• Journal of the Korea Society of Computer and Information
• /
• v.26 no.8
• /
• pp.23-30
• /
• 2021

In this study, we present a comparative analysis of major autoencoder(AE)-based anomaly detection methods for quality determination in the manufacturing process and a new anomaly discrimination criterion. Due to the characteristics of manufacturing site, anomalous instances are few and their types greatly vary. These properties degrade the performance of an AI-based anomaly detection model using the dataset for both normal and anomalous cases, and incur a lot of time and costs in obtaining additional data for performance improvement. To solve this problem, the studies on AE-based models such as AE and VAE are underway, which perform anomaly detection using only normal data. In this work, based on Convolutional AE, VAE, and Dilated VAE models, statistics on residual images, MSE, and information entropy were selected as outlier discriminant criteria to compare and analyze the performance of each model. In particular, the range value applied to the Convolutional AE model showed the best performance with AUC PRC 0.9570, F1 Score 0.8812 and AUC ROC 0.9548, accuracy 87.60%. This shows a performance improvement of an accuracy about 20%P(Percentage Point) compared to MSE, which was frequently used as a standard for determining outliers, and confirmed that model performance can be improved according to the criteria for determining outliers.

• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.1-11
• /
• 2018

Multivariate process control has become important in various applied fields. For instance, there are many situations in which the simultaneous monitoring of multivariate quality characteristics is necessary for the manufacturing industry. Despite its importance, its practical usage is not as convenient because it is difficult to identify the source of the out-of-control signal in a multivariate control chart. In this paper, we will introduce how to detect the source of the out-of-control by using confidence intervals for new observations, and will discuss the identification and interpretation of the out-of-control variable through simulation studies.

• The Korean Journal of Applied Statistics
• /
• v.31 no.6
• /
• pp.811-823
• /
• 2018

Multivariate control chart based on Hotelling's $T^2$ statistic is a powerful tool in statistical process control for identifying an out-of-control process. It is used to monitor multiple process characteristics simultaneously. Detection of the out-of-control signal with the $T^2$ chart indicates mean vector shifts. However, these multivariate signals make it difficult to interpret the cause of the out-of-control signal. In this paper, we review methods of signal interpretation based on the Mason, Young, and Tracy (MYT) decomposition of the $T^2$ statistic. We also provide an example on how to implement it using R software and demonstrate simulation studies for comparing the performance of these methods.

• The Korean Journal of Applied Statistics
• /
• v.37 no.2
• /
• pp.191-207
• /
• 2024

Many studies have been conducted to quickly detect out-of-control situations in autocorrelated processes. The most traditionally used method is a residual control chart, which uses residuals calculated from a fitted time series model. However, many procedures for monitoring autocorrelated processes using statistical learning methods have recently been proposed. In this paper, we propose a monitoring procedure using the latent vector of LSTM Autoencoder, a deep learning-based unsupervised learning method. We compare the performance of this procedure with the LSTM Autoencoder procedure based on the reconstruction error, the RNN classification procedure, and the residual charting procedure through simulation studies. Simulation results show that the performance of the proposed procedure and the RNN classification procedure are similar, but the proposed procedure has the advantage of being useful in processes where sufficient out-of-control data cannot be obtained, because it does not require out-of-control data for training.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2010.08a
• /
• pp.241-241
• /
• 2010

