Authors : Özkan Bartu Leylek; Ömür Şansal Çenberli; Muhammed Kürşad Uçar

Volume/Issue : Volume 10 - 2025, Issue 12 - December

Google Scholar : https://tinyurl.com/3d2unbun

Scribd : https://tinyurl.com/bdhrme2h

DOI : https://doi.org/10.38124/ijisrt/25dec776

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 30 to 40 days to display the article.

Abstract : Objective: With the advent of Industry 4.0, the complexity of production systems has increased, making the early prediction of machine failures critical for operational efficiency and continuity. In this study, a two-stage artificial intelligence-based predictive maintenance model is developed to detect machine failures and identify failure types in industrial systems.  Methodology: The AI4I 2020 Predictive Maintenance dataset was used in this research. In the first stage, the most significant input features were identified through feature selection based on the Spearman correlation coefficient, followed by the application of a systematic sampling method to address data imbalance. During the fault detection stage, various machine learning algorithms (SVM, Ensemble Trees, Artificial Neural Networks, etc.) were comparatively evaluated. In the second stage, a partial least squares (PLS)-based modeling approach was employed for the classification of failure types.  Results: In the first-stage results, the SVM and Ensemble Trees models demonstrated the highest performance, achieving accuracy rates above 92% and an AUC value of 0.98. In the second stage, the PLS based model achieved classification accuracies exceeding 95%, particularly for datasets consisting of six and seven features.  Conclusion: The proposed two-stage predictive maintenance model offers a practical artificial intelligence solution that can contribute to the optimization of maintenance planning, enhancement of operational continuity, and reduction of maintenance costs in industrial systems. Owing to its modular structure, the model can be adapted to different production lines and is considered a decision-support tool that can be integrated into Industry 4.0 infrastructures.

Keywords : Predictive Maintenance; Artificial Intelligence; Industry 4.0; Machine Learning; Fault Prediction.

References :

