• 산업융합연구
• /
• 제2권2호
• /
• pp.47-68
• /
• 2004

The term "social marketing" was first introduced in 1971 to describe the use of marketing principles and techniques to advance a social cause, idea, or behavior. Social Marketing is a strategy for changing behavior. It utilizes concepts of market segmentation, consumer research, product concept development and testing, directed communication, facilitation, incentives, and exchange theory to maximize the target adopters' response. Social marketing requires knowledge of each target-adopter group, including its 1. social demographic characteristics, 2. psychological profile and 3. behavioral characteristics. To know the target adopters in these three related ways enables social marketer to make more accurate predictions. In addition to differentiating among and selecting target adopter groups, the social marketer will identify influence-holding groups, or influentials, who can affect a program's success. Great changes and opportunities exist to produce changes in the ways that individuals and groups think and behave and in meeting human needs. The balance of the scales of social change, we hope, will shift away from the use of force and violence to the use of persuasion and voluntary action. I trust that this dissertation will be useful highlighting the strategies and means of peaceful, planned social change designed to elevate the quality of life.

• Tribology and Lubricants
• /
• 제39권3호
• /
• pp.118-122
• /
• 2023

In our study, we develop a finite element model based on Archard's wear law to predict the cumulative wear and the evolution of the tool profile in friction stir welding (FSW) applications. Our model considers the rotational and translational behaviors of the tool, providing a comprehensive description of the wear process. We validate the accuracy of our model by comparing it against experimental results, examining both the predicted cumulative wear and the resulting changes to the tool profile caused by wear. We perform a detailed comparison between the predictions of the model and experimental data by manipulating non-dimensional coefficients comprising model parameters, such as element sizes and time increments. This comparison facilitates the identification of a specific non-dimensional coefficient condition that best replicates the experimentally observed cumulative wear. We also directly compare the worn tool profiles predicted by the model using this specific non-dimensional coefficient condition with the profiles obtained from wear experiments. Through this process, we identify the model settings that yield a tool wear profile closely aligning with the experimental results. Our research demonstrates that carefully selecting non-dimensional coefficients can significantly enhance the predictive accuracy of finite element models for tool wear in FSW processes. The results from our study hold potential implications for enhancing tool longevity and welding quality in industrial applications.

• Journal of the Korean Data and Information Science Society
• /
• 제22권4호
• /
• pp.755-764
• /
• 2011

기업은 종종 과거의 필드 고장 데이터를 이용하여 미래에 필드에서 고장이 얼마나 일어날 것인지 예측한다. 특히 이런 예측은 필드에서 예기치 않던 고장모드 (failure mode)가 발견될 때 더욱 하고 싶어진다. 왜냐하면 기업은 이런 예측을 통해 미래에 품질보증 비용이 얼마나 될 것인지 파악하고, 고장 난 부품을 재빨리 수리하는데 필요한 여유 부품의 수를 파악하고 싶기 때문이다. 본 연구에서는 기업에서 생길 수 있는 요약 데이터를 사용하여 미래 필드에서 고장이 얼마나 발생할 것인지 예측하고, 이런 요약 데이터이외에 또 어떤 데이터가 생길 수 있으며 이때 분석결과가 어떻게 나올 수 있는지 알아본다.

• 한국게임학회 논문지
• /
• 제22권1호
• /
• pp.33-42
• /
• 2022

온라인 게임에서 다른 플레이어와의 상호작용은 플레이어의 만족도에 영향을 미친다. 따라서, 비슷한 수준의 플레이어를 매치시켜 원활한 상호작용을 도모하는 것은 플레이어의 게임 경험을 위해 중요하다. 그러나, 게임의 최종승패로만 플레이어의 평가점수를 증감시키는 현재의 평가 방식으로는 신규 및 복귀 플레이어의 원활한 매칭이 불가능하다. 본 연구에서는 스타크래프트II의 리플레이를 활용하여 매치메이킹 개선을 위한 기계학습 활용방안을 제시한다. 매치메이킹의 기준이 되는 플레이어의 MMR 점수를 예측하는 기계학습 모델을 생성하고 성능을 평가하였다. 모델의 오차는 리그 평균 MMR 점수 범위의 40.4% 수준으로, 제안된 방식을 통해서 플레이어를 실력과 근접한 리그에 즉시 배치할 수 있음을 확인하였다. 또한, 결과에 대한 플레이어의 수용도를 높일 수 있도록 예측의 근거를 도출하는 방안도 제시되었다.

• 대한토목학회논문집
• /
• 제9권2호
• /
• pp.115-123
• /
• 1989

종래의 탄성 고체법을 이용하여 말뚝의 부마찰력을 산정하는 전산프로그램을 개발하였다. 그리고 한계 상대 변위 개념을 도입하여 이 전산프로그램의 수정을 시도하였다. 끝으로 한계 상대변위로서 말뚝과 흙사이의 미끄러짐 발생 여부를 결정하고 모아-쿨롱의 파괴 방정식을 이용하여 부마찰력을 산정하는 단순 전이함수법을 개발하였다. 이 세가지 방법에 의한 결과는 모두 현장 측정치와 잘 일치하였다. 그러나, 이들이 원심력을 이용한 모형실험 결과를 예측할 때에는 각기 다른 결과를 나타내었다. 종합적으로 보면, 이 논문에서 제안한 단순전이 함수법이 말뚝 부마찰력 산정시 그 결과의 정확성과 계산상의 효율성들을 고려할 때 가장 능률적이라고 판단된다.

