• Smart Media Journal
• /
• 제13권4호
• /
• pp.48-56
• /
• 2024

In the livestock industry, detecting environmental outliers and predicting data are crucial tasks. Outliers in livestock environment data, typically gathered through time-series methods, can signal rapid changes in the environment and potential unexpected epidemics. Prompt detection and response to these outliers are essential to minimize stress in livestock and reduce economic losses for farmers by early detection of epidemic conditions. This study employs two methods to experiment and compare performances in setting thresholds that define outliers in livestock environment data outlier detection. The first method is an outlier detection using Mean Squared Error (MSE), and the second is an outlier detection using a Dynamic Threshold, which analyzes variability against the average value of previous data to identify outliers. The MSE-based method demonstrated a 94.98% accuracy rate, while the Dynamic Threshold method, which uses standard deviation, showed superior performance with 99.66% accuracy.

• 월간산업보건
• /
• 통권8호
• /
• pp.4-30
• /
• 1988

Examination of working environments was conducted to get more detailed data about harmful working environments and to make a contribution to more effective management. Study was carried out on 722 factories located in Taegu city and eight counties in Kyungpook Province, Korea, for a period of one year, from February 1 to December 30, 1986. The total number and proportion of workers exposed to harmful material was 37,697, 45.2% among 83,368 workers. The results according to exposed material were as follows: 1. In the case of noise, proportion of exceeding the 8-hour TLV was 59%, Included were nail-cutting in assembly metal manufacturing industry and weaving process in textile. 2. Dust in mill process of coal manufacturing industries exceeded the TLV of second class of dust at all parts and exceeded the TLV at 6% as a whole.: 3. The fields of industry lower than 70 lux of illumination were storage equipment of food, auto-winder of textile, painting of wood wares and coal mixing, and 44% of all cases was lower than standard. 4. As a result of temperature index investigation(WBGT), about 12% of all sujects exceeded limit value. Included parts were rolling machine and reducing room. 5. In the case of organic solvents, TLV was exceeded at about 8%, The parts exceeded TLV according to materials belonged to this category were as follows. 1) Toluene: adhesive work in assembly metal manufacturing 2) Xylene: printing and paint mixing in chemical manufacturing 3) Methyl ethytl ketone: paint mixing in all parts examined and coating machine partially in chemical manufacturing 4) Methyl isobutyl ketone: printing in chemical manufacturing 5) Acetone: vapor polishing in assembly metal manufacturing 6. Among specified chemical materials, the concentration of HC1 in the air in metal assembly manufacturing factory exceeded TLV. in one of three assembly metal manufacturing examined. Others, such as benzene, acetic acid, formic acid, sodium hydroxide, formalin, ammonia, copper, chromate etc. were lower than TLV in its indoor atmospheric concentration. As a whole, the proportion of exceeding TLV was about 0.8% 7. The concentrations of inorganic lead were lower than TLV in all parts examined. The results of this investigation show the fact that current management of working environments is not satisfactory, and so more active management is needed.

• Journal of Welding and Joining
• /
• 제30권2호
• /
• pp.59-64
• /
• 2012

Welding is an essential process in the automotive industry. Most welding processes that are used for auto body are spot welding and $CO_2$ welding are used in a small part. In production field, $CO_2$ welding process is decreased and spot welding process is increased due to welding quality is poor and defects are occurred in $CO_2$ welding process frequently. But $CO_2$ welding process should be used at robot interference parts and closed parts where spot welding couldn't. Because of the 0.65mm ~ 2.0mm thickness steel sheet were used in the automotive industry, poor quality of welding area such as burn through and under fill were happened frequently in $CO_2$ process. In this paper, we will study about the penetration depth which gives a huge impact on burn through changing a degree of base metal, welding position and torch angle. Voltage, current and welding speed were fixed but degree of base metal, welding position and torch angle were changed. And Cold- Rolled(CR) steel sheet was used. Penetration depth was analysed by multiple regression analysis to derive approximate calculations. And reliability of approximate calculations were confirmed through additional experiments. As the results of this research, we confirmed the effect of torch and plate angle to bead shape. And we present a possibility that can simulate more accurate to weld geometry, as deduced the verification equations that has tolerance of less than 21.69%.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 한국근적외분광분석학회 2001년도 NIR-2001
• /
• pp.1510-1510
• /
• 2001

