AI-Based Predictive Maintenance System for Bearings

Abstract

Bearing failures in rotating machinery can lead to unplanned downtime, increased maintenance costs, and reduced operational reliability. To address these challenges, this paper proposes an artificial intelligence (AI)-based predictive maintenance framework for bearing condition monitoring using temperature, vibration, and electrical-current measurements. The proposed methodology comprises sensor-data acquisition, preprocessing, feature extraction, machine-learning model development, and condition classification. Candidate machine-learning algorithms will be evaluated using labelled data to distinguish normal bearing operation from abnormal conditions. Performance will be assessed using accuracy, precision, recall, F1-score, and confusion-matrix analysis. Where suitable datasets are available, individual sensor inputs will be compared with combined sensor inputs to investigate the potential benefits of sensor fusion. The framework is intended to support condition monitoring by presenting the estimated bearing status and providing a basis for maintenance decisions. Its effectiveness will depend on data quality, representative operating conditions, reliable fault labelsand independent model validation. This work establishes a proposed methodology for investigating the feasibility of integrating multiple sensor measurements with machine learning for bearing predictive maintenance.

Introduction

This research focuses on developing an AI-based predictive maintenance system for bearing-condition monitoring in rotating machinery such as electric motors, pumps, and gearboxes. Bearing faults can reduce machine performance, cause unexpected breakdowns, increase maintenance costs, and lead to unplanned downtime. Therefore, early detection of abnormal bearing conditions is essential for improving equipment reliability and operational efficiency.

Traditional maintenance methods, such as scheduled and corrective maintenance, may not detect faults at an early stage. Predictive maintenance provides an alternative by using condition-monitoring data to identify developing faults and support timely maintenance decisions.

The proposed system uses three sensor inputs: vibration, temperature, and electrical-current measurements. These measurements will be collected, preprocessed, and used to extract relevant features for machine-learning-based classification. Different algorithms, including Logistic Regression, Support Vector Machine (SVM), Random Forest, and Gradient Boosting, will be evaluated to classify bearing conditions as normal or faulty.

The literature review highlights the importance of vibration analysis, signal processing, artificial intelligence, and machine learning in bearing-fault diagnosis. Previous studies have explored techniques such as convolutional neural networks, support vector machines, artificial neural networks, wavelet analysis, and deep learning for fault detection and remaining useful life prediction.

The methodology involves the following stages:

1. Data Collection: Collect vibration, temperature, and motor-current measurements with corresponding bearing-condition labels.

2. Data Preprocessing: Handle missing values, remove noise where appropriate, synchronize sensor data, extract features, and normalize measurements.

3. Model Development: Develop and train suitable supervised machine-learning models.

4. Model Evaluation: Evaluate model performance using accuracy, precision, recall, F1-score, and confusion matrices.

5. Sensor Comparison: Compare individual sensor inputs with combined sensor measurements to investigate whether sensor integration improves classification performance.

6. Condition Prediction: Classify bearing conditions as normal or faulty and display the predicted condition to support maintenance decisions.

The study also considers challenges such as sensor reliability, insufficient fault data, variations in operating conditions, and model generalization across different machines.

The expected outcome is a machine-learning-based bearing-monitoring framework capable of identifying abnormal operating conditions at an early stage. The system aims to improve predictive maintenance, reduce unexpected equipment failures, support maintenance planning, and provide a foundation for future development of intelligent condition-monitoring systems.

Conclusion

This study proposes a machine learning-based predictive maintenance framework for bearing-condition monitoring using vibration, temperatureand motor-current measurements. The proposed approach aims to identify abnormal bearing conditions by analyzing sensor data and learning patterns associated with normal and faulty operating states. Machine learning models such as Logistic Regression, Support Vector Machine (SVM)and Random Forest will be considered to investigate their suitability for bearing-fault classification.The integration of machine learning into bearing-condition monitoring has the potential to support early fault identification, improve equipment reliabilityand assist maintenance planning. However, the effectiveness of the proposed framework will depend on the quality of the collected data, the selection of relevant featuresand the model\'s ability to generalize across different operating conditions. Experimental implementation and performance evaluation are required to establish its actual accuracy and practical benefits.

