Neural networks for equipment failure prediction based on unstructured production data

Mykola Demchyna
Received 05.04.2025, Revised 01.10.2025, Accepted 24.12.2025
Abstract

Industrial enterprises accumulate large volumes of unstructured data from sensors, event logs, and technical reports; however, their potential for failure prediction often remains underutilised. The aim of this study was to demonstrate how modern neural networks transform such heterogeneous data streams into early failure warnings and how the integration of these models into maintenance systems reduces downtime and increases equipment reliability. The methodology combined a unified data preparation pipeline – including cleaning, timestamp matching, and event-based and semantic feature extraction from report texts – with a comparative evaluation of three classes of architectures: convolutional networks for vibration signals and images, recurrent networks capable of modelling long-term dependencies in time series, and attention-based models for multimodal learning. The main results focused on forecast quality and operational impact. Models based on attention mechanisms consistently outperformed other approaches in terms of weighted average accuracy by 5-9 percentage points, providing a warning horizon of 24-72 hours. Convolutional networks demonstrated the highest sensitivity to high-frequency vibration patterns, while recurrent models were more effective in tracking slow degradation trends. Integration of the best-performing model into a maintenance information system – with automated request generation and work prioritisation – reduced downtime by 14-21%, increased mean time to failure by 10-15%, and decreased the proportion of false alarms to a level acceptable to dispatchers. Available production cases were summarised, and theoretical application scenarios were outlined for industries in which empirical data remain limited. The practical significance of the study lies in a reproducible approach to processing and interpreting unstructured data, as well as in the demonstrated contribution to reducing downtime and increasing the reliability of the equipment fleet

Keywords:
current sensors; event logs; technical reports; cross-channel attention; data preparation pipeline; timestamp matching; semantic text parsing

