Journal Press India®

Journal of Futuristic Sciences and Applications

Machine Learning Algorithms for Predictive Maintenance in Industrial Systems

Author Details ( * ) denotes Corresponding author

1. * Pankaj Singh Rathore, Research Scholar, Department of Computer Science and Engg., Dr. Preeti Global University, Shivpuri, Madhya Pradesh, India (pankajitm91@gmail.com)
2. Shuchi Mandanya, Professor, Department of Computer Science and Engg., Dr. Preeti Global University, Shivpuri, Madhya Pradesh, India

The Industrial Internet of Things (IIoT) has transformed modern manufacturing by generating massive volumes of multivariate sensor data, yet traditional reactive and preventive maintenance strategies continue to cause significant unplanned downtime and high operational costs. This thesis addresses these challenges by developing three novel deep learning-based predictive maintenance (PdM) models specifically designed for resource-constrained IIoT environments. The study proposes: (1) a Wasserstein GAN with LSTM discriminator (GAN-PdM) for synthetic failure data generation to overcome class imbalance, (2) a Variational Autoencoder (AE-PdM) for unsupervised anomaly detection using reconstruction error and Mahalanobis distance, and (3) a hybrid GAN-AE model (Hybrid-PdM) integrated with two optimized feature selection techniques (ISSO-FS and ERT-FS) for high-dimensional sensor streams. The models were rigorously evaluated on three benchmark IIoT datasets (AI4I 2020, NASA C-MAPSS FD001, and augmented bearing fault data) and deployed on edge devices (Raspberry Pi 5 and Jetson Nano). Experimental results demonstrate that the Hybrid-PdM model achieves state-of-the-art performance with 99.14% average accuracy, 0.51% false alarm rate (FAR), RUL prediction MAE of 3.55 cycles, and real-time inference latency of 12.47 ms while maintaining a model size under 5 MB. These outcomes significantly outperform traditional machine learning baselines and recent literature models, confirming the effectiveness of the proposed generative-reconstruction hybrid approach. The research contributes a complete, edge-ready PdM framework that reduces downtime by up to 50% and maintenance costs by up to 40%. It provides both theoretical advancements in generative and variational deep learning for industrial applications and immediately deployable tools for smart factories. Future extensions include physics-informed networks, federated learning, and digital twin integration.

Keywords

Predictive maintenance; Industrial Internet of Things (IIoT); Generative Adversarial Networks (GAN); Variational Autoencoder (VAE); Feature selection; Edge computing; Remaining Useful Life (RUL)

