It is important to measure the performance of group project but also very important to evaluate the contribution of individual members fairly. The degree of contribution of group members can be assessed by pair-wise comparison method of the Analytic Hierarchy Process. The degree of contribution of group members can be biased in a way that is advantageous to evaluator oneself during the pair-wise comparison process. In this paper, we will examine whether there is a difference in the contribution weight vectors obtained when including evaluator and excluding oneself in the pair-wise comparison. To do this, the experimental data was obtained by making pair-wise comparison in two ways for 15 5-person groups that perform term projects in university classes and 15 pairs of weight vectors for contribution were obtained. The results of the nonparametric test for these 15 pairs of weights vectors are given.

이 연구는 코르크보드를 보강하여 건축부재 및 놀이기구의 안전부재 등으로 폭넓게 활용할 것을 목적으로 코르크보드의 중층에 금속, 유리섬유, 탄소섬유를 삽입하여 보강한 3종의 코르크복합보드를 제조하였고, 코르크복합보드의 수분흡수에 따른 치수안정성 및 접착층 박리성능을 조사하였다. 코르크복합보드의 흡수율은 0.37% - 0.45%의 범위에 있었고, 코르크보드에 비해 0.61배 - 0.74배의 낮은 값을 나타내었다. 코르크복합보드의 두께팽창률은 0.92% - 1.58%의 범위에 있었고, 코르크보드 보다 1.4 - 2.4배의 높은 값을 나타내었다. 그러나 이 값들은 일반 목질보드보다 현저히 낮았고, KS규격의 12%이하를 하회하는 것이 확인되었다. 코르크복합보드의 준내수 및 내수침지박리시험후의 접착층박리율은 0%로 전혀 접착층의 박리가 일어나지 않아 우수한 내수성을 나타내었고, 흡수율과 흡수두께팽창률은 상온침지에 비해 다소 증가하였으나, 목질보드에 관한 KS규격을 하회하는 우수한 치수안정성을 나타내는 것이 확인되었다.

Lightweight face recognition models, as one of the most popular and long-standing topics in the field of computer vision, has achieved vigorous development and has been widely used in many real-world applications due to fewer number of parameters, lower floating-point operations, and smaller model size. However, few surveys reviewed lightweight models and reimplemented these lightweight models by using the same calculating resource and training dataset. In this survey article, we present a comprehensive review about the recent research advances on the end-to-end efficient lightweight face recognition models and reimplement several of the most popular models. To start with, we introduce the overview of face recognition with lightweight models. Then, based on the construction of models, we categorize the lightweight models into: (1) artificially designing lightweight FR models, (2) pruned models to face recognition, (3) efficient automatic neural network architecture design based on neural architecture searching, (4) Knowledge distillation and (5) low-rank decomposition. As an example, we also introduce the SqueezeFaceNet and EfficientFaceNet by pruning SqueezeNet and EfficientNet. Additionally, we reimplement and present a detailed performance comparison of different lightweight models on the nine different test benchmarks. At last, the challenges and future works are provided. There are three main contributions in our survey: firstly, the categorized lightweight models can be conveniently identified so that we can explore new lightweight models for face recognition; secondly, the comprehensive performance comparisons are carried out so that ones can choose models when a state-of-the-art end-to-end face recognition system is deployed on mobile devices; thirdly, the challenges and future trends are stated to inspire our future works.

Recently, unmanned logistics delivery systems, such as UAV (Unmanned Aerial Vehicle, written as drone below) and autonomous robot delivery systems, have been implemented in many countries due to the rapid development of autonomous driving technology. The development of these new types of advanced unmanned logistics delivery systems is essential not only to become a leading logistics company but also to secure national competitiveness. In this paper, the application of the unmanned logistics delivery system was investigated in terms of market trends, overall technology level of last mile delivery drone and autonomous delivery robot. The direction of response to changes in the last mile delivery service market was checked through a comparison of the technological level between domestic companies that produce last mile devices and advanced foreign companies. As a result of this technology level analysis, the difference between domestic companies and advanced companies was shown using tables and figures to show their relative levels. The results of this analysis reflect the opinions of experts in the field of last-mile delivery technology. In addition, the technology level of unmanned logistics delivery systems for each country was analyzed based on the number of related technology patents. Lastly, insights for the technology level analysis of unmanned last mile delivery systems were proposed as a conclusion.

