ABSTRACT
The basic principle of experimental design is to plan experiments so as to be able to study the effects of certain factors on some specific results, and to identify the important influential factors with a given level of confidence. The effects of the factors and their interactions are evaluated by computing F-values, which are the ratio of effect-mean-square vs. error-mean-square. Randomization and minimization of error are two important mathematical principles for efficient selection of truly influential factors. Randomization is useful in objectively detecting uncontrollable errors, thereby avoiding the inclusion of errors arising from man-made causes in the true errors. Sometimes “blocking” is required during randomization when there is an unavoidable need for handling many factors in an experimental plan. Blocking should be made based on a controllable condition (factor), such as date, so that the influence of that factor can be determined later if required.
