Optimization of a Ga and within a Ga for a 2-dimensional Layout Problem

A GA’s performance on a specific problem is related to many factors, such as genetic operators and corresponding parameter settings and the representation of the problem on the chromosome. Optimization of these factors to improve the speed and robustness of search is essential to successful application of a GA. The work reported here uses a two-level GA system (DAGA2) to solve a practical problem: the conceptual layout of machines on a factory floor (not the detailed design), based on a matrix of positive and negative relationships of various strengths between machines. This is a difficult, high-dimensionality problem. Performances of a traditional parallel GA system and our DAGA2 system are compared. The later one is compared with and without the use of several different representations for the problem at various times and in various subpopulations, demonstrating the strong contribution which the use of multiple representations makes to solution of the problem. The authors argue the necessity of optimizing many aspects of the GA in order to obtain useful solutions on such difficult, real problems.