목적 : 본 연구는 20대가 선호하는 안경집 디자인의 핵심 요소를 도출하고자 실시되었다.
방법 : 서울지역 거주 대학생 집단을 대상으로 3단계에 걸쳐 설문조사를 시행하였다. 다양한 안경집 디자인의 예제들을 수집하고 선호도 조사를 진행한 다음, 도출된 안경집 디자인을 대상으로 SD법을 바탕으로 하는 감성평가 및 요인분석을 실시하여 그 결과를 분석하였다.
결과 : 분석 결과 고유값 1을 기준으로 8개의 감성어휘가 도출되어 2개의 축으로 분류되었다. 선호도 조사에서 상위를 차지한 안경집 디자인들에서는 실용적인, 단순한, 가벼운, 단단한, 깔끔한 등의 감성언어에 대응되는 제 1 요인과 고급스러운, 둥근, 무광의 등의 감성언어에 대응되는 제 2요인이 높게 나타났으며 반대로 하위를 차지한 안경집 디자인에서는 제 1요인과 제 2요인이 낮게 나타났다.
결론 : 20대 소비자 집단은 기능성과 심미성을 주요소로 하여 안경집 디자인을 선택하는 경향이 존재하는 것으 \로 나타났다.
목적: 본 연구는 연령증가에 따른 안경굴절검안에 있어서 조절반응과 조절지체 간의 변화량의 영향 관계를 검정하는데 있다.
방법 : 연령증가에 따른 원점과 근점의 자극에 대응하는 조절 반응점 (ARf, ARn)을 연구하기 위해서, 선조 검영기를 사용하여 정시 74개 (남: 37, 여: 37)의 단안을 대상으로 굴절검사를 진행하였다. MEM 0.4 시력카드의 스넬린 (Snellen chart) E 단일문자를 검영기헤드 (No 18235, WelchAllyn, USA) 전면에 부착한 선조 검영기를 사용하여 조절 반응을 확인하였다.
결과 : 연령증가에 따른 원점 조절반응의 경우Ⅰ>Ⅱ, Ⅲ, Ⅳ의 유의한 결과를 보였고, 나머지 근점 조절반응과 조절지체는 Ⅰ, Ⅱ, Ⅲ<Ⅳ의 결과를 보여주었다. 연령증가에 따른 원점 조절반응과 조절지체 간의 통계적 조절효 과는 Ⅰ·Ⅱ〉Ⅰ·Ⅲ〉Ⅰ·Ⅳ의 유의한 결과를 보였다.
결론 : 40대 이후 연령의 증가에 따른 시축상의 원점 조절반응의 절댓값과 조절지체는 각각 감소하고 증가함에 따라 원거리 안경굴절검사를 위한 안와 내 조절반응의 기준점을 달리 적용하여야 한다.
본 논문은 필리핀 로컬 스페셜티 커피에 대한 소비자의 선호도를 밝히고 일반커피 대비 스페셜티 커피에 대해 소비자들이 부여하는 가치를 추정하는 것에 목적이 있다. 특히 소비자 잉여를 극대화하는 스페셜티 커피의 최적 가격을 분석하여 커피 농가를 포함한 생산자, 투자자들의 이해를 높이고자 하였다. 주 요 연구 결과는 다음과 같다. 1. 응답자들의 연령 및 성별에 따라 소비하는 커피의 형태가 다르게 나타남. 그러나 공통적으로 주로 아침에, 집에서, 습관적으로 커피를 섭취함. 가장 선호하는 커피의 맛은 단맛과 쓴맛임. 대부분의 소비자들은 양조 커피, 압착 커피를 소비함. 2. 고품질의 원두를 엄선하여 만들어지는 스페셜티 등급의 커피 한 잔에 대한 소비자의 평균 지불의사가격은 일반 커피 (42페소) 대비 271% 높은 156페소(약 3.2달러)로 분석됨. 연령이 어릴수록, 학력과 소득이 높을수록, 스페셜티 커피에 대한 사전적 인지도가 높을수록, 향후 스페셜티 커피에 대한 구매 의사가 강할수록 스페셜티 커피에 보다 높은 지불의사가격을 표현함. 3. 로컬 스페셜티 커피는 원두 구매, 가공 및 로스팅, 등급 평가 비용이 발생하여 적절한 가격 인상이 요구됨. 기존 일반 커피 가격대비 40% 인상까지 소비자 후생이 증가하는 것으로 나타나는 반면, 스페셜티 커피 가격이 일반커피 가격대비 50% 이상 인상되는 경우에는 소비자들의 후생이 감소하는 것으로 나타남. 4. 필리핀 소비자들이 지역에서 생산된 스페셜티 커피에 대한 높은 선호를 갖고 있다는 점을 고려하면 현지 스페셜티 커피에 대한 대대적인 홍보가 로컬 스페셜티 커피에 대한 인지도를 높이고 시장 확대로 이어질 수 있음. 이는 필리핀 커피 로드맵, 필리핀 농무부의 지속 가능한 발전 계획 목표 달성에 기여할 것임.
