Level: Intermediate
【Presenter】
Akira Ogawa, Visiting Researcher, Faculty of Business Administration, Mejiro University
Takenori Takahashi, Visiting Professor, Keio University
[Announcement summary]
In order to achieve the results of process improvement by utilizing the design of experiments, first reduce the error in the target process, then plan and execute the experiment that suits the purpose, analyze the obtained data, and obtain the optimum solution. Ask. This is applied to the process to confirm the improvement, and if the improvement is insufficient, the optimum solution is corrected and the result is surely achieved. The above efforts are not just a design of experiments, but a comprehensive design of experiments that comprehensively summarizes a series of efforts. In order to learn this, it is important to experience PDCA for planning, execution, analysis, and improvement even if it is virtual, and in the past, teaching materials such as paper helicopters and table golf were used. However, these have the drawback that they cannot be learned efficiently in a short time by themselves.
In recent years, as JMP's add-in software, a flying ball simulator that can flexibly generate process data including errors has been developed, and even one person can efficiently learn a comprehensive design of experiments. Therefore, in this research, we devised an educational program to experience and learn PDCA of the comprehensive experimental planning method on the premise of custom planning (optimal planning) by linking the flying ball simulator and JMP's experimental design method platform. In particular, it is characterized by adopting multi-level iterative experiments in order to perform regression correction of the optimum solution in response to non-linear responses.
