Explainable Machine Learning and 3D Visualization for Rotary Tube Bending: Expert Evaluation of a Web-Based Tool

Abstract:

When it comes to predicting quality-relevant outcomes of rotary draw bending procedures, like springback and geometric errors, machine learning algorithms have demonstrated encouraging results. However, the challenges associated with understanding these models’ predictions still restrict their actual application in industrial contexts. A web-based 3D visualization designed to help with interactive exploration and explainability of machine learning predictions in rotary draw bending is evaluated through an expert-centered study. Based on an earlier Random Forest regression model, the visualization lets users change important process parameters and view the projected tube geometry and springback in real time. Sixteen experts participated in a structured online survey that combined openended comments, subjective agreement scores, and interactive parameter modification tasks. Results show that while multi-objective optimization remained difficult, participants with different degrees of machine learning knowledge and tube-bending experience were generally able to identify appropriate parameter settings in single-objective problems. Subjective assessment and qualitative feedback from the participants also highlight that the visualization could be used to assist in understanding model behavior and in early process design and training situations. Overall, our study suggests that experts in tube bending applications find benefit from the interactive 3D visualization of the predicted geometry and as a useful interface for exploring machine learning models’ predictions.

• Periodical: Key Engineering Materials (Volume 1050)
• Pages: 43-57
• DOI: https://doi.org/10.4028/p-g76Hu9
• Online since: April 2026
• Authors: Alireza Yazdani*, Jonas Knoche, Bernd Engel, Kristof Van Laerhoven
• Keywords: Explainable AI, Machine Learning, Rotary Tube Bending, User Evaluation

References

[1] J. Rajpal, S.G. Thorat, and B.S. Kothavale: Finite element analysis and mathematical calculation of springback in rotary draw tube bending, Int. J. Innov. Technol. Res. Vol. 3 (2015), No. 3, p.2086–2100.

[2] L.Sözen M.A. Güler, D.Bekar,and E.Acar: Investigation and prediction of spring back in rotary draw tube bending process using finite element method, Proc. IMechE Part C: J. Mechanical Engineering Science Vol. 226 (2012), No. 12, p.2967–2981.

DOI: 10.1177/0954406212440672

[3] G.Y. Zhao Y.L. Liu,H.Yang C.H.Lu,andR.J.Gu:Three-dimensionalfinite-element modeling and simulation of rotary-draw bending process for thin-walled rectangular tube, Mater. Sci. Eng. AVol. 499 (2009), p.257–261.

DOI: 10.1016/j.msea.2007.11.127

[4] R.L. Amaral D.J. Cruz, M. Almeida, M.Barros A.D. Santos,and A.Reis:Analysis of springback in rotary draw bending process:numerical modelling with experimental validation, MATEC Web Conf. Vol. 408 (2025), Art. No. 01077.

DOI: 10.1051/matecconf/202540801077

[5] G. Naderi, S.E. Moussavi Torshizi, and S.H. Dibajian: Experimental-numerical study of wrin kling in rotary-draw bending of tight fit pipes, Thin-Walled Struct. Vol. 183 (2023), Art. No. 110428.

DOI: 10.1016/j.tws.2022.110428

[6] R.Safdarian: Experimental and numerical investigation of wrinkling and tube ovality in therotary draw bending process, Proc. IMechEPart C:J. MechanicalEngineering Science Vol. 233 (2019), No. 16, p.5568–5584.

DOI: 10.1177/0954406219850857

[7] P.A. Prates and A.F.G. Pereira: Recent advances and applications of machine learning in metal forming processes, Metals Vol. 12 (2022), Art. No. 1342.

DOI: 10.3390/met12081342

[8] S.J. Plathottam, A. Rzonca, R. Lakhnori, and C.O. Iloeje: A review of artificial intelligence ap plications in manufacturing operations, J. Adv. Manuf. Process. Vol. 5 (2023), Art. No. e10159.

DOI: 10.1002/amp2.10159

[9] J. Ma, H. Li, G.Y. Chen, T. Welo, and G.J. Li: Machine learning-based prediction and compen sation of springback for tube bending, in: Forming the Future, The Minerals, Metals & Materials Series (2021), p.167–176.

DOI: 10.1007/978-3-030-75381-8_13

[10] Z. Wang, T. Zhou, S. Zhang, C. Sun, J. Li and J. Tan: Bo-LSTM-based cross-sectional profile sequence progressive prediction method for metal tube rotary draw bending, Adv. Eng. Inform. Vol. 58 (2023), Art. No. 102152.

DOI: 10.1016/j.aei.2023.102152

[11] A. Yazdani, J. Knoche, B. Engel, and K. Van Laerhoven: Tube geometry prediction in rotary draw bending process using Random Forestregression, Automatisierungstechnik Vol. 73 (2025), No. 4, p.223–231.

DOI: 10.1515/auto-2024-0131

[12] N. Samadiani, A.S. Barnard, D. Gunasegaram, and N. Fayyazifar: Best practices for machine learning strategies aimed at process parameter development in powder bed fusion additive manufacturing, J. Intell. Manuf. Vol. 36 (2025), p.4477–4517.

DOI: 10.1007/s10845-024-02490-4

[13] C.V. Goldman, M. Baltaxe, D. Chakraborty, and J. Arinez: Explaining learning models in manufacturing processes, Procedia Comput. Sci. Vol. 180 (2021), p.259–268.

DOI: 10.1016/j.procs.2021.01.163

[14] G. Tzionis, P. Mouratidis, G. Kougka, I. Gialampoukidis, S. Vrochidis, I. Kompatsiaris, and M. Vlachopoulou: A review of explainable AI methods and their application in manufacturing systems, Discov. Appl. Sci. Vol. 14 (2025), Art. No. 898.

DOI: 10.1007/s42452-025-07908-z

[15] A. Gerling, C. Seiffer, H. Ziekow, U. Schreier, A. Hess and D. Ould Abdeslam: Evaluation of visualization concepts for explainable machine learning methods in the context of manufactur ing, Proc. 5th Int. Conf. On Computer-Human Interaction Research and Applications (CHIRA) (2021), p.189–201.

DOI: 10.5220/0010688900003060

[16] S. Yoo and N. Kang: Explainable artificial intelligence for manufacturing cost estimation and machining feature visualization, Expert Syst. Appl. Vol. 183 (2021), Art. No. 115430.

DOI: 10.1016/j.eswa.2021.115430

[17] K. Kucher, E. Zohrevandi and C.A.L. Westin: Towards Visual Analytics for Explainable AI in Industrial Applications, Analytics Vol. 4 (2025), No. 1, Article 7.

DOI: 10.3390/analytics4010007