Earticle

Purpose: The purpose of this study was to assess the fitness of anterior and posterior interim crowns fabricated by three different additive manufacturing technologies. Methods: The working model was digitized, and single crowns (maxillary right central incisor and maxillary right first molar) were designed using computer-aided design software (DentalCad 2.2; exocad). On each abutment, interim crowns (n=60) were fabricated using three types of additive manufacturing technologies. Then, the abutment appearance and internal scan data of the interim crown was obtained using an intraoral scanner. The fitness of the interim crowns were evaluated by using the superimposition of the three-dimensional scan data (Geomagic Control X; 3D Systems). The one-way analysis of variance and Tukey posterior test were used to compare the results among groups (α=0.05). Results: A significant difference was found in the fitness of the interim crowns according to the type of additive manufacturing technology (p<0.05). The posterior interim crown showed smaller root mean square value than the anterior interim crown. Conclusion: Since the fitness of the posterior interim crown produced by three types of additive manufacturing technology were all within clinically acceptable range (<120 μm), it can be sufficiently used for the fabrication of interim crowns.

This study has evaluated the genomic estimated breeding value (GEBV) of the commercial Hanwoo population using the genomic best linear unbiased prediction (GBLUP) method and genomic information. Furthermore, it analyzed the accuracy and realized accuracy of the GEBV. 1,740 heads of the Hanwoo population which were analyzed using a single nucleotide polymorphism (SNP) Chip has selected as the test population. For carcass weight (CWT), eye muscle area (EMA), back fat thickness (BFT), and marbling score (MS), the mean GEBVs estimated using the GBLUP method were 3.819, 0.740, -0.248, and 0.041, respectively and the accuracy of each trait was 0.743, 0.728, 0.737, and 0.765, respectively. The accuracy of the breeding value was affected by heritability. The accuracy was estimated to be low in EMA with low heritability and high in MS with high heritability. Realized accuracy values of 0.522, 0.404, 0.444, and 0.539 for CWT, EMA, BFT, and MS, respectively, showing the same pattern as the accuracy value. The results of this study suggest that the breeding value of each individual can be estimated with higher accuracy by estimating the GEBV using the genomic information of 18,499 reference populations. If this method is used and applied to individual selection in a commercial Hanwoo population, more precise and economical individual selection is possible. In addition, continuous verification of the GBLUP model and establishment of a reference population suitable for commercial Hanwoo populations in Korea will enable a more accurate evaluation of individuals.

Address verification is a critical and challenging task for businesses worldwide, given that every country has its address formatting conventions. This is particularly relevant in industries such as imported food safety, where prompt safety measures are vital to safeguard people when food safety issues arise. Hence, we developed a deep learning and Google Geocoding-based technique to parse, validate, and standardize the address accurately. Our proposed model utilizes a deep learning algorithm to identify an address's components, such as the street name, district, city, and state/province then validate them using Google Geocoding. However, our initial model did not yield great results for countries with well-established postal/zip code systems. To resolve this issue, we improved the model by adding an extra administrative level - the Postal Code - specifically for such countries. To evaluate the efficacy of our new model, we compared multiple metrics such as accuracy, precision, recall, and F1 score of the old and new models. The dataset used to test the new model comprised addresses from manufacturers in the United Kingdom, the United States, and Australia. The results showed that our new model outperformed the original model with better accuracy when applied to real address data. With the addition of this extra administrative level, the performance of the verification process for global addresses is improved significantly.

The purpose of this study was to develop and evaluate Point Cloud Data (PCD) deep learning models and a rule-based system for segmenting tree structures (stems and crowns) using fixed terrestrial LiDAR data. The dataset comprised 48 Larix Kaemferi trees, which were collected and preprocessed. For the PCD deep learning models, three downsampled datasets consisting of 1024, 4096, and 16384 points were constructed from the original data. The data was divided into training (70%) and validation (30%) sets. Models were built using PointNet and PointNet++ architectures, resulting in a total of 12 tree structure segmentation models for accuracy comparison. The rule-based system was developed using the original data, applying techniques such as verticality checks, cylindrical structure detection, and slice-based circular fitting to detect the stem. It then segmented the stem through repetitive circle fitting and validation processes based on height. The average accuracy of the PCD deep learning tree structure segmentation models was approximately 95%, with the PointNet++ model using 16384 points achieving the highest classification accuracy of about 98%. The rule-based system achieved high classification accuracy of over 99% for both tree species. This study is expected to contribute to precise measurement and efficient management of forest resources by presenting automated methods for tree structure segmentation using AI technology and rule-based approaches. It is anticipated that this research will serve as a foundation for the advancement of forest digitalization, precision forest management technologies, forest structure analysis, and timber production estimation in various fields.

