ENHANCING TEXTILE MANUFACTURING EFFICIENCY THROUGH GRADIENT BOOSTING CLASSIFIER

Abstract

The textile industry encompasses a range of processes, including upstream, midstream, and downstream operations,all aimed at converting raw materials into final fabric products. Traditional manufacturing approaches often dependon trial-and-error methods, which can result in inefficiencies and excessive resource consumption. This researchpresents a machine learning-based methodology that utilizes Gradient Boosting Machines (GBM), Random Forest,and XGBoost to enhance textile production efficiency. These models, trained on historical production data, helpoptimize decision-making by uncovering patterns that impact fabric quality. Among these, XGBoost achieved anoutstanding 99.88% precision rate. By deploying these models as APIs, real-time analytics and automated notificationswere integrated, leading to significant improvements in defect detection and production efficiency, fostering thedevelopment of more intelligent manufacturing systems.