Analysis of orthogonal array experiments using the multivariate orthogonal regression method

In this article, a multivariate orthogonal regression method is presented for analyzing the data of orthogonal array experiments. It is based on fitting the data with an orthogonal polynomial regression model by using the least squares method. The proposed method can obtain the same statistically testing results as that of the conventional method. In addition, the regression model can be used to determine the optimum combination of levels. Compared to the discrete-type conditions determined by the conventional method, the proposed method can produce the continuous-type conditions of optimum combination of levels. The regression model can also be used for prediction, process optimization, and process control. A numerical example is included to illustrate the procedure.