프리지아 ‘Sunny Gold’는 농촌진흥청 국립원예특작과학원 에서 2010년 노랑색 반겹꽃 프리지아 육성계통 ‘036010’을 모본으로 진노란색 홑꽃 ‘Golden Flame’을 부본으로 교배하여 획득한 종자로부터 2011년 진노란색 겹꽃의 향기가 강한 프리지아 계통을 선발하여 품종화 하였다. 2011년부터 2016년까지 개화 생육특성검을 수행하였으며 핵심수요자의 기호도 평가를 통해 선발되어 2017년 ‘Sunny Gold’ 로 명명되었다. ‘Sunny Gold’는 RHS color chart YO17B의 노란색 겹꽃 프리지아 품종으로 화폭은 6.7cm로 대조품종 ‘Golden Flame’ 6.1cm에 비해 크고, 분지수는 6.5로 다수확성 품종이다. 초장이 101.9cm로 초세가 강하다. ‘Sunny Gold’의 소화수 및 소화장은 각각 13.0개, 9.3cm이며 개화소요일수는 137.7일이다. 이 품종의 절화수명은 약 9일이며 자구번식력은 5.3배로 대조 품종 ‘Golden Flame’ 4.3배에 비해 우수하다. 전자코를 이용한 PCA분석결과 PC1과 PC2는 각각 99.3%와 0.6%로 전체 변이량의 99.9%를 반영하고 있다. Rader plot 분석결과 총 6개 센서에서 모두 ‘Sunny Gold’의 센서값이 향기가 강한 상용품종 ‘Yvonne’의 값에 비해 높게 나타나 ‘Sunny Gold’의 향기가 더 강한 것으로 나타났다.
본 연구는 차광과 온도가 분화 국화의 생육 및 개화에 미치는 영향을 조사하기 위하여, 차광(무차광, 30% 차광, 50% 차광)과 온도(주간/야간 28/20℃, 32/23℃, 36/26℃)가 조절된 인공기상챔버에서 자연단일상태로 재배된 ‘오렌지에그’의 생육 및 개화특성을 비교하였다. 그 결과, 30% 차광 조건에서 분화 국화 ‘오렌지에그’의 초장은 36/26℃에서 24.6cm로 가장 길었으며, 32/23℃에서는 23.0cm, 28/20℃에서는 19.6cm로 생육온도가 증가할수록 초장 신장이 촉진되었으나, 차광정도에 대해서는 유의적인 차이를 보이지 않았다. 분화 국화 ‘오렌지에그’의 개화반응 역시 주간/야간 온도 28/20℃에서는 단일 처리 23.4일 후에 발뢰하여 49.1일만에 100% 개화되었으나, 32/23℃와 36/26℃로 생육온도가 증가할수록 28/20℃에 비해 발뢰가 각각 4.1일과 11.4일, 개화는 각각 8.2일과 16.1일 지연되었으나, 차광정도에 대해서는 유의적인 차이를 보이지 않았다. 이와 같이 분화 국화 ‘오렌지에그’의 생육 및 개화반응에 미치는 영향은 차광에 비해 고온으로 인한 피해가 심각하게 나타날 수 있는 만큼 고품질의 분화 국화를 생산하기 위해서는 생산 시기별 온도 관리가 매우 중요하며, 특히 하계 고온기에는 적극적으로 차광하여 재배온실의 온도를 적정 생육온도에 가깝게 유지하는 것이 필요할 것으로 생각된다.
