• 산업경영시스템학회지
• /
• 제47권1호
• /
• pp.9-19
• /
• 2024

Deep learning-based computer vision anomaly detection algorithms are widely utilized in various fields. Especially in the manufacturing industry, the difficulty in collecting abnormal data compared to normal data, and the challenge of defining all potential abnormalities in advance, have led to an increasing demand for unsupervised learning methods that rely on normal data. In this study, we conducted a comparative analysis of deep learning-based unsupervised learning algorithms that define and detect abnormalities that can occur when transparent contact lenses are immersed in liquid solution. We validated and applied the unsupervised learning algorithms used in this study to the existing anomaly detection benchmark dataset, MvTecAD. The existing anomaly detection benchmark dataset primarily consists of solid objects, whereas in our study, we compared unsupervised learning-based algorithms in experiments judging the shape and presence of lenses submerged in liquid. Among the algorithms analyzed, EfficientAD showed an AUROC and F1-score of 0.97 in image-level tests. However, the F1-score decreased to 0.18 in pixel-level tests, making it challenging to determine the locations where abnormalities occurred. Despite this, EfficientAD demonstrated excellent performance in image-level tests classifying normal and abnormal instances, suggesting that with the collection and training of large-scale data in real industrial settings, it is expected to exhibit even better performance.

• 대한용접접합학회:학술대회논문집
• /
• 대한용접접합학회 2002년도 Proceedings of the International Welding/Joining Conference-Korea
• /
• pp.366-371
• /
• 2002

Because of the environmental problems, automotive companies are trying to reduce the weight of car body. Therefore, TRIP(TRansformation Induced Plasticity) steels, which have high strength and ductility have been developed. Welding process is a complex process; therefore deciding the optimal welding conditions on the basis of experimental data is an effective method. However, trial-and-error method to decide the optimal conditions requires too many experiments. To overcome these problems, response surface methodology was used. Response surface methodology is a collection of mathematical and statistical techniques that are used in the modeling and analysis of problems in which a response of interest is influenced by several variables and the objective is to optimize this response. This method was applied to the resistance spot welding process of the TRIP steel to optimize the welding parameters.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2015년도 춘계 학술논문 발표대회
• /
• pp.17-18
• /
• 2015

In manufacturing industry, image scanning technique has made enormous progress in past decades. 3D models have been also very important to continuously monitor the related spatial information for freeform buildings. The process of shape making of 3D scanning is as follows: mesh surface segmentation, NURBS surface generation, and parametric solid model generation. We will review the process and applying process. Especially in the construction industry, 3D data collection by laser scanning has become an high quality 3D models. Therefore, in this research, we have an effort to review construction of reverse design process for freeform envelope using 3D scanning. The technology enables many 3D shape engineering and design parameterization of reverse engineering in the construction site.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2015년도 추계학술발표대회
• /
• pp.899-902
• /
• 2015

제조업체의 품질관리 및 공정에 대한 개선을 위하여 현장에서 생산되는 데이터를 수집 및 분석하여 작업자 또는 경영자에게 의미 있는 의사결정을 할 수 있는 정보 제공이 필요하다. 본 논문에서는 제조 공정 개선을 위한 데이터 수집 및 분석 시스템을 제안한다. 이를 통해서 소규모 제조 현장에서 원자재 배합의 정확도 향상과 작업시간 단축을 가능하게 해 주고, 건당 처리시간의 감소로 인한 생산 납품시간 단축, 불량률 감소 등에 따른 원가 절감을 가져올 수 있다.

• International Journal of Korean Welding Society
• /
• 제2권2호
• /
• pp.47-50
• /
• 2002

Because of the environmental problems, automotive companies are trying to reduce the weight of car body. Therefore, TRIP(TRansformation Induced Plasticity) steels, which have high strength and ductility have been developed. Welding process is a complex process; therefore deciding the optimal welding conditions on the basis of experimental data is an effective method. However, trial-and-error method to decide the optimal conditions requires too many experiments. To overcome these problems, response surface methodology was used. Response surface methodology is a collection of mathematical and statistical techniques that are used in the modeling and analysis of problems in which a response of interest is influenced by several variables and the objective is to optimize this response. This method was applied to the resistance spot welding process of the TRIP steel to optimize the welding parameters.

• 문화기술의 융합
• /
• 제9권6호
• /
• pp.1001-1006
• /
• 2023

전 세계적인 에너지 비용 상승과 에너지 수요 증가는 공급망 전반에 걸쳐 상당한 양의 에너지를 소비하는 식품가공·제조업에 있어 중요한 문제이다. 이에 에너지 사용을 최적화할 수 있는 실시간 에너지 모니터링 및 분석 시스템의 개발이 필요하다. 본 연구에서는 식품공장에 설치된 IoT를 기반으로 각 설비별 모니터링과 열처리 공정에 관한 에너지 공급·사용량 모니터링, 검색 기능이 포함된 식품 공장 에너지 모니터링 시스템이 제안되었다. 본 시스템은 식품가공공장의 IoT 센서를 기반으로 PLC, 데이터베이스 서버, OPC-UA서버, UI 서버 및 API 서버와 CIMON 사의 HMI 등으로 구성된다. 시스템은 식품공장의 빅데이터 구축과 함께 수집 기능을 통한 설비별 모니터링과 열처리 공정에 관한 에너지 공급·사용량 모니터링 및 검색 서비스 기능을 제공한다. 본 데이터 수집 기반 에너지 모니터링 시스템은 에너지 절감을 위한 중소기업형 공장 에너지 모니터링·관리 시스템 개발을 위한 가이던스 역할이 될 것이다. 향후, 본 시스템을 활용하여 에너지 사용량 파악 및 분석을 통한 공정 작업이 최적화된 정량적인 에너지 절감 방안이 마련될 수 있다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.15-19
• /
• 2001

