|
5.
Process Improvement
5.5. Advanced topics 5.5.9. An EDA approach to experimental design 5.5.9.10. DEX contour plot
|
|||
| Need to define "best" |
We need to identify the theoretical value of the response that would
constitute "best". What value would we like to have seen for the
response?
For example, if the response variable in a chemical experiment is percent reacted, then the ideal theoretical optimum would be 100%. If the response variable in a manufacturing experiment is amount of waste, then the ideal theoretical optimum would be zero. If the response variable in a flow experiment is the fuel flow rate in an engine, then the ideal theoretical optimum (as dictated by engine specifications) may be a specific value (e.g., 175 cc/sec). In any event, for the experiment at hand, select a number that represents the ideal response value. |
||
| Optimal value for this example | For the defective springs data, the response (percentage of acceptable springs) ranged from Y = 52 to 90. The theoretically worst value would be 0 (= no springs are acceptable), and the theoretically best value would be 100 (= 100% of the springs are acceptable). Since we are trying to maximize the response, the selected optimal value is 100. | ||
