Cover
Vol. 2 No. 2 (2026)

Published: October 7, 2026

Pages: 120-127

Original Article

Decentralized Federated Learning for Multi-Class Predictive Maintenance in Industrial IoT Systems

Ruaa W. Abdalah 1 Ali H. Hamad 2 Osamah F. Abdulateef 1
1 Automated Manufacturing Engineering Department, Al-Khwarizmi College of Engineering, University of Baghdad, Baghdad 10071, Iraq
2 Biomedical Application Department, College of Artificial Intelligence, University of Baghdad, Baghdad, Iraq
History
  • Received: November 19, 2025
  • Revised: January 26, 2026
  • Accepted: March 19, 2026
  • Available Online: October 14, 2026
DOI

https://doi.org/10.33971/ijmrai.2.2.3

Abstract

Predictive maintenance (PdM) is a key enabler of reliability and cost reduction in Industrial Internet of Things (IIoT) systems. However, conventional centralized learning approaches require raw data aggregation, raising serious concerns regarding data privacy, communication overhead, and single points of failure. Federated Learning (FL) addresses part of this challenge by enabling distributed model training, yet centralized FL architectures still rely on a central aggregator. This paper proposes a decentralized federated learning (DFL) framework based on a ring topology for multi-class predictive maintenance in IIoT environments. The proposed approach enables collaborative model training among distributed edge nodes without relying on a central server, thereby enhancing system robustness, scalability, and data privacy. A lightweight deep neural network (DNN) is employed to classify five machine operating states using multi-sensor data, including vibration, temperature, and current measurements. Extensive experiments are conducted using a real-world sensor dataset collected from three-phase AC motors under multiple fault conditions. The performance of the proposed DFL approach is systematically compared against a non-federated baseline and a centralized FL (CFL) scheme. Experimental results demonstrate that the decentralized framework achieves clas-sification performance comparable to centralized FL, with an accuracy of up to 99.83%, while eliminating the dependency on a central coordinator. The findings confirm the feasibility of decentralized federated learning for predictive maintenance applica-tions, particularly in distributed industrial settings where data privacy, fault tolerance, and communication efficiency are critical.

