5. As Pearson Correlation of TFC and TQC is Very High
So if we want to reduce TQC then we have to reduce TFC
significantly. And this further can be achieved by
optimizing PC and AC for least TFC.
6. Regression Analysis: TFC versus PC, AC
The following terms cannot be estimated and were removed:
AC*AC*AC*AC*AC
Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value
Regression 9 56.1014 6.2335 4.61 0.191
PC 1 13.7403 13.7403 10.16 0.086
AC 1 7.3297 7.3297 5.42 0.145
PC*PC 1 14.6937 14.6937 10.86 0.081
AC*AC 1 7.5177 7.5177 5.56 0.142
PC*PC*PC 1 15.6906 15.6906 11.60 0.076
AC*AC*AC 1 7.7375 7.7375 5.72 0.139
PC*PC*PC*PC 1 16.5557 16.5557 12.24 0.073
AC*AC*AC*AC 1 7.9580 7.9580 5.88 0.136
PC*PC*PC*PC*PC 1 17.2378 17.2378 12.74 0.070
Error 2 2.7053 1.3526
Lack-of-Fit 1 0.2853 0.2853 0.12 0.789
Pure Error 1 2.4200 2.4200
Total 11 58.8067
Model Summary
S R-sq R-sq(adj) R-sq(pred)
1.16303 95.40% 74.70% *
12. Response Optimization: EF, IF, TFC
Parameters
Response Goal Lower Target Upper Weight Importance
EF Target 0.45 0.5 4.8 1 1
IF Target 0.45 0.5 4.6 1 1
TFC Minimum 2.0 9.4 1 1
Solution
EF IF TFC Composite
Solution PC AC Fit Fit Fit Desirability
1 0.6 0.888775 0.527091 0.708593 1.23568 0.980677
Multiple Response Prediction
Variable Setting
PC 0.6
AC 0.888775
Réponse Fit SE Fit 95% CI 95% PI
EF 0.527 0.666 (-2.340, 3.394) (-3.544, 4.598)
IF 0.709 0.492 (-1.408, 2.825) (-2.297, 3.714)
TFC 1.24 1.15 ( -3.73, 6.20) ( -5.81, 8.28)
13.
14. At the derived optimized values of PC and AC the TFC
(IF+EF) will be minimum. To verify this now run the given
unit on theses suggested values and check whether TFC and
TQC are coming as per given model or not?
Error of ±10% is admissible…