The global pandemic has exposed deficiencies in supply chains in rapidly responding to urgent medical needs and demonstrated the challenge of promptly acquiring medical supplies during emergencies. Even under policy tools and using existing e-commerce alternatives, being able to promptly acquire medical supplies, including personal protective equipment and medical devices, is still a significant concern for future emergency situations. Hence, the details for an open, extensible, service-oriented product marketplace platform that enables rapid matchmaking between consumers and manufacturers in such emergency situations are presented in this work. This marketplace is especially designed based on a microservice architecture, where each service operates independently and communicates through message exchanges, optimizing the supply chain for items according to customers' preferences and constraints regarding budget and lead time. In addition, the results of implementing a simplified prototype of the platform, where a healthcare worker is using the online product marketplace to order a replacement part of a respirator mask, are also demonstrated. By using this online product marketplace, manufacturers will benefit from improved accessibility to their products, services, and manufacturing resources (also under normal operations), while consumers will gain unprecedented access to existing products and manufacturing resources.

Timely evaluation of risk factors is essential for building resilient supply chains and requires continuous monitoring and updating. This paper proposes a conceptual model for supply chain risk management that uses large language models (LLMs) to detect a disruption trigger event from news data, determines its impact, and develops a proper risk mitigation strategy. Even though the conceptual model can accommodate many kinds of supply chain risks, this study focuses on natural hazard risks affecting global supply chains. Because natural hazards are related to the locations of the suppliers, the model uses country level natural hazard risk scores to evaluate suppliers. Here, a country’s risk score determined by the Index for Risk Management (INFORM) is used as a baseline risk score for each country, which are then, updated by the disruption trigger event’s features captured by the LLM. Lastly, a supply chain optimization model is used to determine alternative suppliers to mitigate the impact of risk on disruption. The conceptual model is demonstrated through the case study of a bicycle company. Preliminary results indicate that the proposed model can be used to improve the resilience of supply chains against natural hazards.

Manufacturing Process Selection (MPS) plays a critical role in ensuring product quality, cost efficiency, and sustainability. A great deal of knowledge and experience is needed to determine the most appropriate set of manufacturing processes to realize a product. Thus, effectively capturing, organizing, and utilizing related knowledge is essential. Traditional MPS methods rely on predetermined criteria and frameworks, limiting their adaptability to new knowledge and flexible process selection needs. To address this limitation, we propose an adaptive MPS framework that integrates Knowledge Graphs (KG) with manufacturing processes knowledge, along with MPS rules to enable explainable and flexible process selection. The framework retrieves relevant knowledge from KGs and applies decision rules to guide the execution of MPS tasks without the need for human expert intervention. It includes an automated knowledge extraction and integration module leveraging Large Language Models to ensure continuous updates to manufacturing knowledge. The proposed framework was validated using a case study, integrating a new manufacturing process into the KGs and identifying the optimal manufacturing processes for a respirator mask facepiece. This study demonstrates the feasibility of using graph topology-based knowledge and decision rules for adaptive and automated MPS tasks. Future work will focus on expanding the KG scheme with more types of MPS knowledge and rules and improving knowledge extraction and integration techniques for broader industrial applications.

Manufacturing process selection (MPS) is the process of identifying the most suitable manufacturing process based on various criteria such as production demands. Rapid selection is particularly crucial when sudden manufacturing chain disruptions occur, such as those caused by a disaster or a pandemic, where there is an urgent need to find new or alternative manufacturing processes. Challenges emerge due to the need for in-depth knowledge of manufacturing processes across various engineering domains, which is difficult to master, and the lack of easily accessible MPS knowledge. To address these challenges, we propose a process selection approach called the Open Manufacturing Process Knowledge Graph (OMPKG). The OMPKG is designed to comprehensively capture interdisciplinary engineering knowledge, while offering flexibility and scalability to handle complex and evolving MPS data and ensuring expression consistency for sharing. The OMPKG can be integrated with a general selection method, called screening and ranking, to determine the appropriate manufacturing process rapidly. In this paper, we describe the development rationale and logic of the OMPKG for specific manufacturing processes, such as casting, molding, and additive manufacturing, and demonstrate its use in a case study of producing respiratory mask facepieces.