References

[1] R. Singh, R. Ojha, and S. P. Harsha, “Intelligent fault diagnosis of roller bearings using a transfer learning-based dual-input spectrogram–scalogram convolutional neural network model,” Proceedings of the Institution of Mechanical Engineers, Part K: Journal of Multi-body Dynamics, 2026. doi: 10.1177/14644193261438346. [2] A. Mathur, P. Kumar, and S. P. Harsha, “Ranked feature-based data-driven bearing fault diagnosis using support vector machine and artificial neural network,” Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, vol. 237, no. 9, pp. 1602–1619, 2023. doi: 10.1177/09596518231162892. [3] X. Zhang, B. Zhao, and Y. Lin, “Machine learning-based bearing fault diagnosis using the Case Western Reserve University data: A review,” IEEE Access, vol. 9, pp. 155598–155608, 2021. doi: 10.1109/ACCESS.2021.3128669. [4] Y. Lei, B. Yang, X. Jiang, F. Jia, N. Li, and A. K. Nandi, “Applications of machine learning to machine fault diagnosis: A review and roadmap,” Mechanical Systems and Signal Processing, vol. 138, article 106587, 2020. doi: 10.1016/j.ymssp.2019.106587. [5] R. Liu, B. Yang, E. Zio, and X. Chen, “Artificial intelligence for fault diagnosis of rotating machinery: A review,” Mechanical Systems and Signal Processing, vol. 108, pp. 33–47, 2018. [6] M. Cerrada, R.-V. Sánchez, C. Li, F. Pacheco, D. Cabrera, J. Valente de Oliveira, and R. E. Vásquez, “A review on data-driven fault severity assessment in rolling bearings,” Mechanical Systems and Signal Processing, vol. 99, pp. 169–196, 2018. doi: 10.1016/j.ymssp.2017.06.012. [7] Y. Lei, N. Li, L. Guo, N. Li, T. Yan, and J. Lin, “Machinery health prognostics: A systematic review from data acquisition to RUL prediction,” Mechanical Systems and Signal Processing, vol. 104, pp. 799–834, 2018. doi: 10.1016/j.ymssp.2017.11.016. [8] W. Zhang, G. Peng, C. Li, Y. Chen, and Z. Zhang, “A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals,” Sensors, vol. 17, no. 2, article 425, 2017. doi: 10.3390/s17020425. [9] L. Guo, N. Li, F. Jia, Y. Lei, and J. Lin, “A recurrent neural network-based health indicator for remaining useful life prediction of bearings,” Neurocomputing, vol. 240, pp. 98–109, 2017. doi: 10.1016/j.neucom.2017.02.045. [10] X. Guo, L. Chen, and C. Shen, “Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosis,” Measurement, vol. 93, pp. 490–502, 2016. doi: 10.1016/j.measurement.2016.07.054. [11] W. A. Smith and R. B. Randall, “Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study,” Mechanical Systems and Signal Processing, vols. 64–65, pp. 100–131, 2015. [12] R. B. Randall and J. Antoni, “Rolling element bearing diagnostics: A tutorial,” Mechanical Systems and Signal Processing, vol. 25, no. 2, pp. 485–520, 2011. doi: 10.1016/j.ymssp.2010.07.017. [13] P. K. Kankar, S. C. Sharma, and S. P. Harsha, “Fault diagnosis of ball bearings using machine learning methods,” Expert Systems with Applications, vol. 38, no. 3, pp. 1876–1886, 2011. doi: 10.1016/j.eswa.2010.07.119. [14] P. K. Kankar, S. C. Sharma, and S. P. Harsha, “Fault diagnosis of ball bearings using continuous wavelet transform,” Applied Soft Computing, vol. 11, no. 2, pp. 2300–2312, 2011. [15] A. K. S. Jardine, D. Lin, and D. Banjevic, “A review on machinery diagnostics and prognostics implementing condition-based maintenance,” Mechanical Systems and Signal Processing, vol. 20, no. 7, pp. 1483–1510, 2006. doi: 