Autors: Atanasova, R. P., Yumer, F. I., Ivanova, M. Y.
Title: Methodology for Evaluating Sewing Process Quality Indicators for AI-Adaptive Optimisation
Keywords: Sewing quality, defect analysis, AI-adaptive sewing, process

Abstract: This paper presents a methodology for the quantitative assessment of the sewing process and defect performance indicators, developed to enable future comparative analysis between conventional and AI-adaptive sewing machines. The methodology encompasses the definition of quality metrics, structuring of defect data, collection of production data in a real factory environment, and preliminary statistical processing. Data from trousers production were utilized, covering 28 defect types, controlled production volumes over a period of 62 working weeks, and process stability indicators by production lines and time periods. A framework for the analysis of defect frequency, distribution, and criticality is introduced, alongside an algorithm for evaluating process repeatability and stability. The results obtained through the proposed methodology provide a robust foundation for the development of AIsupported optimisation models aimed at reducing defects in sewing production.

References

    Issue

    Innovations in Energy and the Environment (InnoEE) Conference 2026, issue 1661, pp. 1-10, 2026, Bulgaria, IOP Conf. Series: Earth and Environmental Science, DOI 10.1088/1755-1315/1661/1/012033

    Copyright IOP Publishing Limited

    Вид: публикация в национален форум с межд. уч., публикация в реферирано издание