Predictive maintenance has been one of important applications of data science technology that creates a predictive model by collecting numerous data related to management targeted equipment. It does not predict equipment failure with just one or two signs, but quantifies and models numerous symptoms and historical data of actual failure. Statistical methods were used a lot in the past as this predictive maintenance method, but recently, many machine learning-based methods have been proposed. Such proposed machine learning-based methods are preferable in that they show more accurate prediction performance. However, with the exception of some learning models such as decision tree-based models, it is very difficult to explicitly know the structure of learning models (Black-Box Model) and to explain to what extent certain attributes (features or variables) of the learning model affected the prediction results. To overcome this problem, a recently proposed study is an explainable artificial intelligence (AI). It is a methodology that makes it easy for users to understand and trust the results of machine learning-based learning models. In this paper, we propose an explainable AI method to further enhance the explanatory power of the existing learning model by targeting the previously proposedpredictive model [5] that learned data from a core facility (Hyper Compressor) of a domestic chemical plant that produces polyethylene. The ensemble prediction model, which is a black box model, wasconverted to a white box model using the Explainable AI. The proposed methodology explains the direction of control for the major features in the failure prediction results through the Explainable AI. Through this methodology, it is possible to flexibly replace the timing of maintenance of the machine and supply and demand of parts, and to improve the efficiency of the facility operation through proper pre-control.

There have been a lot of studies in the past for the method of predicting the failure of a machine, and recently, a lot of researches and applications have been generated to diagnose the physical condition of the machine and the parts and to calculate the remaining life through various methods. Survival models are also used to predict plant failures based on past anomaly cycles. In particular, special machine that reflect the fluid flow and process characteristics of chemical plants are connected to hundreds or thousands of sensors, so there are not many factors that need to be considered, such as process and material data as well as application of derivative variables. In this paper, the data were preprocessed through time series anomaly detection based on unsupervised learning to predict the abnormalities of these special machine. Next, clustering results reflecting clustering-based data characteristics were applied to produce additional variables, and a learning data set was created based on the history of past facility abnormalities. Finally, the prediction methodology based on the supervised learning algorithm was applied, and the model update was confirmed to improve the accuracy of the prediction of facility failure. Through this, it is expected to improve the efficiency of facility operation by flexibly replacing the maintenance time and parts supply and demand by predicting abnormalities of machine and extracting key factors.

Recently, a number of researchers have produced research and reports in order to forecast more exactly air quality such as particulate matter and odor. However, such research mainly focuses on the atmospheric diffusion models that have been used for the air quality prediction in environmental engineering area. Even though it has various merits, it has some limitation in that it uses very limited spatial attributes such as geographical attributes. Thus, we propose the new approach to forecast an air quality using a deep learning based ensemble model combining temporal and spatial predictor. The temporal predictor employs the RNN LSTM and the spatial predictor is based on the geographically weighted regression model. The ensemble model also uses the RNN LSTM that combines two models with stacking structure. The ensemble model is capable of inferring the air quality of the areas without air quality monitoring station, and even forecasting future air quality. We installed the IoT sensors measuring PM2.5, PM10, H2S, NH3, VOC at the 8 stations in Jeonju in order to gather air quality data. The numerical results showed that our new model has very exact prediction capability with comparison to the real measured data. It implies that the spatial attributes should be considered to more exact air quality prediction.

K1A1 tank commander’s primary thermal sight is a device that enables tank commanders to detect, identify, aim and track the target by observing targets in all directions during day, night and in situations of smokescreen and fog through 360° rotation independent from the gunner’s primary thermal sight and stabilizing the line of sight even under the vibrations occurring when the tank is standstill and moving. The main function of this device is to detect and process visible and thermal images and deliver the final images to the tank commander. One of the core parts to that end is the observation window (daytime/ thermal image window). This core part is mounted at the entrance of the optical path for observing the target and plays the role of making visible light during the daytime and infrared light during the night pass through the target and transmitting the resultant images to the internal optical system of the tank commander’s primary thermal sight. Such core parts have been selected as depot maintenance items so that they are replaced by new parts instead of being recycled when they are subjected to maintenance in most cases. That is, the military budget is wasted because such parts are replaced by new parts despite that they can be recycled for maintenance. Therefore, this study proposed a mounting tool for polishing and coating observation windows (daytime and thermal image window) using planar polishing equipment and DLC (Diamond-Like Carbon) coating equipment. In addition, this study presented an amendment (proposal) of the Depot Maintenance Work Request (DMWR) already published to verify the performance of recycled products including the establishment of inspection standards for recycling processes.

