Utilities for Refining and Extending the Analysis
Variability Analysis
For designs with more than one replicate, DOE++ gives you the ability to determine the variability of the response(s) across runs and to analyze that standard deviation information. This offers valuable insight into the sources of variation within the experimental data.
Diagnostics for a Closer Look
The Diagnostics window provides run-by-run results including fitted values, residuals, leverage and Cook’s distance measure to help you assess the model at each data point for potential outliers and influential observations.
Response Prediction for Untested Treatments
DOE++'s Prediction window allows you to enter your own combinations of factor settings and returns response values, complete with confidence metrics, based on the fitted model.
Optimization for Your Desired Outcome
DOE++ can help you determine the best factor settings to achieve your goals for the response(s) in your analysis. For each response, you can:
- Specify whether you want to minimize the response, maximize it or bring it as close as possible to a particular target value.
- Designate maximum, minimum and/or target values per the analysis objective.
- Specify how much emphasis is placed on these values when determining the desirability of solutions (i.e. how narrowly to define "desirable").
- Specify how important the optimization of each response is with respect to the optimization of other responses included in the analysis (i.e. which response is most important to optimize).
In addition, you can set boundaries or constraints on each factor to be used in the optimization process. DOE++ will search for the most effective combinations of factor settings and will rank the solutions according to their desirability.
