Design of Experiments (DOE) with Geometric Process Control technology

Details

Published on Wednesday, 24 March 2010 12:52

Geometric Process Control (GPC) technology can be successfully applied to DOE through the C Response Surface Visualiser (CRSV) tool. CRSV allows Experimenters or Formulators to perform their own Response Surface Analysis at their workbench in a fraction of the time currently required. It replaces the numerically intensive finding of multi-variable regression equations to describe the Response Surface with a simple visual method that anyone can use.

For the first time experimenters can see the complete set of relationships within their response surface. This technique is perfectly suited to Formulation / Design of Experiments (DOE) problems where the number of data points available is very small, perhaps as small as 8 to 10. The ability to move values around and watch the effects on other parameters is invaluable. Parameters needing further investigation can be quickly identified, thereby eliminating unnecessary trials and improving both ROI and speed to market/trial. The affect of changing parameter values and seeing the consequences opens up a new view on how to progress and where to focus within your process.

In the display shown, the experimental variables are on the leftmost 9 axes and the result variables are on the right. The user has begun to explore the Surface by fixing values for the first 4 variables. These are shown as blue dots. The remaining variable values are now indicated to lie within the green ranges. The model has correctly deduced that these experimental variables are independent of each other so any value can be chosen from the entire green range. This is a powerful tool, especially to users of the very small datasets that are commonly found in Formulation and Design-of-Experiments activities. The Surface Model was built in seconds from the whole set of 12 points without requiring any mathematical knowledge.

What results will I get if I perform this experiment?

Change the first four variable values in the above display by simply clicking and dragging the blue dots. See the results instantaneously, without waste or inconvenience. Explore the entire surface and apply and check domain knowledge and possibly eliminate an Experiment or Formulation altogether for extra benefit.

What experiments should I perform to achieve my desired result?

In this picture, the result variables are on the left and the experimental variables on the right. Insert the desired results and see how to operate in order to achieve them. Save valuable time and money using GPC.

• < Prev
• Next >