The objective of this study was to prove the effect of pig slurry application with charcoal on nitrogen use efficiency (NUE), feed value and ammonia (NH3) emission from maize forage. The four treatments were applied: 1) non-pig slurry (only water as a control), 2) only pig slurry application (PS), 3) pig slurry application with large particle charcoal (LC), 4) pig slurry application with small particle charcoal (SC). The pig slurry was applied at a rate of 150 kg N ha-1, and the charcoal was applied at a rate of 300 kg ha-1 regardless of the size. To determine the feed value of maize, crude protein, dry matter intake, digestible dry matter, total digestible nutrient, and relative feed value were investigated. All feed value was increased by charcoal treatment compared to water and PS treatment. Also, the NUE for plant N was significantly higher in charcoal treatments (LC and SC) compared to PS treatment. On the other hand, there is no significant difference for feed value and NUE between LC and SC. The NH3 emission was significantly reduced 15.2% and 27.9% by LC and SC, respectively, compared to PS. Especially, SC significantly decreased NH3 emission by 15% compared to LC. The present study clearly showed that charcoal application exhibited positive potential in nitrogen use efficiency, feed value and reducing N losses through NH3 emission.
The present study was aimed to estimate the effect of ensiling period and bacterial inoculants on chemical compositions and fermentation characteristics on rye silage harvested at delayed stage. Rye (Secale cereale L.) was harvested after 20 days of heading stage (29.4% dry matter, DM). The harvested rye forage was applied with different inoculants following: applications of distilled water (CON), Lactobacillus brevis (LBB), Leuconostoc holzapfelii (LCH), or mixture of LBB and LCH at 1:1 ratio (MIX). Each forage was ensiled into 20 L mini bucket silo (5 kg) for 50 (E50D) and 100 (E100D) days in triplicates. The E50D silages had higher in vitro digestibilities of DM (IVDMD, p<0.001) and neutral detergent fiber (IVNDFD, p=0.013), and lactate (p=0.009), and acetate (p=0.011) than those of E100D, but lower pH, lactic acid bacteria (LAB), and yeast. By inoculant application, LCH had highest IVDMD and IVNDFD (p<0.05), while MIX had highest lactate and lowest pH (p<0.05). The CON and LCH in E50D had highest LAB and yeast (p<0.05), whereas LBB in E100D had lowest (p<0.05). Therefore, this study concluded that LCH application improved the nutrient digesbility (IVDMD and IVNDFD) of lignified rye silage, and longer ensiling period for 100 days enhanced the fermentation characteristics of silage compared to ensiling for 50 days.
Cymbidium is one of the most popular and economically important species cultivated as a commercial ornamental crop. The objectives of this study were to determine the appropriate electrical conductivity (EC) treatments of nutrient solution, which gives the highest spike production and quality. Three-year-old Cymbidium ‘Lovely Smile’ plants were grown in the environmentally controlled Information and Communication Technology (ICT) smart greenhouse at Seoul Women’s University. The EC of the nutrient solution was changed in three distinct stages: vegetative, flower initiation, and flower development. The EC treatments were 1-0-1 (dS·m-1, EC101), 1-1-1 (dS·m-1, EC111), 2-1-2 (dS·m-1, EC212), 2-2-2 (dS·m-1, EC222), 3-2-3 (dS·m-1, EC323), 3-3-3 (dS·m-1, EC333) and the pH was adjusted to 6.0–6.5. Pseudobulb diameter increased in the plants treated with EC 101 and EC111 compared to the plants treated with EC 2.0–3.0 dS·m-1 at the reproductive stage 28 weeks after nutrient solution treatment. Flower spike production per pot and pseudobulb showed the highest values in the plants treated with EC111 of 3.3 and 1.4, respectively. Flower spikes length was the highest in the plants treated with EC 1.0 dS·m-1 and stem thickness, number of flowers, and fresh weight were the largest in the plants with EC 1.0 dS·m-1 among the EC treatments. Flower spikes had the worst quality (e.g., plant growth and flowering quality) in the plants treated with EC 3.0 dS·m-1 among the EC treatments. Floral bud and flower development took place 1–2 weeks earlier in the plants treated with EC 101, 111, and 212 than the other treatments. Flower diameter showed the highest values in the plants treated with EC 1.0 dS·m-1 among the EC treatments and flower color showed higher L* and b* values and lower a* values in the plants treated with EC 3.0 dS·m-1 compared to EC 1.0 and 2.0 dS・m-1. Nutrient solution of EC 1.0 dS·m-1 (EC111) can be recommended to improve flower spike quality and advanced flower development of Cymbidium.
In supply chain, there are a variety of different uncertainties including demand, service time, lead time, and so forth. The uncertainty of demand has been commonly studied by researchers or practitioners in the field of supply chain. However, the uncertainty of upstream supply chain has also increased. A problem of uncertainty in the upstream supply chain is the fluctuation of the lead time. The stochastic lead time sometimes causes to happen so called the order crossover which is not the same sequences of the order placed and the order arrived. When the order crossover happens, ordinary inventory policies have difficult to find the optimal inventory solutions. In this research, we investigate the lead time distribution in case of the order crossover and explore the resolutions of the inventory solution with the order crossover.
