Authors

Elo A. Fredrick

Centre for Information and Telecommunication Engineering, University of Port Harcourt, Rivers State, Nigeria

Bourdillon O. Omijeh

Professor, University of Port Harcourt, Rivers State, Nigeria

Crescent O. Omeje

University of Port Harcourt, Rivers State, Nigeria

Abstract

Predictive maintenance (PdM) is a critical application of Artificial Intelligence (AI) in high-stakes industrial environments like the midstream oil and gas sector. However, the efficacy of AI is often limited by the performance of standalone models, which struggle to handle the complexity and imbalance of real-world machinery data. This paper introduces and validates a novel, multi-stage hybrid AI framework designed to overcome these limitations. The research begins by establishing the performance of baseline models, including XGBoost classifiers and a standalone Long Short-Term Memory (LSTM) network for Remaining Useful Life (RUL) prediction. It is demonstrated that the standalone RUL model fails to learn the degradation patterns effectively (RMSE > 2500 hours). To solve this, a "Super Hybrid" classifier is developed, which fuses traditional time-domain features with spectral features from Fast Fourier Transform (FFT) and an anomaly score from a Variational Autoencoder (VAE). After calibration, this classifier achieves near-perfect performance (100% Precision, 100% Recall). The core contribution is a two-stage, classifier-gated architecture where this high-performance classifier first identifies a "degrading" state, which then activates a specialized, deep LSTM model trained with a custom weighted-loss function. This final gated RUL model successfully anticipates the terminal failure phase, providing a robust and reliable prognostic tool. This work presents a significant methodological advancement, proving that an intelligently structured, multi-stage hybrid framework is superior to individual models for complex industrial PdM tasks.

Keywords

Predictive Maintenance Hybrid AI Classifier-Gated Anomaly Detection LSTM XGBoost Variational Autoencoder

Citation of this Article

Elo A. Fredrick, Bourdillon O. Omijeh, & Crescent O. Omeje. (2025). A Classifier-Gated Hybrid AI Framework for Predictive Maintenance in Critical Industrial Systems. International Current Journal of Engineering and Science (ICJES), 4(10), 16-22. Article DOI: https://doi.org/10.47001/ICJES/2025.410004

Licence Copyright (c) 2026 International Current Journal of Engineering and Science. This work is licensed under a Creative Commons Attribution Non Commercial 4.0 International Licence.

