Determination of optimal mutation interval for µga leased on the performance analysis of GA and µGA

In order to provide the fastest search for optimal solution, both the structure and the parameters of genetic algorithms (GA) should be optimized. The advantage of recently introduced micro-genetic algorithms (μGA) is that they operate with smaller genetic populations than standard GA. This paper describes the method of determining an optimal (best possible) mutation interval for μGA based on the analysis of the influence of mutation on the performance of GA and μGA. The results obtained analytically and numerically for a selected vector optimization problem show the practical value of the method.