Design of Experiments
Design of Experiments is something of a buzzword; any experiment that you put thought into planing is, in some sense, designed.
The term Design of Experiments (DOE) as used in industry usually refers to a set of defined statistical methods used to design sets of experiments and analyze the results. The available designs can be broadly classified into three groups, comparative designs, screening designs, and response surface designs; which you should use depends on the objective of your experiment.
- Use a Comparative Design to choose between alternatives.
- Use a Screening Design to identify which factors or effects are important.
- Use a Response Surface Design to
- hit a target,
- maximize or minimize a response,
- reduce variation by locating a region where a process is easier to manage,
- make a process robust.