Response optimization helps you identify the combination of variable settings that jointly optimize a single response or a set of responses. This is useful when you need to evaluate the impact of multiple variables on a response.
You must fit a model before you can use the response optimizer. If you want to optimize multiple responses, you must fit a model for each response separately. Response optimizer does not use the data in the worksheet. Instead, Minitab looks in the worksheet for a stored model(s) to obtain the necessary information.
For example, a soft drink manufacturer is creating a new all-natural lemonade. They want to determine the proportion of lemons, water, and sugar that will maximize flavor ratings for the new drink. However, to increase profits, they also want to minimize the cost of the ingredients. Adding more water and less sugar can decrease cost, but might adversely affect flavor. To obtain both goals, they use response optimization to determine the proportion of ingredients that produces the best results within an acceptable range for each response (flavor and cost).
Minitab calculates an individual desirability for each response and weights each by the importance you assign it. These values are combined to determine the composite, or overall, desirability of the multi-response system. An optimal solution occurs where composite desirability obtains its maximum. Using an optimization plot, you can adjust the variable settings and determine how the changes affect the response.
Response optimization is most effective when interpreted in conjunction with relevant subject matter expertise, including background information, theoretical principles, and knowledge obtained through observation or previous experimentation.