Response Optimization Overview
see also  

Many designed experiments involve determining optimal conditions that will produce the "best" value for the response. Depending on the design type (factorial, response surface, or mixture), the operating conditions that you can control may include one or more of the following design variables: factors, components, process variables, or amount variables.

For example, in product development, you may need to determine the input variable settings that result in a product with desirable properties (responses). Since each property is important in determining the quality of the product, you need to consider these properties simultaneously. For example, you may want to increase the yield and decrease the cost of a chemical production process. Optimal settings of the design variables for one response may be far from optimal or even physically impossible for another response. Response optimization is a method that allows for compromise among the various responses.

Minitab provides two commands to help you identify the combination of input variable settings that jointly optimize a set of responses. These commands can be used after you have created and analyzed factorial designs, response surface designs, and mixture designs.

·    Response Optimizer - Provides you with an optimal solution for the input variable combinations and an optimization plot. The optimization plot is interactive; you can adjust input variable settings on the plot to search for more desirable solutions.

·    Overlaid Contour Plot - Shows how each response considered relates to two continuous design variables (factorial and response surface designs) or three continuous design variables (mixture designs), while holding the other variables in the model at specified levels. The contour plot allows you to visualize an area of compromise among the various responses.