Distribution and logistics industries contribute some of the biggest GDP(gross domestic product) in South Korea and the number of related companies are quarter of the total number of industries in the country. The number of retail tech companies are quickly increased due to the acceleration of the online and untact shopping trend. Furthermore, major distribution and logistics companies try to achieve integrated data management with the fulfillment process. In contrast, small and medium distribution companies still lack of the capacity and ability to develop digital innovation and smartization. Therefore, in this paper, a deep learning-based demand forecasting & recommendation model is proposed to improve business competitiveness. The proposed model is developed based on real sales transaction data to predict future demand for each product. The proposed model consists of six deep learning models, which are MLP(multi-layers perception), CNN(convolution neural network), RNN(recurrent neural network), LSTM(long short term memory), Conv1D-BiLSTM(convolution-long short term memory) for demand forecasting and collaborative filtering for the recommendation. Each model provides the best prediction result for each product and recommendation model can recommend best sales product among companies own sales list as well as competitor’s item list. The proposed demand forecasting model is expected to improve the competitiveness of the small and medium-sized distribution and logistics industry.
A machine learning-based algorithms have used for constructing species distribution models (SDMs), but their performances depend on the selection of backgrounds. This study attempted to develop a noble method for selecting backgrounds in machine-learning SDMs. Two machine-learning based SDMs (MaxEnt, and Random Forest) were employed with an example species (Spodoptera litura), and different background selection methods (random sampling, biased sampling, and ensemble sampling by using CLIMEX) were tested with multiple performance metrics (TSS, Kappa, F1-score). As a result, the model with ensemble sampling predicted the widest occurrence areas with the highest performance, suggesting the potential application of the developed method for enhancing a machine-learning SDM.
Our research team has developed a gamma ray detector which can be distributed over large area through air transport. Multiple detectors (9 devices per 1 set) are distributed to measure environmental radiation, and information such as the activity and location of the radiation source can be inferred using the measured data. Generally, radiation is usually measured by pointing the detector towards the radioactive sources for efficient measurement. However, the detector developed in this study is placed on the ground by dropping from the drone. Thus, it does not always face toward the radiation source. Also, since it is a remote measurement system, the user cannot know the angle information between the source and detector. Without the angle information, it is impossible to correct the measured value. The most problematic feature is when the backside of the detector (opposite of the scintillator) faces the radiation source. It was confirmed that the measurement value decreased by approximately 50% when the backside of the detector was facing towards the radiation source. To calibrate the measured value, we need the information that can indicate which part of the detector (front, side, back) faces the source. Therefore, in this study, we installed a small gamma sensor on the backside of the detector to find the direction of the detector. Since this sensor has different measurement specifications from the main sensor in terms of the area, type, efficiency and measurement method, the measured values between the two sensors are different. Therefore, we only extract approximate direction using the variation in the measured value ratio of the two sensors. In this study, to verify the applicability of the detector structure and measurement method, the ratio of measured values that change according to the direction of the source was investigated through MCNP simulation. The radioactive source was Cs-137, and the simulation was performed while moving in a semicircular shape with 15 degree steps from 0 degree to 180 degrees at a distance of 20 cm from the center point of the main sensor. Since the MCNP result indicates the probability of generating a pulse for one photon, this value was calculated based on 88.6 μCi to obtain an actual count. Through the ratio of the count values of the two sensors, it was determined whether the radioactive source was located in the front, side, or back of the probe.
