Akira Ogawa, Visiting Researcher, Faculty of Management, Mejiro University Takenori Takahashi, Visiting Professor, Graduate School of Health Management, Keio University
In experimental design, it is important to narrow down the factors by a small number of experiments, then model the characteristics using the narrowed down factors and determine the level at which the desired characteristics can be obtained. As statistical quality control methods, the former is called screening and the latter is called modeling, and they are treated differently. However, in practice, both are often carried out as a series of approaches, and in order to achieve the purpose of the experiment, screening experiments and modeling experiments need to be planned from a comprehensive perspective. In this presentation, we will review the experimental plan and analysis results of screening and modeling in JMP applied to improve processing processes. The key point of the review is that it is necessary to take into account specific goals, constraints to be considered, recommended plans, etc. when formulating a comprehensive experimental plan. After the experiment is conducted, the degree of achievement of the goal, cost-effectiveness, and even evaluation from stakeholders are also included. In this presentation, we will summarize the issues from a comprehensive perspective of experimental design and show the outlook for future experimental design.