노후화 시설물의 증가에 따라 선제적 유지관리의 중요성은 점차 증대되고 있다. 선제적 유지관리는 시설물의 응답 계측으로부터 시작되기 때문에 높은 정밀도를 가지는 응답을 획득하는 것이 중요하다. 국부적인 응답 중 변형률은 균열 감지 및 피로 진전 예측 등에 활용가능하다. 변형률 센서는 크게 이산형 및 분포형 센서로 구분된다. 이산형 센서의 대표적인 예가 광섬유 브래그 격자(FBG)와 전기 저항식 게이지이다. 이산형 센서는 높은 정확성과 재현성(고 정밀)을 가지지만, 측정점이 제한된다는 한계를 가진다. 브릴루앙 산란 기반 광섬유 변형률 계측 시스템 중 하나인 Brillouin Optical Correlation Domain Analysis (BOCDA)은 대표적인 분포형 센서이며, 5 cm 라는 높은 공간 분해능을 가진다. BOCDA는 투영된 광원에서 발생하는 산란파를 이용하여 광섬유 전 구간의 변형률을 계측한다. 측정점이 많아지는 장점이 있으나, 이산형 센서에 낮은 정확도와 재현성을 가진다. 본 연구에서는 고 정밀 데이터(이산형 센서)와 저 정밀 데이터(분포형 센서) 각각의 장점을 융합하는 후처리 기법을 제안하였으며, 이에 대한 가능성을 검증 실험을 통해 확인했다.

As aging infrastructures increase along with time, the efficient maintenance becomes more significant and accurate responses from the sensors are pre-requisite. Among various responses, strain is commonly used to detect damage such as crack and fatigue. Optical fiber sensor is one of the promising sensing techniques to measure strains with high-durability, immunity for electrical noise, long transmission distance. Fiber Bragg Grating (FBG) is a point sensor to measure the strain based on reflected signals from the grating, while Brillouin Optic Correlation Domain Analysis (BOCDA) is a distributed sensor to measure the strain along with the optical fiber based on scattering signals. Although the FBG provides the signal with high accuracy and reproducibility, the number of sensing points is limited. On the other hand, the BOCDA can measure a quasi-continuous strain along with the optical fiber. However, the measured signals from BOCDA have low accuracy and reproducibility. This paper proposed a multi-fidelity data-fusion method based on Gaussian Process Regression to improve the fidelity of the strain distribution by fusing the advantages of both systems. The proposed method was evaluated by laboratory test. The result shows that the proposed method is promising to improve the fidelity of the strain.

하천 수위 예측의 주요 목적 중 하나는 홍수예경보 발령을 위한 기준으로 활용하는 것이다. 본 연구에서는 딥러닝 기반의 하천 수위 예측 모델을 홍수예경보 측면에서 효과적으로 활용하기 위해 학습데이터를 최적화하고, 딥러닝 모델의 정확도 향상을 평가하기 위해 딥러닝 모델의 자동 설계 및 최적화를 지원하는 AutoKeras를 활용하여 인위적인 요인을 배제한 모델을 구축하였다. 한탄강 상류유역을 대상지역으로 선정하고, 3개의 수위관측소와 유역평균강우 데이터를 구축하였고, 구축된 데이터를 이용하여 수위 변화 여부와 관계없이 강우가 발생한 모든 학습데이터 셋을 사용한 모델(Model 1)과 일정 수준 이상의 수위 상승 변화가 있는 학습데이터 셋을 사용한 딥러닝 모델(Model 2)을 개발하여 한탄강 상류 한탄대교의 수위 및 홍수 예측 성능을 평가하였다. 실시간 하천 홍수예측 결과, 시계열 수위 예측에서 Model 1이 더 많은 데이터를 활용함으로써 상관계수와 평균제곱근오차(RMSE)에서 다소 우수한 성능을 보였다. 반면, Model 2는 홍수 예측에서 재현율(recall), F1-score, 임계성공지수(CSI) 등의 지표에서 더 뛰어난 성과를 보였다. 본 결과는 학습데이터의 특성과 구성 방식이 딥러닝 모델의 예측 능력에 큰 영향을 미친다는 것을 보여주며, 홍수와 같은 특정 사건을 예측하려면 수위 상승과 같은 핵심 요인 위주의 데이터를 더 집중적으로 학습시킬 필요가 있음을 시사한다.

