Prediction of Rotating-Bending Fatigue Strength in Steel at 10⁷ Cycles: A Machine Learning Framework Applied to The NIMS Experimental Database‎

Hasan Abbas Flayyih, Ahmed Ali Farhan Ogaili, Alaa Abdulhady Jaber, Lutif.A. Al-Haddad, Abdul Rasool Kareem Jweri, Emad Kadum Njim

Abstract

Precise prediction of fatigue strength in steel is essential but challenging because the effects of chemical composition, microstructure, and processing conditions are complex and nonlinear. Conventional experimental fatigue testing is resource-intensive, motivating the development of data-driven approaches. The present study is specifically focused on rotating bending fatigue strength at 10⁷ cycles using the National Institute for Materials Science (NIMS) experimental database. Therefore, the findings may not be directly applicable to other loading modes or cycle counts without further validation. In this paper, various machine learning models, including Linear Regression (LR), Support Vector Regression (SVR), K-Nearest Neighbors (KNN), Artificial Neural Network (ANN), and Gradient Boosting (GB), were trained and compared using evaluation metrics on a large fatigue dataset. The coefficient of determination (R²), root mean squared error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) were used to assess model performance. Gradient Boosting achieved the best predictive performance (R² = 0.9984), with a MAPE of only 1.07% and an RMSE of 3.51 MPa, substantially outperforming all other models. Linear Regression achieved a respectable R² of 0.9493, whereas ANN (R² = 0.603) and SVR (R² = 0.605) demonstrated limited performance on this high-dimensional tabular dataset, highlighting the advantage of tree-based ensemble methods. These findings demonstrate the strong potential of nonlinear learning algorithms for data-driven fatigue behavior modeling. The proposed framework improves fatigue prediction accuracy and provides a scalable and interpretable approach for materials design and reliability assessment. However, the results are specific to rotating bending fatigue at 10⁷ cycles based on the NIMS database, and broader generalization requires additional experimental validation.

Keywords

References

[1] J. A. Bannantine, J. J. Comer, and J. L. Handrock, Fundamentals of Metal Fatigue Analysis. Englewood Cliffs, NJ: Prentice Hall, 1990.

[2] J. Schijve, Fatigue of Structures and Materials, 2nd ed. Dordrecht, Netherlands: Springer, 2009, doi: 10.1007/978-1-4020-6808-9.

[3] L. M. Nassir et al., “Integrated experimental, statistical, and finite element analysis of nanoparticle-reinforced polymer composites for advanced structural applications completed with bibliometric analysis,” ASEAN Journal for Science and Engineering in Materials, vol. 5, no. 2, pp. 287–310, 2025. [Online]. Available: https://ejournal.bumipub likasinusantara.id/index.php/ajsem/article/view/849

[4] H. Wang, X. Liu, M. Zhang, Y. Wang, and X. Wang, “Prediction of material fatigue parameters for low alloy forged steels considering error circle,” International Journal of Fatigue, vol. 121, pp. 135–145, 2019, doi: 10.1016/j.ijfatigue.2018.12. 002.

[5] G. Mi, G. Sun, S. Yang, X. Liu, S. Chen, and W. Kang, “Machine learning-based prediction of fatigue fracture locations in 7075-T651 aluminum alloy friction stir welded joints,” Metals, vol. 15, no. 5, p. 569, 2025, doi: 10.3390/met15050569.

[6] S. Farhadi, S. Tatullo, and F. Ferrian, “Comparative analysis of ensemble learning techniques for enhanced fatigue life prediction,” Scientific Reports, vol. 15, Art. no. 11136, 2025, doi: 10.10 38/s41598-024-79476-y.

[7] R. Ramprasad, R. Batra, G. Pilania, A. Mannodi-Kanakkithodi, and C. Kim, “Machine learning in materials informatics: recent applications and prospects,” npj Computational Materials, vol. 3, no. 1, art. no. 54, 2017, doi: 10.1038/s41524-01 7-0056-5.

[8] A. A. Nayeeif, E. S. Al-Ameen, N. A. Jebur, A. A. Farhan Ogaili, Z. K. Hamdan, and E. K. Njim, “Investigation of the effects of unbalance and bearing wear on shaft vibration in a natural gas turbine plant,” Applied Science and Engineering Progress, vol. 18, no. 4, Art. no. 7863, Oct. 2025, doi: 10.14416/j.asep.2025.07.0 12.

[9] A. Agrawal, P. D. Deshpande, A. Cecen, G. P. Basavarsu, A. N. Choudhary, and S. R. Kalidindi, “Exploration of data science techniques to predict fatigue strength of steel from composition and processing parameters,” Integrating Materials and Manufacturing Innovation, vol. 3, no. 1, pp. 90–108, 2014, doi: 10.1186/2193-9772-3-8.

