Improvement of process conditions in acrylic fiber dyeing using gray-based Taguchi-neural network approach

The main aim of this study was to enhance the product quality by improving dyeing process conditions of acrylic fiber used as raw material in a factory-produced carpet. There are three quality characteristics consisting of desired (nominal) color strength, maximized acrylic fiber strength and minimized dyestuff in dye bath. Dyeing temperature, fixation duration, softener, antistatic, amount of material (fiber), pH, retarder and dispergator, which have an influence on dyeing, were chosen as control factors. Dyeing temperature and antistatic were seemed to be significant factors on dyeing for 95 % confidence interval statistically. Optimal dyeing process conditions were determined by hybrid gray-based Taguchi-artificial neural network (ANN) method. Gray relational grade as a performance evaluation index obtained from gray relational analysis reduces the number of quality characteristics. Gray relational grade was found as 0.6630 for existing conditions and improved as 0.7749 by gray-based Taguchi ANN method. The suggested methodology improves quality of dyed acrylic fiber, reduces defective products and provides dyeing operations much more efficient. All of these translate into significant cost savings.