In recent years, the urban heat island phenomenon in the Tokyo ward area has become a significant problem. To determine the regional characteristics of the urban heat island, a high spatial density meteorological observation system is required. For this purpose, Tokyo Metropolitan Research Institute for Environmental Protection installed a high-density meteorological observation system called METROS in the Tokyo ward area in July 2002. However, missing values are often present in the observed data. Analyzing data without handling missing values can cause loss of precision and biased estimates. Thus, an imputation method for missing values that considers the temperature characteristics of the Tokyo ward area is required. In this study, we propose two imputation methods considering temperature fluctuations with high locality in central Tokyo. Both methods are modification of the inverse distance weight (IDW) method, which is a general imputation method for spatial data. Our proposed methods correct surrounding observed values before imputing unobserved value by their weighted average. Different approaches are considered for correcting surrounding observed values. A simulation study based on various missing data was conducted. Simulations revealed that our proposed methods exhibited higher performance in different settings of missing data than the IDW method.

Taguchi’s T-method belong to the MT System, which is a representative method in quality engineering. Another relevant technique is multiple single regression (MSR), which incorporates the concept of the resistance line. These methods are used to make predictions when there is one output. In addition, multivariate multiple regression analysis is used as a prediction method when there are multivariate outputs.

This study proposes methods that apply the T-method and MSR to multiple outputs. The proposed methods are: "independent multivariate T-method" (iMvT), "multivariate T-method" (MvT), "simple multivariate T-method" (MvT+), "independent multivariate multiple single regression" (iMMSR), "multivariate multiple single regression" (MMSR), and "simple multivariate multiple single regression" (MMSR+). iMvT and iMMSR combine predictions simply, MvT and MMSR apply generalized least squares, and MvT+ and MMSR+ apply Moore–Penrose’s generalized inverse matrix. We compared the accuracy of the proposed methods with that of multivariate multiple regression analysis. Artificial data analysis was performed by changing the number of training data and correlation. We also performed analyses on the actual data. The results indicate that the proposed methods are more accurate than the existing methods.

Japan is one of the world's most disaster-prone countries. The continuation and restoration of business are important in such circumstances. The inability to provide healthcare services, in particular, has a severe impact on disaster areas. Therefore, hospitals need to establish a Business Continuity Management System (BCMS) to improve business continuity. BCMS requires regular evaluation and improvement of the management system. However, since BCMS is targeted for business continuity in an emergency, it cannot be evaluated using the results of daily work. Therefore, it is necessary to perform disaster exercises, evaluate the results, and improve BCMS. However, there are cases where it does not lead to improvement of BCMS. Additionally, disaster exercises also have the purpose of raising the staff's awareness of the disaster. However, disaster exercises that are far from reality may reduce motivation and awareness. This study aims to propose a method for planning exercises based on events that occurred in hospitals in the disaster area during past disasters. The study makes it possible to reproduce the previous disaster by exercise and raise awareness of the disaster. Moreover, this study enables us to carry out an evidence-based evaluation of disaster exercise by using past records to provide answers.

Evaluating measurement methods is important to perform quality assurance of measurement results. This study aims to estimate the measurement precision of ordinal categorical data. ISO 5725-1 states that precision of measurement methods can be quantified by conducting precision evaluation experiments, and ISO 5725-2 provides a methodology to evaluate the measurement precision for quantitative data. Regarding qualitative data, some previous studies exist on quantifying the precision of binary and multinomial data. However, no reasonable methodologies have yet been established. Thus, this study proposes the analysis method for the data obtained from a collaborative study on intratracheal administration testing for assessing effects in rat lungs. A primary objective of the collaborative study was to demonstrate dose-response relationships. To investigate the dose-response relationships, hypothesis tests such as a Cochran–Armitage trend test and a cumulative chi-squared test were performed, which are applicable even for ordinal categorical data. Moreover, logistic regression analysis was performed to examine the dose-response relationships and the precision among laboratories. This result was expressed by 50% efficient dose ( ) and area under the curve (AUC). Then, we obtained that a dose-response relationship was observed for each laboratory, and the differences in the relationships among laboratories expressed precision.

The Mahalanobis Taguchi (MT) and the recognition Taguchi (RT) methods are typical methods of the MT system and are, used for discrimination of multivariate data such as bankruptcy judgement of a company, medical checkup, and character recognition. They have a common problem regarding how missing data are handled. The use of these methods requires the assumption that normal samples are obtained from a homogenous space known as a unit space. Therefore, it is necessary to prepare sufficient samples for analysis.

We first apply multiple imputation to the MT method, the MI-MT method. However, the MT method has a restriction that analysis cannot be performed if the sample size is smaller than the number of variables. Even if the sample size is larger, it is reported that if the sample size is not sufficient, the results will be unstable. On the other hand, the RT method reduces to two variables, so it is not easily affected by the sample size. Therefore, we apply multiple imputation to the RT method, the MI-RT method. Monte Carlo simulation, is used to investigate the accuracy of MI-MT and MI-RT. We conclude that MI-MT and MI-RT are better than MT and RT using other imputation methods.