This paper presents a new methodology to solve multi-response statistical optimization problems. This methodology integrates desirability function and simulation approach with a genetic algorithm. The desirability function is responsible for modeling the multi-response statistical problem, the simulation approach generates required input data from a simulated system, and finally the genetic algorithm tries to optimize the model. This methodology includes four methods. The methods differ from each other in controlling the randomness of the problem. In the first and second methods, replications control this randomness while in the third and fourth methods we control the variation by statistical tests. Furthermore, we compare the performances of these methods by numerical examples and designed experiments and report the results.