• 한국건축시공학회지
• /
• 제11권5호
• /
• pp.459-467
• /
• 2011

공동주택 건설공사는 많은 기술자들이 참여하여 작성한 설계 도서를 토대로 다양한 공종이 연계되어 발생되며, 이로 인해 예기치 못한 설계상 실수나 자재 결함 및 공사 중의 잘못이 중첩되어 하자가 발생하게 된다. 건설업체는 준공된 건축물을 일정기간 동안 하자보수를 실시해야 하며, 이를 위해 하자보수비용을 효율적으로 예측하여 사업계획을 수립하게 된다. 하자발생은 정확한 예측이 어렵기 때문에 실적자료를 기반으로 예측하게 된다. 국내 공동주택의 경우 하자보수비용 관련 자료가 미흡하여 이를 예측하는 방안 등이 거의 없는 실정이다. 따라서 본 연구에서는 준공후 10년의 실적자료를 기반으로 공급유형 및 지역별 하자보수비용을 예측할 수 있는 모형을 개발하고자 한다.

• 품질경영학회지
• /
• 제49권3호
• /
• pp.421-431
• /
• 2021

Purpose: For improving outgoing quality, this study presents a novel sampling framework based on predictive analytics. Methods: The proposed framework is composed of three steps. The first step is the variable selection. The knowledge-based and data-driven approaches are employed to select important variables. The second step is the model learning. In this step, we consider the supervised classification methods, the anomaly detection methods, and the rule-based methods. The applying model is the third step. This step includes the all processes to be enabled on real-time prediction. Each prediction model classifies a product as a target sample or random sample. Thereafter intensive quality inspections are executed on the specified target samples. Results: The inspection data of three Samsung products (mobile, TV, refrigerator) are used to check functional defects in the product by utilizing the proposed method. The results demonstrate that using target sampling is more effective and efficient than random sampling. Conclusion: The results of this paper show that the proposed method can efficiently detect products that have the possibilities of user's defect in the lot. Additionally our study can guide practitioners on how to easily detect defective products using stratified sampling

• 산업경영시스템학회지
• /
• 제45권4호
• /
• pp.157-166
• /
• 2022

In the case of a die-casting process, defects that are difficult to confirm by visual inspection, such as shrinkage bubbles, may occur due to an error in maintaining a vacuum state. Since these casting defects are discovered during post-processing operations such as heat treatment or finishing work, they cannot be taken in advance at the casting time, which can cause a large number of defects. In this study, we propose an approach that can predict the occurrence of casting defects by defect type using machine learning technology based on casting parameter data collected from equipment in the die casting process in real time. Die-casting parameter data can basically be collected through the casting equipment controller. In order to perform classification analysis for predicting defects by defect type, labeling of casting parameters must be performed. In this study, first, the defective data set is separated by performing the primary clustering based on the total defect rate obtained during the post-processing. Second, the secondary cluster analysis is performed using the defect rate by type for the separated defect data set, and the labeling task is performed by defect type using the cluster analysis result. Finally, a classification learning model is created by collecting the entire labeled data set, and a real-time monitoring system for defect prediction using LabView and Python was implemented. When a defect is predicted, notification is performed so that the operator can cope with it, such as displaying on the monitoring screen and alarm notification.

• 품질경영학회지
• /
• 제52권2호
• /
• pp.185-199
• /
• 2024

Purpose: The improvement of yield and quality in product manufacturing is crucial from the perspective of process management. Controlling key variables within the process is essential for enhancing the quality of the produced items. In this study, we aim to identify key variables influencing product defects and facilitate quality enhancement in CNC machining process using SHAP(SHapley Additive exPlanations) Methods: Firstly, we conduct model training using boosting algorithm-based models such as AdaBoost, GBM, XGBoost, LightGBM, and CatBoost. The CNC machining process data is divided into training data and test data at a ratio 9:1 for model training and test experiments. Subsequently, we select a model with excellent Accuracy and F1-score performance and apply SHAP to extract variables influencing defects in the CNC machining process. Results: By comparing the performances of different models, the selected CatBoost model demonstrated an Accuracy of 97% and an F1-score of 95%. Using Shapley Value, we extract key variables that positively of negatively impact the dependent variable(good/defective product). We identify variables with relatively low importance, suggesting variables that should be prioritized for management. Conclusion: The extraction of key variables using SHAP provides explanatory power distinct from traditional machine learning techniques. This study holds significance in identifying key variables that should be prioritized for management in CNC machining process. It is expected to contribute to enhancing the production quality of the CNC machining process.

• 분석과학
• /
• 제37권2호
• /
• pp.114-122
• /
• 2024

A transmission Raman spectroscopy-based quantitative model, which can analyze the content of a drug product containing naproxen sodium as its active pharmaceutical ingredient (API), was developed. Compared with the existing analytical method, i.e., high-performance liquid chromatography (HPLC), Raman spectroscopy exhibits high test efficiency owing to its shorter sample pre-treatment and measurement time. Raman spectroscopy is environmentally friendly since samples can be tested rapidly via a nondestructive method without sample preparation using solvent. Through this analysis method, rapid on-site analysis was possible and it could prevent the production of defective tablets with potency problems. The developed method was applied to the assays of the naproxen sodium of coated tablets that were manufactured in commercial scale and the content of naproxen sodium was accurately predicted by Raman spectroscopy and compared with the reference analytical method such as HPLC. The method validation of the new approach was also performed. Further, the specificity, linearity, accuracy, precision, and robustness tests were conducted, and all the results were within the criteria. The standard error of cross-validation and standard error of prediction values were determined as 0.949 % and 0.724 %, respectively.

