Statistical Modeling for weld quality assessment using AI SAW Welding of Mild Steel

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2026 | Volume : 14 | 02 | Page :
    By

    Mirza Farhatulla Baig,

  • Dharmendra Dubey,

  1. Research Scholar, Department of Mechanical Engineering Bhagwant University, Ajmer, Rajasthan, India
  2. Professor, Department of Mechanical Engineering, Shree Dhanvantary College of Engineering and Technology, Surat, Gujarat, India

Abstract

The main issue to the industries that apply Submerged Arc Welding (SAW) is quality assurance since the structural integrity dictates safety and the performance of the industry. The existing system of checking manuals is not only time consuming but also has human errors that make it mandatory to deploy automated intelligent systems. This study carries out an extensive comparison of the leading approaches based on the use of Artificial Intelligence (AI) that integrates statistical models with logistic regression and deep learning in the form of Convolutional Neural Networks (CNN) to transform the measurement of weld quality. The statistical model was demonstrated to be reliable and accurate with 98.89% cross-validation accuracy because the data utilized consisted of SAW weld images that are annotated with defects. CNN model performed well in pattern recognition of complex defect analysis based on the results of its evaluation that had a training accuracy of 87.85% and test accuracy of 55.06%. Statistical models as well as CNN models prove to be the strongest ones in the current study since the statistical models cause orderly feature analyses and CNN models suggest the potential of various possibilities of automated visual defects identifications. The study presents a basic base on which future hybrid evaluation techniques and real time inspection of welds technology can be developed. The high accuracy of the statistical model demonstrates that feature-based approaches can serve as reliable, low-cost tools for real-time weld inspection, while the CNN results reveal the potential and current challenges of automated image-based defect detection. These insights help manufacturers decide which model is operationally suitable for quality assurance and provide direction for developing more robust, hybrid AI-driven inspection systems for SAW processes.

Keywords: Weld Quality Assessment, Submerged Arc Welding, Artificial Intelligence, Statistical Modeling, Deep Learning

How to cite this article:
Mirza Farhatulla Baig, Dharmendra Dubey. Statistical Modeling for weld quality assessment using AI SAW Welding of Mild Steel. Journal of Polymer & Composites. 2026; 14(02):-.
How to cite this URL:
Mirza Farhatulla Baig, Dharmendra Dubey. Statistical Modeling for weld quality assessment using AI SAW Welding of Mild Steel. Journal of Polymer & Composites. 2026; 14(02):-. Available from: https://journals.stmjournals.com/jopc/article=2026/view=240987


References

  1. Gook S, El-Sari B, Biegler M, Rethmeier M. Application of AI-based welding process monitoring for quality control in pipe production. Paton Weld J. 2024;6:3–8.
  2. Kesse MA. Artificial intelligence: a modern approach to increasing productivity and improving weld quality in TIG welding. Forthcoming 2021.
  3. Kumar V, Parida MK, Albert SK. State-of-the-art methodologies for quality analysis of arc welding process using weld data acquisition and analysis techniques. Int J Syst Assur Eng Manag. 2022;13(1):34–56.
  4. Wang B, Hu SJ, Sun L, Freiheit T. Intelligent welding system technologies: state-of-the-art review and perspectives. J Manuf Syst. 2020;56:373–91.
  5. Nadeau F, Thériault B, Gagné MO. Machine learning models applied to friction stir welding defect index using multiple joint configurations and alloys. Proc Inst Mech Eng Part L J Mater Des Appl. 2020;234(5):752–65.
  6. Shah MAUD, Maekai IA. Statistical modelling and optimization of clad characteristics in SAW welding of SS-304. Weld Int. 2023;37(5):237–53.
  7. Devaraj J, Ziout A, Qudeiri JEA. Grey-based Taguchi multiobjective optimization and AI-based prediction of dissimilar gas metal arc welding process performance. Metals. 2021;11(11):1858. doi:10.3390/met11111858.
  8. Martinez RT, Alfaro SCA. Data analysis and modeling techniques of welding processes: state-of-the-art. Welding-Modern Topics. Forthcoming 2020.
  9. Wordofa TN, Perumalla JR, Sharma A. An artificial intelligence system for quality level–based prediction of welding parameters for robotic gas metal arc welding. Int J Adv Manuf Technol. 2024;132(7):3193–212.
  10. Stavropoulos P, Papacharalampopoulos A, Sabatakakis K. Robust and secure quality monitoring for welding through platform-as-a-service: a resistance and submerged arc welding study. Machines. 2023;11(2):298. doi:10.3390/machines11020298.
  11. Vasan V, Sridharan NV, Balasundaram RJ, Vaithiyanathan S. Ensemble-based deep learning model for welding defect detection and classification. Eng Appl Artif Intell. 2024;136:108961. doi:10.1016/j.engappai.2023.108961.
  12. Amirafshari P, Kolios A. Estimation of weld defect size distributions, rates and probability of detection in fabrication yards using a Bayesian theorem approach. Int J Fatigue. 2022;159:106763. doi:10.1016/j.ijfatigue.2022.106763.
  13. Liu Y, Liu F, Zhang W, Ding X, Arai F. Prediction and optimization of joint quality in laser transmission welding using serial ANNs integrated with Markov decision process. J Laser Appl. 2024;36(3). doi:10.2351/7.0000866.
  14. Sada SO. Use of multi-objective genetic algorithm (MOGA) in optimizing and predicting weld quality. Cogent Eng. 2020;7(1):1741310. doi:10.1080/23311916.2020.1741310.
  15. Stavropoulos P, Papacharalampopoulos A, Sabatakakis K. Data attributes in quality monitoring of manufacturing processes: the welding case. Appl Sci. 2023;13(19):10580. doi:10.3390/app131910580.
  16. Kumar A, Mishra M. A review on welding quality prediction using artificial intelligence techniques. Forthcoming.
  17. Mehta A, Vasudev H. Advances in welding sensing information processing and modelling technology: an overview. J Adhes Sci Technol. 2024;1–45. doi:10.1080/01694243.2024.2333654.
  18. Kumar A, Mishra M. Welding quality prediction using advanced artificial intelligence techniques. Forthcoming.
  19. Devaraj J. Minimization of weld distortion by weld sequence optimization using artificial intelligence. Forthcoming 2021.
  20. Aswal VK, Jain JK, Sonia P. Review on the behavior of various parameters on heat distribution in the SAW process. Mater Today Proc. 2021;47:6734–9.
  21. Patra PP, Pathak DPK. Artificial intelligence: analysis of various agent programming languages. Int J Res Adv Comput Sci Eng. 2024;10(1):1–5. doi:10.53555/cse.v10i1.2406.
  22. Say, D., Qaisar, S. M., Krichen, M., & Zidi, S. (2024). Artificial Intelligence Assistive Non-Destructive Testing of Welding Joints: A Review.
  23. Vasan, V., Sridharan, N. V., Balasundaram, R. J., & Vaithiyanathan, S. (2024). Ensemble-based deep learning model for welding defect detection and classification. Engineering Applications of Artificial Intelligence, 136, 108961.
  24. Murzin, S. P. (2024). Artificial intelligence-driven innovations in laser processing of metallic materials. Metals, 14(12), 1458.
Ahead of Print Subscription Original Research
Volume 14
02
Received 17/11/2025
Accepted 27/01/2026
Published 25/04/2026
Publication Time 159 Days
108

Login


My IP

PlumX Metrics