Applying regression analysis to improve dyeing process quality: a case study

Abstract

The main purpose of this paper is to propose a regression-based new approach to pH control in open beck dyeing to improve its process quality. Specifically, we examine the current practice of adjusting pH of dye liquor at a large manufacturer of automotive carpets in the United States and point out several drawbacks that lead to substandard products. To properly address the issues, we theorize casual relationships among the key variables in the dyeing process and seek to identify them by using a multiple linear regression model based on the data gathered from actual production dyeings at the company. Our findings show that a dyer should set the pH of the solution in the beck before the dyes are added so that the dyebath pH after the addition of the dyes will be in the target range for best coloring results with minimum or no adjustment. The suggested approach enhances the quality of the dyed carpets, reduces laboratory work required to fix defective products, and renders the operations much more efficient. All of these translate into significant cost savings.