• 전자공학회논문지SC
• /
• 제44권4호통권316호
• /
• pp.67-73
• /
• 2007

광디스크 드라이브에서 일반적으로 광 픽업은 포커싱과 트래킹 서보 시스템의 결합된 동력한 시스템을 갖는다. 이 결합된 동력학 시스템은 서로의 구동간섭에 의하여 광디스크 드라이브의 성능을 저하시킨다. 특히 디스크의 표면에 결함이 존재할 경우에는 포커싱과 트래킹 서보 시스템의 구동간섭으로 인해 시스템이 안정화하는데 많은 시간이 소요되고 결함구간이 길어질 경우 광학 렌즈가 추종하던 트랙을 이탈할 수 있다. 본 논문에서는 이 문제점을 극복하기 위하여, 트래킹 에러와 포커싱 에러의 관측을 기반으로 광디스크 드라이브의 새로운 제어 방법을 제안한다. 결함이 존재할 경우 결합된 동력학 시스템에 대한 보상을 통하여 시스템의 안정화 시간을 감소시킨다. 동일한 표면 결함에 대하여 제안된 제어 방법의 개선된 성능은 실험을 통하여 검증하였다.

• 산업경영시스템학회지
• /
• 제44권4호
• /
• pp.12-22
• /
• 2021

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.

• 대한조선학회논문집
• /
• 제55권1호
• /
• pp.66-74
• /
• 2018

In order to evaluate the precision degree of the blocks on the dock, the shipyards recently started to use the point cloud approaches using the 3D scanners. However, they hesitate to use it due to the limited time, cost, and elaborative effects for the post-works. Although it is somewhat traditional instead, they have still used the electro-optical wave devices which have a characteristic of having less dense point set (usually 1 point per meter) around the contact section of two blocks. This paper tried to expand the usage of point sets. Our approach can estimate the rework time to weld between the Pre-Erected(PE) Block and Erected(ER) block as well as the precision of block construction. In detail, two algorithms were applied to increase the efficiency of estimation process. The first one is K-mean clustering algorithm which is used to separate only the related contact point set from others not related with welding sections. The second one is the Concave hull algorithm which also separates the inner point of the contact section used for the delayed outfitting and stiffeners section, and constructs the concave outline of contact section as the primary objects to estimate the rework time of welding. The main purpose of this paper is that the rework cost for welding is able to be obtained easily and precisely with the defective point set. The point set on the blocks' outline are challenging to get the approximated mathematical curves, owing to the lots of orthogonal parts and lack of number of point. To solve this problems we compared the Radial based function-Multi-Layer(RBF-ML) and Akima interpolation method. Collecting the proposed methods, the paper suggested the noble point matching method for minimizing the rework time of block-welding on the dock, differently the previous approach which had paid the attention of only the degree of accuracy.

• 한국정보전자통신기술학회논문지
• /
• 제12권3호
• /
• pp.242-250
• /
• 2019

본 논문에서는 인공신경망(Artificial Neural Network) 모델 중, 시계열 데이터의 변환을 위한 모델인 Seq2Seq(Sequence to Sequence) 모델을 이용한 산업용 로봇 고장 예지 기술에 대하여 제안한다. 제안 방법은 고장 예지를 위한 추가적인 센서의 부착 없이 로봇 자체적으로 측정 가능한 관절 별 전류와 각도 값을 데이터로 사용하였고, 측정된 데이터를 모델이 학습할 수 있도록 전처리한 후, Seq2Seq 모델을 통해 전류를 각도로 변환하도록 지도 학습 하였다. 고장 진단을 위한 이상 정도(Abnormal degree)는 예측 각도와 실제 각도 간의 단위시간 동안의 RMSE(Root Mean Squared Error)를 사용하였다. 제안 방법의 성능평가는 로봇의 정상 및 결함 조건을 달리한 상태에서 측정한 테스트 데이터를 이용하여 수행되었고 이상 정도가 임계값 넘어가면 고장으로 분류하게 하여, 실험으로부터 96.67% 고장 진단 정확도를 보였다. 제안 방법은 별도의 추가적인 센서 없이 고장 예지 수행이 가능하다는 장점이 있으며, 로봇에 대한 깊은 전문지식을 요구하지 않으면서 수행할 수 있는 방법으로 높은 진단 성능과 효용성을 실험으로부터 확인하였다.

• 구강회복응용과학지
• /
• 제24권1호
• /
• pp.1-17
• /
• 2008

이 연구의 목적은 고정성 보철물의 제거와 지대치의 생존율을 위한 원인을 조사하는 것이었다. 총 192개의 고정성 보철물이 조선대 치과병원에서 제거되었고 308개의 지대치가 조사되었다. 제거의 가장 빈번한 이유는 치주 문제였고(30.7%), 그 다음이 우식(29.7%), 그리고 치근단 문제(18.8%) 였다. 금속도재관에서 치주 문제는 가장 빈번한 제거의 원인이었다. 완전주조관에서는 우식이 제거의 가장 흔한 이유였다. 치근단과 치주의 문제는 각각 40대 이하와 50대 이상의 사람들에서 가장 흔한 원인이었다. 308개의 지대치 중 생활치와 실활치는 각각 135(43.8%)와 173(56.2%) 이었다. 135개의 생활치 중에서 39(28.9%)개가 발거되었고, 173개의 실활치 중에서 85(49.1%)개가 발거되었다. 고정성 보철물의 제거와 지대치의 발거를 위한 각각의 위험 요소는 최종 보철물과 지대치의 예상과 진단을 위해 더 명확히 평가되어야만 한다.