[10] T. Shiraiwa, Y. Miyazawa, and M. Enoki, “Prediction of fatigue strength in steels by linear regression and neural network,” Materials Transactions, vol. 60, no. 2, pp. 189–198, 2019, doi: 10.2320/matertrans.ME201714.

[11] M. YI et al., “Machine learning for predicting fatigue properties of additively ‎manufactured materials,” Chinese Journal of Aeronautics, vol. 37, no. 4, pp. 1–22, ‎Apr. 2024, doi: 10.1016/ j.cja.2023.11.001.

[12] N. Nagashima, M. Hayakawa, H. Masuda, and K. Nagai, “Estimating the S–N curve by machine learning random forest method,” Materials Transactions, vol. 65, no. 4, pp. 428–433, 2024, doi: 10.2320/matertrans.MT-Z2023006.

[13] Z. Huang et al., “Deep learning-based fatigue strength prediction for ferrous alloy,” Processes, vol. 12, no. 10, p. 2214, 2024, doi: 10.3390/pr12102214.

[14] X. Zhang, F. Liu, M. Shen, D. Han, Z. Wang, and N. Yan, “Ultra-high-cycle fatigue life prediction of metallic materials based on machine learning,” Applied Sciences, vol. 13, no. 4, p. 2524, 2023, doi: 10.3390/app13042524.

[15] C. Liu, X. Wang, W. Cai, J. Yang, and H. Su, “Prediction of the fatigue strength of steel based on interpretable machine learning,” Materials, vol. 16, no. 23, p. 7354, 2023, doi: 10.3390/ma16237354.

[16] ‎C. D. Naiju, M. Adithan, and P. Radhakrishnan, “An investigation of process variables influencing fatigue properties of components produced by direct metal laser sintering,” Applied Science and Engineering Progress, vol. 4, no. 1, pp. 63–69, Sep. 2013. [Online]. Available: https://ph02.tci-thaijo.org/index. php/ijast/article/view/67381.

[17] B. D. S. Deeraj, K. Joseph, J. S. Jayan, and A. Saritha, “Dynamic mechanical performance of natural fiber reinforced composites: A brief review,” Applied Science and Engineering Progress, vol. 14, no. 4, pp. 614–623, Oct. 2021, doi: 10.14416/j.asep. 2021. 06.003.

[18] P. Thomas, C. W. Lai, and M. R. Johan, “Prospective of magnesium and alloy-based composites for lightweight railway rolling stocks,” Applied Science and Engineering Progress, vol. 15, no. 2, p. 5716, May 2022, doi: 10.14416/j.asep. 2022.02.006.

[19] S. Walzer, M. Liewald, N. Simon, J. Gibmeier, H. Erdle, and T. Böhlke, “Improvement of sheet metal properties by inducing residual stresses into sheet metal components by embossing and reforming,” Applied Science and Engineering Progress, vol. 15, no. 1, p. 5437, Oct. 2021, doi: 10.14416/j.asep.2021.09.006.

[20] A. A. F. Ogaili, Z. T. Al-Sharify, A. A. Jaber, and F. A. Abdulla, “Vibration-based fault detection and classification in ball bearings using statistical analysis and random forest,” in Proceedings of the 5th International Conference on Green Energy, Environment, and Sustainable Development (GEESD 2024), vol. 13279, 2024, p. 1327921, doi: 10.1117/12.3041850.

[21] A. S. Abdul-Zahra, E. Ghane, A. Kamali, and A. A. Farhan Ogaili, “Power forecasting in continuous extrusion of pure titanium using Naïve Bayes algorithm,” Terra Joule Journal, vol. 1, no. 1, p. 2, 2024, doi: 10.64071/3080-5724.1000.

[22] A. A. F. Ogaili, Z. T. Al-Sharify, A. A. Jaber, D. A. Farhan, and S. M. Al-Khafaji, “Effective ball bearing fault diagnosis leveraging ANN and statistical feature integration,” in Proceedings of the 10th Scholar’s Yearly Symposium of Technology, Engineering and Mathematics (SYSTEM 2024), 2024, pp. 49–60. [Online]. Available: https://ceur-ws.org/Vol-3870/p06.pdf

[23] S. A. Sarow, H. A. Flayyih, M. Bazerkan, L. A. Al-Haddad, Z. T. Al-Sharify, and A. A. F. Ogaili, “Advancing sustainable renewable energy: XGBoost algorithm for the prediction of water yield in hemispherical solar stills,” Discover Sustainability, vol. 5, no. 1, p. 510, 2024, doi: 10.1007/s43621-024-00782-6.

[24] A.-R. K. Jweri, L. A. Al-Haddad, A. A. F. Ogaili, A. A. Jaber, and M. I. Al-Karkhi, “Edge-intelligent leak detection in water distribution systems using CatBoost: A sustainable solution for reducing infrastructure losses,” Clean Energy Science and Technology, vol. 3, no. 4, p. 398, 2025, doi: 10.18686/cest398.