K-1계열 전차의 전차장 열상조준경은 주야간 및 연막, 안개 등의 상황에서 포수조준경과 독립적으로 360°회전을 통한 전 방향의 표적 관측과 전차가 정지 및 기동 간 발생하는 진동에서도 조준선을 안정화하여 전차장이 표적의 감지, 식별, 조준 및 추적 할 수 있는 장치이다. 이 장치 의 주요기능 중 하나인 가시상 및 열상을 감지하고 처리하여 최종 영상을 전차장에게 전달하는 것으로 이를 위한 핵심 부품은 주간 및 열상 창이다. 이 핵심 부품은 목표물을 관측하는 광행 로 입구에 장착되어 있으며, 목표물에 대해서 주간에는 가시광, 야간에는 적외선을 통과하여 전 차장 열상조준경의 내부 광학계통으로 전달하는 기능을 수행한다. 이와 같은 핵심부품에 대한 정비는 창 정비 품목으로 선정되어 대부분 재생정비가 아닌 신품 교환 정비를 하고 있는 실정 이다. 즉, 재생정비가 가능한 품목임에도 불구하고 신품교환에 따라 군 예산이 낭비되고 있다. 따라서 본 연구는 평면연마장비와 DLC(diamond-like carbon) 코딩장비를 활용하여 주간 및 열상 창을 연마·코팅할 수 있도록 장착치구를 개발하였다. 또한 재생공정에 대한 검사기준 정립을 포함하여 재생품에 대한 성능검증을 위해 기 발간된 창 정비작업요구서(DMWR) 수정(안)을 제시 하였다.

The mortality rate in industrial accidents in South Korea was 11 per 100,000 workers in 2015. It’s five times higher than the OECD average. Economic losses due to industrial accidents continue to grow, reaching 19 trillion won much more than natural disaster losses equivalent to 1.1 trillion won. It requires fundamental changes according to industrial safety management. In this study, We classified the risk of accidents in industrial complex of Ulju-gun using spatial analytics and data mining. We collected 119 data on accident data, factory characteristics data, company information such as sales amount, capital stock, building information, weather information, official land price, etc. Through the pre-processing and data convergence process, the analysis dataset was constructed. Then we conducted geographically weighted regression with spatial factors affecting fire incidents and calculated the risk of fire accidents with analytical model for combining Boosting and CART (Classification and Regression Tree). We drew the main factors that affect the fire accident. The drawn main factors are deterioration of buildings, capital stock, employee number, officially assessed land price and height of building. Finally the predicted accident rates were divided into four class (risk category-alert, hazard, caution, and attention) with Jenks Natural Breaks Classification. It is divided by seeking to minimize each class’s average deviation from the class mean, while maximizing each class’s deviation from the means of the other groups. As the analysis results were also visualized on maps, the danger zone can be intuitively checked. It is judged to be available in different policy decisions for different types, such as those used by different types of risk ratings.

Recently, the material industry in the world has started appreciating the value of new materials that can overcome the limitation of steel material. In particular, new materials are expected to play a very important role in the future industry, demonstrating superior performance compared to steel in lightweight materials and ability to maintain in high temperature environments. Carbon materials have recently increased in value due to excellent physical properties such as high strength and ultra lightweight compared to steel. However, they have not overcome the limitation of productivity and price. The carbon materials are classified into various composites depending on the purpose of use and the performance required. Typical composites include carbon-glass, carbon-carbon, and carbon-plastic composites. Among them, carbon-carbon composite technology is a necessary technology in aviation and space, and can be manufactured with high investment cost and technology. In this paper, in order to find the optimal conditions to achieve productivity improvement and cost reduction of carbon material densification process, the correlation between each process parameters and results of densification is first analyzed. The main process parameters of the densification process are selected by analyzing the correlation results. And then a certain linear relationship between major process variables and density of carbon materials is derived by performing a regression analysis based on the historical production result data. Using the derived casualty, the optimal management range of major process variables is suggested. Effective process operation through optimal management of variables will have a great effect on productivity improvement and manufacturing cost reduction by shortening the lead time.

