Predicting the finished fabric width and areal density (Grams per Square Meter) of commercially produced plain Single Jersey (100% Cotton) Knitted Fabric using Fuzzy Inference System (FIS)

Abstract

The purpose of this research is to predict Finished Fabric Width (FW) & Areal Density (GSM) of 100% cotton plain single jersey knitted fabric by building a fuzzy inference model incorporating key input parameters such as Stitch Length (SL), Yarn Fineness or Count (YC) and Machine Diameter (D). More than 30,000 mass production-grade data points have been used to generate the model with remarkable precision. Once the model was prepared, it was verified using new experimental data. The Coefficient of Determination (R 2 ), Mean Absolute Percentage Error (MAPE), and Root Mean Square Error (RMSE) between the actual and the predicted FW were found to be 0.979, 1.214%, 1.103, respectively. For GSM the corresponding metrics were 0.940, 1.661%, 3.892, respectively. Both prediction outcomes showed excellent precision, justifying the model's applicability in the textile industry for predicting two important knit fabric parameters namely FW and GSM. The system's reliability was ensured by using a large set of industry standard data. This, combined with the adaptation of carefully designed fuzzy logic rules based on proven scientific method, significantly contributed to producing more accurate results. Together, all these aspects make the system stand out from similar studies, offering a practical and trustworthy approach for real world textile application with enhanced process optimization.