1. A. Bahador et al., “Condition Monitoring for Predictive Maintenance of Machines and Processes in ARTC Model Factory,” Intelligent Systems Reference Library, vol. 202, pp. 113–141, 2021.
2. K. Shukla, S. Nefti-Meziani, and S. Davis, “A heuristic approach on predictive maintenance techniques: Limitations and scope,” Advances in Mechanical Engineering, vol. 14, no. 6, Jun. 2022.
3. S. Maataoui, G. Bencheikh, and G. Bencheikh, “Predictive Maintenance in the Industrial Sector: A CRISP-DM Approach for Developing Accurate Machine Failure Prediction Models,” 2023 5th International Conference on Advances in Computational Tools for Engineering Applications, ACTEA 2023, pp. 223–227, 2023.
4. T. Frost, N. Moser, J. Nöcker, and L. Zhukov, “Getting Value from Predictive Maintenance,” Mechanisms and Machine Science, vol. 105, pp. 178–185, 2021.
5. N. Es-sakali, M. Cherkaoui, M. O. Mghazli, and Z. Naimi, “Review of predictive maintenance algorithms applied to HVAC systems,” Energy Reports, vol. 8, pp. 1003–1012, Nov. 2022.
6. H. M. Hashemian and W. C. Bean, “State-of-the-art predictive maintenance techniques,” IEEE Trans Instrum Meas, vol. 60, no. 10, pp. 3480–3492, Oct. 2011.
7. P. Lu, H. Liu, C. Serratella, and X. Wang, “Assessment of Data-Driven, Machine Learning Techniques for Machinery Prognostics of Offshore Assets,” Proceedings of the Annual Offshore Technology Conference, vol. 5, pp. 3651–3672, May 2017.
8. F. E. Bezerra et al., “Impacts of Feature Selection on Predicting Machine Failures by Machine Learning Algorithms,” Applied Sciences 2024, Vol. 14, Page 3337, vol. 14, no. 8, p. 3337, Apr. 2024.
9. S. Y. Baroud, N. A. Yahaya, and A. M. Elzamly, “Cutting-Edge AI Approaches with MAS for PdM in Industry 4.0: Challenges and Future Directions,” Journal of Applied Data Sciences, vol. 5, no. 2, pp. 455–473, May 2024.
10. J. Karina Fernández Salazar et al., “Machine Learning for Predictive Maintenance in Industry 4.0: A Systematic Review of Algorithms and Implementation Cases,” in Proceedings of the LACCEI international Multi-conference for Engineering, Education and Technology, Mexico City,: Latin American and Caribbean Consortium of Engineering Institutions, 2025.
11. S. Xie, “Advancing Predictive Maintenance Research Trends: Using Artificial Intelligence for Enhanced Industrial Reliability,” Proceedings - 2024 IEEE International Conference on Future Machine Learning and Data Science, FMLDS 2024, pp. 283–288, 2024.
12. A. Belounnasa, F. Brissaud, and P. Faly Ba, “Predictive Maintenance of Natural Gas Regulators by Forecasting Output Pressure with Artificial Intelligence Algorithms,” in Proc. of the 31st European Safety and Reliability Conference (ESREL 2021), B. Castanier, M. Cepin, D. Bigaud, and C. Berenguer, Eds., Singapore: Research Publishing, Sep. 2021, pp. 2150–2150.
13. A. Sinha and D. Das, “Data-Driven Techniques for Fault Diagnosis and Predictive Maintenance,” in Applied Artificial Intelligence and Machine Learning Techniques for Engineering Applications, CRC Press, 2025, pp. 30–49.
14. S. Madasamy, B. P. Shankar, R. K. Yadav, and K. P. Jayalakshmi, “A Machine Learning Approach in Predictive Maintenance in the IoT Enabled Industry 4.0,” Proceedings of the 4th International Conference on Smart Electronics and Communication, ICOSEC 2023, pp. 418–423, 2023.
15. H. Meriem, H. Nora, and O. Samir, “Predictive Maintenance for Smart Industrial Systems: A Roadmap,” Procedia Comput Sci, vol. 220, pp. 645–650, Jan. 2023.
16. V. M. Caldana, F. D. G. da Silva, R. A. de Oliveira, and J. F. Borin, “Internet of Things and Artificial Intelligence applied to predictive maintenance in Industry 4.0: A systematic literature review,” in Proceedings of the International Conference on Industrial Engineering and Operations Management, Michigan, USA: IEOM Society International, Apr. 2021, pp. 1387–1398.
17. S. Matzka, “Explainable Artificial Intelligence for Predictive Maintenance Applications,” Proceedings - 2020 3rd International Conference on Artificial Intelligence for Industries, AI4I 2020, pp. 69–74, Sep. 2020.
18. “AI4I 2020 Predictive Maintenance Dataset - UC Irvine Machine Learning Repository.” doi: 10.24432/C5HS5C.
19. R. Alpar, Applied Statistic and Validation - Reliability. Detay Publishing, 2010. [Online]. Available: https://books.google.com.tr/books/about/Uygulamalı_istatistik_ve_gecerlik_guv.html?id=ITk1MwEACAAJ&pgis=1
20. M. K. Uçar, M. Nour, H. Sindi, and K. Polat, “The Effect of Training and Testing Process on Machine Learning in Biomedical Datasets,” Math Probl Eng, vol. 2020, pp. 1–17, 2020.
21. M. Fernández-Delgado, E. Cernadas, S. Barro, D. Amorim, and A. Fernández-Delgado, “Do we Need Hundreds of Classifiers to Solve Real World Classification Problems?,” Journal of Machine Learning Research, vol. 15, no. 90, pp. 3133–3181, 2014, Accessed: Dec. 11, 2025.
22. H. Herv´ and H. Abdi, “Partial least squares regression and projection on latent structure regression (PLS Regression),” Wiley Interdiscip Rev Comput Stat, vol. 2, no. 1, pp. 97–106, Sep. 2010.
23. T. Fawcett, “An introduction to ROC analysis,” Pattern Recognit Lett, vol. 27, no. 8, pp. 861–874, Jun. 2006.