Level: Intermediate
Kevin Gallagher, Scientist, PPG Industries
A quick look at the JMP Custom Design platform reveals that there are several types of factors (continuous, discrete numeric, categorical, blocking, covariate and mixture). However, little attention has been paid to a very important class of experimental factors: order. There are many situations in which the order that events happen has an influence on the outcome or response. This is especially true for formulated products such as paints, adhesives, foods, cosmetics, etc., for which the order-of-addition of the ingredients can often be the difference between a successful or unsuccessful product. There are many other situations in which one may want to study the influence of order – for example, the order in which glasses of wine are presented to judges in wine taste tests or the order in which straps are tightened when putting on a knee brace. In this presentation the concept of a “pairwise ordering factor” will be described along with case study examples (development of an automotive paint formulation with premium appearance) that illustrate how to use JMP statistical software to both design and analyze experiments with order factors.