An effective robust genetic algorithm based on reduced evaluation

Robust optimization searches for optimal solutions which remain feasible under the effect of uncertainties. The conventional robust optimization, however, often requires great computational cost to find robust solutions. This paper introduces a novel robust optimization method based on genetic algorithm. In this method, individuals with "good" performance are selected and tested for their robustness for reduction of computational time. It is shown that the present method works in shorter time in comparison with the conventional robust optimization.