Culturally, it is wonderful that they try to integrated their ordinary life on the earth into heavenly spiritual life through their funeral rituals and song so called Yeong-Jang-Sori as same context as in the Pacific. During life, they eat traditional bread ( Bing-Teok) and drink traditional liquor (O-Me-Gi Ssol) and wear persimmon-dyeing Gal-ot healthily and they meet each other in the heaven eternally through happy dying. We are evidently sensitive to the unique cultural heritage of the island. UNESCO’s mission is the promotion of peace amongst peoples and its cultural heritage is also dedicated to working with indigenous peoples’ on their eco-friendlly knowledge of a sustainable planet. As Jeju Island is inscribed in part as both a World Heritage Site and a UNESCO Biosphere, it is important to give due consideration to the natural and cultural heritage of the island, as seen and experienced by the islanders. As such, if there is something useful we can discuss, for example the recognition of the special relationship between Jeju islanders and sacred natural sites on the island.
As Deepfakes phenomenon is spreading worldwide mainly through videos in web platforms and it is urgent to address the issue on time. More recently, researchers have extensively discussed deepfake video datasets. However, it has been pointed out that the existing Deepfake datasets do not properly reflect the potential threat and realism due to various limitations. Although there is a need for research that establishes an agreed-upon concept for high-quality datasets or suggests evaluation criterion, there are still handful studies which examined it to-date. Therefore, this study focused on the development of the evaluation criterion for the Deepfake video dataset. In this study, the fitness of the Deepfake dataset was presented and evaluation criterions were derived through the review of previous studies. AHP structuralization and analysis were performed to advance the evaluation criterion. The results showed that Facial Expression, Validation, and Data Characteristics are important determinants of data quality. This is interpreted as a result that reflects the importance of minimizing defects and presenting results based on scientific methods when evaluating quality. This study has implications in that it suggests the fitness and evaluation criterion of the Deepfake dataset. Since the evaluation criterion presented in this study was derived based on the items considered in previous studies, it is thought that all evaluation criterions will be effective for quality improvement. It is also expected to be used as criteria for selecting an appropriate deefake dataset or as a reference for designing a Deepfake data benchmark. This study could not apply the presented evaluation criterion to existing Deepfake datasets. In future research, the proposed evaluation criterion will be applied to existing datasets to evaluate the strengths and weaknesses of each dataset, and to consider what implications there will be when used in Deepfake research.
This study attempted to interpret the meaning of the U.S.-China hegemony competition based on liberal theory, and predicted future U.S.-China relations by analyzing the hegemony confrontation between the U.S. and China. During the U.S.-China hegemony competition, countries around China, which are directly threatened by China, were divided into continental and maritime national groups, and the characteristics, concerns, and responses of those countries were considered. It proposed a new maritime-new continent global bridge economic block system, suggested strengthening economic cooperation, technical cooperation, cultural cooperation and liberal international order as its effectiveness, and suggested the need of a new international order during the Fourth Industrial Revolution.
Oral squamous cell carcinoma (OSCC) metastasis is characterized by distant metastasis and local recurrence. Combined chemotherapy with cisplatin and 5-fluorouracil is routinely used to treat patients with OSCC, and the combined use of gefitinib with cytotoxic drugs has been reported to enhance the sensitivity of cancer cells in vitro . However, the development of drug resistance because of prolonged chemotherapy is inevitable, leading to a poor prognosis. Therefore, understanding alterations in signaling pathways and gene expression is crucial for overcoming the development of drug resistance. However, the altered characterization of Ca2+ signaling in drug-resistant OSCC cells remains unclear. In this study, we investigated alterations in intracellular Ca2+ ([Ca2+]i) mobilization upon the development of gefitinib resistance in human tongue squamous carcinoma cell line (HSC)-3 and HSC-4 using ratiometric analysis. This study demonstrated the presence of altered epidermal growth factor- and purinergic agonist-mediated [Ca2+]i mobilization in gefitinib-resistant OSCC cells. Moreover, Ca2+ content in the endoplasmic reticulum, store-operated calcium entry, and lysosomal Ca2+ release through the transient receptor potential mucolipin 1, were confirmed to be significantly reduced upon the development of apoptosis resistance. Consistent with [Ca2+]i mobilization, we identified modified expression levels of Ca2+ signaling-related genes in gefitinib-resistant cells. Taken together, we propose that the regulation of [Ca2+]i mobilization and related gene expression can be a new strategy to overcome drug resistance in patients with cancer.
