Pattern design and optimization of yarn-dyed plaid fabric using modified interactive genetic algorithm

Ning Zhang,Ruru Pan,Lei Wang,Yang Wu,Weidong Gao

DOI: https://doi.org/10.1080/00405000.2020.1738617

2020-03-13

Abstract:With the advancing consumption level, it is difficult to shorten the product cycle and meet consumer's demands during the design process of yarn-dyed plaid fabric. In order to extract consumers' preferences and obtain timely feedbacks, a novel method using modified interactive genetic algorithm (IGA) is proposed for pattern design and optimization of yarn-dyed plaid fabric in this paper. Initially, the pattern of plaid fabric was encoded using the natural number code. The population was initialized based on common colour schemes. Then, survival of the fittest was performed as the selection operator to extinguish unsatisfactory patterns and preserve satisfactory patterns. Subsequently, common crossover operator and mutation operator, special mutation operator including addition and deletion were implemented to generate offspring. Moreover, users can manipulate individuals directly for convergence acceleration during the evolutionary process. Finally, the experiment was carried out by twenty-four users of different age groups and genders. Experimental results indicate that the application of IGA is feasible and efficient, which can supply design references to the designer.