Experimental Design

Design of experiments includes the design of all information-gathering exercises where variation is present, whether under the full control of the experimenter or not. (The latter situation is usually called an observational study.) Often the experimenter is interested in the effect of some process or intervention (the 'treatment') on some objects (the 'experimental units'), which may be people. Design of experiments is thus a discipline that has very broad application across all the natural and social sciences.
See more at Wikipedia.org...

Nutrition science advances by observation (for example, nutritional epidemiologists observe statistical associations between what people eat and their patterns of health and disease) and by experiment, in which subjects are given specifically formulated diets and certain effects are noted. For example, to test the metabolic effects of components of diets, experiments are performed in which a diet containing the test substance (treatment) is compared against a reference diet (control) which, if possible, is the same in all respects other than the test substance and in which the test substance is substituted by a suitable placebo. Subjects should be allocated randomly to diets so that factors other than those tested, and which may otherwise confound the results, are equalised between treatments. Ideally, neither the subjects nor the experimenters should be aware of the identity of treatment and placebo groups (double-blind).

a set of research experimental procedures specifying: the test units, sampling procedures, independent variables, dependent variables and how external variables are to be controlled.

In industrial settings, Experimental design (DOE) techniques apply analysis of variance principles to product development. The primary goal is usually to extract the maximum amount of unbiased information regarding the factors affecting a production process from as few (costly) observations as possible. In industrial settings, complex interactions among many factors that influence a product are often regarded as a "nuisance" (they are often of no interest; they only complicate the process of identifying important factors, and in experiments with many factors it would not be possible or practical to identify them anyway). Hence, if you review standard texts on experimentation in industry (Box, Hunter, and Hunter, 1978; Box and Draper, 1987; Mason, Gunst, and Hess, 1989; Taguchi, 1987) you will find that they will primarily discuss designs with many factors (e.g., 16 or 32) in which interaction effects cannot be evaluated, and the primary focus of the discussion is how to derive unbiased main effect (and, perhaps, two-way interaction) estimates with a minimum number of observations.
For more information, see the Experimental Design chapter.