One of the primary objectives of predicting river water levels is to establish criteria for issuing flood warnings and alerts. This study aims to optimize the training data for a deep learning-based river water level prediction model and enhance its accuracy by utilizing AutoKeras, which supports automatic design and optimization of deep learning models, to develop models that minimize artificial influences. The upper basin of the Hantan River was selected as the study area, and datasets were constructed using water level data from three observation stations and mean areal rainfall data. Based on these datasets. Based on these datasets, Two models were developed: Model 1 was trained on datasets that included all recorded rainfall events, while Model 2 was trained on datasets capturing significant water level increases. Predictions for Hantan Bridge indicated that Model 1 achieved higher accuracy in time-series water level estimation, as evidenced by a higher correlation and lower RMSE. In contrast, Model 2 exhibited superior flood detection capability, showing higher recall, F1-score, and CSI. These results highlight the importance of selecting appropriate training data when developing deep learning models, particularly for flood prediction. Emphasizing critical factors such as water level rises can enhance model performance, enabling more effective early warning systems and improving disaster preparedness.

PURPOSES : The travel times of expressway buses have been estimated using the travel time data between entrance tollgates and exit tollgates, which are produced by the Toll Collections System (TCS). However, the travel time data from TCS has a few critical problems. For example, the travel time data include the travel times of trucks as well as those of buses. Therefore, the travel time estimation of expressway buses using TCS data may be implicitly and explicitly incorrect. The goal of this study is to improve the accuracy of the expressway bus travel time estimation using DSRC-based travel time by identifying the appropriate analysis period of input data. METHODS : All expressway buses are equipped with the Hi-Pass transponders so that the travel times of only expressway buses can be extracted now using DSRC. Thus, this study analyzed the operational characteristics as well as travel time patterns of the expressway buses operating between Seoul and Dajeon. And then, this study determined the most appropriate analysis period of input data for the expressway bus travel time estimation model in order to improve the accuracy of the model. RESULTS : As a result of feasibility analysis according to the analysis period, overall MAPE values were found to be similar. However, the MAPE values of the cases using similar volume patterns outperformed other cases. CONCLUSIONS : The best input period was that of the case which uses the travel time pattern of the days whose total expressway traffic volumes are similar to that of one day before the day during which the travel times of expressway buses must be estimated.

As global warming becoming an environmentally serious issue, more attention is drawn to fuel consumption which is the direct source of green house gas emission. The fuel consumption by aircraft operation is not an exception. Motivated by the societal and environmental context, this paper explains a method for estimation of aircraft fuel consumed during their flights as well as the computational process using real flight track data. Applying so-called 'Total Energy Model' along with aircraft specific parameters provided in EUROCONTROL's Base of Aircraft Data (BADA) to aircraft radar track data, we estimate fuel consumption of individual aircraft flown between Gimpo and Jeju airports. We then assess the estimation accuracy by comparing the estimated fuel consumption with the actual one collected from an airline. The computational results are quite encouraging in that the method is able to estimate the actual fuel consumption within ${\pm}6{\sim}11%$ of error margin. The limitations and possible enhancements of the method are also discussed.

수도권 지하철 AFC자료는 승객의 환승정보가 누락되어 있다. AFC자료는 통행수, 통행시간 및 통행거리의 통행지표를 TagIn 단말기ID를 기준으로 할당한다. 따라서 AFC자료는 승객의 실제 통행궤적을 반영하지 못하며, 이는 통행지표 추정의 오류로 작용되고 있다. 본 연구는 TagIn 및 TagOut 단말기ID를 연결하는 통행경로 파악을 통하여 통행지표를 산정하는 방법론을 제안한다. 이를 위해 승객은 차내시간, 환승보행시간, 배차간격을 고려한 최소통행시간경로를 통행한다고 가정한다. 이 방법은 승객이 이동한 통행궤적을 따라 환승을 반영하기 때문에 승객이 이동한 경로에 포함된 지하철 운영기관의 통행관련자료가 통행지표에 정확하게 반영된다. 제안된 방법론은 기존 AFC자료보다 1.47배가 증가한 통행을 산정하여 교통정책을 위한 지표산정방안으로 평가될 수 있음을 보여준다.

