In general, the choice of the location of the evaluation points is important in the process of response surface generation. In this scenario, Design of Experiments is used to help the task of point location through the design space. One popular technique is the so called Latin Hypercube. This is a fast-to-generate design, however, due the random nature of the generation process, it can present some disadvantages, such as, the possibility of a bad design in terms of fitting a meta model. To overcome this difficulty, one can use the Optimal Latin Hypercube. The big deal with this design is the computational cost associated with its generation. Therefore, solving this problem requires an optimization technique for searching the design space. This paper describes a method for generating the Optimal Latin Hypercube design of experiments. Following, two key-points to speed up the process are presented. Finally, numerical results are reported, illustrating the success of using the methodology presented in generation of the Optimal Latin Hypercube.