Design of Experiments (DOE) is a formal way in which to investigate and identify the relationships between key input factors and the response of a process resulting in improved optimisation and control. The methodology enables an organisation to conduct a series of tests and the resultant analysis of pre-defined outcomes can then be assessed in a controlled manner.
In order to reveal an interaction or dependency the traditional approach of ‘one change at a time’ analysis relied heavily upon the process engineer taking the tests in the appropriate direction, which in itself can result in a flawed outcome. The DOE approach on the other hand plans for all possible dependencies in the first place, and prescribes exactly what data sets are required for assessment i.e. whether input variables change the response on their own, when combined, or not at all. The exact length and size of a given experiment are set in the design phase before the testing begins.
DOE methodology provides the ability to succinctly find solutions to the key contributing factors to a given problem and to aid in the determination of factors and actions which when implemented will result in a reduction in variation.