Individual passenger transfer information is not included in Seoul metropolitan subway Automatic Fare Collection (AFC) data. Currently, basic data such as travel time and distance are allocated based on the TagIn terminal ID data records of AFC data. As such, knowledge of the actual path taken by passengers is constrained by the fact that transfers are not applied, resulting in overestimation of the transport index. This research proposes a method by which a transit path that connects the TagIn and TagOut terminal IDs in AFC data is determined and applied to the transit index. The method embodies the concept that a passenger's line of travel also accounts for transfers, and can be applied to the transit index. The path selection model for the passenger calculates the line of transit based on travel time minimization, with in-vehicle time, transfer walking time, and vehicle intervals all incorporated into the travel time. Since the proposed method can take into account estimated passenger movement trajectories, transport-related data of each subway organization included in the trajectories can be accurately explained. The research results in a calculation of 1.47 times the values recorded, and this can be evaluated directly in its ability to better represent the transportation policy index.

우리나라의 경우 강수량의 2/3 정도가 하절기에 집중되는 강우특성상 해마다 여름철 홍수기의 탁수 문제가 다양하게 발생하고 있다. 이상강우와 기상이변에 의한 집중강우가 증가 추세이며, '02년 태풍 루사', '03년 태풍 매미', '06년 에위니아'부터 20년 마이삭, 하이선 까지 장마와 태풍에 의한 유입량이 급증하는 시기 탁수의 유입으로 수중 탁도가 급상승하며 댐 저수지 내 탁수 문제가 발생하였다. 특히 연 평균 물사용량의 대부분을 하천 및 댐 저수지를 이용하는 우리나라의 경우 탁수 문제가 장기화될 경우 댐 하류 해당 지역 농업, 공업, 수생태 등 사회적, 환경적으로 많은 문제를 발생시킨다. 이러한 탁수 예측을 통한 대응을 위해 탁수 모델링에 대한 연구가 활발히 진행되고 있다. 탁수 현황을 모의하기 위해서는 유량, 수온, SS 데이터가 필요하다. 이를 위해 국가측정망에서 하천 및 댐 저수지 내 SS를 측정하여 탁수를 측정 하고 있으나 설비가 미흡하여 데이터 해상도가 낮다는 한계점이 있으며 주요 댐 저수지 내에서는 수자원공사에서 관리하는 자동 측정기기를 활용하여 높은 데이터 해상도를 유지 하고 있으나 댐 별, 기상 조건에 따라 미측정 기간이 존재한다. 탁도를 측정을 위한 센서로는 Optical Backscatter Sensor (OBS), YSI 등이 있으며 SS를 측정하기 위한 센서는 레이저부유사측정기(Laser In-Situ Scattering and Transmissometry, LISST) 등의 장비를 이용하고 있다. 하지만 이런 첨단 센서의 경우 또한 수중에 고정하여 측정하기에는 장비의 안정성 등의 이유로 한계가 있다. 따라서, 취득된 유량, 수온, SS, 탁도 데이터를 기반으로 분석을 통해 미측정 기간이 존재함으로 입력자료에 활용되는 SS를 산정하기 위해 관계식 개발을 필요로한다. 본 연구에서는 댐 방류구 인근 지점 측정 데이터를 기반으로 개발된 탁도-SS 관계식을 통해 수자원 공사 SURIAN 시스템에서 활용되고 있는 AEM3D 모델을 이용하여 탁수 발생 예측 정확도 개선을 하고자 하였다.