The coronavirus pandemic has exposed deficiencies of global supply chains in rapidly responding to urgent medical needs and demonstrated the challenge of promptly acquiring medical supplies during emergencies. Even under policy tools and using existing online alternatives, being able to promptly acquire medical supplies, including personal protective equipment and medical devices, is still a concern for future emergency situations. Hence, the details for an open, extensible, service-oriented cloud platform that enables the rapid matchmaking between consumers and manufacturers in emergency situations is presented in this work. More specifically, the system architecture deployed for the platform, along with the details of its main components are discussed. In addition, the results of implementing a simplified prototype of the platform, where a healthcare worker is using the online product marketplace to make an order for a replacement part of a respirator mask, are also briefly demonstrated. By using this online product marketplace, manufacturers will benefit from improved accessibility to their products, services, and manufacturing resources (even under normal operations), while consumers will gain unprecedented access to existing products and manufacturing resources. The platform also has the potential to improve the nation's ability to prevent, prepare for, and respond to future pandemics or other emergency events.

Resistance sport welding (RSW) research usually suffers a lack of data to build an accurate estimation model. Besides physical experiment data, the text knowledge for welding mechanisms would be additional data. However, the text knowledge is often incomplete. Knowledge completion (KC) is expected to solve the problem. It aims to derive the missing (or unknown) knowledge from explicitly expressed (or known) knowledge. So far, the KC study has focused on web-based knowledge like DBLP and Freebase. Research on RSW KC is rare. The characteristic of RSW knowledge is different from that of web knowledge. Web knowledge uses triples for its representation and triples are a type of graph. Thus, Graph Neural Networks (GNNs) has recently become a promising KC algorithm. To assess whether the GNN-based KC is feasible for RSW KC, this paper performs three tasks: assessing RSW knowledge types; building possible scenarios for RSW KC; discussing the applicability of the GNN-based KC method to RSW KC by using GNN-centered problem formulations to resolve the scenarios. This paper highlights two considerable challenges. The first is that RSW knowledge requires causal relations in addition to the structural relations indicated in web knowledge. The second is that, although the problem of RSW KC could be formally formulated the same as web knowledge KC, additional research works are needed to confirm that the formulation goes in a sufficient level because the amount of data issue for GNNs remains.

Resistance spot welding (RSW) is a critical joining method in sheet-metal industries. The machine-learning technique fueled by the historical experimental data of the existing materials has been used to build the data-driven model (DDM). The DDM is expected to be a promising tool to investigate a new material and its welding behavior because DDM can narrow the range of the test matrix and can thus reduce the number of necessary physical experiments and the cost. However, one of crucial data quality problems with machine learning is that training data sets' lack of descriptability for test sets causes poor prediction. This research starts by indicating that such data quality problems that exist in the context of weldment design. To resolve this problem, the presented study introduces a novel approach named Similar Weldment Case Selection (SWCS), which predicts the key parameter, the nugget size, of spot welding results of a new material by selecting the most similar one among the existing welding cases and then constructing a prediction model to generate the results. In order to overcome the difficulties with defining the selection criteria only with the material properties and geometric features, this study has come up with another factor, nugget-size weld-current series (NWS), to consider; the NWS is a factor that describes the shape of the relation between weld-current and nugget size. The similarity between two NWSs of different materials is calculated (quantified) with the dynamic time warping (DTW) method. Initially, the twelve conventional algorithms are tested for varying degrees of descriptability between the two weldment designs for test and train datasets; the prediction accuracies are found to be proportional to the train set's descriptability on the test set. The results are then compared with those from the SWCS. The SWCS yields superior accuracy than the twelve algorithms do when the two materials are similar or different. However, the superiority disappears when the two are the same.