The distribution characteristics of rock fractures determine the hydro-mechanical behavior of natural barriers. Rock fractures are defined by various parameters, which are analyzed as the probability distribution from observation results by surveying the exposed rock surface or borehole. The size is known to have the most uncertainty among the fracture parameters because it cannot be directly measured. Therefore, various estimation methods have been proposed for fracture size distribution using the fracture traces observable on the rock surface. However, most methods are based on a planar survey area, limiting their applicability to the underground research laboratory (URL) excavated in the form of tunnels. This study aims to review a method that can be applied to estimate the size distribution of fractures in deep rock masses at the URL site. The estimation method using the joint center volume (JCV) has recently been extended to be applicable regardless of the geometry of the survey area, which means that it can be applied to the URL site with complex structures. To apply the JCV-based estimation method to non-planar survey areas, JCV calculation using Monte Carlo simulation and estimation of fracture size distribution using the maximum likelihood method are required. In this study, we applied the JCV-based estimation method to a tunnel-shaped survey area to examine its applicability to the URL site. The error rates were analyzed when there were fracture sets with various orientations, size distributions, and maximum fracture sizes in the rock mass, and it was found to be less than 10% in all cases. This result indicates that the JCV-based estimation method can be used to estimate the fracture size distribution of the surrounding rock mass if accompanied by a reliable survey of fracture traces on the tunnel surface inside the URL site. Also, since there are no restrictions on the geometry of the survey area, we can continuously update the estimation results during the URL excavation process to increase reliability. The fracture size distribution is essential for constructing the discrete fracture network (DFN) model of the rock mass units at the URL site. In the future, the uncertainty for the fracture size in the DFN model is expected to be reduced by applying the JCV-based estimation method.
This paper reports the results of an experimental examination using X-rays to test annealing materials for lapped bearing steel (STB2), to confirm the validity of the weighted averaging analysis method. The distribution behavior for the sin diagram and the presence or absence of differences in the peak method, half-value breadth method, and centroid method were investigated. When lapping the annealed bearing steel (STB2) material, a residual stress state with a non-directional steep gradient appeared in the surface layer, and it was found that the weighted averaging analysis method was effective. If there is a steep stress gradient, the sin diagram is curved and the diffraction intensity distribution curve becomes asymmetric, resulting in a difference between the peak method, half-value breadth method, and centroid method. This phenomenon was evident when the stress gradient was more than 2~3 kg/mm2/μm. In this case, if the position of the diffraction line is determined using the centroid method and the weighted averaging analysis method is applied, the stress value on the surface and the stress gradient under the surface can be obtained more accurately. When the stress gradient becomes a problem, since the curvature of the sin diagram appears clearly in the region of sin > 0.5, it is necessary to increase the inclination angle as much as possible. In the case of a lapping layer, a more accurate value can be obtained by considering in the weighted averaging analysis method. In an isotropic biaxial residual stress state, the presence or absence of can be determined as the presence or absence of strain for sin≈0.4.
The water supply facilities of Korea have achieved a rapid growth, along with the other social infrastructures consisting a city, due to the phenomenon of urbanization according to economic development. Meanwhile, the level of water supply service demanded by consumer is also steadily getting higher in keeping with economic growth. However, as an adverse effect of rapid growth, the quantity of aged water supply pipes are increasing rapidly, Bursts caused by pipe aging brought about an enormous economic loss of about 6,161 billion won as of 2019. These problems are not only worsening water supply management, also increasing the regional gap in water supply services. The purpose of this study is to classify hazard evaluation indicators and to rank the water distribution network hazard by cluster using the TOPSIS method. In conclusion, in this study, the entropy-based multi-criteria decision-making methods was applied to rank the hazard management of the water distribution network, and the hazard management ranking for each cluster according to the water supply conditions of the county-level municipalities was determined according to the evaluation indicators of water outage, water leakage, and pipe aging. As such, the hazard ranking method proposed in this study can consider various factors that can impede the tap water supply service in the water distribution network from a macroscopic point of view, and it can be reflected in evaluating the degree of hazard management of the water distribution network from a preventive point of view. Also, it can be utilized in the implementation of the maintenance plan and water distribution network management project considering the equity of water supply service and the stability of service supply.
