cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar
Optimizing Mixtures When the Response Is a Nonlinear Curve

Bernd Heinen, JMP Systems Engineer, SAS
James S. Dailey, PhD, Global Technical Key Account Manager, BASF 


Sometimes the result of an experiment is not just a set of measurements; it is a curve. How can the optimal mixture be derived if a set of curves needs to be compared? In this situation many experimenters tend to choose a somehow representative X-value and use the associated Y-value as the outcome of the experiment. Thus they ignore a wide set of valuable information. The solution here is to understand what the optimal curve is, to fit a parameterized curve for each experiment, to interpret the parameters as the outcome of the individual experiments and to optimize the mixture with respect to those parameters. The analytical process uses Graph Builder, the Nonlinear platform and Fit Model to exploit all available information and to find the optimal result. The whole solution is motivated and demonstrated with data from a chemical mixture optimization problem.

Comments
volinoca

I enjoyed your talk at the JMP Discovery Summit in San Diego. This seems like a logical solution to an old problem, and I appreciate your ideas and approach.

Discovery Summit 2015 Resources

Discovery Summit 2015 is over, but it's not too late to participate in the conversation!

Below, you'll find papers, posters and selected video clips from Discovery Summit Europe 2015.