Level: Advanced
Hyper design is an analytical design that uses a hyper structure function, which is a function with a double structure based on hyper factors. This method increases the possibilities of design and enables advanced designs that were not possible before. However, the hyper design developed in the early days required that all intervals could be approximated with a single function with high accuracy. However, in recent years, the number of experiments with a large number of levels has increased, and when the number of levels increases from around 10 to more, it becomes difficult to accurately approximate all intervals with a single function. In such cases, it is reasonable to use a composite approximation that divides the entire interval into small intervals and approximates each interval with high accuracy. However, this results in a significantly larger number of parameters (coefficients) to be handled in the design, so it is necessary to deal with this. This study introduces a method for doing so. In addition, flexible and advanced design is difficult to achieve with an analysis of variance approach, but is possible with mathematical programming. The important thing here is an approach that runs the "PDCA cycle of design". In this study, we consider design as consensus building among the parties involved and propose a method to achieve this with an approach that runs the "PDCA cycle of design" based on mathematical programming. As an advanced application, we also discuss flexible design that allows hyper factors to be freely specified.
For nearly 50 years, he has been researching Total Quality Management (TQM), Statistical Quality Management (SQM), and design theory. Since the beginning of the 21st century, he has proposed a new design paradigm called Hyper Design, developed its mathematical theory, HOPE theory (Hyper Optimization for Prospective Engineering), and is co-developing its support software, HOPE-Add-in, with SAS. Through this effort, he has realized a new design method that combines the concept of Hyper Design, the statistical theory of HOPE theory, and the support tool, HOPE-Add-in for JMP. As a social scientific extension of this theory, he has proposed selective two-sided causal analysis (multi-group principal component regression analysis as obverse causal analysis + factor regression analysis as reverse causal analysis), and is currently proposing reverse causal analysis (a causal analysis method equivalent to quasi-SEM), which is an advanced approach following selective two-sided causal analysis.
Go to the collection site
https://community.jmp.com/t5/Discovery-Summit-Japan-2022/Discovery-Summit-Japan-2022-%E8%AB%96%E6%96%87%E9%9B%86/ta-p/568396
