A Designed Experiment is a test or series of tests in which purposeful changes are made to the input variables of a process so that we can measure corresponding changes in the output response.

The figure is a representation of the process as a box with inputs and outputs represented by the arrows. By using statistical concepts to chose the experimental settings it is possible to maximise the information from a small set of experimental runs.

Objectives of a DoE may include: Determine which parameters are most influential on the quality characteristics. Identify ideal settings for parameters to optimise the quality characteristic. Minimise the effect of noise on the quality characteristic In traditional experimentation only one parameter at a time would be varied. In Designed Experiments parameters are allowed to vary all together in a controlled way.