Transforming Fatigue Life Prediction in Additive Manufacturing: Synergies of Surrogate Modeling, Transfer Learning, and Bayesian Inference

Abstract:

Accurately modeling the fatigue life and strength of additively manufactured (AM) components ensures their reliability and performance in critical applications. However, this task is hindered by the complexities of AM processes, including material defects, anisotropy, residual stress, and surface roughness. This review explores how integrating surrogate modeling, transfer learning, and Bayesian inference can address these challenges and elevate predictive capabilities to new levels of accuracy and robustness. Surrogate modeling offers computationally efficient approximations of the intricate relationships between AM process parameters and fatigue behavior, enabling rapid exploration and optimization of design spaces. Transfer learning facilitates the adaptation of knowledge across different machines and process conditions, improving predictions even in low-data scenarios. Bayesian inference adds a layer of reliability by incorporating uncertainty quantification and prior knowledge into the modeling process. Together, these advanced methodologies present a transformative opportunity to improve the quality, efficiency, and robustness of fatigue life predictions for AM components, setting the stage for their broader adoption in high-performance applications

• Periodical: Key Engineering Materials (Volume 1035)
• Pages: 85-92
• DOI: https://doi.org/10.4028/p-2rwV2I
• Online since: December 2025
• Authors: Mustafa Mamduh Mustafa Awd*, Frank Walther
• Keywords: Additive Manufacturing, Bayesian Inference, Fatigue Life Prediction, Surrogate Modeling, Transfer Learning

References

[1] Wei Li, Ning Gao, Hongqiao Zhao, and Xinxin Xing. Crack initiation and early growth behavior of TC4 titanium alloy under high cycle fatigue and very high cycle fatigue. Journal of Materials Research, 33(8):935-945, 2018.

DOI: 10.1557/jmr.2017.476

[2] Emre Akgun, Xiang Zhang, Tristan Lowe, Yanhui Zhang, and Matthew Doré. Fatigue of laser powder-bed fusion additive manufactured Ti-6Al-4V in presence of process-induced porosity defects. Engineering Fracture Mechanics, 259, 2022.

DOI: 10.1016/j.engfracmech.2021.108140

[3] Ziyang Zhang, Qingyang Liu, and Dazhong Wu. Predicting stress-strain curves using transfer learning: Knowledge transfer across polymer composites. Materials & Design, 218, 2022.

DOI: 10.1016/j.matdes.2022.110700

[4] Alessio Centola, Alberto Ciampaglia, Andrea Tridello, and Davide Salvatore Paolino. Machine learning methods to predict the fatigue life of selectively laser melted Ti6Al4V components. Fatigue & Fracture of Engineering Materials & Structures, 46(11):4350-4370, 2023.

DOI: 10.1111/ffe.14125

[5] Arnaud Doucet, Simon Godsill, and Christian Robert. Marginal maximum a posteriori estimation using Markov chain Monte Carlo. Statistical Computing, 12(1):77-84, 2002. Publisher: Springer.

DOI: 10.1023/a:1013172322619

[6] Anton du Plessis, Ina Yadroitsava, and Igor Yadroitsev. Effects of defects on mechanical properties in metal additive manufacturing: A review focusing on X-ray tomography insights. Materials & Design, 187:108385, 2020.

DOI: 10.1016/j.matdes.2019.108385

[7] Mustafa Awd and Frank Walther. AI-Powered Very-High-Cycle Fatigue Control: Optimizing Microstructural Design for Selective Laser Melted Ti-6Al-4V. Materials, 2025.

DOI: 10.3390/ma18071472

[8] Lang Cheng, Zimeng Jiang, Hesai Wang, Chenguang Ma, Aoming Zhang, Honghong Du, Canneng Fang, Kai Wu, and Yingjie Zhang. Low-rank adaptive transfer learning based for multilabel defect detection in laser powder bed fusion. Optics and Lasers in Engineering, 184:108683, 2025.

DOI: 10.1016/j.optlaseng.2024.108683

[9] Erik Westphal and Hermann Seitz. A machine learning method for defect detection and visualization in selective laser sintering based on convolutional neural networks. Additive Manufacturing, 41:101965, 2021.

DOI: 10.1016/j.addma.2021.101965

[10] Omid Sedehi, Costas Papadimitriou, and Lambros S. Katafygiotis. Hierarchical Bayesian uncertainty quantification of Finite Element models using modal statistical information. Mechanical Systems and Signal Processing, 179:109296, 2022.[11] A. Das and N. Debnath. A Bayesian finite element model updating with combined normal and lognormal probability distributions using modal measurements. Applied Mathematical Modelling, 61:457-483, 2018.

DOI: 10.1016/j.ymssp.2022.109296

[12] Mustafa Awd, Shafaqat Siddique, and Frank Walther. Microstructural damage and fracture mechanisms of selective laser melted Al-Si alloys under fatigue loading. Theoretical and Applied Fracture Mechanics, 106:102483, 2020.

DOI: 10.1016/j.tafmec.2020.102483

[13] Mustafa Awd, Shafaqat Siddique, Jan Johannsen, Claus Emmelmann, and Frank Walther. Very high-cycle fatigue properties and microstructural damage mechanisms of selective laser melted AlSi10Mg alloy. International Journal of Fatigue, 124:55-69, 2019.

DOI: 10.1016/j.ijfatigue.2019.02.040

[14] Saurabh Gairola, Jayaganthan Rengaswamy, and Raviraj Verma. A study on XFEM simulation of tensile, fracture toughness, and fatigue crack growth behavior of Al 2024 alloy through fatigue crack growth rate models using genetic algorithm. Fatigue & Fracture of Engineering Materials & Structures, 2023.

DOI: 10.1111/ffe.13987