Maritime monitoring requirements have been beyond human operators capabilities due to the broadness of the coverage area and the variety of monitoring activities, e.g. illegal migration, or security threats by foreign warships. Abnormal vessel movement can be defined as an unreasonable movement deviation from the usual trajectory, speed, or other traffic parameters. Detection of the abnormal vessel movement requires the operators not only to pay short-term attention but also to have long-term trajectory trace ability. Recent advances in deep learning have shown the potential of deep learning techniques to discover hidden and more complex relations that often lie in low dimensional latent spaces. In this paper, we propose a deep autoencoder-based clustering model for automatic detection of vessel movement anomaly to assist monitoring operators to take actions on the vessel for more investigation. We first generate gridded trajectory images by mapping the raw vessel trajectories into two dimensional matrix. Based on the gridded image input, we test the proposed model along with the other deep autoencoder-based models for the abnormal trajectory data generated through rotation and speed variation from normal trajectories. We show that the proposed model improves detection accuracy for the generated abnormal trajectories compared to the other models.
Nypa fruticans Wurmb (NFW) contains a large amount of phenolic acid and flavonoids, and is popular as a superfood in Myanmar. NFW has various biological activities, such as anti-inflammatory, anti-oxidant, and neuroprotective properties; however, the anti-cancer effect of NFW have not been reported. In this study, we investigated the anticancer activity of water extracts of NFW (WeNFW) and the underlying mechanism in human FaDu hypopharyngeal squamous carcinoma cells. The WeNFW inhibited FaDu cell growth in a dose-dependent manner without affecting normal cells (L929), as determined by an MTT assay and Live and Dead assay. In addition, the concentrations of WeNFW without cytotoxicity (0.025, 0.05, and 0.1 mg/mL) inhibited wound healing and colony formation. Furthermore, WeNFW significantly induced apoptosis through the proteolytic cleavage of caspase-3 and -9, poly (ADP-ribose) polymerase, and downregulation of Bcl-2 and upregulation of Bax in FaDu cells, as determined by DAPI staining, FACS analysis, and western blot analysis. Taken together, these results suggest that WeNFW exhibits potent anti-cancer effects by suppressing the growth of oral cancer cells, wound healing and colony formation activity. Via mitrochondrial-dependent apoptotic pathways in human FaDu hypopharyngeal squamous carcinoma cells. Therefore, WeNFW can provide a natural chemotherapeutic drug for oral cancer in humans.
COVID-19 has been spreading all around the world, and threatening global health. In this situation, identifying and isolating infected individuals rapidly has been one of the most important measures to contain the epidemic. However, the standard diagnosis procedure with RT-PCR (Reverse Transcriptase Polymerase Chain Reaction) is costly and time-consuming. For this reason, pooled testing for COVID-19 has been proposed from the early stage of the COVID-19 pandemic to reduce the cost and time of identifying the COVID-19 infection. For pooled testing, how many samples are tested in group is the most significant factor to the performance of the test system. When the arrivals of test requirements and the test time are stochastic, batch-service queueing models have been utilized for the analysis of pooled-testing systems. However, most of them do not consider the false-negative test results of pooled testing in their performance analysis. For the COVID-19 RT-PCR test, there is a small but certain possibility of false-negative test results, and the group-test size affects not only the time and cost of pooled testing, but also the false-negative rate of pooled testing, which is a significant concern to public health authorities. In this study, we analyze the performance of COVID-19 pooled-testing systems with false-negative test results. To do this, we first formulate the COVID-19 pooled-testing systems with false negatives as a batch-service queuing model, and then obtain the performance measures such as the expected number of test requirements in the system, the expected number of RP-PCR tests for a test sample, the false-negative group-test rate, and the total cost per unit time, using the queueing analysis. We also present a numerical example to demonstrate the applicability of our analysis, and draw a couple of implications for COVID-19 pooled testing.
