In the field of humanitarian logistics, facility and stock positioning are receiving more research attention than they had in the past because of the steep increase in natural disasters around the world. In such studies, objective functions are among the most important criteria used to classify facility location models. A review of related literature has determined that most optimization models formulate their objective functions from numerous different aspects, and that the two most popular objective functions are cost and time minimization. Among the time performance aspects, models designated as type (A) generally refer to those designed to minimize the total expected response time between demand points and facilities over the whole logistics system. However, type (B), consisting of fewer models, attempts to minimize the maximum response time or distance between demand points and facilities rather than attempting to address average system performance. The goal of this research is to propose two facility location models for humanitarian relief logistics by focusing on response time objective aspects, and to analyze the structure of those two models by conducting extensive numerical experiments. Illustrative examples are provided to show how the proposed models can be used to optimize stock facility locations in order to address relief supply chain problems. Furthermore, comparisons of the performance and solution for the two models are well described from several perspectives related to emergency logistics.
This paper presents a distribution model for emergency relief supplies by considering route availability in a specific period of time. This model considers a single distribution center, multiple disaster areas, a homogenous fleet of vehicles, multi-items and multi-periods. First, we develop an algorithm to generate all possible path combinations using binary code that would lead to the determination of the number of all possible scenarios. Afterward, the algorithm generates the available routes for each scenario in each planning period, and then the probability of route availability for each scenario is calculated. These two outputs, route availability and its probability for each scenario become the two important inputs for the next stage, mathematical model formulation. Second, we formulate our mathematical model as a mixed-integer programming model whose objective is to maximize the amount of relief supplies sent to disaster areas for each scenario. The optimum number of vehicles required for the distribution center in each planning period is determined simultaneously. To understand this proposed model better, we give an illustration that demonstrates a large-scale observation including all possible scenarios. Our model is intended to help the government and/or decision-maker to prepare and respond quickly as the disaster strikes.
As the birthrate decline becomes increasingly serious in Japan and other developed countries, the childrearing generation is actively suggesting that fathers also take care of their children. Such men are called “ikumen” in Japan. However, daily childcare in the home certainly puts a burden on their daily lives since these men have jobs outside of the home where they often have long work hours during the weekdays and even sometimes on the weekend and during holidays, especially in Japan. If the physical activity quantity burden can be measured quantitatively in real-time for childcare, then it might improve (Kaizen) childcare work using industrial engineering concepts. Additionally, the childcare services for supporting a ikumen might be innovated with this quantitative measurement and analysis of activities. This research measures ikumen activities using large physical exercise quantity with a real-time triaxial accelerometer as a life log of a daily life, analyzes the characteristics and trends of the quantity of physical activity during childcare, and discusses supporting ikumen using industrial engineering concepts. First, the triaxial accelerometer (OMRON HEALTHCARE) (Oshima, 2011) is attached to ikumen and non-ikumen, and their life logs are recorded for an extended period of time. Next, using the accumulated life logs, how childcare affects the body from the perspective of the quantity of physical activity is analyzed. Finally, based on the measurements and analysis, the results for supporting ikumen with industrial engineering concepts are discussed.
Due to the customer needs of reducing cost and shortening delivery dates, the ability to quickly change production plans has become increasingly more important. In the multi-period system (e.g., production lines) where target processing time exists, production, idle and delay risks occur repeatedly for multiple periods. In such situations, a delay in one process may influence the delivery date of an entire process. In this paper, we discuss the minimum expected cost of this type of case; where the risk depends on a previous situation and occurs repeatedly for multiple periods. We introduce two limited-cycle models with multiple periods, and then a simple optimal switching frequency policy based on a non-reset model. We also introduce a simple optimal assignment policy based on the reset model. Finally, we discuss our research regarding some optimal policies on non-reset and reset models using numerical experiments.
Accidents due to the misuse of consumer products are the main concern in this study. Currently, the traditional risk assessment processes such as the ISO/IEC 51 guide have been proposed for industrial machines. However, designers and safety managers for consumer products tend to relay on their experiences to conduct risk assessment. The reason is that the workload could be high because, with the traditional risk assessment processes, all risks including acceptable level of risk are evaluated for the targeted product. To solve this problem, we have developed the analysis method for worst accidents reasons (AMWAR), which is based on the concept of sabotage analysis focusing only on high-risk accidents. A case study was conducted using people with little experience in risk assessment to determine whether or not they could find out accident scenarios with AMWAR. A company survey was also conducted to evaluate AMWAR by people in charge of product safety. During the case study, the participants managed to forecast many accident scenarios. In the company survey, AMWAR was approved as a risk assessment method that could help reduce the workload. It was also suggested that AMWAR could be used as an educational tool for risk assessment and could be applied in all manufacturing stages of the product; especially those which newly developed without an accident history at the market.
Due to the development of information technologies, there is a huge amount of text data posted on the Internet. In this study, we focus on distance metric learning, which is one of the models of machine learning. Distance metric learning is a method of estimating the metric matrix of Mahalanobis squared distance from training data under an appropriate constraint. Mochihashi et al. proposed a method which can derive the optimal metric matrix analytically. However, the vector space for document data is normally very high dimensionally and sparse. Therefore, when this method is applied to document data directly, over-fitting may occur because the number of estimated parameters is in proportion to the square of the input data dimensions. To avoid the problem of over-fitting, a regularization term is introduced in this study. The purpose of this study is to formulate the regularized estimation of the metric matrix in which the optimal metric matrix can be derived analytically. To verify the effectiveness of the proposed method, document classification using a Japanese newspaper article is conducted.
The present paper demonstrates the potential of an eye movement-based approach to cognitive task analysis in a dialysis room. An analysis framework composed of a task classification scheme and a series of steps for data processing was developed. The framework makes it possible to trace the cognitive aspects of clinical engineer behavior in real work situations. The framework developed was applied to observation data (including eye movement data) obtained from eight clinical engineers working at a hospital. In the eye movement data, we found a slight tendency for engineers having long careers to show relatively superior behavior among all engineers observed in terms of their attention allocation processes toward key information. Based on the analysis results, we discuss the feasibility of the proposed analysis framework and some implications for improving the eye movement-based approach for practical use.