• 한국안전학회지
• /
• 제32권2호
• /
• pp.46-51
• /
• 2017

This paper presents classification of domestic fire detector using response time index. Response time is measured using fire detector distributed in Korea, and the response time index is estimated. Plunge test prescribed by FM is conducted to measure response time of fire detector. The detector used to test is fixed temperature type(thermistor and bimetal type) and rate of rise temperature type(thermistor and pneumatic type). The nominal operation temperature of fixed temperature type detector is $70^{\circ}C$ and rate of rise temperature is $15^{\circ}C/min$. The fixed temperature type is measured 7 products, and the rate of rise temperature type is measured 5 products. The results show that in case of fixed temperature type(thermistor) is classified "Quick" or "Standard" and fixed temperature type(bimetal) is not classified. The rate of rise temperature type(thermistor) is classified "Fast" or "Ultra Fast" and the rate of rise temperature type(pneumatic) is classified "Very Fast" or "Ultra Fast". The pneumatic type shows more fast response than thermistor type. Also these results indicate the fixed temperature type(bimetal) is not suitable for early stage fire detection.

• 한국CDE학회논문집
• /
• 제10권3호
• /
• pp.188-198
• /
• 2005

There are two types of mold parts, ready-made standard parts and ETO (Engineered-to-Order) parts, the latter are of increasing importance to manufacturers. However, the ETO parts require more engineering support and communication than the ready-made standard parts. Existing e-Catalog modules provide classification structures of products that allow customers to select products based on their needs, and the trade begins with the provided specification. However, machine parts or mold parts have different purchasing patterns. Customers do not purchase the ready-made standard parts offered by an e-Catalog. They usually (1) add own options to the provided specifications or (2) change specification items such as length. To support these trades, a new e-Catalog system is proposed. The proposed system is based on the product design process and the specification selection process in addition to the parts classification structure.

• 펄프종이기술
• /
• 제35권2호
• /
• pp.58-64
• /
• 2003

Because of its tendency of making strong Hoc on the fiber surface with fines, Meogujaengi has not been valued as a material of Hanji. As an attempt to manufacture high value-added products using the material made from Jumchihanji, this study performed morphological classification and chemical component analysis and selection of pulping of Meogujaengi method. As a result, it can be concluded as follows, 1. Meogujaengi is assumed to be a local variety of Broussonetia karinoki and its outward appearance is distinguished from Broussonetia kazinoki. 2. The bast fiber of Meogujaengi is longer and thinner than that of Broussonetia papyrifera or Broussonetia kazinoki. However, because of the coarse linear of fiber tissue, there are many clusters. 3. The cluster phenomenon of Meogujaengi is nothing to do with its chemical components. Although the contents of its chemical components are different from those of Broussonetia kazinoki, no component was found that obstructs pulping. 4. The pretreatment for suppressing the occurrence of clusters of Meogujaengi was effective, and it was necessary to do secondary beating using hollander beater after beating mixed with PAM using knife beater.

• Design & Manufacturing
• /
• 제15권3호
• /
• pp.1-6
• /
• 2021

An artificial neural network model based on a deep learning algorithm is known to be more accurate than humans in image classification, but there is still a limit in the sense that there needs to be a lot of training data that can be called big data. Therefore, various techniques are being studied to build an artificial neural network model with high precision, even with small data. The transfer learning technique is assessed as an excellent alternative. As a result, the purpose of this study is to develop an artificial neural network system that can classify burr images of light guide plate products with 99% accuracy using transfer learning technique. Specifically, for the light guide plate product, 150 images of the normal product and the burr were taken at various angles, heights, positions, etc., respectively. Then, after the preprocessing of images such as thresholding and image augmentation, for a total of 3,300 images were generated. 2,970 images were separated for training, while the remaining 330 images were separated for model accuracy testing. For the transfer learning, a base model was developed using the NASNet-Large model that pre-trained 14 million ImageNet data. According to the final model accuracy test, the 99% accuracy in the image classification for training and test images was confirmed. Consequently, based on the results of this study, it is expected to help develop an integrated AI production management system by training not only the burr but also various defective images.

• 산업경영시스템학회지
• /
• 제45권4호
• /
• pp.86-98
• /
• 2022

Recently, many studies have been conducted to improve quality by applying machine learning models to semiconductor manufacturing process data. However, in the semiconductor manufacturing process, the ratio of good products is much higher than that of defective products, so the problem of data imbalance is serious in terms of machine learning. In addition, since the number of features of data used in machine learning is very large, it is very important to perform machine learning by extracting only important features from among them to increase accuracy and utilization. This study proposes an anomaly detection methodology that can learn excellently despite data imbalance and high-dimensional characteristics of semiconductor process data. The anomaly detection methodology applies the LIME algorithm after applying the SMOTE method and the RFECV method. The proposed methodology analyzes the classification result of the anomaly classification model, detects the cause of the anomaly, and derives a semiconductor process requiring action. The proposed methodology confirmed applicability and feasibility through application of cases.

