A new desirability function based method for multi-response robust parameter design

Parameter design is an effective way to improve product or process quality, and desirability function method is the most popular approach for multi-response parameter design optimization, however, the variations due to the noise factors and the fluctuation of the design parameters are ignored in this approach. Therefore a real robust solution can not be obtained using this method. Thus a new desirability function method is presented. The proposed method combines variances of the responses and the influence of the variations of the design parameters. The implementation and effectiveness of the proposed method are illustrated by an example from the literature, and the sensitivity of the response to the changes of controllable factors was analyzed to check if the solution is robust to the fluctuation of the design parameters. The result shows, compared with traditional desirability function approach, the variances obtained by the new desirability function based method are reduced, and the responses are more robust to the variations of the design parameters. As a result, the proposed method can yield better results than traditional desirability function approach.