NIRS uses reflectance signals resulting from bending and stretching vibrations in chemical bonds between carbon, nitrogen, hydrogen, sulfur and oxygen. These reflectance signals are used to measure the concentration of major chemical composition and other descriptors of homogenized and freeze-dried whole broiler carcasses. Six strains of chicken were analyzed and the NIRS model predictions compared to reference data. The results of this comparison indicate that NIRS is a rapid tool for predicting dry matter (DM), fat, crude protein (CP) and ash content in the broiler carcass. Males and females of six commercial strain crosses of broiler chicken (Gallus domesticus) were used in this study (6$\times$2 factorial design). Each strain was grown to 16 weeks of age, and duplicate serial samples were taken for body composition analysis. Each whole carcass was pressure-cooked, homogenized, and a representative sample was freeze-dried. Body composition determined as follows: DM by oven dried method at 105$^{\circ}C$ for 3 hours, fat by Mojonnier diethyl ether extraction, CP by measuring nitrogen content using an auto-analyzer with Kjeldhal digest and ash by combustion in a muffle furnace for 24 hour at 55$0^{\circ}C$. These homogenized and freeze-dried carcass samples were then scanned with a Foss NIR Systems 6500 visible-NIR spectrophotometer (400-2500nm) (Foss NIR Systems, Silver Spring, MD., US) using Infra-Soft-International, ISI, WinISl software (ISI, Port Matilda, US). The NIRS spectra were analyzed using principal component (PC) analysis. This data was corrected for scatter using standard normal “Variate” and “Detrend” technique. The accuracy of the NIRS calibration equations developed using Partial Least Squares (PLS) for predicting major chemical composition and carcass descriptors- such as body mass (BM), bird dry matter and moisture content was tested using cross validation. Discrimination analysis was also used for sex and strain identification. According to Dr John Shenk, the creator of the ISI software, the calibration equations with the correlation coefficient, $R^2$, between reference data and NIRS predicted results of above 0.90 is excellent and between 0.70 to 0.89 is a good quantifying guideline. The excellent calibration equations for DM ($R^2$= 0.99), fat (0.98) and CP (0.92) and a good quantifying guideline equation for ash (0.80) were developed in this study. The results of cross validation statistics for carcass descriptors, body composition using reference methods, inter-correlation between carcass descriptors and NIRS calibration, and the results of discrimination analysis for sex and strain identification will also be presented in the poster. The NIRS predicted daily gain and calculated daily gain from this experiment, and true daily gain (using data from another experiment with closely related broiler chicken from each of the six strains) will also be discussed in the paper.

• The Journal of the Korea Contents Association
• /
• 제20권10호
• /
• pp.294-307
• /
• 2020

The global automotive industry has suffered various crises such as products defects and unethical company management. In order to examine the effectiveness of crisis communication strategy of an automotive company with Chinese consumers who occupy the largest proportion in the global market, this research analyzed the influences of crisis responsibility, crisis types, and crisis communication strategy on attitudes to a company, intentions of negative communication, and intentions to participate in a boycott. A 2(crisis responsibility: high/low) × 2(crisis types: corporate ability/corporate crisis responsibility) × 2(strategy: defensive/accommodative) experimental design was conducted with 1,600 Chinese students in Seoul. High crisis responsibility and corporate social responsibility crises were related to unfavorable attitudes to a company, higher intentions of negative communication, and higher intentions to participate in a boycott. Crisis responsibility and communication strategy showed a significant interaction. When crisis responsibility was high, the accommodative strategy was more effective than the defensive strategy. When crisis responsibility was low, there was no difference between the strategies. Corporate social responsibility crises found no difference between the strategies regardless of the crisis responsibility level. In the case of corporate ability crises, the accommodative strategy was more effective for the high crisis responsibility crisis.

• Resources Recycling
• /
• 제14권3호
• /
• pp.16-36
• /
• 2005

The quantity of passenger cars in industrial countries has been significantly increased in recent years. According to prognoses, this tendency is likely to continue in the forthcoming future. As a direct consequence, an increase of End-of Life-Vehicles (ELV) will confront us with the problem of "ELV-Recycling". In order to cope with this situation, the European regulation for the treatment of End-of-Life-Vehicles (09/2000) has been transferred to national law in Germany (ELV-Regulation from 1 July 2002). The long term aim is to reduce residues from the ELV-treatment to less than 5 wt% from 30 wt% within the next 10 years (2015). For that reason, there is a need for innovative and more efficient recycling techniques tailored to future materials in automobiles. The design process at automotive industry is continuously changing due to the strong demand on optional equipment and new technical solutions for fuel saving. Light materials, such as aluminum and plastics, consequently become more important and cause a decrease of ferrous metals. Since plastic materials are often used as compounds, a separation into initial material types by means of mechanical recycling methods is not possible. For that reason, efficient recycling can only be realized by introducing recycling-friendly car designs. In the end an integrated approach of auto makers and recycling industry is of decisive significance for the fulfillment of future regulations.