Riveting is a prominent joining method due to the capability of easily assembling an innovated material (e.g., dissimilar materials with light-weight and stronger performance) to an enhanced structure (e.g., body-in-white). It receives attention recently in the transportation industry and the packing manufacturing industry (e.g., white goods). The quality prediction of riveting can increase the efficiency of the riveting process and design. Even though there is much potential to data mining technique for the prediction, the mining approach is rarely used for riveting. At this perspective, we carried out a survey to make an accessible bibliography of riveting prediction for helping researchers. In this research, we searched for the past 25 year's works of literature related to the quality measure and prediction method of riveting with special interests to self-piercing riveting (SPR). To do this, we retrieve research papers indexed at Engineering Village Database. Firstly, we categorize riveting-quality measures and prediction methods. Secondly, we analyze the topic and trend of riveting research from the collected papers. Finally, we conclude with the remaining challenges for the riveting quality prediction as well as in a data mining perspective.

Resistance spot welding (RSW) is one of the critical joining methods in sheet metal-based industries. The nugget size is a crucial factor, which determines the stoutness of final products. In this research, we present a hybrid prediction method based on a theoretical physical model and a machine learning method (i.e., spline-based interpolation). The theoretical physical model expects to show relatively higher precision in prediction; however, it needs an expert knowledge to find proper dynamic parameters (e.g., resistance and heat energy) of the model. On the other hand, the machine learning model is likely to find the model automatically. However, it seems to be overfitted for the specific training dataset (especially with inconsistent RSW dataset) and it needs massive datasets. To overcome these data inconsistency and hidden physical patterns, this paper proposes a hybrid method based on the theoretical model and the interpolation with the spline method. The theoretical physical model builds a mathematical model for the nugget growth and its formation. The spline-based interpolation method analyzes the correlation between the input parameters (e.g., stack-up characteristics, welding force, welding current, and welding time) and the output parameter (i.e., nugget diameter). The experiment results show that the proposed hybrid method provides more meaningful prediction than the existed prediction algorithms with the RSW dataset and can be an error-data endurable method.

The understanding of stakeholder's behavior is essential to design a system because the system should satisfy and support stakeholders for the stakeholders to adopt the system (Jiao & Chen, 2006).Understanding of stakeholders' behavior requires knowledge about how they work in the designed system and how they respond to the designed product.Furthermore, if we can understand why the stakeholders work or respond in such a way, we can predict the behavior of stakeholders.The causality refers to the relationship between causes and effects.The causality is essential to stakeholder behavior analysis.The causality analysis of the stakeholder behaviour contributes to the system design and analysis by providing knowledge on three perspectives (i.e., the prediction of stakeholder behavior to the new system, the motivation of the new system design, and the new system itself).Specific examples for these three perspectives are following: first, we can build a stakeholder response model.The model can be a structuralhypothetic model in social science (Biddle & Marlin, 1987;Bagozzi & Yi, 1988) and customer's cognition model for a product (Khalid & Helander, 2004; Li, 2004).Second, stakeholder's dissatisfaction inferred by (or evaluated from) the model can be a motivation for a new system.Lastly, the causality of stakeholder's behavior can be implemented as an intelligent system itself.For instance, the causality can be converted into a mathematical model like operations research model (Shannon et al., 1980).

For the sake of building an intelligent patient-centered decision support system (PACDSS), we propose a novel approach for preparing causal knowledge from the clinical dataset by using GBN (General Bayesian Network). For this purpose, we used clinical datasets from the stem-cell therapy, and applied GBN mechanism to induce causal knowledge and implement an inference engine as a result. What-if analysis results were convincing enough to conclude that our proposed approach can secure design of an intelligent PACDSS embedded in causal knowledge.

Objectives: By using six years of KNHANES dataset (2008~2013) about 60 ages older people, we analyzed how the depression prevalence rate in the elderly is influenced by disease and activity limit. Especially, to add a sense of more reality, we adopted stress experience as a control variable to see how the depression prevalence rate in the elderly is influenced by disease and activity limit depending on the stress experience. Methods : We adopted six years of KNHANES dataset, indicating that our results were based on long period of time capable of considering temporal patterns in the depression prevalence rate in the elderly. Total 1,160 elderly people in KNHANES were selected for our empirical analyses. Dependent variable is either 0 or 1 depending on whether the elderly people feel depression. Main explanatory variables for our study include disease and activity limit. Logistic regression analysis was applied for two group such as stress experience and non-experience. Results : According to the empirical results, stress factor is found to be significant in explaining the depression in the elderly. Depression prevalence rate increased when the elderly has stress experience: chronical disease(OR=1.650), chronical disease with activity limit(OR=3.388), non-chronical disease with stress(OR=11.841) chronical disease with stress (OR=13.561) and chronical disease with activity limit and stress(OR=28.691). Conclusions: The finding suggest that the Countermeasures of elderly's depression alleviation should include stress management.