분자동역학에서의 원자들의 유도전하를 계산하기 위해서는 유도전하를 미지수로 하는 선형방정식을 풀어야 하는데 원자들의 위치가 변화할 때마다 필요한 계산이므로 상당한 계산비용이 요구된다. 따라서 효율적인 유도전하 계산 방법은 다양한 시스템을 해석 하기 위해서 필수적이다. 본 연구에서는 constraints가 존재하는 Lagrange 방정식의 해에 대한 선형 시스템, 즉 saddle point를 가지는 문제를 해결하기 위해서 Uzawa method를 도입하였다. Uzawa 매개변수가 수렴 속도에 영향을 미치는 단점을 극복하고 행렬 연산의 효율성을 위해서 Schur complement와 preconditioned conjugate gradient (PCG) 방법을 통해 계산의 효율성을 극대화하는 가속 Uzawa algorithm을 적용한다. 두 금속 나노입자가 전기장에 놓여진 분자동역학 수치모델을 통해서 제시된 방법이 유도전하계산의 수렴성, 효율성 측면에서 모두 향상된 결과를 도출함을 확인하였다. 특히 기존의 가우스 소거법에 의한 계산보다 약 1/10으로 계산비용이 절 감되었고, 기본 Uzawa method에 비하여 conjugate gradient (CG)의 높은 수렴성이 입증되었다.
Waterworks facilities inevitably experience some amount of leakage even if there is a lot of investment or state-of-the-art technology that is applied such as DMA(District Metered Area) system construction, leakage detection, repair, pipe rehabilitation, etc. The primary reason is the leakage is naturally restored over time. In the UK, this restoration characteristic is defined as NRR(Natural rate of rise of leakage) and used to decision making for prioritizing active leakage control of DMAs. However, this restoration characteristic is well recognized, but researches on NRR in the water distribution system are insufficient in Korea. In this study, the estimation method of NRR was developed suitable for applicating in Korea considering of SCADA data, water infrastructure, and water usage patterns by modification of the UK's NRR method. The proposed method was applied to 9 DMAs and verified it's applicability by comparing with the other water loss performance indicators. It is expected that the proposed method can be used to support decision making for sustainable NRW(Nor-revenue water) management in the water distribution system.
2017년, 스트리밍 서비스의 매출이 세계 음악 소비의 38.4%, 66억 달러에 이름으로써 스트리밍 서비스가 처음으로 음악 산업 최대 수입원이 되었다. 스트리밍 방식이 음악 저작물 소비 경향의 주류가 되자, 저작권자 및 실연자가 정당한 몫을 받지 못하고 있다는 문제가 더욱 부각되었다. 소비자가 음원 파일 자체를 영구적으로 소유하게 되는 다운로드 방식과 달리, 일정한 기간 동안 재생하기만 하는 스트리밍 방식의 이용료가 상대적으로 저렴한 까닭에 자연히 음원 1회 재생당 발생하는 매출액 자체가 낮아지므로, 스트리밍 서비스 이용 비중이 높아질수록 저작권자와 실연자에게 돌아가는 몫은 더 적어지기 때문이다. 현재는 음원 창작에 직접적으로 관여하지 않은 음원서비스업체가 음원 스트리밍 서비스로 인한 수익을 가장 많이 가져가고 있어, 음원 유통 및 수익 분배 구조에 문제가 있다는 견해가 많은 것으로 보인다. 현재 40%인 음원서비스업체에게 돌아가는 수익 분배율을 낮춤으로써, 음원서비스업체 스스로 음원 서비스 제공에 소요되는 비용을 절감할 수 있는 새로운 형태의 플랫폼을 개발하거나, 블록체인 기술 등을 적용한 대안적 플랫폼이 등장 하도록 유도하여 작곡⋅작사가, 실연자의 수익분 배율을 높이는 방안이 실용적인 해결책이 될 수 있을 것이다.