• 한국컴퓨터정보학회논문지
• /
• 제12권6호
• /
• pp.89-96
• /
• 2007

유한고장수를 가진 비동질적인 포아송 과정에 기초한 모형들에서 잔존 오류 1개당 고장 발생률은 일반적으로 상수, 혹은 단조증가 및 단조 감소 추세를 가지고 있다. 소프트웨어 제품의 정확한 인도시기를 예측하거나 효용성 및 신뢰성을 예측하기 위해서는 소프트웨어 테스팅 과정에서 중요한 요소인 테스트 커버리지를 이용하면 보다 효율적인 테스팅 작업을 할 수 있다. 본 논문에서는 기존의 소프트웨어 신뢰성 모형인 지수 커버리지 모형과 S-커버리지 모형을 적용하고 이 분야에 적용 될 수 있는 변형 커버리지 모형(중첩모형 및 혼합모형) 비교 문제를 제안하였다. 고장 간격시간으로 구성된 자료를 이용한 모수추정 방법은 최우추정법과 수치해석 방법인 이분법을 사용하여 모수 추정을 실시하고 효율적인 모형 선택은 편차자승합(SSE)을 이용하였다.

• Environmental Engineering Research
• /
• 제24권3호
• /
• pp.404-411
• /
• 2019

In this study, we designed a data-driven model to predict chlorophyll-a using M5P model tree and extreme learning machine (ELM). The Juksan weir in the Youngsan River has high chlorophyll-a, which is the primary indicator of algal bloom every year. Short-term algal bloom prediction is important for environmental management and ecological assessment. Two models were developed and evaluated for short-term algal bloom prediction. M5P is a classification and regression-analysis-based method, and ELM is a feed-forward neural network with fast learning using the least square estimate for regression. The dataset used in this study includes water temperature, rainfall, solar radiation, total nitrogen, total phosphorus, N/P ratio, and chlorophyll-a, which were collected on a daily basis from January 2013 to December 2016. The M5P model showed that the prediction model after one day had the highest performance power and dropped off rapidly starting with predictions after three days. Comparing the performance power of the ELM model with the M5P model, it was found that the performance power of the 1-7 d chlorophyll-a prediction model was higher. Moreover, in a period of rapidly increasing algal blooms, the ELM model showed higher accuracy than the M5P model.

• Nuclear Engineering and Technology
• /
• 제54권1호
• /
• pp.30-35
• /
• 2022

In a nuclear reactor, gamma radiation is the dominant energy deposition in non-fuel regions. Heat is generated upon gamma deposition and consequently affects the mechanical and thermal structure of the material. Therefore, the safety of samples should be carefully considered so that their integrity and quality can be retained. To evaluate relevant parameters, an in-core gamma thermometer (GT) was used to measure gamma heating (GH) throughout the operation of the McMaster nuclear reactor (MNR) at four irradiation sites. Additionally, a Monte Carlo reactor physics code (Serpent-2) was utilized to model the MNR with the GT located in the same irradiation sites used in the measurement to verify its predictions against measured GH. This research aids in the development of modeling, calculation, and prediction of the GH utilizing Serpent-2 as well as implementing a new GH measurement at the MNR core. After all uncertainties were quantified for both approaches, comparable GH profiles were observed between the measurements and calculations. In addition, the GH values found in the four sites represent a strong level of radiation based on the distance of the sample from the core. In this study, the maximum and minimum GH values were found at 0.32 ± 0.05 W/g and 0.15 ± 0.02 W/g, respectively, corresponding to 320 Sv/s and 150 Sv/s. These values are crucial to be considered whenever sample is planned to be irradiated inside the MNR core.

• 대한산업공학회지
• /
• 제38권1호
• /
• pp.46-56
• /
• 2012

The purpose of virtual metrology (VM) in semiconductor manufacturing is to predict every wafer's metrological values based on its process equipment data without an actual metrology. In this paper, we propose novelty detection-based reliability estimation models for VM in order to support flexible utilization of VM results. Because the proposed model can not only estimate the reliability of VM, but also identify suspicious process variables lowering the reliability, quality control actions can be taken selectively based on the reliance level and its causes. Based on the preliminary experimental results with actual semiconductor manufacturing process data, our models can successfully give a high reliance level to the wafers with small prediction errors and a low reliance level to the wafers with large prediction errors. In addition, our proposed model can give more detailed information by identifying the critical process variables and their relative impacts on the low reliability.

• 한국정보시스템학회지:정보시스템연구
• /
• 제32권4호
• /
• pp.149-162
• /
• 2023

Purpose In the textile industry, fabric defects significantly impact product quality and consumer satisfaction. This research seeks to enhance defect detection by developing a transformer-based deep learning image segmentation model for learning high-dimensional image features, overcoming the limitations of traditional image classification methods. Design/methodology/approach This study utilizes the ZJU-Leaper dataset to develop a model for detecting defects in fabrics. The ZJU-Leaper dataset includes defects such as presses, stains, warps, and scratches across various fabric patterns. The dataset was built using the defect labeling and image files from ZJU-Leaper, and experiments were conducted with deep learning image segmentation models including Deeplabv3, SegformerB0, SegformerB1, and Dinov2. Findings The experimental results of this study indicate that the SegformerB1 model achieved the highest performance with an mIOU of 83.61% and a Pixel F1 Score of 81.84%. The SegformerB1 model excelled in sensitivity for detecting fabric defect areas compared to other models. Detailed analysis of its inferences showed accurate predictions of diverse defects, such as stains and fine scratches, within intricated fabric designs.