본 연구에서는 약용식물 추출물을 함유한 숙취해소 음료가 항산화 및 알코올 분해에 미치는 영향을 조사하기 위하여, 약용식물 추출물의 총 페놀성화합물 함량, 총 플라보노이드 함량 및 DPPH radical 소거활성 및 ALDH(aldehyde dehydrogenase) 활성을 측정하였고, 약용식물 추출물을 함유한 숙취해소 음료의 복용이 음주 후 호흡 중 알코올 농도에 미치는 영향을 조사하였다. 항산화활성은 빼빼목 추출물 에서 높은 수치(total phenolic compound 97.5.6mg/g, total flavonoid 306.5mg/g, DPPH radical 소거능 80.1%)를 나타냈으며 Acetaldehyde(ALDH)에 대한 활성은 헛개 및 울금 추출물에서 306% 및 292%로 가장 높은 활성을 나타내어, 헛개 및 울금 추출물이 ALDH 제거에 효과적인 것으로 나타났다. 헛개 및 울금 추출물이 함유된 숙취해소 음료를 이용하여, 임상실험을 실시한 결과, 시간경과에 따라 대조군에 비해 숙취해소음료 섭취군에서 알코올 농도가 감소되는 경향을 나타냈으며 흡연자에 비해 비 흡연자가, 남성에 비해 여성에게 숙취해소 음료의 섭취가 더 효과적인 것으로 확인되었다.

Korea`s industrial death rate is 13 percent in 2015. It’s five times higher than the OECD average. Economic losses due to industrial accidents continue to grow, reaching 19 trillion won in natural disaster losses equivalent to 1.1 trillion won, requiring fundamental changes in industrial safety levels. In this study, We classified the risk of accidents in industrial complex of Ulju-gun using spacial analysis and decision tree methodologies. We draw the main factors that affect the accident and developed the four risk category(alert, hazard, caution, and attention). It is judged to be available in different policy decisions for different types, such as those used by different types of risk ratings, targeted education, and technical support.

본 연구는 대추나무 생산농가의 합리적인 생산 계획수립을 지원하는 방안으로 2015년 9월부터 10월 사이에 경북 경산시 과원에서 5~30년생 대추나무의 수령별 과실특성, 수체특성 및 생산량을 조사하였 다. 대추나무 수령에 따른 과실 특성에 있어서, 과실 중량, 과실종경, 과실횡경은 5~25년생에서는 유의 성을 나타내지 않았으나, 30년생에서 감소된 수치를 나타냈고, 과형지수는 수령이 증가할수록 감소되는 경향을 나타냈다. 그러나, 과실당도는 23.28~25.46°Brix 범위로 수령에 따른 유의성이 나타나지 않았 다. 대추나무 수령에 따른 수체특성에 있어서, 수고, 근원경, 수관 폭 및 수관면적은 10년생까지 유의성 을 나타내며 생장하였으나, 10년생부터 30년생까지는 유의성을 나타내지 않았다. 생대추 생산량은 25년 생부터 유의하게 증가하는 경향을 나타내어, 대추나무의 수령이 생대추 생산량에 영향을 미치는 것을 확인하였다. 대추나무 수령이 생대추 생산량에 영향을 미치는 인자를 비교하였으며, 조사된 수체특성 인자에 대하여 상관계수를 도출하였다. 그 결과 생대추 생산량에 가장 큰 영향을 미치는 수체특성 인자 는 수관폭으로 나타났다.

Compound logistics is a service aimed to enhance logistics efficiency by supporting that shippers and consigners jointly use logistics facilities. Many of these services have taken place both domestically and internationally, but the joint logistics services for e-commerce have not been spread yet, since the number of the parcels that the consigners transact business is usually small. As one of meaningful ways to improve utilization of compound logistics, we propose a brokerage service for shipper and consigners based on the hybrid recommendation system using very well-known classification and clustering methods. The existing recommendation system has drawn a relatively low satisfaction as it brought about one-to-one matches between consignors and logistics vendors in that such matching constrains choice range of the users to one-to-one matching each other. However, the implemented hybrid recommendation system based brokerage agent service system can provide multiple choice options to mutual users with descending ranks, which is a result of the recommendation considering transaction preferences of the users. In addition, we applied feature selection methods in order to avoid inducing a meaningless large size recommendation model and reduce a simple model. Finally, we implemented the hybrid recommendation system based brokerage agent service system that shippers and consigners can join, which is the system having capability previously described functions such as feature selection and recommendation. As a result, it turns out that the proposed hybrid recommendation based brokerage service system showed the enhanced efficiency with respect to logistics management, compared to the existing one by reporting two round simulation results.