In this study, We studied design method to reduce scattering reflection by adding a three - dimensional structure to light guide plate of LED backlight widely used in small and portable devices. In order to improve the efficiency of the light-incident part, a three-dimensional structure is used in which a repetitive structure of a prism shape and a semi-cylindrical shape is repeatedly added to the light-entering face in light-guide plate. As a result of the study, it was found that uniform luminance values were obtained in the light-incident portion and the backlight when a repetitive pattern of a semi-cylindrical shape was added as compared with a case in which a prism shape was added. Particularly, when a rectangular lens with a semi-cylindrical shape is added, the average value of the total luminance is increased by about 15% from 7278 lm to 8324 lm as compared with a general light guide plate.
Acoustic surveys were conducted in the seas surround the South Korea (South Sea A, South Sea B (waters around the Jeju Island), West Sea and East Sea) in spring and autumn in 2016. First, the vertical and horizontal distributions of fisheries resources animals were examined. In most cases vertical acoustic biomass was high in surface water and mid-water layers other than South Sea A in autumn and West Sea. The highest vertical acoustic biomass showed at the depth of 70-80 m in the South Sea A in spring (274.4 m2/nmi2) and the lowest one was 10-20 m in the West Sea in autumn (0.4 m2/nmi2). With regard to the horizontal distributions of fisheries resources animals, in the South Sea A, the acoustic biomass was high in eastern and central part of the South Sea and the northeast of Jeju Island (505.4-4099.1 m2/nmi2) in spring while it was high in eastern South Sea and the coastal water of Yeosu in autumn (1046.9-2958.3 m2/nmi2). In the South Sea B, the acoustic biomass was occurred high in the southern and western seas of Jeju Island in spring (201.0-1444.9 m2/nmi2) and in the southern of Jeju Island in autumn (203.7-1440.9 m2/nmi2). On the other hand, the West Sea showed very low acoustic biomass in spring (average NASC of 1.1 m2/nmi2), yet high acoustic biomass in the vicinity of 37 N in autumn (562.6-3764.2 m2/nmi2). The East Sea had high acoustic biomass in the coastal seas of Busan, Ulsan and Pohang in spring (258.7~976.4 m2/nmi2) and of Goseong, Gangneung, Donghae, Pohang and Busan in autumn (267.3-1196.3 m2/nmi2). During survey periods, fish schools were observed only in the South Sea A and the East Sea in spring and the West Sea in autumn. Fish schools in the South Sea A in spring were small size (333.2 ± 763.2 m2) but had a strong SV (–49.5 ± 5.3 dB). In the East Sea, fish schools in spring had low SV (–60.5 ± 14.5 dB) yet had large sizes (537.9 ± 1111.5 m2) and were distributed in the deep water depth (83.5 ± 33.5 m). Fish schools in the West Sea in autumn had strong SV (–49.6 ± 7.4 dB) and large sizes (507.1 ± 941.8 m2). It was the first time for three seas surrounded South Korea to be conducted by acoustic surveys to understand the distribution and aggregation characteristics of fisheries resources animals. The results of this study would be beneficially used for planning a future survey combined acoustic method and mid-water trawling, particularly deciding a survey location, a time period, and a targeting water depth.
PURPOSES: This study aims to develop a rational procedure for estimating the pavement roughness index considering vehicle wandering. METHODS: The location analysis of the passing vehicle in the lane was performed by approximately 1.2 million vehicles for verification of the wandering distribution. According to verification result, the distribution follows the normal distribution pattern. The probability density function was estimated using each lane's wandering distribution model. Then the procedure for applying a weighted value into the lane profile was conducted using this function. RESULTS : The modified index, MRIw, with consideration towards applying the wandering weighted value application was computed then compared with MRI. It was found that the Coefficient of Variation for distribution of lateral roughness index in the lane was high in the case of a large difference between each index (i.e., MRIw and MRI) observed. CONCLUSIONS : This result confirms that the new procedure with consideration of the weight factor can successfully improve the lane representative characteristics of the roughness index.