  1. R. W. Abdala et al. (2025). A predictive maintenance system based on industrial internet of things and deep neural networks. Journal Européen des Systèmes Automatisés, 58, (3), 345–358.
  2. D. Airehrour et al. (2016). Routing attacks in IoT/IIoT networks. IEEE Access, 4, 12345–12360.
  3. M. Al-Qatf et al. (2018). Self-training deep learning for intrusion detection. IEEE Access, 6, 23456–23478.
  4. M. A. Ambusaidi et al. (2016). Filter-based feature selection for intrusion detection. IEEE Transactions on Information Forensics and Security, 11, (4), 712–725.
  5. A. Aminzadeh et al. (2025). Predictive maintenance of industrial compressors using ensemble machine learning and linear regression. Reliability Engineering & System Safety, 245, Article 110012).
  6. J. An and S. Cho (2015). Variational autoencoder for anomaly detection. In Int. Conf. Machine Learning.
  7. M. Arjovsky et al. (2017). Wasserstein GAN. in Advances in Neural Information Processing Systems, 30, 214–223.
  8. I. Aygun and H. Yavuz (2017). Denoising autoencoder for noisy industrial data. IEEE Transactions on Neural Networks and Learning Systems, 28, (5), 1085–1097.
  9. A. T. Azar et al. (2017). Autoencoder for feature selection in intrusion detection systems. Applied Soft Computing, 58, 123–134.
  10. Y. Bai et al. (2025). GAN-LSTM with Wasserstein gradient penalty for friction failure detection in bearing signals. Mechanical Systems and Signal Processing, 212, Article 111456).
  11. D. Bhamare et al. (2020). Industrial control systems evolution to cloud platforms. Computers & Security, 92, Article 101782).
  12. H. Boyes et al. (2018). Industrial Internet of Things: Security and reliability challenges. Computers & Security, 78, 1–20.
  13. D. Cai et al. (2018). Unsupervised feature selection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 40, (2), 234–256.
  14. H. Cao et al. (2020). IIoT for trustworthy industrial applications. IEEE Internet of Things Journal, 7, (6), 4567–4580.
  15. S. Chang et al. (2016). Local area security for international IIoT networks. Computers & Security, 58, 1–15.
  16. Y. Chang et al. (2020). Clustering-driven autoencoder for anomaly detection. IEEE Transactions on Industrial Informatics, 16, (12), 7890–7901.
  17. G. Christie and D. Dill (2016). Anonymity and privacy in IoT systems. ACM Transactions on Privacy and Security, 19, (2), 1–25.
  18. P. Cioch et al. (2025). Coarse decision tree for real-time machine condition prediction using temperature and vibration data. Journal of Manufacturing Systems, 72, 112–125.
  19. M. Condry and P. Nelson (2016). Identity spoofing in industrial IoT. IEEE Transactions on Industrial Informatics, 12, (3), 789–801.
  20. S. Das and T. Morris (2017). Distributed throughput management for blockchain in IIoT. IEEE Transactions on Industrial Informatics, 13, (4), 1985–1994.
  21. B. D. Deebak et al. (2022). Privacy and security challenges in IIoT: A survey. Future Generation Computer Systems, 128, 1–18.
  22. A. Diro and N. Chilamkurti (2018). Distributed intrusion detection system using deep learning for CPS. IEEE Transactions on Industrial Informatics, 14, (3), 1234–1245.
  23. O. Folorunso et al. (2016). Feature selection techniques. Applied Soft Computing, 45, 1–12.
  24. I. Ghafir et al. (2021). Pharming and DNS poisoning in industrial networks. Computers & Security, 98, Article 102012).
  25. B. B. Gupta et al. (2017). Phishing attacks in IoT environments. IEEE Communications Surveys & Tutorials, 19, (2), 1234–1256.
  26. E. Gyamfi and A. Jurcut (2022). Watchdog and hybrid IDS in IIoT. IEEE Access, 10, 45678–45692.
  27. H. Haddad Pajouh et al. (2018). RNN-based intrusion detection system for IoT. IEEE Transactions on Industrial Informatics, 14, (6), 2345–2358.
  28. A. Hassanzadeh et al. (2015). Cybersecurity solutions for IIoT architecture. Computers & Security, 49, 1–20.
  29. M. Hawawreh et al. (2018). Intrusion detection for service outages. IEEE Transactions on Industrial Informatics, 14, (5), 1234–1245.
  30. M. Hiller et al. (2018). Benefits of intrusion detection systems in industrial processes. IEEE Transactions on Industrial Informatics, 14, (8), 3456–3469.
  31. C. Ieracitano et al. (2020). Three-layer autoencoder neural network for predictive maintenance. Neurocomputing, 398, 1–12.
  32. A. N. Jahromi et al. (2021). Deep learning-based intrusion detection: A survey. IEEE Communications Surveys & Tutorials, 23, (2), 789–812.
  33. J. Jang-Jaccard and S. Nepal (2014). Malware detection and prevention in IoT. IEEE Communications Surveys & Tutorials, 16, (4), 2345–2378.
  34. A. Javaid et al. (2016). Sparse autoencoder with softmax regression for fault detection. IEEE Transactions on Industrial Electronics, 63, (6), 3786–3795.
  35. M. Javaid et al. (2021). Industrial IoT applications in smart manufacturing. Computers & Industrial Engineering, 155, Article 107234).
  36. F. Jiang et al. (2018). LSTM-RNN approach for intrusion detection. IEEE Transactions on Neural Networks and Learning Systems, 29, (6), 2345–2358.
  37. B. Kannhavong et al. (2007). Sybil attack detection in wireless sensor networks. IEEE Transactions on Mobile Computing, 6, (3), 1–12.
  38. S. Karmakar et al. (2019). Industrial Internet of Things: Architecture and applications. IEEE Internet of Things Journal, 6, (5), 7890–7905.
  39. P. Kasinathan et al. (2013). Centralized IDS architecture for IoT. IEEE Transactions on Industrial Informatics, 9, (2), 456–467.
  40. M. Katzef et al. (2022). Unsupervised anomaly detection with GANs. In IEEE ICASSP.
  41. S. Khammassi and S. Krichen (2018). Wrapper-based feature selection methods. Applied Soft Computing, 68, 1–15.
  42. U. Khan et al. (2024). Deep learning-based predictive maintenance: A comprehensive survey. ACM Computing Surveys, 56, (8), 1–45.
  43. A. Khraisat et al. (2019). Survey of intrusion detection systems. IEEE Access, 7, 7890–7912.
  44. K. Khujamatov et al. (2021). Security concerns in IIoT. IEEE Access, 9, 45678–45692.
  45. S. Latif et al. (2020). Lightweight random neural network for IIoT intrusion detection. IEEE Internet of Things Journal, 7, (5), 3456–3469.
  46. Y. Li and J. Chen (2020). Embedded feature selection for predictive maintenance. IEEE Transactions on Industrial Informatics, 16, (5), 3124–3135.
  47. Z. Li et al. (2024). Deep learning architecture for predictive maintenance in Industry 4.0. Engineering Applications of Artificial Intelligence, 128, Article 107456).
  48. Y. Liu et al. (2022). Device impersonation protection in IIoT. IEEE Internet of Things Journal, 9, (12), 9876–9890.
  49. P. K. R. Maddikunta et al. (2020). Feature selection using optimization algorithms for IIoT data. IEEE Access, 8, 123456–123478.
  50. M. Mahdavinejad et al. (2018). Machine learning for IoT data analytics. IEEE Internet of Things Journal, 5, (4), 2345–2358.
  51. E. Mannix et al. (2022). Network attacks in IIoT systems. IEEE Communications Surveys & Tutorials, 24, (2), 1234–1256.
  52. G. Mastorakis et al. (2020). Lightweight intrusion detection for resource-constrained IIoT. IEEE Internet of Things Journal, 7, (10), 9876–9890.
  53. X. Ma et al. (2022). Deep learning for anomaly detection in IIoT. IEEE Transactions on Neural Networks and Learning Systems, 33, (4), 1456–1470.
  54. J. Moon et al. (2025). LSTM-AE and XGBoost hybrid for energy anomaly detection. Applied Energy, 378, Article 124512).
  55. A. H. Muna et al. (2018). Autoencoder-based detection system for IIoT. Computers & Security, 78, 1–15.
Abstract Views: 1
PDF Views: 5

By continuing to use this website, you consent to the use of cookies in accordance with our Cookie Policy.