Classifiers and imputation methods have played crucial parts in the field of big data analytics. Especially, when using data sets characterized by horizontal scattering, vertical scattering, level of spread, compound metric, imbalance ratio and missing ratio, how to combine those classifiers and imputation methods will lead to significantly different performance. Therefore, it is essential that the characteristics of data sets must be identified in advance to facilitate selection of the optimal combination of imputation methods and classifiers. However, this is a very costly process. The purpose of this paper is to propose a novel method of automatic, adaptive selection of the optimal combination of classifier and imputation method on the basis of features of a given data set. The proposed method turned out to successfully demonstrate the superiority in performance evaluations with multiple data sets. The decision makers in big data analytics could greatly benefit from the proposed method when it comes to dealing with data set in which the distribution of missing data varies in real time. (C) 2015 Elsevier Ltd. All rights reserved.

국내에서 이뤄진 건강관련 삶의 질 관련 연구는 대부분 신체활동과 질환 간의 관계를 살피거나, 스트레스나 우울증과 삶의 질 간의 관계만을 살피는데 한정되어 있었다. 그러나, 스트레스와 같은 감정상태가 신체활동을 통한 건강관련 삶의 질에 미치는 영향에 관한 체계적인 실증연구는 아직 이루어지지 않고 있다. 이에 본 연구에서는 2008~2013년도 국민건강영양조사(KNHANES)자료를 토대로, 스트레스와 신체활동에 따른 건강관련 삶의 질 간의 관련성을 분석하였다. 이를 위해 교차분석과 집단별 로지스틱회귀분석을 수행하였다. 분석결과, 스트레스 비경험 군에서 남녀 모두 고강도 신체활동에 비하여 저강도 신체활동 수행시 삶의 질 수준이 낮아졌다(남: OR=1.15 p<0.001, 여: OR=1.18 p<0.001). 그러나, 스트레스 경험군에서는 남자는 스트레스 비경험군과 동일하게 저강도 신체활동 수행시 삶의 질 수준이 낮아졌지만(OR=1.79 p<0.01), 여자의 경우에는 오히려 삶의 질 수준이 높아졌다(OR=1.18 p<0.05). 이를 통해 스트레스에 따라 신체활동 수준이 건강관련 삶의 질에 미치는 영향에 차이가 있음을 발견하였다. In literature, empirical studies investigating how stress affects HRQOL (health related quality of life) through physical activities are insufficient. In this sense, on the basis of KNHANES dataset for 2008 ~ 2013, we conducted an empirical study. Empirical results revealed that in the male group with stress experience, HRQOL was significantly influenced by age, household income, education, occupation, physical activity level. To do so, we adapt the chi-square analysis and the logistic regression analysis. Meanwhile, in the stress non-experience group, the low level activity has lower HRQOL than the high level activity(male: OR=1.15 p<0.001, female: OR=1.18 p<0.001). In the stress experience group, the male has the same pattern of effect compared to the stress non-experience group(OR=1.79 p<0.01). However, in the female, the low level activity has the higher HRQOL than the high level activity(OR=1.18 p<0.05). Therefore, we confirmed that physical activity has a different effect on HRQOL through stress experience.

