GAs provides a high level of robustness by simulating nature's capability of adapting to many different environments. Through the application of GAs to design optimization problems, the performance characteristics of the GAs are shown to be powerful in solving optimization problems with high dimensional objective functions containing several local minima. As they use only the fitness values, GAs does not require derivatives or any other additional information about the objective function. The performance of GAs can be improved significantly by applying selective pressure and adequately providing a shape modeling technique to control the range of variation for the shape to be optimized.
