Advancing circular Textiles: MLR-Based optimization of tri-blend Melange yarns from sustainable fibers

• Maximal utilization of recycled cotton to reduce the environmental impacts derived from the cultivation and processing of virgin cotton. • Maximal utilization of recycle polyester to promote the circular material flows and resource efficiency. • Employed MATLAB software and MLR model to predict the optimized formulation of mix components while ensuring the desired properties of yarn. • A sustainable approach to utilize recycled polyester and recycled cotton together in yarn. The growing demand for sustainable textiles has prompted research into fiber blends that reduce environmental impact without compromising product performance. This study focuses on creating a sustainable tri-blend yarn consisted of recycled cotton, recycled polyester, and Tencel fibers. Recycled cotton replaces virgin cotton, recycled polyester enables the reuse of PET bottle and textile waste, and Tencel—derived from sustainably sourced wood pulp—enhances softness, comfort and tensile properties. MATLAB was used to perform Multiple Linear Regression (MLR) analysis to determine the optimal blend ratio that maximizes recycled fiber content while ensuring required yarn characteristics for fabric production. Based on Extreme Vertices Design, twenty-nine yarn samples with varying blend ratios were produced. Their quality parameters were evaluated, and the impacts of fiber composition on yarn properties were analyzed using Multiple Linear Regression (MLR) models. The MLR models generated predictive equations for each yarn parameter and performed blend optimization through two approaches: criteria-based optimization, which minimized yarn unevenness and imperfections while maximizing tensile properties, and target-based optimization, which aligned blends with the specific requirements of buyer or consumer. Validation of the optimized blends confirmed the model’s predictive reliability, with deviations between predicted and actual values remaining below 5%.