The following examples use data from [3]. The experiment uses three factors - nitrogen, phosphoric acid, and potash - all ingredients in fertilizer. The effect of the fertilizer on snap bean yield was studied in a central composite design.
The actual units for the -1 and +1 levels are 2.03 and 5.21 for nitrogen, 1.07 and 2.49 for phosphoric acid, 1.35 and 3.49 for potash. To reduce the impact of non-orthogonal terms, Minitab fits the model in coded units.
1 Choose Stat > DOE > Response Surface > Create Response Surface Design.
2 Under Type of Design, choose Central composite.
3 From Number of factors, choose 3.
4 Click Designs. To create the design, click OK.
5 Click Factors. In the Name column, enter Nitrogen Phosphoric Acid Potash in rows one through three, respectively. Click OK in each dialog box.
1 Open the worksheet CCD_EX1.MTW. (The design from the previous step and the response data have been saved for you.)
2 Choose Stat > DOE > Response Surface > Analyze Response Surface Design.
3 In Responses, enter BeanYield.
4 Click Terms.
5 From Include the following terms, choose Linear. Click OK in each dialog box.
Session window output
Response Surface Regression: BeanYield versus Nitrogen, PhosAcid, Potash
Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value Model 3 7.7886 2.5962 1.08 0.387 Linear 3 7.7886 2.5962 1.08 0.387 Nitrogen 1 4.4960 4.4960 1.86 0.191 PhosAcid 1 0.4593 0.4593 0.19 0.668 Potash 1 2.8332 2.8332 1.17 0.295 Error 16 38.5965 2.4123 Lack-of-Fit 11 36.0569 3.2779 6.45 0.026 Pure Error 5 2.5396 0.5079 Total 19 46.3851
Model Summary
S R-sq R-sq(adj) R-sq(pred) 1.55315 16.79% 1.19% 0.00%
Coded Coefficients
Term Effect Coef SE Coef T-Value P-Value VIF Constant 10.198 0.347 29.36 0.000 Nitrogen -1.148 -0.574 0.420 -1.37 0.191 1.00 PhosAcid 0.367 0.183 0.420 0.44 0.668 1.00 Potash 0.911 0.455 0.420 1.08 0.295 1.00
Regression Equation in Uncoded Units
BeanYield = 10.01 - 0.361 Nitrogen + 0.258 PhosAcid + 0.426 Potash
Fits and Diagnostics for Unusual Observations
Obs BeanYield Fit Resid Std Resid 7 8.260 11.163 -2.903 -2.17 R 16 13.190 10.500 2.690 2.03 R
R Large residual |
It is important to check the adequacy of the fitted model, because an incorrect or under-specified model can lead to misleading conclusions. By checking the fit of the linear (first-order) model you can tell if the model is under specified. The small p-value (p = 0.026) for the lack of fit test indicates the linear model does not adequately fit the response surface. The F-statistic for this test is (Adj MS for Lack of Fit) / (Adj MS for Pure Error).
The VIFs are all close to 1, which indicates that the predictors are not correlated. VIF values greater than 5-10 suggest that the regression coefficients are poorly estimated due to severe multicollinearity.
Because the linear model does not adequately fit the response surface, you need to fit a quadratic (second-order) model.