M. Zeydan and G. Toğa
Reducing Variation of Dyeing Process in Textile Manufacturing Industry
2272 - 2280
2011
5
11
International Journal of Materials and Textile Engineering
https://publications.waset.org/pdf/13143
https://publications.waset.org/vol/59
World Academy of Science, Engineering and Technology
This study deals with a multicriteria optimization
problem which has been transformed into a single objective
optimization problem using Response Surface Methodology (RSM),
Artificial Neural Network (ANN) and Grey Relational Analyses
(GRA) approach. GreyRSM and GreyANN are hybrid techniques
which can be used for solving multicriteria optimization problem.
There have been two main purposes of this research as follows.
1. To determine optimum and robust fiber dyeing process
conditions by using RSM and ANN based on GRA,
2. To obtain the best suitable model by comparing models
developed by different methodologies.
The design variables for fiber dyeing process in textile are
temperature, time, softener, antistatic, material quantity, pH,
retarder, and dispergator. The quality characteristics to be evaluated
are nominal color consistency of fiber, maximum strength of fiber,
minimum color of dyeing solution. GRARSM with exact level
value, GRARSM with interval level value and GRAANN models
were compared based on GRA output value and MSE (Mean Square
Error) performance measurement of outputs with each other. As a
result, GRAANN with interval value model seems to be suitable
reducing the variation of dyeing process for GRA output value of the
model.
Open Science Index 59, 2011