본 논문은 종교적 언어가 신을 나타낼 수 있음을 보여주고자 한다. 종교적 체험과 종교적 언어의 관계성에서, 중요한 주제들 중의 하나는 신적 체험을 전달 하는데 사용되는 언어가 인간의 조건과 존재론적으로 명확히 구별되는 신을 나타낼 수 있는가의 문제를 다루는 것이다. 신과 세계의 사이에는 차이 혹은 간격이 있다. 왜냐하면 초월적 존재로서 신은 인간이 이해할 수 없는 존재의 측면을 가지고 있기 때문이다. 신에 대한 이해불가능성은 아마도 인간 언어가 지니고 있는 한계점들을 암시하고 있을 것이다. 즉 인간 언어는 인간이 이해할 수 없는 것을 표현할 수 없는 한계점이 있다. 그러나 세계가 신을 나타낼 수 있는 중요 한 영역들 중의 하나는 바로 경험의 세계이다. 한 개인은 바로 경험의 역동성을 통해서 신을 만나게 된다. 신에 대한 인간의 체험을 전달하는 것은 바로 인간의 언어이다. 그리고 이러한 인간의 언어는 현존하는 것뿐만 아니라 부재하는 것도 드러낼 수 있다. 따라서 본 논문은 어떻게 종교적 언어가 신을 나타낼 수 있는 가에 대해서 보여주고자 한다. 전통과 공동체의 중요성, 종교적 언어의 본질, 그리고 종교적 언어의 변화적 힘 등이 논의될 것이다.
The 17th APEC Future Education Forum (AFEF) and the 19th International ALCoB Conference were held from 28 September (Tuesday) to 1 October (Friday) in a virtual manner. The annual forum and conference serve as an invaluable opportunity to discuss future directions of education and human resources development in the Asia-Pacific region. Since 2005, the forum and conference have served their role as the largest and longest thematic meeting in the APEC Human Resources Development Working Group (HRDWG). The 2021 forum and conference have gathered 780 participants from 20 APEC member economies, guest economies and international institutions. Under the theme “Strengthening lifelong competencies and skills development for individuals’ career, education, training, and life cycles”, the forum and conference reflected the current educational agenda from 21 APEC member economies and other international organizations such as G20, UNESCO. As a way forward, the forum and conference speakers recommend APEC to 1) consider individual and industrial needs for curriculum development, 2) expand lifelong learning policies and practices at both domestic and international level and 3) develop a public-private partnership to prepare its future talents with adequate competencies. This review explains the 17th AFEF and the 19th International ALCoB Conference background, summary and outcomes. The review also briefs on collective actions member economies can take as future steps to continue the forum and conference discussions.
This article suggests the machine learning model, i.e., classifier, for predicting the production quality of free-machining 303-series stainless steel(STS303) small rolling wire rods according to the operating condition of the manufacturing process. For the development of the classifier, manufacturing data for 37 operating variables were collected from the manufacturing execution system(MES) of Company S, and the 12 types of derived variables were generated based on literature review and interviews with field experts. This research was performed with data preprocessing, exploratory data analysis, feature selection, machine learning modeling, and the evaluation of alternative models. In the preprocessing stage, missing values and outliers are removed, and oversampling using SMOTE(Synthetic oversampling technique) to resolve data imbalance. Features are selected by variable importance of LASSO(Least absolute shrinkage and selection operator) regression, extreme gradient boosting(XGBoost), and random forest models. Finally, logistic regression, support vector machine(SVM), random forest, and XGBoost are developed as a classifier to predict the adequate or defective products with new operating conditions. The optimal hyper-parameters for each model are investigated by the grid search and random search methods based on k-fold cross-validation. As a result of the experiment, XGBoost showed relatively high predictive performance compared to other models with an accuracy of 0.9929, specificity of 0.9372, F1-score of 0.9963, and logarithmic loss of 0.0209. The classifier developed in this study is expected to improve productivity by enabling effective management of the manufacturing process for the STS303 small rolling wire rods.
Recently, the importance of preventive maintenance has been emerging since failures in a complex system are automatically detected due to the development of artificial intelligence techniques and sensor technology. Therefore, prognostic and health management (PHM) is being actively studied, and prediction of the remaining useful life (RUL) of the system is being one of the most important tasks. A lot of researches has been conducted to predict the RUL. Deep learning models have been developed to improve prediction performance, but studies on identifying the importance of features are not carried out. It is very meaningful to extract and interpret features that affect failures while improving the predictive accuracy of RUL is important. In this paper, a total of six popular deep learning models were employed to predict the RUL, and identified important variables for each model through SHAP (Shapley Additive explanations) that one of the explainable artificial intelligence (XAI). Moreover, the fluctuations and trends of prediction performance according to the number of variables were identified. This paper can suggest the possibility of explainability of various deep learning models, and the application of XAI can be demonstrated. Also, through this proposed method, it is expected that the possibility of utilizing SHAP as a feature selection method.