• 품질경영학회지
• /
• 제52권1호
• /
• pp.43-56
• /
• 2024

Purpose: The injection molding process, crucial for plastic shaping, encounters difficulties in sustaining product quality when replacing injection machines. Variations in machine types and outputs between different production lines or factories increase the risk of quality deterioration. In response, the study aims to develop a system that optimally adjusts conditions during the replacement of injection machines linked to molds. Methods: Utilizing a dataset of 12 injection process variables and 52 corresponding sensor variables, a predictive model is crafted using Decision Tree, Random Forest, and XGBoost. Model evaluation is conducted using an 80% training data and a 20% test data split. The dependent variable, classified into five characteristics based on temperature and pressure, guides the prediction model. Bayesian optimization, integrated into the selected model, determines optimal values for process variables during the replacement of injection machines. The iterative convergence of sensor prediction values to the optimum range is visually confirmed, aligning them with the target range. Experimental results validate the proposed approach. Results: Post-experiment analysis indicates the superiority of the XGBoost model across all five characteristics, achieving a combined high performance of 0.81 and a Mean Absolute Error (MAE) of 0.77. The study introduces a method for optimizing initial conditions in the injection process during machine replacement, utilizing Bayesian optimization. This streamlined approach reduces both time and costs, thereby enhancing process efficiency. Conclusion: This research contributes practical insights to the optimization literature, offering valuable guidance for industries seeking streamlined and cost-effective methods for machine replacement in injection molding.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1116-1120
• /
• 2004

In the nuclear power plant, there are several cylindrical vessels such as reactor vessel, pressuriser and so on. The vessels are usually constructed by welding large rolled plates, forged sections or nozzle pipes together. In order to assure the integrity of the vessel, these welds should be periodically inspected using sensors such as ultrasonic transducer or visual cameras. This inspection is usually conducted under water to minimize exposure to the radioactively contaminated vessel walls. The inspections have been performed by using a conventional inspection machine with a big structural sturdy column, however, it is so huge and heavy that maintenance and handling of the machine are extremely difficult. It requires much effort to transport the system to the site and also requires continuous use of the utility's polar crane to move the manipulator into the building and then onto the vessel. Setup beside the vessel requires a large volume of work preparation area and several shifts to complete. In order to resolve these problems, we have developed an underwater mobile robot guided by the laser pointer, and performed a series of experiments both in the mockup and in the real reactor vessel. This paper introduces our robotic inspection system and the laser guidance of the mobile robot as well as the results of the functional test.

• 센서학회지
• /
• 제25권4호
• /
• pp.285-290
• /
• 2016

Human sensing is an important but challenging technology. Unlike other methods for sensing humans, a vision sensor has many advantages, and there has been active research in automatic human detection in camera images. The combination of Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) is currently one of the most successful methods in vision-based human detection. However, extracting HOG features from an image is computer intensive, and it is thus hard to employ the HOG method in real-time processing applications. This paper describes an efficient solution to this speed problem of the HOG method. Our method obtains edge information of an image and finds candidate regions where humans very likely exist based on the distribution pattern of the detected edge points. The HOG features are then extracted only from the candidate image regions. Since complex HOG processing is adaptively done by the guidance of the simpler edge detection step, human detection can be performed quickly. Experimental results show that the proposed method is effective in various images.

