PREDICTION OF YARN BREAKAGE PROBABILITY IN CIRCULAR KNITTING MACHINES: A BASELINE FOR INDUSTRIAL AI IN TEXTILES

Circular knitting technology is widely used in textile manufacturing due to its high efficiency and flexibility. Smooth and consistent yarn feeding is essential to maintaining continuous production. Yarn feeding systems play a critical role in reliably delivering yarn to the knitting mechanism, and their proper functioning is crucial for preventing yarn breakages. Such breakages not only disrupt production but also compromise fabric quality. Therefore, accurately predicting yarn breakage events is essential for implementing preventive maintenance strategies and enhancing overall production efficiency. In this study, machine learning methods were applied to estimate the probability of yarn breakages associated with the negative yarn feeding system in circular knitting machines. Operational data collected from these machines were used to train predictive models capable of estimating breakage probabilities. The results demonstrate that the machine learning model employed achieves high accuracy and consistency in predicting yarn breakage likelihood. The prediction model developed in this study lays the groundwork for applying AI-based solutions to enhance maintenance planning and production workflows in textile manufacturing, thereby contributing to smoother and more efficient operations.