Relationship of Fiber Properties to Vortex Yarn Quality via Partial Least Squares

Abstract

The Cotton Quality Research Station (CQRS) of theUSDA-ARS, recently completed a comprehensivestudy of the relationship of cotton fiber properties tothe quality of spun yarn. The five year study, beganin 2001, utilized commercial variety cotton grown,harvested and ginned in each of three major growingregions in the US (Georgia, Mississippi, and Texas).CQRS made extensive measurements of the rawcotton properties (both physical and chemical) of 154lots of blended cotton. These lots were then spuninto yarn in the CQRS laboratory by vortex spinningwith several characteristics of the yarn and spinningefficiency measured for each lot. This studyexamines the use of a multivariate statistical method,partial least squares (PLS), to relate fiber propertiesto spun yarn quality for vortex spinning. Twodifferent sets of predictors were used to forecast yarnquality response variables: one set being only HVI™variables, and the second set consisting of bothHVI™ and AFIS™ variables. The quality ofpredictions was not found to significantly changewith the addition of AFIS™ variables.