Deliberate choices of restrictions in complete randomization

Complete randomization is often emphasized in literature as an important feature in the planning phase of experimental design. However, complete randomization is seldom strictly applied in industrial experimentation, as it often demands resources not available to the experimenter. In this paper, consequences of incomplete randomization are discussed. To illustrate, an experiment conducted on an injection‐moulding process is used. The objective of this paper is to develop a pragmatic and user‐friendly decision tree which supports the experimenter in choosing about restrictions in complete randomization. Furthermore, the decision tree emphasizes the importance of distinguishing between randomization and resetting; increasing the awareness of their different consequences for the experimental result. In summary, the decision tree presented here should be useful in the planning phase of experiments.