Due to the rapid change of global business environment, the growth of China’s steel industry and the inflow of cheap products, domestic steel industry is faced on downward trend. The change of business paradigms from a quantitative growth to a qualitative product is needed in this steel industry. In this environment, it is very important for domestic steel distribution companies to secure their competitiveness by selecting good supply companies through a efficient procurement strategy and effective method. This study tried to find out the success factors of steel distribution industry based on survey research from experts. Weighted values of each factors were found by using AHP (analytic hierarchy process) analysis. The weighted values were applied to DEA(data envelopment analysis) model and eventually the best steel supply company were selected. This paper used 29 domestic steel distribution firms for case example and 5 steps of decision process to select good vendors were suggested.This study used quality, price, delivery and finance as a selection criteria. Using this four criterions, nine variable were suggested. Which were product diversity, base price, discount, payment position, average delivery date, urgency order responsibility and financial condition. These variables were used as a output variable of DEA. Sales and facilities were used as an input variable. Pairwise comparison was conducted using these variables. The weighted value calculated by AHP pairwise comparison were used for DEA analysis. Through the analysis of DEA efficiency process, good DMU (decision making unit) were recommended as a steel supply company. The domestic case example was used to show the effectiveness of this study.
In this study, determination methods of monitoring location in water distribution system were suggested and applied to real test bed. Small block of Gwangtan water distribution system is consisted of 582 pipes, 564 junctions, 1 reservoir, and 1 pump station. Small block of Ho Chi Minh water distribution system is consisted of 162 pipes, 148 junctions, and 1 reservoir. Two small block water distribution systems were analyzed by pressure contribution analysis method to determine the optimum monitoring locations. The pressure change was estimated at each junctions by the additional demand at a junction. From the results, the optimum monitoring location can be determined by rank of pressure contribution index at each junctions due to demand change at a junction.
본 논문에서는 도로교 설계기준이 개편됨에 따라 변화된 활하중과 설계식의 검토를 하려한다. 이를 위해 강합성 플레이트 거더교와 PSC 거더교를 이용한 두 예제를 살펴본다. 기존 도로교설계기준 하중과 AASHTO LRFD 하중 그리고 새로 개편된 도로교설계기준 2015 한계상태설계법에 나온 하중을 적용하여 횡분배 계수법과 유한요소 격자해석을 통해 구한 포락선을 비교․검토한다. 또한 강합성 플레이트 거더교의 경우 중앙지간의 길이 변화에 대한 검토를 하고, PSC 거더교의 경우에는 가로보의 개수와 경간의 개수, 그리고 각 경간장의 변화를 주어 각각 그 차이점을 분석한다.
HS-SPME법을 이용하여 한국 서남해 연안해역에서 Sea-nine 211, Irgarol 1051, Diuron과 같은 방오제의 분포 특성을 검토하였다. 반감기가 짧은 Sea-nine 211은 모든 시료채취지점 및 지역의 해수와 퇴적물에서 아주 낮은 농도 또는 검출한계 이하로 분포하였다. Irgarol 1051은 해수 및 퇴적물에서의 최고 농도가 각각 6.98μg/L, 28.50 ng/g-dry wt로 검출되었으며, 지역별 분포 특성은 나타나지 않았다. 반감기가 가장 길고 생물 농축성이 강한 Diuron은 모든 시료채취 지점에서 Sea-nine 211과 Irgarol 1051보다 높은 농도로 분포하였으며, Mo7(목포)의 퇴적물에서 최고농도(3882.22 ng/g-dry wt)로 분석되었다. Irgarol 1051과 Diuron은 조선산업단지 및 선박정박지가 위치하고 있는 지점에서 높은 농도로 분포하는 것으로 나타났다. 또한 퇴적물에서의 방오제 분포는 내만에서 외만으로 그 농도가 낮아졌다. 이러한 결과로부터 방오제는 내만에 위치하고 있는 항구나 조선산업단지로부터 오염될 가능성이 높을 것으로 판단된다.