In Korea, about two-thirds of the precipitation is concentrated in the summer season, so the problem of turbidity in the summer flood season varies from year to year. Concentrated rainfall due to abnormal rainfall and extreme weather is on the rise. The inflow of turbidity caused a sudden increase in turbidity in the water, causing a problem of turbidity in the dam reservoir. In particular, in Korea, where rivers and dam reservoirs are used for most of the annual average water consumption, if turbidity problems are prolonged, social and environmental problems such as agriculture, industry, and aquatic ecosystems in downstream areas will occur. In order to cope with such turbidity prediction, research on turbidity modeling is being actively conducted. Flow rate, water temperature, and SS data are required to model turbid water. To this end, the national measurement network measures turbidity by measuring SS in rivers and dam reservoirs, but there is a limitation in that the data resolution is low due to insufficient facilities. However, there is an unmeasured period depending on each dam and weather conditions. As a sensor for measuring turbidity, there are Optical Backscatter Sensor (OBS) and YSI, and a sensor for measuring SS uses equipment such as Laser In-Situ Scattering and Transmissometry (LISST). However, in the case of such a high-tech sensor, there is a limit due to the stability of the equipment. Therefore, there is an unmeasured period through analysis based on the acquired flow rate, water temperature, SS, and turbidity data, so it is necessary to develop a relational expression to calculate the SS used for the input data. In this study, the AEM3D model used in the Water Resources Corporation SURIAN system was used to improve the accuracy of prediction of turbidity through the turbidity-SS relationship developed based on the measurement data near the dam outlet.

Non-Parallelism the axial direction occurs during grinding process of long slender shafts. The reason for the axial error is due to elastic deformation of the components, accumulation phenomenon of the grinding and wheel wear during the grinding process. The accumulation phenomenon, the size generation mechanism and the wheel wear process during traverse grinding result in complicated process at each step on the wheel surface. The grinding system stiffness obtained from the stiffness of the center on the tailstock and the workpiece varing according to the relative position of the wheel and the workpiece. Further more, the value of wheel wear increases as the grinding process advances. The above mentioned issues make the shape generation process during traverse grinding quite complicated. This research analyzes the shape generation process in the direction of the work spindle. First, the formulation of the grinding system stiffness was conducted and the simulation analysis method of the traverse grinding was established. Also, a measuring system for assessing the dimensinal accuracy of the workpiece has been developed.

기후변화로 인한 극한 강우 사상의 증가로 하천에서의 신속하고 정확한 홍수 수위 예측의 중요성이 높아지고 있다. 본 연구에서는 LSTM (Long Short-Term Memory) 기반 딥러닝 모형을 활용하여 금강의 지류하천인 유등천과 갑천에 위치한 4개 수위관측소(복수교, 인창교, 만년교, 원촌교)를 대상으로 학습데이터 구성 방법에 따른 홍수기 수위 예측 정확도를 비교 분석하였다. 학습데이터는 전체 관측 기간 수위 자료(Case 1), 12시간 무강우 기간을 제외한 수위 자료(Case 2), 상류 관측소 데이터를 조정한 수위 자료(Case 3)로 구성하였다. 2025년 전체 기간을 대상으로 한 분석에서는 Case 1의 NSE가 가장 높게 나타났으나, 최대 홍수 사상을 대상으로 한 분석에서는 Case 2가 우수한 예측 성능을 보였다. 만년교의 경우 1시간 후 예측 기준 Case 2의 NSE가 0.94, RMSE가 0.06 m였으며, Case 1의 NSE 0.86, RMSE 0.09m보다 개선되었다. 복수교의 경우 1시간 후 예측 기준 Case 2의 NSE가 0.99, RMSE가 0.03 m로 나타났다. 또한 만년교에 대해 상류 학습데이터를 조정한 Case 3은 Case 2 대비 Peak Error가 최대 0.35 m 개선되어, 4시간 후 예측에서도 0.04m 이내의 첨두 수위 오차를 달성하였다. 본 연구의 결과는 LSTM 기반 홍수 수위 예측 모형 개발 시 학습데이터 구성 방법의 선택 기준을 제시하는 기초 자료로 활용될 수 있으며, 향후 데이터 기반 수위 예측 모형의 정확도 향상 및 적용성 확대를 위한 연구에 기여할 수 있을 것으로 기대된다.

