In a designed experiment we investigate the effects of predictor variables, that we can
control and set to specific values, on some response. The design of the experiment is
the choice of values for these predictor variables. The predictor variables are known as
factors, and the chosen values for the factors are refered to as levels. This chapter
deals with factorial experiments and their extensions including the central composite
design. Multiple regression analysis is used for the analysis of these experiments. The
strategy of experimenting by making small changes to a process during routine production,
known as evolutionary operation (EVOP), is introduced.
|