• Journal of Dairy Science and Biotechnology
• /
• 제25권1호
• /
• pp.47-54
• /
• 2007

The labeling for functionality of food or food components are recognized globally and are essential for the growth of dairy product industry. To expand the functional claims of dairy product especially fermented milk product, the regulations restricting the labeling of functionality should be amended as soon as possible. The labeling regulations of functionality and efficacy for processed and functional foods in related Acts and subordinated statues such as definition, scope of claims, etc. were reviewed and the problems existed in the system were identified. The definition and classification of efficacy for processed food in Food Sanitation Act were analyzed to revise the labeling regulation of dairy product. A draft of amended labeling regulation for the functionality of dairy product is proposed, which provides consumers with appropriate information of beneficial effects on health purpose for human body structure and function. In order to develop the dairy products and related industries, these contents should be considered when the annexed Table 14 in the Enforcement Rule of the Processing of Livestock Products Act is amended.

• 한국식물병리학회:학술대회논문집
• /
• 한국식물병리학회 1998년도 Proceedings of International symposium on Recent Technology of Chemical Control of Plant Diseases
• /
• pp.50-69
• /
• 1998

Dimethomorph is a novel fungicide with a high level of activity against diseases induced by certain Oomycetes, including fungal populations that are resistant to other products. In several ways, this fungicide illustrates the opportunities and challenges presented by many modern pesticides. The specific mode of action, which affects cell wall formation, is associated with a very high level of performance and low dose rates under field conditions. These low dose rates, combined with a low level of toxicity to non-target organisms present an outstanding safety profile. This same highly-specific mode of action, however, limits the spectrum of activity and suggests the need for a resistance management plan, both of which must be addressed in new product development. In addition, the biological and physiochemical properties of this, and other new products are not adequately described by the traditional classification of fungicides into“protectant”and“systemic”types. These unique profiles provide novel and useful products for disease control.

• 대한인간공학회:학술대회논문집
• /
• 대한인간공학회 1997년도 추계학술대회논문집
• /
• pp.372-375
• /
• 1997

Classivication scheme of usability element for the evaluation of consumer electronic product was developed in this study. Using hierachical structuring and clustering methods, usability element of consumer products interface is developed both for the performance perspective and for the image/appeal perspective of a product. Perfoormance element included variables such as simplicity, directness, learn- ability, flexibility, user support and effectiveness. Image/Appeal element included variables such as sensibility, descriptive impression, evaluation of appeal, and attitude towards the product. The classifi- cation scheme developed in this study is found to be comprehensive and robust relative to other existing paradigms. They can be effectively used and applied for the usability evaluation of consumer electronic products.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2013년도 춘계 학술논문 발표대회
• /
• pp.164-165
• /
• 2013

Currently, the process of selecting products such as construction materials and equipment in the design phase proceeds based on the reuse of products used in the previous projects or samples from design companies as well as information on the limited products provided from the website of several manufactures. In particular, it is urgently required to formulate a standard product information management system that can be utilized in the BIM-based design environment these days. In Korea, domestic construction information classification system was developed, but its utilization has been underestimated since continuous studies on the practical usability are still lacking. Therefore, this paper derived problem of domestic construction materials information system by comparing with abroad material information system. Issues and implication is derived in terms of (1) Input information (2) Provide information, and (3) Utilize Information.

• 식품과학과 산업
• /
• 제50권1호
• /
• pp.57-66
• /
• 2017

Home meal replacements (HMR) are kinds of convenient foods as cooked or semi-cooked foods, which are produced outside of home, to eat directly or after simple cooking in substitution for traditional home meals. Recently the market size for HMR is expanding rapidly around the world due to the changes of global consumer trends, growth of single-person household, increase in economic participation of women, aging population, and so on. The Europe takes over 52.4% of the global market share for HMR in global HMR market, and North America, Asia-Pacific, and Africa-Middle East are following. The most popular HMR products in US and Europe are frozen foods, whereas the market share of chilled products in Asia including Japan, South Korea, and Vietnam are much higher than that in US and Europe. Currently, the trends in HMR is focus on the expansion of the list of products that replace for meals with simple cooking, but it is expected that nutrition-enforced HMR product for aged persons or patients who live alone are requested for the further growth of the HMR market.