• 패션비즈니스
• /
• 제27권2호
• /
• pp.108-123
• /
• 2023

Recent increase of eco-conscious trends and pleasure from Do It Yourself (DIY) activities have led to a surge in sales of package products bundling together clothing patterns and raw materials. However, a well-structured market system is yet to be established. We surveyed 460 women with sewing as a hobby who had purchased these DIY clothing pattern packages. The survey revealed that majority of respondents had their hobby for over five years. Choosing the right fabric to match clothing patterns presented a common challenge. Most participants owned a sewing machine and an overlocker, with price being the primary concern when purchasing a package. For guidance during the sewing process, participants preferred print materials featuring real-life images. Those with less sewing experience leaned towards video tutorials. Items of interest or those commonly created included blouses, shirts, and dresses. Desire for further learning in sewing and pattern-making was prominent, with a clear preference for online classes. Several strategies are recommended to enhance the appeal of DIY clothing package products, including broadening range of packages that incorporate fabric, offering supplementary educational resources to improve users' skills, implementing affordable pricing structures, supplying comprehensive creation guidelines, and making available design modification guides. These considerations could significantly boost customer satisfaction. This research intends to lay groundwork for understanding DIY clothing creation market, ultimately fostering production of highly desirable products. Insights of this study will prove instrumental in refining product development and devising effective marketing tactics, leading to a more rewarding consumer experience.

• 한국산업융합학회 논문집
• /
• 제21권5호
• /
• pp.247-255
• /
• 2018

In this research, the effect of normal load, sliding velocity, and texture density on thefriction coefficient of surfaces micro-textured on AISI 4140 under paraffin oil lubrication were investigated. The predicted tribological behavior by numerical calculation can be serves as guidance for the designer during the machine development stage. Therefore, in this research friction coefficient prediction model based on response surface methodology (RSM), support vector machine (SVM), and artificial neural network (ANN) were developed. The experimental result shows that the variation of load, speed and texture density were influence the friction coefficient. The RSM, ANN and SVM model was successfully developed based on the experimental data. The ANN model can effectively predict the tribological characteristics of micro-textured AISI 4140 in paraffin oil lubrication condition compare to RSM and SVM.

• Journal of Radiation Protection and Research
• /
• 제47권1호
• /
• pp.8-15
• /
• 2022

Background: We developed a machine vision technology program that tracks patients' real-time breathing and automatically analyzes their breathing patterns. Materials and Methods: To evaluate its potential for clinical application, the image tracking performance and accuracy of the program were analyzed using a respiratory motion phantom. Changes in the stability and regularity of breathing were observed in healthy adult volunteers according to whether the breathing pattern mirrored the breathing guidance. Results and Discussion: Displacement within a few millimeters was observed in real-time with a clear resolution, and the image tracking ability was excellent. This result was consistent even in the sections where breathing patterns changed rapidly. In addition, the respiratory gating method that reflected the individual breathing patterns improved breathing stability and regularity in all volunteers. Conclusion: The findings of this study suggest that this technology can be used to set the appropriate window and the range of internal target volume by reflecting the patient's breathing pattern during radiotherapy planning. However, further studies in clinical populations are required to validate this technology.

• 인터넷정보학회논문지
• /
• 제24권6호
• /
• pp.49-55
• /
• 2023

본 연구에서는 유튜브 먹방 콘텐츠의 인기를 예측하는 모형을 제안하고 사후 분석을 통하여 먹방 콘텐츠의 인기에 영향을 주는 요인들을 식별하였다. 이를 위해 API와 Pretty Scale을 활용하여 구독자수 상위 먹방 채널들로부터 22,223개 콘텐츠의 정보를 수집하고 Random Forest, XGBoost 및 LGBM 등의 머신러닝 알고리즘을 기반으로 조회수와 좋아요수 예측모델을 구축하였다. SHAP 분석 결과 조회수 예측 모형에서는 구독자수가 예측에 가장 큰 영향을 미치는 반면, 좋아요수 예측 모형에서는 크리에이터의 매력도가 중요변수로 도출되는 등 콘텐츠 조회와 좋아요 반응에 대한 선행요인이 다름을 확인할 수 있었다. 본 연구는 대량의 온라인 콘텐츠를 분석하여 실증 분석을 진행하였다는 점에서 학술적 의의가 있으며 먹방 크리에이터들에게 시청자들의 콘텐츠 소비 경향을 알려주고 상품성 높은 콘텐츠 제작의 가이드를 제공한다는 점에서 실무적인 의의를 지닌다.

