Full Factorial Experiments

Figure above illustrates a "full factorial experiment" for four control factors at 2 levels. In full factorial method, all possible combinations of control factor levels are investigated (16 in this case).

If time & resources allow, Full Factorial is the best type of experimentation as it provides a comperehensive basis for detailed understanding of the process. However, it is not uncommon to find processes with as many as 12 control factors with say 3 levels. A full factorial study in such a case would mean 531,441 experimental runs. This is obviously impractical for any type of environment, be it manufacturing or research. A fractional factorial D.O.E in such a case can provide useful & often adequate knowledge about the process in just 27 runs!

 

► Click here to go back to the list of "Possible routes for Experimentation"