DOEs 201: Screening and Reduced Designs
A short time ago we covered DOE 101, the basics of Design of Experiments in an Insider Series Webinar You can watch this webinar here. DOE 201 is a follow on Insider Series Webinar on advanced DOE techniques, mostly what are called screening and reduced designs. Screening designs are performed when there are so many factors to analyze that performing a DOE with all of the factors at a high number of factor levels is not practical. As an example, suppose there is a DOE with 6 factors, each at 4 levels. This DOE would require at least 46 = 4096 runs.
In cases like this one, screening experiments are often performed to determine which factors are important. In a screening experiment, each factor is only evaluated at two levels. So in this case the number of data points, or runs, required is only 26 = 64. In some cases, it is possible to reduce the number of runs further by employing a fractional factorial design.
This webinar will discuss how to perform these screening and fractional factorial DOEs and the risks in doing so. In the webinar, several screening and fractional factorial DOEs will be analyzed with Minitab. The data will be given to the attendees so that they can practice on their own. A 30 day trial version of Minitab is available for free.Less