적외선 센서는 빛의 유무에 관계없이 물체 또는 인체에서 발산하는 적외선을 감지한다. 이러한 센서를 전자 및 디스플레이 시스템과 연동하면 열영상 시스템이 되는데, 이는 전방 감시, 플랜트 감시, 보안, 방범용으로 많이 사용되며, 특히 자동차 야간 운전자 보조용으로 사용되어 최첨단, 고부가가치를 지니고 있는 핵심부품이다. 비냉각형 적외선 센서인 마이크로볼로미터는 상온에서 작동하므로 극저온 Cooler가 불필요하며, 무게와 부피가 작아 각종 시스템에 부착가능하다. 특히 볼로미터형 적외선 센서는 용량이 적은 TE cooler로 상온으로 안정화를 시키며, 진공으로 유지되는 금속 또는 세라믹 패키지를 사용하게 된다. 본 연구에서는 마이크로 볼로미터용 진공패키지를 제작하여 패키지 조립 및 측정기술에 대해 조사하였다. 패키지는 금속재질인 kovar를 사용하여 제작되었고, 내부에 TE Cooler와 장수명 진공유지를 위한 getter, 그리고 온도센서 및 볼로미터 센서 칩을 장착하여 조립하였다. 패키지 Cap ass'y와 base envelop의 솔더링 공정은 약 $200^{\circ}C$에서 수행하였으며, evacuation system을 이용하여 5일 동안 패키지 bake-out 공정을 수행하였다. 이 후 getter를 활성화시키고, seal-off 공정으로 진공 기밀을 유지하였다. 진공 패키지의 기밀성은 $6{\times}10^{-9}\;std.cm^3/sec$로 기밀성을 유지하였다. 볼로미터 센서의 반응도는 $10^2\;V/W$ 이상을 나타내었으며, 탐지도는 $2{\times}10^8\;cm-Hz^{1/2}/W$를 나타내었다.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2010.02a
• /
• pp.252-252
• /
• 2010

적외선 센서는 빛의 유무에 관계없이 주 야간 전방의 물체에서 발산하는 미약한 적외선(열선)을 감지하여 영상으로 재현하는 열상시스템은 자동차 야간 운전자 보조용 나이트 비젼, 핵심 시설의 감시 관리, 군수 등의 분야에 적용되어지고 있는 최첨단, 고부가가치를 지니고 있는 기술이다. 양자형은 센서 특성은 좋으나 냉각기(작동온도: $-196^{\circ}C$) 및 고진공 패키지인 dewar를 사용하는 반면에, 열형은 대부분 상온에서 동작되는 온도안정화를 위한 전자냉각모듈만을 구비하면 되므로 저가형으로 제작이 가능한 비냉각형 적외선 센서이다. 본 연구에서는 적외선 센서용 진공패키지 조립공정 및 패키지된 센서의 측정기술을 개발하였다. 금속 메탈패키지를 제작하였으며, 금속 진공패키지는 소자냉각용 TE Cooler와 장수명 진공유지를 위한 getter, 그리고 센서칩, 온도센서를. 장착하여 칩을 조립하였다. Cap ass'y와 base envelop의 솔더링 공정을 수행하였으며, 진공패키지의 진공유지를 위해 TMP를 이용하여 진공을 유지하고, 약 5일동안 패키지 bake-out을 수행하였다. 진공압력은 $10^{-7}\;torr$ 이하를 유지하였으며, getter를 활성화시키고, pinch-off 공정으로 조립 ass'y를 완성하였다. 진공 패키지의 기밀성은 He leak tester를 이용하여 측정하였으며, ${\sim}10^{-9}\;std.cm^3/sec$로 기밀성을 유지하였다. TE cooler를 작동한 온도안정성은 0.05 K 이하였다. 볼로미터 센서의 반응도는 $10^2\;V/W$ 이상을 나타내었으며, 탐지도는 $2{\times}10^8cm-Hz^{1/2}/W$를 나타내었다. 본 연구를 통하여 얻어진 결과는 향후 2차원 열영상용 어레이 검출기 및 웨이퍼수준의 패키징 공정에 유용하게 응용될 것으로 판단된다.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.36 no.7
• /
• pp.813-822
• /
• 2012

Abnormal operating sounds radiated from a moving transfer robot in LCD (liquid crystal display) product lines have been used for the fault detection line of a robot instead of other source signals such as vibrations, acoustic emissions, and electrical signals. Its advantage as a source signal makes it possible to monitor the status of multiple faults by using only a microphone, despite a relatively low sensitivity. The wavelet packet transform for feature extraction and the artificial neural network for fault classification are employed. It can be observed that the abnormal operating sound is sufficiently useful as a source signal for the fault diagnosis of mechanical components as well as other source signals.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.2
• /
• pp.82-89
• /
• 2013

Novelty detection (ND) is an effective technique that can be used to determine whether a future observation is normal or not. In the present study we propose a novelty detection algorithm that can handle a situation where the distributions of target (normal) observations are inhomogeneous. A simulation study and a real case with the TFT-LCD process demonstrated the effectiveness and usefulness of the proposed algorithm.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1228-1229
• /
• 2023

