Robust parameter design (RPD) is used in the Taguchi Method to reduce variability in system outputs and enhance quality. The technology development process utilizing RPD comprises an evaluation component for assessing robustness and a mechanism analysis component for devising new systems if the objectives are unmet. In this study, we focus on the Causality Search T-Method (CS-T method), an efficient method for mechanism analysis, and its development method, Knowledge Search-Instrumental variables (KS -IV). The CS-T method is a practical technique that can improve the efficiency of system selection while maintaining the efficiency of an RPD. However, the conditions under which these methods can be used with high accuracy and validity remain unclear. In this study, we propose an extension of the CS-T and KS-IV methods for dynamic parameter design. In addition, we discuss the differences between the data aggregation methods of the two methods and the relationship between the aggregation contents and results, and based on these discussions, we improve the accuracy and interpretability of the methods.

View full abstract

Securing medical personnel in outpatient or inpatient wards in the medical field is difficult. In particular, the 2024 reform will apply overtime limits to physicians’ work. Reducing the burden on physicians and their working hours is a crucial task. Therefore, the demand for nurses who can perform specific actions has increased to reduce the burden on physicians. However, many challenges exist, such as the fact that physicians cannot be entrusted with all their duties, and the specific actions to be performed by nurses are vague and unclear. It is necessary to resolve these issues and establish appropriate work systems to effectively activate specific types of nurses. This study identified issues in activating specific nurses through a review of previous studies and a case study of Hospital I. Measures to enhance their effective activation were proposed, along with a roadmap outlining steps to establish appropriate work systems. Additionally, methods for evaluating the competence of specific nurses and verifying their effectiveness were suggested. The findings are expected to serve as a practical guide for hospitals, providing actionable measures to optimize the use of specific nurses and reduce physicians’ workloads while improving operational efficiency.

View full abstract

In recent years, with the development of the information society, the accumulation and utilization of diverse data has become increasingly important. Among them, there is a growing need to analyze high-dimensional data that is difficult for humans to interpret. However, it is difficult to analyze high-dimensional data as it is, and extracting and visualizing important information from the enormous dimensions leads to understanding the overall trends of the data and the relationships between data points. Against this background, methods have been studied to reduce high-dimensional data to low-dimensional space and extract important information that is interpretable by humans. Among them, probabilistic neighborhood embedding methods that consider nonlinear relationships are widely used. For example, t-SNE, which visualizes based on the local similarity between data points, and vMF-SNE, which expresses global data similarity using angles in low-dimensional space, are known. By considering the similarity of labels between data in these dimensionality reduction methods, it becomes possible to perform classification tasks in reduced low-dimensional space and verify the characteristics of adjacent data. In fact, supervised t-SNE has been proposed as an extension of t-SNE, but because embedding is based on local information, when label information is taken into account, only data with the same label tend to be gathered as adjacent points. In this study, we propose an embedding method for vMF-SNE when category labels are given. By performing embedding based on angles that consider both global data similarity and label information, it becomes possible to evaluate the similarity between data with different labels. Finally, we perform visualization of high-dimensional data and evaluate the model by comparing visualization and classification with conventional methods.

View full abstract