This study aims to develop a quantitative skill evaluation system for sewing operators, addressing the limitations of traditional subjective evaluation methods in the garment industry.
A sewing evaluation framework was designed, incorporating three dimensions: power consumption, hand movement and sewing accuracy. The system integrates hardware and software to collect data from sewing tasks and automatically extracts relevant metrics. Ten participants with varying sewing experience performed four standardized sewing tasks. Their data were used to compute overall skill scores, which were then compared with human-rated evaluations for validation.
The system effectively distinguished operator skill levels across the three dimensions. Skilled operators demonstrated higher efficiency in machine use, more stable hand movements and improved sewing accuracy. The results suggest observable differences between skilled and unskilled operators and individual variation within groups. The skill scores generated by the system show alignment with human ratings, supporting its applicability as an objective evaluation tool.
This study proposes a structured approach to addressing the limitations of subjective sewing skill evaluation by combining custom hardware, analytical software and empirical validation. Unlike previous studies that focused on isolated criteria, this study identifies three dimensions closely related to the characteristics of sewing tasks and proposes an integrated system. While the technique cannot reliably quantify the degree of sewing skill yet, it represents an important advance toward automation of sewing skill evaluation.
