A Model for Predicting a Yarn's Knittability

In this study, we attempt to determine the knittability of a yarn before knitting. Thirty different yarns are knitted into 1 × 1 rib structures with three different loop length values. The knitting process is observed and the number of machine stops, yarn breaks, and holes are recorded. The results of experiments are evaluated by charts and statistical analyses. With a multiple regression analysis, two equations that depend on yarns characteristics can be used to predict yarn-needle and yarn-yarn friction values. In the last stage, we use a binary logistic regression analysis to indicate knittability, depending on the yarn-needle and yarn-yarn friction values, the machine tightness factor, and machine speed values. © 2004, Sage Publications. All rights reserved.