of Computer Applications, GH Raisoni University, Amravati, Maharashtra, India

Abstract

Estimating the Remaining Useful Life (RUL) of industrial machinery in real-time is now vital for both operational safety and smart resource management. In the aviation industry, turbofan engines deal with constantly shifting flight conditions, making traditional, scheduled maintenance both expensive and prone to error. This paper addresses the flaws in common “point-prediction” AI models, which offer a single failure date without any margin for error, by introducing a new, uncertainty-aware framework. Our approach combines the predictive power of XGBoost with a specialized safety layer known as Conformal Prediction. We tested this system using the complex NASA C-MAPSS FD004 dataset, which involves multiple failure modes and six different operating environments. To handle this complexity, we applied piecewise RUL targets and regime – specific normalization to maintain stability in our date features. The results show that our XGBoost ensemble reaches a competitive RMSE of 28.62. More significantly, the system generates dynamic safety windows of ±21.6 cycles, providing the maintenance teams with a reliable “buffer” for decision-making. Our financial analysis suggests that this approach could cut maintenance costs by 76.2%, potentially saving millions in operational expenses. Ultimately, this research demonstrates that pairing gradient-boosted machines with statistical safety intervals provides the transparency and economic proof needed for real-world industrial use.

Keywords: Predictive Maintenance, XGBoost, Remaining Useful Life (RUL), Conformal Prediction, NASA C-MAPSS, Industrial IoT, Uncertainty Quantification, Cost Optimization, Aviation Safety

How to cite this article:
Taufique khan Asadulla Khan, Sonali Nimbhorkar. Industrial Prognostics via Ensemble Machine Learning: An Uncertainty Aware Framework for RUL Estimation on NASA FD004 Telemetry. Journal of Aerospace Engineering & Technology. 2026; 16(02):-.
How to cite this URL:
Taufique khan Asadulla Khan, Sonali Nimbhorkar. Industrial Prognostics via Ensemble Machine Learning: An Uncertainty Aware Framework for RUL Estimation on NASA FD004 Telemetry. Journal of Aerospace Engineering & Technology. 2026; 16(02):-. Available from: https://journals.stmjournals.com/joaet/article=2026/view=249916


References

  1. Saxena A, Goebel K. Turbofan engine degradation simulation data set. NASA ames prognostics data repository. 2008 Jan;18:878-87.
  2. Chen T, Guestrin C. Xgboost: A scalable tree boosting system. InProceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining 2016 Aug 13 (pp. 785-794).
  3. Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Advances in neural information processing systems. 2017;30.
  4. Shafer G, Vovk V. A tutorial on conformal prediction. Journal of machine learning research. 2008 Mar 1;9(3).
  5. Vovk V, Gammerman A, Shafer G. Algorithmic learning in a random world. Boston, MA: Springer US; 2005 Jan.
  6. Deepika J, Reddy PM, Murari K, Rahul B. Predictive maintenance of aircraft engines: Machine learning approaches for remaining useful life estimation. In2025 6th International Conference on Mobile Computing and Sustainable Informatics (ICMCSI) 2025 Jan 7 (pp. 1679-1686). IEEE.
  7. Heimes FO. Recurrent neural networks for remaining useful life estimation. In2008 international conference on prognostics and health management 2008 Oct 6 (pp. 1-6). IEEE.
  8. Li X, Ding Q, Sun JQ. Remaining useful life estimation in prognostics using deep convolution neural networks. Reliability Engineering & System Safety. 2018 Apr 1;172:1-1.
  9. Kumar A, Manas MP, Pradeep AM. Identification and classification of operating flow regimes and prediction of stall in a contra-rotating axial fan using machine learning. The Aeronautical Journal. 2022 Aug;126(1302):1351-69.
  10. Wang B, Lei Y, Li N, Li N. A hybrid prognostics approach for estimating remaining useful life of rolling element bearings. IEEE Transactions on reliability. 2018 Dec 14;69(1):401-12.
  11. Ensarioğlu K. Hierarchical RUL Prediction for Turbofan Engines Based on Health Stage Classification and Change Point-Guided Data Augmentation. IEEE Access. 2025 Jul 28.
Ahead of Print Subscription Review Article
Volume 16
02
Received 10/07/2026
Accepted 13/07/2026
Published 16/07/2026
Publication Time 6 Days
37

Login


My IP

PlumX Metrics