현대 도시 모빌리티의 중요한 구성 요소로 자리 잡은 전동킥보드는 편리한 이동 수단으로 인기를 얻고 있으나, 이에 따른 안전사고 증가로 운전자와 보행자의 안전이 심각하게 위협받고 있다. 본 논문에서는 전동킥보드 운전 중에 발생한 사고의 원인을 객관적으로 분석하고, 사고가 운전자의 부주의로 인한 것인지를 판별하며, 이로 인한 배상 책임을 정확하게 결정하기 위한 모델을 제안한다. 운전 중 수집된 센서 데이터를 활용하여 Deep SVDD (Deep Support Vector Data Description) 모델을 구축하고, 이상치 탐지를 통해 운전 패턴을 분류하며 운전자의 부주의로 인한 사고를 파악한다. 이를 통해, 정확하고 공정한 배상 책임 판단을 지원하며, 도시 모빌리티 분야에서 안전사고 감소에 기여할 것으로 기대된다.

• Journal of Intelligence and Information Systems
• /
• v.20 no.3
• /
• pp.93-108
• /
• 2014

To support business decision making, interests and efforts to analyze and use transaction data in different perspectives are increasing. Such efforts are not only limited to customer management or marketing, but also used for monitoring and detecting fraud transactions. Fraud transactions are evolving into various patterns by taking advantage of information technology. To reflect the evolution of fraud transactions, there are many efforts on fraud detection methods and advanced application systems in order to improve the accuracy and ease of fraud detection. As a case of fraud detection, this study aims to provide effective fraud detection methods for auction exception agricultural products in the largest Korean agricultural wholesale market. Auction exception products policy exists to complement auction-based trades in agricultural wholesale market. That is, most trades on agricultural products are performed by auction; however, specific products are assigned as auction exception products when total volumes of products are relatively small, the number of wholesalers is small, or there are difficulties for wholesalers to purchase the products. However, auction exception products policy makes several problems on fairness and transparency of transaction, which requires help of fraud detection. In this study, to generate fraud detection rules, real huge agricultural products trade transaction data from 2008 to 2010 in the market are analyzed, which increase more than 1 million transactions and 1 billion US dollar in transaction volume. Agricultural transaction data has unique characteristics such as frequent changes in supply volumes and turbulent time-dependent changes in price. Since this was the first trial to identify fraud transactions in this domain, there was no training data set for supervised learning. So, fraud detection rules are generated using outlier detection approach. We assume that outlier transactions have more possibility of fraud transactions than normal transactions. The outlier transactions are identified to compare daily average unit price, weekly average unit price, and quarterly average unit price of product items. Also quarterly averages unit price of product items of the specific wholesalers are used to identify outlier transactions. The reliability of generated fraud detection rules are confirmed by domain experts. To determine whether a transaction is fraudulent or not, normal distribution and normalized Z-value concept are applied. That is, a unit price of a transaction is transformed to Z-value to calculate the occurrence probability when we approximate the distribution of unit prices to normal distribution. The modified Z-value of the unit price in the transaction is used rather than using the original Z-value of it. The reason is that in the case of auction exception agricultural products, Z-values are influenced by outlier fraud transactions themselves because the number of wholesalers is small. The modified Z-values are called Self-Eliminated Z-scores because they are calculated excluding the unit price of the specific transaction which is subject to check whether it is fraud transaction or not. To show the usefulness of the proposed approach, a prototype of fraud transaction detection system is developed using Delphi. The system consists of five main menus and related submenus. First functionalities of the system is to import transaction databases. Next important functions are to set up fraud detection parameters. By changing fraud detection parameters, system users can control the number of potential fraud transactions. Execution functions provide fraud detection results which are found based on fraud detection parameters. The potential fraud transactions can be viewed on screen or exported as files. The study is an initial trial to identify fraud transactions in Auction Exception Agricultural Products. There are still many remained research topics of the issue. First, the scope of analysis data was limited due to the availability of data. It is necessary to include more data on transactions, wholesalers, and producers to detect fraud transactions more accurately. Next, we need to extend the scope of fraud transaction detection to fishery products. Also there are many possibilities to apply different data mining techniques for fraud detection. For example, time series approach is a potential technique to apply the problem. Even though outlier transactions are detected based on unit prices of transactions, however it is possible to derive fraud detection rules based on transaction volumes.