As extreme rainfall events intensify due to climate change, the need for rapid and accurate flood-stage water level prediction in small and medium rivers has become increasingly critical. This study develops Long Short-Term Memory (LSTM)-based deep learning models for four water level stations (Boksugyo, Inchanggyo, Mannyeongyo, and Wonchongyo) located along Yudeungcheon and Gapcheon, tributaries of the Geum River, and comparatively evaluates prediction accuracy according to different training data set construction methods. Three training data configurations were examined: complete observed water level records spanning the entire observation period (Case 1), records excluding 12-hour dry periods (Case 2), and an adjusted data set with reduced upstream station data (Case 3). While Case 1 yielded higher NSE values over the full 2025 evaluation period, Case 2 demonstrated superior performance during peak flood events. For Mannyeongyo, Case 2 achieved an NSE of 0.94 and RMSE of 0.06 m at a 1-hour lead time, compared to NSE of 0.86 and RMSE of 0.09 m for Case 1. For Boksugyo, Case 2 attained an NSE of 0.99 and RMSE of 0.03 m at a 1-hour lead time. Furthermore, the adjusted training data set (Case 3) at Mannyeongyo reduced peak water level error by up to 0.35 m relative to Case 2, achieving peak errors within 0.04 m even at a 4-hour lead time. The findings provide practical guidance on training data selection for LSTM-based flood forecasting models and are expected to contribute to the development of real-time AI-based flood warning systems.

본 연구는 데이터의 품질이 인공지능(AI) 성능에 미치는 영향을 검토한다. 이를 위해, 데이터 특성변수(Feature)의 유사도와 클래스(Class) 구성의 불균형을 고려한 모의실험(Simulation)을 통해 라벨링 오류 수준이 인공지능의 성능에 미치 는 영향을 비교 분석하였다. 그 결과, 특성변수 간 유사성이 높은 데이터에서는 특성 변수 간 유사성이 낮은 데이터에 비해 라 벨링 정확도에 더 민감하게 반응하였으며, 클래스 불균형이 증가함에 따라 인공지능 정확도가 급격히 감소되는 경향을 관찰 하였다. 이는 인공지능 학습데이터의 품질평가 기준 및 관련 연구를 위한 기초자료가 될 것이다.

The study investigates the impact of data quality on the performance of artificial intelligence (AI). To this end, the impact of labeling error levels on the performance of artificial intelligence was compared and analyzed through simulation, taking into account the similarity of data features and the imbalance of class composition. As a result, data with high similarity between characteristic variables were found to be more sensitive to labeling accuracy than data with low similarity between characteristic variables. It was observed that artificial intelligence accuracy tended to decrease rapidly as class imbalance increased. This will serve as the fundamental data for evaluating the quality criteria and conducting related research on artificial intelligence learning data.

합리적인 방법으로 적정공사비를 예상하는 것은 성공적인 건설사업 수행을 위한 가장 중요한 것 중 하나이다. 이에 적정공사비를 예측하기 위해 많은 연구가 진행되었다. 하지만 기존의 공사비 예측연구는 입력변수 즉, 사업초기단계에서 획득되는 정보가 사업마다 그리고 사업이 진행됨에 따라 변하지만 이러한 입력변수 변화에 따른 연구는 미흡하다. 따라서 본 연구는 사업초기단계에 단계별로 획득 가능한 입력변수가 변함에 따른 적정 예측기법 제시를 목적으로 한다. 이를 위해 3년간 공동주택 공사비 실적자료를 수집하고 주성분분석을 통해 객관적인 방법으로 입력변수 수준을 분류하였다. 분류된 수준마다 회귀분석, 신경망, 의사결정나무분석을 사용한 예측모형 구축 후, 예측정확도와 발주자 의사결정지원측면에서 적절한 공사비 예측기법을 제시하였다.

To estimate a reasonable cost in early stage of a project is one of the most important works to implement construction project successfully. It is necessary for estimating reasonable cost to predict cost considering procurable input variables which is different depend on each project and a time of prediction. Therefore, the purpose of study is to present an adequate predicting method by level of procurable input variables. This study classifies level of procurable information in an early stage of project by previous studies and actual cost data of apartments, and presents adequate methods on each level, based on characteristic of estimating method and accuracy of estimating model in Data Mining. The result of this study, If it goes through process of variable transformation and variable selection, cost prediction using regression analysis at all of the levels would suitable. And Decision tree, which has lower accuracy but able to easily inform frequency distribution of cost based on causes of transformation, would be more useful, if we consider the purpose of cost prediction is accuracy of prediction and to understand easily making a decision of owner.