본 연구의 목적은 신체활동수준과 스트레스 수준에 따른 직군별 Vitamin D 겹필률 차이를 조사하기 위함이다. 본 연구는 KNHANES 2008~2013 데이터를 이용하여 실증연구를 수행하였다. 이를 위해서 교차분석, 로지스틱분석을 사용하였으며, 스트레스의 간접효과 분석을 위해서 SOBEL 테스트를 사용하였다. 분석결과 Vitamin D 결핍은 옥외활동이 많은 직군에서 상대적으로 낮게 나타났다. 또한 신체활동수준이 높을수록 결핍률이 낮게 나타났다. 스트레스는 신체활동수준을 매개변수로 하여 Vitamin D 결핍에 영향을 미치고, 스트레스가 높을수록 Vitamin D 결핍이 높은 것으로 나타났다. 이러한 분석 결과 옥외활동이 많은 사무종사자 등의 직군에서는 근무자가 주기적으로 햇볕에 노출될 수 있도록 해야 하며, 이를 위해서 신체활동수준을 향상 시킬 수 있다면 적절한 해결책이 될 수 있는 것으로 파악되었다. 또한 직장 내 스트레스를 관리하기 위하여 근로자의 신체활동 수준을 유지하는 것도 고려될 필요가 있다. The purpose of the study is to investigate the occupational difference of Vitamin D deficiency according to the level of physical activity and the level of stress. For this purpose, We performed empirical approaches and adopted KNHANES 2008~2013 dataset, to which were applied such methods as crosstabulation analysis, logistic regression and SOBEL test. As a result, we found that Vitamin D deficiency was higher in the group of people with indoor-working-job than in the group of people with out-door working job. Besides, Vitamin D deficiency tends to increase in the people with lower physical activity and higher stress experience. As a result, those employees highly related with in-door activities must be ensured that they need to be exposed to sunlight on a regular basis. Moreover, improving the physical activity levels of employees could be one of appropriate solutions to alleviate Vitamin D deficiency problems. Besides, lowering stress levels in workplace needs to be seriously considered in order not to drop the physical activity levels of workers.

In the healthcare literature, happiness has been neglected as an important research issue. However, it seems clear that happiness becomes more important in research agenda as societies grow older. It is well known that happiness is related with various kinds of factors. Therefore, managing happiness requires careful handling of a number of related factors in a very systematic and organized way. In this sense, this study proposes General Bayesian Network (GBN) approach in order to extract causal knowledge about happiness from the analysis of the related factors. By using Friends-and-Family dataset, we performed GBN-based scenario analyses to answer sophisticated questions about happiness. The empirical results showed that GBN proved to have huge potentials in handling well-ness problems in the field of health informatics.

In a ubiquitous environment, high-accuracy data analysis is essential because it affects real-world decision-making. However, in the real world, user-related data from information systems are often missing due to users’ concerns about privacy or lack of obligation to provide complete data. This data incompleteness can impair the accuracy of data analysis using classification algorithms, which can degrade the value of the data. Many studies have attempted to overcome these data incompleteness issues and to improve the quality of data analysis using classification algorithms. The performance of classification algorithms may be affected by the characteristics and patterns of the missing data, such as the ratio of missing data to complete data. We perform a concrete causal analysis of differences in performance of classification algorithms based on various factors. The characteristics of missing values, datasets, and imputation methods are examined. We also propose imputation and classification algorithms appropriate to different datasets and circumstances.

A novel activity recognition method is proposed based on acoustic information acquired from microphones in an unobtrusive and privacy-preserving manner. Behavior detection mechanisms may be useful in context-aware domains in everyday life, but they may be inaccurate, and privacy violation is a concern. For example, vision-based behavior detection using cameras is difficult to apply in a private space such as a home, and inaccuracies in identifying user behaviors reduce acceptance of the technology. In addition, activity recognition using wearable sensors is very uncomfortable and costly to apply for commercial purposes. In this study, an acoustic information-based behavior detection algorithm is proposed for use in private spaces. This system classifies human activities using acoustic information. It combines strategies of elimination and similarity and establishes new rules. The performance of the proposed algorithm was compared with that of commonly used classification algorithms such as case-based reasoning, k-nearest neighbors, support vector machine, and multiple regression.

Context-aware applications, which consist of a sensor system, a reasoning system and service artifacts such as mobile devices, kiosks and robots, require data from the sensors to be queried on a continuous basis. The smaller the sensing interval and the greater the amount of service time, the more accurate the service, but the more energy is consumed. Thus, use of context-aware applications always involves a trade-off. In this paper, we propose an automatic method of optimizing the level of personalization involving the sensing cycle and service time of a personalized application. The method proposes a quadratic form of total cost curve which demonstrated that the minimum identified value is always the global optimum. This eliminates the necessity of an exhaustive search for the minimum value for all levels of personalization. (C) 2014 Elsevier Ltd. All rights reserved.