In order to distinguish something about a particular process or system, experiments or tests are performed in virtually all fields of inquiry. A designed experiment is a test or series of tests in which the purposeful changes are made to the input variables of a process or system so that we may observe and identify the reason for changes in the output response.
R. Fisher was the pioneer in the use of mathematical statistics in the experimental design during 1918~1939. He first introduced the analysis of variance (ANOVA) for the tests of significance in the agricultural experiments. Unlike the industrial experimenters, it was very important for agronomist that they have been more often concerned with uniformity trials. They should produce essentially equivalent strains of seeds and new variety of seeds, not just on the experimental farm but all region, in the other states where the crop is grown, even in other countries. On the other side, the industrial experimenters were concerned with the basic problem of finding optimum conditions for operating chemical process. G. E. P. Box (1951) introduced the response surface method (RSM) to estimate and optimize the performance of chemical process as a black box with controllable factors.
Hence, the design of experiments is classified with their application purposes. For the effect analysis and ANOVA, the level balanced design is recommended for complete experiments. However, for the RSM, the rotatable design is recommended to minimize the variance of estimated responses.