Two genetic algorithms to solve a layout problem in the fashion industry

Genetic Algorithms (GAs) have proven to be a valuable method for solving a variety of hard combinatorial optimization problems. In this paper, we develop a pair of GAs to solve a layout problem in the fashion industry. Over the past years, a number of integer programming (IP) models have been constructed that are capable of solving small, real life, layout cases in an acceptable amount of time. However, when the dimensions of the problem cases increase and approach the complexity of some large layout instances in the fashion industry, these IP models fail to offer a flexible solution to the layout problem in general. Moreover, optimality is not always a primary concern for large cases, and a satisfactory solution to a particular layout problem can be provided by a heuristic such as a GA. The GAs in our paper differ from each other in that they are based on two alternative IP models for the layout problem. The aim of this paper is then (1) “build a GA that generates optimal or near optimal solutions on small problem instances, and that is capable of solving large, real life layout problems in the fashion industry in an acceptable amount of time”, and (2) “to investigate which problem formulation is better (in terms of accuracy and computation time) to solve the layout problem by a GA”. We investigate the ability of both GAs to find optimal or near optimal solutions. Also, we study the importance of their genetic operators and investigate why the GAs behave differently. Finally, we compare computation times of both GAs on a variety of large real life layout instances.