Statistical model for predicting the air permeability of polyester/cotton-blended interlock knitted fabrics

The aim of this study is to model the effect of knitting parameters on the air permeability (AP) of polyester/cotton interlock fabrics. Fabric samples of areal densities ranging from 105 to 654 g/m(2) were knitted using yarns of three different polyester/cotton blends, each of the three different linear densities by systematically varying knitting loop lengths for obtaining different cover factors. It was found that changing the polyester/cotton blend ratio from 65/35 to 52/48 and 40/60 did not have a statistically significant effect on the fabric AP. AP sharply decreased with decrease in knitting loop length owing to increase in fabric areal density. Increase in yarn linear density (tex) resulted in a decrease in AP due to increase in fabric thickness as well as the areal density. It was concluded that response surface regression modeling could adequately model the effect of knitting parameters on the fabric AP. The model was validated by unseen data-set and found that predicted and actual values were in good agreement with each other with less than 5% absolute error. Sensitivity analysis was also performed to determine the relative contribution of each input variable on the AP of the interlock fabrics.