References

  1. D. Kornack and P. Rakic, “Cell Proliferation without Neurogenesis in Adult Primate Neocortex,” Science, vol. 294, Dec. 2001, pp. 2127-2130, doi:10.1126/science.1065467.
  2. M. Young, The Technical Writer’s Handbook. Mill Valley, CA: University Science, 1989.
  3. Abbasi, T., Lim, K., Soomro, T., Ismail, I., & Ali, A. (2020). Condition Based Maintenance of Oil and Gas Equipment: A Review. 2020 3rd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), 1-9. https://doi.org/10.1109/iCoMET48670.2020.9073819.
  4. Dabbousi, R., Sekitani, Y., Alqahtani, K., & Balfaqih, H. (2023). Management of Electric Motors Lifecycle At Oil & GAS Industrial Facilities. 2023 IEEE IAS Petroleum and Chemical Industry Technical Conference (PCIC), 223-228. https://doi.org/10.1109/PCIC43643.2023.10414327.
  5. Taşcı, B., Omar, A., & Ayvaz, S. (2023). Remaining useful lifetime prediction for predictive maintenance in manufacturing. Comput. Ind. Eng., 184, 109566. https://doi.org/10.2139/ssrn.4344017.
  6. Zhang, W., Yang, D., & Wang, H. (2020). Data-driven methods for predictive maintenance of industrial equipment: A survey. IEEE Systems Journal, 13(3), 2213–2227. https://doi.org/10.1109/JSYST.2018.2867180
  7. Orrù, P., Zoccheddu, A., Sassu, L., Mattia, C., Cozza, R., & Arena, S. (2020). Machine Learning Approach Using MLP and SVM Algorithms for the Fault Prediction of a Centrifugal Pump in the Oil and Gas Industry. Sustainability. https://doi.org/10.3390/su12114776.
  8. Lei, Y., Li, N., Guo, L., Li, N., Yan, T., & Lin, J. (2018). Machinery health prognostics: A systematic review from data acquisition to RUL prediction. Mechanical Systems and Signal Processing, 104, 799–834. https://doi.org/10.1016/j.ymssp.2017.10.016
  9. Jones, M., Smith, P., & Brown, R. (2023). The cost of downtime in the oil and gas industry. Journal of Petroleum Engineering, 45(3), 125-140.
  10. Lazzaro, A., D'Addona, D., & Merenda, M. (2022). Comparison of Machine Learning Models for Predictive Maintenance Applications., 657-666. https://doi.org/10.1007/978-3-031-16281-7_62.
  11. S. Namuduri, B. N. Narayanan, V. S. P. Davuluru, L. Burton, and S. Bhansali, “Review—Deep Learning Methods for Sensor Based Predictive Maintenance and Future Perspectives for Electrochemical Sensors,” Journal of The Electrochemical Society, vol. 167, no. 3, p. 37552, Jan. 2020, doi: 10.1149/1945-7111/ab67a8.
  12. Gopalakrishnan, M., & Skoogh, A. (2021). Maintenance strategy development within Industry 4.0. Procedia CIRP, 93, 1244–1249. https://doi.org/10.1016/j.procir.2021.01.172.
  13. Çınar, Z. M., Abdussalam Nuhu, A., Zeeshan, Q., Korhan, O., Asmael, M., & Safaei, B. (2020). Machine learning in predictive maintenance towards sustainable smart manufacturing in industry 4.0. Sustainability, 12(19), 8211.
  14. Bono, F., Cinquemani, S., Chatterton, S., & Pennacchi, P. (2022). A deep learning approach for fault detection and RUL estimation in bearings., 12049, 1204908 - 1204908-13. https://doi.org/10.1117/12.2607084.
  15. Khan, U., Cheng, D., Setti, F., Fummi, F., Cristani, M., & Capogrosso, L. (2025). A Comprehensive Survey on Deep Learning-based Predictive Maintenance. ACM Transactions on Embedded Computing Systems. https://doi.org/10.1145/3732287.
  16. Akyaz, T., & Engin, D. (2024). Machine Learning-Based Predictive Maintenance System for Artificial Yarn Machines. IEEE Access, 12, 125446-125461. https://doi.org/10.1109/ACCESS.2024.3454548.
  17. Wang, H., Zhang, W., Yang, D., & Xiang, Y. (2023). Deep-Learning-Enabled Predictive Maintenance in Industrial Internet of Things: Methods, Applications, and Challenges. IEEE Systems Journal, 17, 2602-2615. https://doi.org/10.1109/JSYST.2022.3193200.
  18. Hanifi, S., Alkali, B., Lindsay, G., Waters, M., & McGlinchey, D. (2024). Advancements in predictive maintenance modelling for industrial electrical motors: Integrating machine learning and sensor technologies. Measurement: Sensors, 101473.
  19. Bouabdallaoui, Y., Lafhaj, Z., Yim, P., Ducoulombier, L., & Bennadji, B. (2021). Predictive Maintenance in Building Facilities: A Machine Learning-Based Approach. Sensors (Basel, Switzerland), 21. https://doi.org/10.3390/s21041044.
  20. Liu, C., Zhu, H., Tang, D., Nie, Q., Zhou, T., Wang, L., & Song, Y. (2022). Probing an intelligent predictive maintenance approach with deep learning and augmented reality for machine tools in IoT-enabled manufacturing. Robotics and Computer-Integrated Manufacturing, 77, 102357. https://doi.org/10.1016/j.rcim.2022.102357.