Retrieved from Volume 23, No. 2, 2025

Pages 104–114

References Suggested citation
  1. Abbasi, T., Lim, K.H., & Yam, K.S. (2019). Predictive maintenance of oil and gas equipment using recurrent neural network. IOP Conference Series: Materials Science and Engineering, 495, article number 012067. doi: 10.1088/1757-899X/495/1/012067.
  2. Abidi, M.H., Mohammed, M.K., & Alkhalefah, H. (2022). Predictive maintenance planning for Industry 4.0 using machine learning for sustainable manufacturing. Sustainability, 14(6), article number 3387. doi: 10.3390/ su14063387.
  3. Bampoula, X., Siaterlis, G., Nikolakis, N., & Alexopoulos, K. (2021). A deep learning model for predictive maintenance in cyber-physical production systems using LSTM autoencoders. Sensors, 21(3), article number 972. doi: 10.3390/s21030972.
  4. Carvalho, T.P., Soares, F.A., Vita, R., de Francisco, R.P., Basto, J., & Alacala, S. (2019). A systematic literature review of machine learning methods applied to predictive maintenance. Computers & Industrial Engineering, 137, article number 106024. doi: 10.1016/j.cie.2019.106024.
  5. Chen, X. (2024). A novel transformer-based DL model enhanced by position-sensitive attention and gated hierarchical LSTM for aero-engine RUL prediction. Scientific Reports, 14(1), article number 10061. doi: 10.1038/ s41598-024-59095-3.
  6. Contreras-Valdes, A., Amezquita-Sanchez, J., Granados-Lieberman, D., & Valtierra-Rodriguez, M. (2020). Predictive data mining techniques for fault diagnosis of electric equipment: A review. Applied Sciences, 10(3), article number 950. doi: 10.3390/app10030950.
  7. Cui, J., Kuang, W., Geng, K., & Jiao, P. (2025). Intelligent fault diagnosis and operation condition monitoring of transformer based on multi-source data fusion and mining. Scientific Reports, 15(1), article number 7606. doi: 10.1038/s41598-025-91862-8.
  8. Dairi, A., Fouzi, H., Khaldi, B., & Ying, S. (2024). Graph neural networks-based spatiotemporal prediction of photovoltaic power: A comparative study. Neural Computing and Applications, 37(6), 4769-4795. doi: 10.1007/ s00521-024-10751-9.
  9. Deng, S., & Zhou, J. (2024). Prediction of remaining useful life of aero-engines based on CNN-LSTM-attention. International Journal of Computational Intelligence Systems, 17, article number 232. doi: 10.1007/s44196-02400639-w.
  10. Fan, Y., Tang, Q., Guo, Y., & Wie, Y. (2024). BiLSTM-MLAM: A multi-scale time series prediction model for sensor data based on Bi-LSTM and local attention mechanisms. Sensors, 24(12), article number 3962. doi: 10.3390/ s24123962.
  11. Fordal, J.M., Schjølberg, P., Helgetun, H., Skjermo, T., Wang, Y., & Wang, C. (2023). Application of sensor data based predictive maintenance and artificial neural networks to enable Industry 4.0. Advances in Manufacturing, 11(4), 248-263. doi: 10.1007/s40436-022-00433-x.
  12. Hadadi, F., Dawes, J.H., Shin, D., Bianculli, D., & Briand, L. (2024). Systematic evaluation of deep learning models for log-based failure prediction. Empirical Software Engineering, 29(5), article number 105. doi: 10.1007/ s10664-024-10501-4.
  13. Kang, H., & Kang, P. (2024). Transformer-based multivariate time series anomaly detection using intervariable attention mechanism. Knowledge-Based Systems, 290(2), article number 111507. doi: 10.1016/j. knosys.2024.111507.
  14. Karimi, A., Wu, Y., Koyuturk, M., & French, R.H. (2021). Spatiotemporal graph neural network for performance prediction of photovoltaic power systems. Proceedings of the AAAI Conference on Artificial Intelligence, 35(17), 15323-15330. doi: 10.1609/aaai.v35i17.17799.
  15. Kumar, A. (2024). Machine learning algorithms for predictive maintenance in industrial environments: A comparative study. Journal of Artificial Intelligence General Science, 2(1), 149-150. doi: 10.60087/jaigs.v2i1.p150.
  16. Leukel, J., Gonzalez, J., & Riekert, M. (2021). Adoption of machine learning technology for failure prediction in industrial maintenance: A systematic review. Journal of Manufacturing Systems, 61, 87-96. doi: 10.1016/j. jmsy.2021.08.012.
  17. Li, T., Zhou, Z., Li, S., Sun, C., Yan, R., & Chen, X. (2022). The emerging graph neural networks for intelligent fault diagnostics and prognostics: A guideline and a benchmark study. Mechanical Systems and Signal Processing, 168, article number 108653. doi: 10.1016/j.ymssp.2021.108653.
  18. Lu, D., Zhou, X., & Li, S. (2024). Spatio-temporal attention-based hybrid deep network for time series prediction of industrial process. Applied Intelligence, 55, article number 169. doi: 10.1007/s10489-024-06033-5.
  19. Muthukumar, G., & Philip, J. (2024). CNN-LSTM hybrid deep learning model for remaining useful life estimation. International Journal for Innovative Research in Multidisciplinary Field, 10(54), 38-53. doi: 10.2015/IJIRMF/ ICSETI-2024/P04.
  20. Nor, K., Pedapati, S.R., Muhammad, M., & Leiva, V. (2022). Abnormality detection and failure prediction using explainable Bayesian deep learning: Methodology and case study with industrial data. Mathematics, 10(4), article number 554. doi: 10.3390/math10040554.
  21. Nota, G., Postiglione, A., & Carvello, R. (2022). Text mining techniques for the management of predictive maintenance. Procedia Computer Science, 200, 778-792. doi: 10.1016/j.procs.2022.01.276.
  22. Rodríguez, M.L., Kubler, S., de Giorgio, A., Cordy, M., Robert, J., & Le Traon, Y. (2022). Multi-agent deep reinforcement learning based Predictive Maintenance on parallel machines. Robotics and Computer-Integrated Manufacturing, 78(1), article number 102406. doi: 10.1016/j.rcim.2022.102406.
  23. Serradilla, O., Zugasti, E., & Zurutuza, U. (2020). Deep learning models for predictive maintenance: A survey, comparison, challenges and prospect. ArXiv. doi: 10.48550/arXiv.2010.03207.
  24. Siraskar, R., Kumar, S., Patil, S., Bongale, A., & Kotecha, K. (2023). Reinforcement learning for predictive maintenance: A systematic technical review. Artificial Intelligence Review, 56(11), 12885-12947. doi: 10.1007/ s10462-023-10468-6.
  25. Supramaniam, A., Ahmad, S.S.S., & Yusoh, Z.I.M. (2024). Predictive maintenance using deep reinforcement learning. In Proceedings of the international conference on artificial intelligence in engineering and technology (pp. 671-676). Kota Kinabalu: IEEE. doi: 10.1109/IICAIET62352.2024.10730350.
  26. Tsallis, C., Papageorgas, P., Piromalis, D., & Munteanu, R.A. (2025). Application-wise review of machine learning-based predictive maintenance: Trends, challenges, and future directions. Applied Sciences, 15(9), article number 4898. doi: 10.3390/app15094898.
  27. Wang, L., You, P., Zhang, X., Jiang, L., & Li, Y. (2025). Adaptive adjustment graph representation learning method for rotating machinery fault diagnosis under noisy signals. Frontiers of Mechanical Engineering, 20(1), article number 2. doi: 10.1007/s11465-024-0818-y.
  28. Wang, Y., Qian, C., & Qin, J. (2023a). Attention-mechanism based DiPLS-LSTM and its application in industrial process time series big data prediction. Computers & Chemical Engineering, 176, article number 108296. doi: 10.1016/j.compchemeng.2023.108296.
  29. Wang, Y., Wu, M., Jin, R., Li, X., Xie, L., & Chen, Z. (2023b). Local-global correlation fusion-based graph neural network for remaining useful life prediction. IEEE Transactions on Neural Networks and Learning Systems, 36(1), 753-766. doi: 10.1109/TNNLS.2023.3330487.
  30. Yadav, D.K., Kaushik, A., & Yadav, N. (2024). Predicting machine failure using machine learning and deep learning algorithms. Sustainable Manufacturing and Service Economics, 3, article number 100029. doi: 10.1016/j. smse.2024.100029.
  31. Zhang, W., Yang, D., & Wang, H. (2019). Data-driven methods for predictive maintenance of industrial equipment: A survey. IEEE Systems Journal, 13(3), 2213-2227. doi: 10.1109/JSYST.2019.2905565.
  32. Zheng, L., Xue, W., Chen, F., Guo, P., Chen, J., & Chen, B. (2019). A fault prediction of equipment based on CNNLSTM network. In Proceedings of the international conference on energy internet (pp. 537-541). Nanjing: IEEE. doi: 10.1109/ICEI.2019.00101.
Demchyna, M. (2025). Neural networks for equipment failure prediction based on unstructured production data. Journal of Kryvyi Rih National University, 23(2), 104-114. https://doi.org/10.31721/2306-5451-2025-2-23-104-114
en