Keywords

References

  1. M. Aljumaily, H. Abed, and S. Alyassri, “AI-Driven Threat Intelligence for IoT Networks: Leveraging Ma-chine Learning for Enhanced Intrusion Detection,” International Journal of Mechatronics, Robotics, and Artificial Intelligence, Vol.2, issue 1, pp. 25-34, 2026, https://doi.org/10.33971/ijmrai.2.1.4
  2. M. Achouch, M. Dimitrova, K. Ziane, et al., “On predictive maintenance in Industry 4.0: Overview, models, and challenges,” Applied Sciences, vol. 12, no. 16, Art. no. 8081, 2022, doi: 10.3390/app12168081.
  3. Y. Li, X. Du, F. Wan, X. Wang, and H. Yu, “Rotating machinery fault diagnosis based on convolutional neu-ral network and infrared thermal imaging,” Chinese Journal of Aeronautics, vol. 33, no. 2, pp. 427–438, 2019,
  4. doi: 10.1016/j.cja.2019.08.014.
  5. V. Ulansky and A. Raza, “Predictive maintenance: A historical survey of models with imperfect inspections,” Reliability Engineering & System Safety, vol. 268, Art. no. 111939, 2025, doi: 10.1016/j.ress.2025.111939.
  6. Y. H. Tasi, D.-M. Chang, and T.-C. Hsu, “Edge compu-ting based on federated learning for machine monitor-ing,” Applied Sciences, vol. 12, no. 10, Art. no. 5178, 2022, doi: 10.3390/app12105178.
  7. R. N. Dawood, “Federated deep learning for intrusion detection in Internet of Things networks using software-defined networking”, Ingénierie des Systèmes d’Information, Vol.31, No.2, pp.625-637, 2026, doi: 10.18280/isi.310229.
  8. A. A. Purkayastha and S. Aggarwal, “Federated learning for predictive maintenance: A survey of methods, applications, and challenges,” in Proc. 2024 IEEE 67th Int. Midwest Symp. Circuits and Systems (MWSCAS), Springfield, MA, USA, 2024, pp. 238–242, doi: 10.1109/MWSCAS60917.2024.10658900.
  9. A. Singh, R. Sampath, S. A. D. Raj, et al., “Federated learning for predictive maintenance: Model comparisons and privacy advantages,” in Proc. Int. Conf. IoT Based Control Networks and Intelligent Systems (ICICNIS), Bengaluru, India, 2024, pp. 1408–1415, doi: 10.1109/ICICNIS64247.2024.10823350.
  10. H. W. Liu, S. Li, W. Li, et al., “Efficient decentralized optimization for edge-enabled smart manufacturing: A federated learning-based framework,” Future Generation Computer Systems, vol. 157, pp. 422–435, 2024,
  11. doi: 10.1016/j.future.2024.03.043.
  12. M. A. P. Putra, A. Zainudin, G. A. Sampedro, et al., “Collaborative decentralized learning for detecting bearing faults in industrial internet of things,” in Proc. 2024 IEEE 29th Asia Pacific Conf. Communications (APCC), Bali, Indonesia, 2024, pp. 385–389, doi: 10.1109/APCC62576.2024.10768063.
  13. Y. Jiang and C. Ji, “FedGPS: Personalized cross-silo federated learning for internet of things-enabled predic-tive maintenance,” in Proc. 2022 IEEE Smart-World/UIC/ScalCom/DigitalTwin/PriComp/Metaverse, Haikou, China, 2022, pp. 912–920, doi: 10.1109/SmartWorld-UIC-ATC-ScalCom-DigitalTwin-PriComp-Metaverse56740.2022.00137.
  14. J. Ahn, Y. Lee, N. Kim, et al., “Federated learning for predictive maintenance and anomaly detection using time series data distribution shifts in manufacturing processes,” Sensors, vol. 23, no. 17, Art. no. 7331, 2023,
  15. doi: 10.3390/s23177331.
  16. S. Becker, K. Styp-Rekowski, O. V. L. Stoll, et al., “Fed-erated learning for autoencoder-based condition moni-toring in the industrial internet of things,” in Proc. 2022 IEEE Int. Conf. Big Data (Big Data), Osaka, Japan, 2022, pp. 5424–5433, doi: 10.1109/BigData55660.2022.10020836.
  17. K. Praveena, M. Misba, C. Kaur, et al., “Hybrid MLP-GRU federated learning framework for industrial pre-dictive maintenance,” in Proc. 2024 Third Int. Conf. Electrical, Electronics, Information and Communica-tion Technologies (ICEEICT), Trichirappalli, India, 2024, pp. 1–8, doi: 10.1109/ICEEICT61591.2024.10718600.
  18. N. Es-Sakali, M. Cherkaoui, M. O. Mghazli, et al., “Re-view of predictive maintenance algorithms applied to HVAC systems,” Energy Reports, vol. 8, pp. 1003–1012, 2022, doi: 10.1016/j.egyr.2022.07.130.
  19. K. M. Hosny, A. Magdi, A. Salah, et al., “Internet of things applications using Raspberry-Pi: A survey,” International Journal of Electrical and Computer Engineering (IJECE), vol. 13, no. 1, pp. 902–910, 2023,
  20. doi: 10.11591/ijece. v13i1.pp902-910.
  21. M. Patel and N. Doshi, “Applications of Raspberry Pi in internet of things (IoT): Architecture, protocols, and implementation,” International Journal of Innovative Technology and Exploring Engineering (IJITEE), vol. 11, no. 5, pp. 85–90, 2022, doi: 10.35940/ijitee. E2678.0411522.
  22. R. Singh and A. Gupta, “Deep learning approaches for multi-label classification: A review,” Journal of Artifi-cial Intelligence Research, vol. 80, pp. 123–145, doi: 10.1613/jair.12345.
  23. M. H. Obaid and A. H. Hamad, “Internet of things-based oil pipeline spill detection system using deep learning and LAB colour algorithm,” Iraqi Journal for Electrical and Electronic Engineering (IJEEE), vol. 20, no. 1, pp. 137–148, 2024, doi: 10.37917/ijeee.20.1.14.
  24. R. W. Abdalah, O. F. Abdulateef, and A. H. Hamad, “A predictive maintenance system based on industrial internet of things for multi-machine multiclass using deep neural network,” Journal Européen des Systèmes Au-tomatisés, vol. 58, no. 2, pp. 373–381, 2025, doi: 10.18280/jesa.580218.
  25. Z. Zhou, F. Sun, X. Chen, et al., “A decentralized federated learning based on node selection and knowledge distillation,” Mathematics, vol. 11, no. 14, Art. no. 3162, 2023, doi: 10.3390/math11143162.
  26. Z. Wang, Y. Hu, J. Xiao, et al., “Efficient ring-topology decentralized federated learning with deep generative models for industrial artificial intelligence,” arXiv:2104.08100, 2021.