• Journal of the Korean Society of Safety
• /
• v.38 no.2
• /
• pp.104-111
• /
• 2023

Rolling stock maintenance, which focuses on preventive maintenance, is typically implemented considering the potential harm that may be inflicted to passengers in the event of failure. The cost of preventive maintenance throughout the life cycle of a rolling stock is 60%-75% of the initial purchase cost. Therefore, ensuring stability and reducing maintenance costs are essential in terms of economy. In particular, private railroad operators must reduce government support budget by effectively utilizing railroad resources and reducing maintenance costs. Accordingly, this study analyzes the reliability characteristics of components using field data. Moreover, it resolves the problem of determining an economical replacement interval considering the timing of scrapping railroad vehicles. The procedure for determining the optimal replacement interval involves five steps. According to the decision model, the optimal replacement interval for the onboard signal device components of the "A" line train is calculated using field data, such as failure data, preventive maintenance cost, and failure maintenance cost. The field data analysis indicates that the mileage meter is 9 years, which is less than the designed durability of 15 years. Furthermore, a life cycle in which the phase signal has few failures is found to be the same as the actual durability of 15 years.

• Journal of Platform Technology
• /
• v.11 no.3
• /
• pp.45-55
• /
• 2023

This paper presents a model that supports decision-makers in the Korean film industry to maximize the success of online movies. To achieve this, we collected historical box office movies and clustered them into types to propose a model predicting each type's online box office performance. We considered various features to identify factors contributing to movie success and reduced feature dimensionality for computational efficiency. We systematically classified the movies into types and predicted each type's online box office performance while analyzing the contributing factors. We used automated machine learning (AutoML) techniques to automatically propose and select machine learning algorithms optimized for the problem, allowing for easy experimentation and selection of multiple algorithms. This approach is expected to provide a foundation for informed decision-making and contribute to better performance in the film industry.

• Journal of Korean Biological Nursing Science
• /
• v.25 no.3
• /
• pp.143-159
• /
• 2023

Purpose: The purpose of this study was to provide evidence for future research by analyzing the contents, delivery methods, and educational effects of educational programs for hemodialysis patients. Methods: A literature review was conducted in the following order: problem identification, literature search and data collection, data evaluation, data analysis, and data presentation. In total, 936 documents were retrieved from the search, and 24 documents were finally included in this study. Results: The contents of the educational programs were comprehensive hemodialysis-related knowledge education in seven studies, exercise-related education in 10 studies, diet-related education in four studies, drug-related education in one study, and complex education in two studies. The delivery methods were pamphlets in six studies, audio-visual materials in four studies, direct guidance by researchers in six studies, and multiple methods in seven studies. The effects of the educational program were divided into physical indicators, emotional indicators, and knowledge and performance related to hemodialysis. Sixteen out of 20 articles using physical indicators showed significant outcomes, and 10 out of 12 articles using emotional indicators derived positive results. Hemodialysis-related knowledge and performance were measured in eight and 10 studies, respectively, and meaningful results were found in six studies for knowledge and seven studies for performance. Conclusion: Future research will require the development of a systematic and standardized educational program with comprehensive content that can be provided to all hemodialysis patients, as well as a highly accessible education delivery method to enhance the effectiveness of education.

• Journal of Platform Technology
• /
• v.10 no.4
• /
• pp.91-96
• /
• 2022

Recently, the importance of Research and Development(R&D) security as well as R&D investment is emphasized in the flow of technology hegemony competition, where technology is directly related to national competitiveness.However, despite the enormous impact of the R&D security failure results, research output leakage accidents continue to occur.To solve this problem, this study analyzed leakage accidents and cases of R&D output and concluded that it is priory to develop regulations to raise security awareness at the field researcher level rather than the macroscopic security management system. In addition, in order to design the direction of the researcher security measures, observational study was conducted at the university research site, and four directions were presented, including case analysis and integration. The direction for designing researcher security measures will be used as a basis for developing security regulations specialized in future research sites and security management systems for research institutes.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.3
• /
• pp.89-100
• /
• 2023

Recently, on-device artificial intelligence (AI) solutions using mobile devices and embedded edge devices have emerged in various fields, such as computer vision, to address network traffic burdens, low-energy operations, and security problems. Although vision transformer deep learning models have outperformed conventional convolutional neural network (CNN) models in computer vision, they require more computations and parameters than CNN models. Thus, they are not directly applicable to embedded edge devices with limited hardware resources. Many researchers have proposed various model compression methods or lightweight architectures for vision transformers; however, there are only a few studies evaluating the effects of model compression techniques of vision transformers on performance. Regarding this problem, this paper presents a performance evaluation of vision transformers on embedded platforms. We investigated the behaviors of three vision transformers: DeiT, LeViT, and MobileViT. Each model performance was evaluated by accuracy and inference time on edge devices using the ImageNet dataset. We assessed the effects of the quantization method applied to the models on latency enhancement and accuracy degradation by profiling the proportion of response time occupied by major operations. In addition, we evaluated the performance of each model on GPU and EdgeTPU-based edge devices. In our experimental results, LeViT showed the best performance in CPU-based edge devices, and DeiT-small showed the highest performance improvement in GPU-based edge devices. In addition, only MobileViT models showed performance improvement on EdgeTPU. Summarizing the analysis results through profiling, the degree of performance improvement of each vision transformer model was highly dependent on the proportion of parts that could be optimized in the target edge device. In summary, to apply vision transformers to on-device AI solutions, either proper operation composition and optimizations specific to target edge devices must be considered.

