The Experiment tab of the DOE Builder dialog is used to specify which Design of Experiments (DOE) analysis to run.
| Experimental Design | Description |
|---|---|
| One Variable | This experiment calculates the effect of a single variable on a set of criteria that define the part quality. The variable is selected from a list in the Variables tab, which is based on the molding process and analysis sequence selected. The set of criteria is selected from the Quality Criteria tab. The result is an XY plot of the effect of the chosen variable on each of the selected quality criteria. This analysis is recommended if you are only interested in a single, known variable. |
| Variable Influences (Taguchi) | This experiment uses a Taguchi model to rank the impact of several variables on a set of criteria that define part quality. The variables are selected from a list in the Variables tab, which is based on the molding process and analysis sequence selected. The set of criteria is selected from the Quality Criteria tab. The result is a list in the DOE log of the selected quality criteria and the percent impact each chosen variable has on that criterion. This analysis is recommended if you need to know which variables have the greatest impact on part quality. |
| Variable Responses (Face Centered Cubic) | The Variable Responses analysis uses a face centered composite design to look at non-linear interactions between input variables and a set of criteria that define part quality. The variables are selected from a list in the Variables tab, which is based on the molding process and analysis sequence selected. The set of criteria is selected from the Quality Criteria tab. The result is a response surface plot, which shows how sensitive each selected quality criterion is to the selected input variable. A large number of experiments is needed for this analysis. |
| Variable Influences then Responses | This analysis runs the Taguchi analysis as a screening method to rank the importance of the input variables on overall quality. The variables with the most effect on quality are then used in a factorial analysis. The variables are selected from a list in the Variables tab, which is based on the molding process and analysis sequence selected. The set of criteria is selected from the Quality Criteria tab. This analysis is recommended if you would like to perform an extensive test to see how sensitive select quality criteria are those variables that have greatest impact on part quality, but are unsure which variables have the greatest impact. |