Purpose: The purpose of this study was to compare the accuracy of the scan data acquired by the extra-oral and intra-oral scanner. Methods: The maxillary right first molar was made of polymethyl methacrylate (PMMA)specimen. This PMMA specimen was scanned with a engineering scanner and intra-oral scanner. Meanwhile, extra-oral scanner scanned stone die duplicated from PMMA master die. Trueness and precision of scan datas was measuredby3-dimensinal inspection. Independent t-test was conduct to analysis the significant difference.(a=0.05) Results: In the trueness analysis, mean of discrepancies were13.82um for intra oral scanner and16.84 um for extra-oral scanner. In the precision analysis, mean of discrepancies were 11.72 for inta-oral scanner and 9.2 for extra-oral scanner. Both trueness and precision showed a statistically significant difference (Table 1, p<0.05). Conclusion: Intra-oral scanner can show higher trueness than extra-oral scanner, it has lower precision.

To investigate how training data bias impacts recommendation quality, this study experimented with simulated data to observe changes in recommendation performance and result bias across varying levels of data bias. The experiment revealed that while imbalances in popularity and preference distributions within the training data did not significantly affect recommendation accuracy, they did strongly influence recommendation result bias, leading to amplification effects. Specifically, increased preference distribution imbalance intensified recommendation result bias, whereas a certain level of popularity imbalance had no effect on recommendation result bias nor caused amplification. These findings challenge previous research suggesting that data bias degrades recommendation accuracy but support claims that data bias amplifies recommendation result bias. This study is notable for providing insights into how data bias mechanisms impact recommendation quality, offering valuable guidance for future research focused on bias mitigation.

As project environments become more complex and uncertain, the usefulness of deep learning-based schedule forecasting depends not only on algorithms but also on how project data are managed and processed. Drawing on organizational information processing theory, this study examines the associations among project data management competency, information processing quality, project characteristics, and schedule forecasting accuracy. PLS-SEM was applied to 283 survey responses from project practitioners and managers; model learning quality and project state representation were analyzed using a subsample of 127 respondents with experience in deep learning-based forecasting systems. The survey outcome represents respondents’ perceived forecasting accuracy, not error metrics calculated from their projects. The results show that data quality management, data operations management maturity, and data richness and diversity are positively associated with perceived forecasting accuracy, partly through information processing quality. Project complexity and uncertainty strengthen selected relationships, whereas the moderating role of project management methodology is limited. Separately, exploratory LSTM experiments using RCPLIB and DSLIB report objective performance through MAE and RMSE and provide supplementary evidence that forecasting errors vary with data conditions. The findings distinguish perceived from objective accuracy and suggest that data management competency and information processing capability are important conditions for AI-enabled project schedule forecasting.

본 연구는 가축을 대상으로 이표용 맥파 압력센서, 전도성 섬유센서, 광센서 중 사육 가축에 적용 할 최적의 맥파 센서를 분석하기 위하여 소10마리, 돼지 10마리를 대상으로 맥파 압력센서, 전도성 섬유센서, 광센서를 적용한 결과, 소는 광센서가 압력센서와 전도성 섬유센서에 비하여 평균값에 대한 표준편차가 작은 것으로 나타나 광센서가 가장 안정적인 것으로 나타났고, 돼지는 압력센서, 전도성 섬유센서, 광센서 모두 안정적인 맥파수를 나타냈다. 따라 서 소와 같이 신체부의의 굴곡이 많고 털이 길고 조밀한 가축의 경우, 광센서를 이용한 맥파 측정에 가장 효율적인 것으로 나타났다.

In this research, we aimed to investigate the most optimum pulse wave sensor to ear label of live stocks among pulse wave piezo film sensor, conductive textile sensor, photo sensor. As a result of this research with application to 10 cattle, 10 pigs objects with pulse wave piezo film sensor, conductive textile sensor, photo sensor, photo sensor shows less standard deviation to average value than piezo film sensor or conductive textile sensor which means it is the most stable for the cattle. With pigs, piezo film sensor, conductive textile sensor and photo sensor all show stable pulse rate. Thus, to take pulse rate of livestock with curved body and long and dense coat such as cow, photo sensor will be considered as the most efficient mean..