• Annual Conference of KIPS
• /
• 2010.04a
• /
• pp.581-584
• /
• 2010

Based on superspeed nation-wide communication infrastructure, high speed and high quality ALL IP communication network is created. It is a point in time where such infrastructure allows residential and commercial phones to be ported from PSTN to IP network. The problem experienced on IP phones in ALL IP environment and IMS server products are not easily approachable and dealt with. Therefore, this paper discusses approaches used in analyzing and resolving problems before they actually occur. This paper first discusses briefly the IP phones used in IMS network and their related problems. Then next topic deals with characteristics of these problems and current status. Last but not least, the causes of the problems are analyzed and solutions to these problems are introduced. Also discussed are solutions to possible future issues followed by conclusion.

• Journal of Korea Society of Industrial Information Systems
• /
• v.14 no.4
• /
• pp.1-7
• /
• 2009

This paper presents a process routing (PR) algorithm with multiple objectives. PR determines the optimum sequence of operations for transforming a raw material into a completed part within the available machining resources. In any computer aided process planning (CAPP) system, selection of the machining operation sequence is one of the most critical activities for manufacturing a part and for the technical specification in the part drawing. Here, the goal could be to generate the sequence that optimizes production time, production cost, machine utilization or with multiple these criteria. The Pareto Stratum Niche Cubicle (PS NC) GA has been adopted to find the optimum sequence of operations that optimize two conflicting criteria; production cost and production quality. The numerical analysis shows that the proposed PS NC GA is both effective and efficient to the PR problem.

• Journal of Korean Society of Rural Planning
• /
• v.29 no.3
• /
• pp.53-67
• /
• 2023

In this study, the problem of the idleness of public buildings in rural areas is seriously emerging, but there are no clear standards for assessing the utilization of these buildings. Therefore, this study endeavors to investigate the actual state of idleness of buildings created by rural area development projects based on the criteria for each type. As part of this study, rural facilities were selected in two locations in Sangju City and Miryang City among the target areas of the Rural Agreement. Utilization evaluation criteria were developed to identify the conditions of underutilization. In order to determine the utilization pattern for each facility type, an in-depth interview with the operator and an inspection of the facility were conducted. Based on the analysis, the degree of utilization (e.g., low use, non-use, etc.) was different for each facility type, and among the four facility types (rural tourism, culture and welfare, exercise and recreation, income-based), rural tourism was rated as the least efficient. It has been found that the operating entity's capacity has a significant influence on the utilization of facilities. Socioeconomic factors, such as a decrease in utilization rates, are greatly influenced by the operating entity's capacity. Therefore, support from local governments as well as the national level is required to recycle idle facilities. Lastly, this study suggests the need for different standards for utilization, depending on the type of facility. This moves beyond checking the degree of idleness by the same standard that was previously implemented.

• Journal of Internet Computing and Services
• /
• v.24 no.4
• /
• pp.1-13
• /
• 2023

As federated learning brings a large paradigm to modern artificial intelligence research, efforts are being made to incorporate federated learning into research in various fields. However, researchers who apply federated learning face the problem of choosing a federated learning framework and benchmark tool suitable for their situation and purpose. This study aims to present guidelines for selection of federated learning frameworks and benchmark tools considering the circumstances of researchers who apply federated learning in practice. In particular, there are three main contributions in this study. First, it generalizes the situation of the researcher applying federated learning by combining it with the goal of federated learning and proposes guidelines for selecting a federated learning framework suitable for each situation. Second, it shows the suitability of selection by comparing the characteristics and performance of each federated learning framework to the researcher. Finally, the limitations of the existing federated learning framework and a plan for real-world federated learning operation are proposed.

• Analyses & Alternatives
• /
• v.7 no.1
• /
• pp.147-170
• /
• 2023

The purpose of this study is to examine the socal policy of the Yoon Seok-Yeol government from the perspective of both policy and theory. Theoretically, the analysis reveals that the Yoon government attempts to build and implement a welfare system model that was distinctly different from the previous Moon government. The newly elected government adopted a welfare model known as 'welfare for the vulnerable.' This model inherits the welfare ideology of the conservative parties in Korea, which is the self-reliance welfare. The Yoon government continues to expand welfare and family support, which were promoted by the Moon government. However, in most other areas, the Yoon government pursues different policies than the previous government. In terms of policy, this study demonstrates that the social security policies of the Yoon government is insufficient to solve the welfare blind spot problem that is widespread in the Korean welfare system. Among the Yoon government's policies, there are only a few policies that can eliminate welfare blind spots, such as improving social assistance systems, introducing parental allowance, and promoting sickness allowances.