• 한국컴퓨터정보학회논문지
• /
• 제28권12호
• /
• pp.23-32
• /
• 2023

현대 사회의 새로운 스포츠로 정의되는 e-스포츠는 세계적으로 많은 사랑을 받는 스포츠로 자리매김했다. 그 중, E-sports를 대표하는 AOS(Aeon of Strife) 장르의 게임은 플레이어 개개인과 팀의 운영이 승패를 좌우하는 요소가 된다는 특징을 가진다. 본 논문은 실제 유저들의 게임 데이터를 수집하고 데이터를 통계적 기법으로 분석하여 정보를 제공한다. 또한, 수집한 데이터를 활용해 머신러닝 기법을 이용하여 승패 예측 모형을 설계하고 실험한다. 5개의 머신러닝 알고리즘이 사용되었고, 평균적으로 개인 데이터 모형에서는 Accuracy 80%, 팀 데이터 모형에서는 Accuracy 95%의 성능을 보인다. 본 연구에서 모형 설계 시 사용된 데이터는 개인 데이터 1,149,950건, 팀 데이터 230,234건으로 규모가 크고 일반 유저들의 플레이 성격을 잘 반영하고 있기 때문에 개발사의 게임 운영이나 일반 유저의 전략 수립 등에 도움이 될 것으로 기대한다. 실험 결과, 개인 데이터 모형과 팀 데이터 모형을 비교하였을 때, 팀 단위 모형의 성능이 상대적으로 매우 좋게 나타났다.

• Computers and Concrete
• /
• 제32권6호
• /
• pp.577-594
• /
• 2023

Engineered cementitious composites with calcined clay limestone cement (LC³-ECC) as a kind of green, low-carbon and high toughness concrete, has recently received significant investigation. However, the complicated relationship between potential influential factors and LC³-ECC compressive strength makes the prediction of LC³-ECC compressive strength difficult. Regarding this, the machine learning-based prediction models for the compressive strength of LC³-ECC concrete is firstly proposed and developed. Models combine three novel meta-heuristic algorithms (golden jackal optimization algorithm, butterfly optimization algorithm and whale optimization algorithm) with support vector regression (SVR) to improve the accuracy of prediction. A new dataset about LC³-ECC compressive strength was integrated based on 156 data from previous studies and used to develop the SVR-based models. Thirteen potential factors affecting the compressive strength of LC³-ECC were comprehensively considered in the model. The results show all hybrid SVR prediction models can reach the Coefficient of determination (R²) above 0.95 for the testing set and 0.97 for the training set. Radar and Taylor plots also show better overall prediction performance of the hybrid SVR models than several traditional machine learning techniques, which confirms the superiority of the three proposed methods. The successful development of this predictive model can provide scientific guidance for LC³-ECC materials and further apply to such low-carbon, sustainable cement-based materials.

• 지질공학
• /
• 제34권1호
• /
• pp.107-115
• /
• 2024

This study assesses forest fire susceptibility in Gangwon-do, South Korea, which hosts the largest forested area in the nation and constitutes ~21% of the country's forested land. With 81% of its terrain forested, Gangwon-do is particularly susceptible to wildfires, as evidenced by the fact that seven out of the ten most extensive wildfires in Korea have occurred in this region, with significant ecological and economic implications. Here, we analyze 480 historical wildfire occurrences in Gangwon-do between 2003 and 2019 using 17 predictor variables of wildfire occurrence. We utilized three machine learning algorithms—random forest, logistic regression, and support vector machine—to construct wildfire susceptibility prediction models and identify the best-performing model for Gangwon-do. Forest and soil map data were integrated as important indicators of wildfire susceptibility and enhanced the precision of the three models in identifying areas at high risk of wildfires. Of the three models examined, the random forest model showed the best predictive performance, with an area-under-the-curve value of 0.936. The findings of this study, especially the maps generated by the models, are expected to offer important guidance to local governments in formulating effective management and conservation strategies. These strategies aim to ensure the sustainable preservation of forest resources and to enhance the well-being of communities situated in areas adjacent to forests. Furthermore, the outcomes of this study are anticipated to contribute to the safeguarding of forest resources and biodiversity and to the development of comprehensive plans for forest resource protection, biodiversity conservation, and environmental management.