10.1016/j.ymssp.2005.09.012. [16] J. Chen, C. Lin, B. Yao, L. Yang, and H. Ge, “Intelligent fault diagnosis of rolling bearings with low-quality data: A Feature significance and diversity learning method,” Reliability Engineering & System Safety, vol. 237, Art. no. 109343, 2023. doi: 10.1016/j.ress.2023.109343. [17] A. Mathur, P. Kumar, and S. P. Harsha, “Ranked feature-based data-driven bearing fault diagnosis using support vector machine and artificial neural network,” Proceedings of the Institution of Mechanical Engineers, Part C: Journal of MechanicalEngineering Science, 2023. doi: 10.1177/09596518231162892. [18] J. Lu, K. Wang, C. Chen, and W. Ji, “A deep learning method for rolling bearing fault diagnosis based on attention mechanism and Gramian angular field,” Sensors, vol. 23, no. 12, Art. no. 5487, 2023. doi: 10.3390/s23125487. [19] “An improved multi-scale branching convolutional neural network for rolling bearing fault diagnosis,” PLOS ONE, 2023. doi: 10.1371/journal.pone.0291353. [20] X. Li, P. Yuan, X. Wang, D. Li, Z. Xie, and X. Kong, “An unsupervised transfer learning bearing fault diagnosis method based on depth-wise separable convolution,” Measurement Science and Technology, vol. 34, Art. no. 095401, 2023. doi: 10.1088/1361-6501/acda55. [21] “A systematic review of rolling bearing fault diagnoses based on deep learning and transfer learning: Taxonomy, overview, application, open challenges, weaknesses and recommendations,” Ain Shams Engineering Journal, vol. 14, no. 4, Art. no. 101945, 2023. doi: 10.1016/j.asej.2022.101945. [22] L. Shi et al., “Bearing fault diagnosis method based on deep learning and health state division,” Applied Sciences, vol. 13, no. 13, Art. no. 7424, 2023. doi: 10.3390/app13137424. [23] “Development of intelligent fault diagnosis technique of rotary machine element bearing: A machine learning approach,” Sensors, vol. 22, no. 3, Art. no. 1073, 2022. doi: 10.3390/s22031073. [24] “Explainable AI for bearing fault prognosis using deep learning techniques,” Micromachines, vol. 13, no. 9, Art. no. 1471, 2022. doi: 10.3390/mi13091471. [25] “Bearing fault diagnosis based on discriminant analysis using multi-view learning,” Mathematics, vol. 10, no. 20, Art. no. 3889, 2022. doi: 10.3390/math10203889. [26] “Deep learning aided data-driven fault diagnosis of rotatory machine: A comprehensive review,” Energies, vol. 14, no. 16, Art. no. 5150, 2021. doi: 10.3390/en14165150. [27] S. Schwendemann, Z. Amjad, and A. Sikora, “A survey of machine-learning techniques for condition monitoring and predictive maintenance of bearings in grinding machines,” Computers in Industry, vol. 125, Art. no. 103380, 2021. doi: 10.1016/j.compind.2020.103380. [28] M. Adamsab, “Machine learning algorithms for rotating machinery bearing fault diagnostics,” Materials Today: Proceedings, vol. 44, pp. 4931–4933, 2021. doi: 10.1016/j.matpr.2020.12.050. [29] S. Zhang, S. Zhang, B. Wang, and T. G. Habetler, “Deep learning algorithms for bearing fault diagnostics—A comprehensive review,” IEEE Access, vol. 8, pp. 29857–29881, 2020. doi: 10.1109/ACCESS.2020.2972859. [30] D.-T. Hoang and H.-J. Kang, “A survey on deep learning-based bearing fault diagnosis,” Neurocomputing, vol. 335, pp. 327–335, 2019. doi: 10.1016/j.neucom.2018.06.078.

Copyright

Copyright © 2026 Dr. N. N. Gotkhindikar, Sarvesh Thorat, Nikhil Jain, Pranav Nyati, Shahid Shah. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.