[25] B. G. Mejbel, S. A. Sarow, M. T. Al-Sharify, L. A. Al-Haddad, A. A. F. Ogaili, and Z. T. Al-Sharify, “A data fusion analysis and random forest learning for enhanced control and failure diagnosis in rotating machinery,” Journal of Failure Analysis and Prevention, vol. 24, no. 6, pp. 2979–2989, Dec. 2024, doi: 10.1007/s11668 -024-02075-6.

[26] L. M. Nassir, A. J. Ramadhan, N. T. Al-Sharify, M. I. Khalaf, A. A. F. Ogaili, A. A. Jaber, and Z. T. Al-Sharify, “Robust multi-state EEG cognitive classification via optimized time-domain features and CatBoost,” International Journal of Robotics and Control Systems, vol. 5, no. 2, pp. 968–989, 2025, doi: 10.31763/ijrcs.v5i2.1340.

[27] A.-R. Jweri et al., “Enhancing predictive maintenance in energy systems using a hybrid Kolmogorov–Arnold network (KAN) with short-time Fourier transform (STFT) framework for rotating machinery,” ASEAN Journal of Science and Engineering, vol. 5, no. 2, pp. 465–494, 2025, doi: 10.17509/ajse.v5i2. 89023.

[28] Z. W. Metteb et al., “Optimization of hybrid core designs in 3D-printed PLA+ sandwich structures: An experimental, statistical, and computational investigation completed with bibliometric literature review,” Indonesian Journal of Science and Technology, vol. 10, no. 2, pp. 207–236, 2025, doi: 10.17509/ijost. v10i2.81743.

[29] A. A. F. Ogaili, K. A. Mohammed, A. A. Jaber, and E. S. Al-Ameen, “Automated wind turbines gearbox condition monitoring: A comparative study of machine learning techniques based on vibration analysis,” FME Transactions, vol. 52, no. 3, pp. 471–485, 2024, doi: 10.5937/fme2403471O.

[30] K. A. Mohammed, A. A. F. Ogaili, A. W. A. Taha, and A. M. Alsayah, “Evaluating spot welds of dissimilar metals via integrated mechanical testing and finite element modeling,” Applied Engineering Letters, vol. 10, no. 2, pp. 77–89, 2025, doi: 10.46793/ aeletters.2025.10.2.2.

[31] G. A. Muthulingam and V. S. Parvathy, “A novel bacterial foraging optimization based multimodal medical image fusion approach,” Applied Science and Engineering Progress, vol. 16, no. 4, p. 6794, Oct. 2023, doi: 10.14416/j.asep.2023.03.004.

[32] A. A. F. Ogaili, A. A. Jaber, and M. N. Hamzah, “Statistically optimal vibration feature selection for fault diagnosis in wind turbine blade,” International Journal of Renewable Energy Research, vol. 13, no. 3, pp. 1082–1092, 2023, doi: 10.20508/ijrer.v1 3i3.14096.

[33] S. T. Kurdi, L. A. Al-Haddad, and A. A. F. Ogaili, “Path optimization for aircraft based on geographic information systems and deep learning,” Automation, vol. 7, no. 1, Art. no. 12, Jan. 2026, doi: 10.3390/automation7010012.

[34] S. T. Bunyan et al, “Intelligent thermal condition monitoring for predictive maintenance of gas turbines using machine learning,” Machines, vol. 13, no. 5, Art. no. 401, May 2025, doi: 10.33 90/machines13050401.

[35] K. J. Mohammed, A. A. F. Ogaili, A. -R. K. Jweri, S. Amin, M. Khalaf, L. Al-Haddad, and A. Jaber, “Intelligent fault detection of UAV propellers through time-domain vibration analysis and ensemble learning,” Automated Systems, vol. 9, pp. 365–381, 2026, doi: 10.1007/s42401-026-00460-7.

[36] E. N. Abbas, E. K. Njim, M. J. Jweeg, R. Madan, A. A. F. Ogaili, and F. T. Al-Maliky, “Experimental and theoretical analysis of mechanical properties of composite materials with diverse reinforcement types,” World Journal of Engineering, vol. 23, no. 3, pp. 576–586, 2026, doi: 10.1108/WJE-07-2024-0433.

[37] N. T. Al-Sharify, A. F. H. AL-Maliki, A. A. F. Ogaili, A. -R. K. Jweri, A. A. Jaber, and L. A. Al-Haddad, “A hybrid XFEM-ML framework for high-fidelity, rapid fracture prediction in cortical bone: From microstructure to clinical translation,” Mesopotamian Journal of Computer Science, vol. 6, no. 1, p. 1, 2026, doi: 10.58496/2958-6631.1070.

Full Text: PDF

DOI: 10.14416/j.asep.2026.07.003

Refbacks

There are currently no refbacks.

Username
Password
Remember me