A dedicated reverse engineering(RE) system for rapid manufacturing of human head in a 3D bust has been developed. The first step in the process is to capture the surface details of a human head and shoulder by three scanners based upon the digital moire fringe technique. Then the multiple scans captured from different angles are aligned and merged into a single polygonal mesh, and the aligned data set is refined by smoothing, subdividing or hole filling process. Finally, the refined data set is sent to a 4-axis computer numerically control(NC) machine to manufacture a replica. In this paper, we mainly describe on the algorithms and software for aligning multiple data sets. The method is based on the recently popular Iterative Closest Point(ICP) algorithm that aligns different polygonal meshes into one common coordinate system. The ICP algorithm finds the nearest positions on one scan to a collection of points on the other scan by minimizing the collective distance between different scans. We also integrate some heuristics into the ICP to enhance the aligning process. A typical example is presented to validate the system and further research work is also discussed.

• 유통과학연구
• /
• 제20권7호
• /
• pp.97-105
• /
• 2022

Purpose: This study aims to analyze and prove the effect of logistic capability and Total Quality Management practices on Micro, Small and Medium Enterprises (SME) performance directly or mediated by non-financial performance. Research design, data and methodology: This study tested the hypothesis using Hierarchical multiple regression analysis, the method of data collection in this study was using questionnaire, the sampling technique was purposive sampling technique, with SME that has been established for more than 5 years and manufacturing. The data analyzed were 180 respondents using SPSS 25. Results: The findings showed that logistic capability has direct and indirect effects on SME financial performance and has a positive effect on SME financial performance mediated by non-financial performance. While the total quality management practices have a positive effect on SME financial performance mediated by non-financial performance. Thus, companies can achieve maximum financial performance if they invest in developing employee knowledge and concerning on non-financial actions, such as employee satisfaction, innovation and proactively seeking market opportunities. Conclusions: In conclusion, one of the main factors that companies need to consider to improve financial performance is non-financial performance in mediating the effect of logistic capability and TQM practices on the financial performance of SMEs.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.92-93
• /
• 2022

스마트 팩토리의 중요성이 강조됨에 따라, 스마트 팩토리 구축을 위해 산업용 Ethernet 기반 장치의 활용이 증가할 것으로 전망되고 있다. PROFINET은 SIEMENS사에서 개발한 산업용 이더넷 프로토콜이며, 현재 다수의 스마트 팩토리가 PROFINET 기반 제품으로 구축되고 있는 실정이다. 이에 따라 산업용 사물인터넷(Industrial IoT) 기반의 다양한 서비스 개발 및 활용을 위해 PROFINET 기반의 제조 장비에서 각종 센서 데이터 및 정보를 수집하여 엣지 컴퓨터로 데이터를 전송할 수 있는 IIoT 디바이스가 필요하며, IIoT의 주요 메시징 프로토콜인 MQTT(Message Queuing Telemetry Transport)를 활용한 데이터 수집 IIoT 디바이스 개발 방안을 제시하고자 한다.

• 한국의상디자인학회지
• /
• 제13권3호
• /
• pp.89-103
• /
• 2011

The purpose of this study is to analyze the formative characteristics of accordion pleats shown in 10/11 S/S London, Paris, New York collection and evaluate the differences of visual image by variations in the width of pleats and the length of the accordion pleats skirt. The stimuli are 21 samples: 7 variations of the width of pleats and 3 variations of the length of skirt. The data has been obtained from 43 fashion design majors. The data has been analyzed by Factor Analysis, Anova, Scheffe's Test. The results of the study are as follows: Accordion pleats was first introduced by Marriano Fortuny. He published delphos dress by creating silk pleats manufacturing technique and it was recognized as the best pleats work. Since then, pleats with synthetic fiber characteristics were used by many fashion designers. Accordion pleats were used in various items including one-peice, skirt and pants in collection of 2010 and 2011. The design of accordion pleats skirt used different width of pleats, length of the skirt and materials. The visual image by the width of pleats and the length of skirt are composed of 3 factors : dearness, attractiveness and elegance. In these factors, dearness factor is estimated by the most important factor. The accordion p1eats skirt is most attractive and dear when the width of pleats is the shortest, 1/8", and the width of 4/8" and 5/8" follows. The accordion pleats skirt is most attractive and dear when the length of skirt is similar to that of mini-skirt, and the skirt was estimated to be elegant as the skirt gets longer. When the length of skirt is in between, the skirts were estimated to be ambiguous in all visual images. The interaction effect of visual image according to the changes in width of pleats and skirt length of accordion pleats skirt was seemed to have significant differences by factors, and the main effect had great differences in all factors.