Genetic Algorithm parameters tuning for Quadratic Assignment Problem based on Experimental Design

QAP (quadratic assignment problem) is one of the practical issues in location and layout problem.However there are some exact solution methods for the QAP but when the size of problem increases, the time of solution will increase exponentially. To eliminate this problem we use Genetic Algorithm (GA) for solving in a reasonable time. One of the most important issues in GA is determining parameters of the algorithm. In this research we tried the algorithm according to the factorial design experiments. After that we applied statistical non parametric test for ensuring about accuracy of the solution method.

Keywords:

Genetic Algorithm , Quadratic Assignment Problem , Design of Experiment (DOE) , Kruskal-Wallis test