• Proceedings of the Korean Vacuum Society Conference
• /
• 한국진공학회 2012년도 제42회 동계 정기 학술대회 초록집
• /
• pp.241-241
• /
• 2012

Semiconductor industry has been taking the advantage of improvements in process technology in order to maintain reduced device geometries and stringent performance specifications. This results in semiconductor manufacturing processes became hundreds in sequence, it is continuously expected to be increased. This may in turn reduce the yield. With a large amount of investment at stake, this motivates tighter process control and fault diagnosis. The continuous improvement in semiconductor industry demands advancements in process control and monitoring to the same degree. Any fault in the process must be detected and classified with a high degree of precision, and it is desired to be diagnosed if possible. The detected abnormality in the system is then classified to locate the source of the variation. The performance of a fault detection system is directly reflected in the yield. Therefore a highly capable fault detection system is always desirable. In this research, time series modeling of the data from an etch equipment has been investigated for the ultimate purpose of fault diagnosis. The tool data consisted of number of different parameters each being recorded at fixed time points. As the data had been collected for a number of runs, it was not synchronized due to variable delays and offsets in data acquisition system and networks. The data was then synchronized using a variant of Dynamic Time Warping (DTW) algorithm. The AutoRegressive Integrated Moving Average (ARIMA) model was then applied on the synchronized data. The ARIMA model combines both the Autoregressive model and the Moving Average model to relate the present value of the time series to its past values. As the new values of parameters are received from the equipment, the model uses them and the previous ones to provide predictions of one step ahead for each parameter. The statistical comparison of these predictions with the actual values, gives us the each parameter's probability of fault, at each time point and (once a run gets finished) for each run. This work will be extended by applying a suitable probability generating function and combining the probabilities of different parameters using Dempster-Shafer Theory (DST). DST provides a way to combine evidence that is available from different sources and gives a joint degree of belief in a hypothesis. This will give us a combined belief of fault in the process with a high precision.

• The Journal of the Convergence on Culture Technology
• /
• 제6권2호
• /
• pp.437-447
• /
• 2020

The purpose of this study is to research the countermeasures and expected effects through the use of drones in the field of disaster prevention as a drone-based smart quarantine performance method. The environmental, market, and technological approaches to the review of the current quarantine performance task and its countermeasures are as follows. First, in terms of the environment, the effectiveness of the quarantine performance business using drone-based control is to broaden the utilization of forest, bird flu, livestock, facility areas, mosquito larvae, pests, and to simplify and provide various effective prevention systems such as AI and cholera. Second, in terms of market, the standardization of livestock and livestock quarantine laws and regulations according to the use of disinfection and quarantine missions using domestic standardized drones through the introduction of new technologies in the quarantine method, shared growth of related industries and discovery of new markets, and animal disease prevention It brings about the effect of annual budget savings. Third, the technical aspects are (1) on-site application of disinfection and prevention using multi-drone, a new form of animal disease prevention, (2) innovation in the drone industry software field, and (3) diversification of the industry with an integrated drone control / control system applicable to various markets. (4) Big data drone moving path 3D spatial information analysis precise drone traffic information ensures high flight safety, (5) Multiple drones can simultaneously auto-operate and fly, enabling low-cost, high-efficiency system deployment, (6) High precision that this was considered due to the increase in drone users by sector due to the necessity of airplane technology. This study was prepared based on literature surveys and expert opinions, and the future research field needs to prove its effectiveness based on empirical data on drone-based services. The expected effect of this study is to contribute to the active use of drones for disaster prevention work and to establish policies related to them.

• Korean Journal of Plant Resources
• /
• 제25권4호
• /
• pp.473-481
• /
• 2012

This study was conducted to investigate the contents of ginsenosides and physiochemical properties of Panax ginseng after 9 times steaming and drying treatment by using the new auto steamer which is more fast and simple than previous report. In the process of steaming and drying, the content of six major ginsenosides such as Rg1, Re, Rb1, Rc, Rb2 and Rd were gradually decreased. On the other hand, the content of seven minor ginsenosides includes Rh1, 20(S)-Rg2, 20(R)-Rg2, 20(S)-Rg3, 20(R)-Rg3, Rk1 and Rg5 were gradually increased. We observed the protopanxadiol ginsenosides such as Rb1, Rb2, Rc and Rd were converted into 20(S)-Rg3, 20(R)-Rg3, Rk1 and Rg5; similarly protopanxatriol ginsenosides of Rg1 and Re were converted into Rh1, 20(S)-Rg2 and 20(R)-Rg2. Based on the result of fresh ginseng, the contents of reducing sugar, acidic polysaccharide and total phenolic compounds were gradually increased and reached to maximum at 7 times repetitive steaming process of the fresh ginseng. Whereas DPPH radical scavenging activities were gradually decreased to 68% at 7 times steaming. New auto 9 repetitive steaming and drying process has similar production with original methods, but content of benzo(a)pyrene were not almost detected comparatively taking less time. The present results suggested that this method is best for the development of value-added ginseng industry related products.

• Journal of Intelligence and Information Systems
• /
• 제27권3호
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.