Determination of the Optimal Percentage of Blending of Animal Fiber

3

Determinationof the Optimal Percentage of Blending of Animal Fiber Length-DiameterGroups Using a Simplex Lattice Blend Design Approach

Determinationof the Optimal Percentage of Blending of Animal Fiber Length-DiameterGroups Using a Simplex Lattice Blend Design Approach

Abstract

Awell established tool known as YarnspecTMwas used to predict the quality of the resultant ring span yarn fromthe length-diameter blended Inner Mongolian cashmere (IMC) fiber andAustralian Superficial wool (ASFW). The effect of the interactionthat occurred between the length-diameter groups (including theMedium-Midrange, Short-Fine, and Large-Coarse) of the fiber werestudied using blend design experiments. Ridge analysis was used tocarry out the optimization of the product, which facilitated thedetermination of optimum blend propositions on the basis of thepredicted tenacity (cN/tex), yarn unevenness (CV %), hairiness (Hvalue), and elongation (%). The highest effect on tenacity value wasnoticed in the short-fine group. The optimum values of theLarge-Coarse, Medium-Mid-range, and Short-Fine groups in the blendwere recorded as 15-100 %, 0-40 %, and 0-85 %, respectively and inaccordance with the features of the resultant yarn. Additionally, theresults of ridge analysis indicated that the supreme blend of thefiber’s length-diameter should include no Large-Coarse, 35 %Medium-Mid-Range, and 65 % Short-Fine in order to obtain a maximumpreference score of 7.24. The hairiness value, elongation, tenacity,and unevenness of the sample required in order to find the overallpreference score of 7.24 were predicted to be 0.435, 0.435 %, 3.650(cN/tex), 0.435 %, respectively.

1.Introduction

Blenddesign is an important methodology that can be used to investigatethe functions of different components of the ring spun yarns. Themethodology also approves the significance of interaction among thecomponents. The blend design methodology was used in the presentstudy to investigate the impact of the interaction betweenLarge-Coarse, Medium-Mid-Range, and Short-Fine groups on the featuresof the ring spun yarn. The study identified the optimum blend’slength-diameter groups in order to generate a product that is mostacceptable with respect to the properties of the resultant yarn.

2.Experimental Design

Theeffect of the fiber’s Large-Coarse (A), Medium-Midrange(B) and Short-Fine (C) groups was elevated on the features of theresultant spun yarn using the simplex lattice blend design (SLMD).The optimum combinations that would yield a product that has thehighest quality score was estimated using the ridge analysis.Propositions of different components were articulated as proportionsof the yarn’s blend with the sum (A+B+C) of one. The three factors,experimental design, and levels, both coded and un-coded, arepresented in Table 1. The 15 components include 9 two-componentblends, 3 single-component blends, and 3 three-component blends asshown in Figure 1.

3.Statistical study and modeling of trial data

Thepolynomial equation presented below was fitted for all factors thatwere evaluated at each of the trial points. The polygonal modeldiffered from the full polynomial models since it did not contain theconstant term that would have served as the intercept equal to zero.The polynomial model equation applied in the present study was

Yrepresented the estimated reply and the following are constantcoefficients β1,β2, β3, β12, β13, β23andβ123.Uncoded unit was used to perform the analysis. The computation (suchas ternary contour graphical representation) was accomplished usingStatistica, 1995. The predicted equations were computed using the JMPstatistical package software. The estimated ridge of the maximum aswell as a least response for the increased radii starting from thecentral point of the initial design was also computed using the JMP(2009). The contour schemes that were made from forecast equationswere place over in order to obtain the best possible region. MINITABWindows Release 13 (Minitab 2000) was used to determine thecorrelation between different parameters. The difference between themean combination values was determined using Duncan Multiple RangeTest (MstatC, 1986).