• Journal of Korean Institute of Industrial Engineers
• /
• v.41 no.6
• /
• pp.540-550
• /
• 2015

There have been many attempts to find knowledge from data using conventional statistics, data mining, artificial intelligence, machine learning and pattern recognition. In those research areas, knowledge is approached in two ways. Firstly, researchers discover knowledge represented in general features for universal recognition, and secondly, they discover exceptional and distinctive features. In process mining, an instance is sequential information bounded by case ID, known as process instance. Here, an exceptional process instance can cause a problem in the analysis and discovery algorithm. Hence, in this paper we develop a method to detect the knowledge of exceptional and distinctive features when performing process mining. We propose a method for anomaly detection named Distance-based Anomaly Process Instance Detection (DAPID) which utilizes distance between process instances. DAPID contributes to a discovery of distinctive characteristic of process instance. For verifying the suggested methodology, we discovered characteristics of exceptional situations from log data. Additionally, we experiment on real data from a domestic port terminal to demonstrate our proposed methodology.

• Journal of Korean Society for Quality Management
• /
• v.50 no.3
• /
• pp.459-471
• /
• 2022

Purpose: The purpose of this study was to increase prediction accuracy for an anomaly interval identified using an artificial intelligence-based time series anomaly detection technique by establishing a pre-processing process. Methods: Significant variables were extracted by applying feature selection techniques, and anomalies were derived using the TadGAN time series anomaly detection algorithm. After applying machine learning and deep learning methodologies using normal section data (excluding anomaly sections), the explanatory power of the anomaly sections was demonstrated through performance comparison. Results: The results of the machine learning methodology, the performance was the best when SHAP and TadGAN were applied, and the results in the deep learning, the performance was excellent when Chi-square Test and TadGAN were applied. Comparing each performance with the papers applied with a Conventional methodology using the same data, it can be seen that the performance of the MLR was significantly improved to 15%, Random Forest to 24%, XGBoost to 30%, Lasso Regression to 73%, LSTM to 17% and GRU to 19%. Conclusion: Based on the proposed process, when detecting unsupervised learning anomalies of data that are not actually labeled in various fields such as cyber security, financial sector, behavior pattern field, SNS. It is expected to prove the accuracy and explanation of the anomaly detection section and improve the performance of the model.

• International Journal of Internet, Broadcasting and Communication
• /
• v.9 no.2
• /
• pp.20-26
• /
• 2017

Process control is crucial in many industries, especially in semiconductor manufacturing. In such large-volume multistage manufacturing systems, a product has to go through a very large number of processing steps with reentrant) before being completed. This manufacturing system has many machines of different types for processing a high mix of products. Each process step has specific quality standards and most of them have nonlinear dynamics due to physical and/or chemical reactions. Moreover, many of the processing steps suffer from drift or disturbance. To assure high stability and yield, on-line quality monitoring of the wafers is required. In this paper we develop a real-time fault detection system on semiconductor manufacturing process. Proposed system is superior to other incremental fault detection system and shows similar performance compared to batch way.

• Proceedings of the KSME Conference
• /
• 2001.11a
• /
• pp.821-826
• /
• 2001

Out of all metal-cutting processes, the hole-making process is the most widely used. It is estimated to be more than 30% of the total metal-cutting process. It is therefore desirable to monitor and detect drill wear during the hole-drilling process. One important aspect in controlling the drilling process is monitoring drill wear status. Accordingly, this paper deals with Basic system and Online system. Basic system comprised of spindle rotational speed, feed rates, thrust, torque and flank wear measured tool microscope. Online system comprised of spindle rotational speed, feed rates, AE signal, flank wear area measured computer vision. On-line monitoring system does not need to stop the process to inspect drill wear. Backpropagation neural networks (BPNs) were used for on-line detection of drill wear. This paper deals with an on-line drill wear monitoring system to fit the detection of the abnormal tool state.

• The Korean Journal of Applied Statistics
• /
• v.34 no.1
• /
• pp.99-114
• /
• 2021

With the Fourth Industrial Revolution based on new technology, the semiconductor manufacturing industry researches various analysis methods such as detecting process abnormalities and predicting yield based on equipment sensor data generated in the manufacturing process. The semiconductor manufacturing process consists of hundreds of processes and thousands of measurement processes associated with them, each of which has properties that cannot be defined by chemical or physical equations. In the individual measurement process, the actual measurement ratio does not exceed 0.1% to 5% of the target product, and it cannot be kept constant for each measurement point. For this reason, efforts are being made to determine whether to manage by using equipment sensor data that can indirectly determine the normal state of each step of the process. In this study, the Functional Data Analysis (FDA) was proposed to define a process abnormality detection process based on equipment sensor data and compensate for the disadvantages of the currently applied statistics-based diagnosis method. Anomaly detection accuracy was compared using machine learning on actual field case data, and its effectiveness was verified.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.4
• /
• pp.9-15
• /
• 2016

Symmetry is everywhere in the world around us from galaxy to microbes. From ancient times symmetry is considered to be a reflection of the harmony of universe. Symmetry is not only a significant clue for human cognitive process, but also useful information for computer vision such as image understanding system. Application areas include face detection and recognition, indexing of image database, image segmentation and detection, analysis of medical images, and so on. The technique used in this paper extracts edges, and the perpendicular bisector of any two edge points is considered to be a candidate axis of symmetry. The coefficients of candidate axis are accumulated in the coefficient space. Then the axis of symmetry is determined to be the line for which the coefficient histogram has maximum value. In this paper, an improved method is proposed that utilizes the directional information of edges, which is a byproduct of the edge detection process. Experiment on 20 test images shows that the proposed method performs 22.7 times faster than the original method. In another test on 5 images with 4% salt-and-pepper noise, the proposed method detects the symmetry successfully, while the original method fails. This result reveals that the proposed method enhances the speed and accuracy of detection process at the same time.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2000.11b
• /
• pp.279-286
• /
• 2000

The objective of this research is to detect the papaya fruits on tree in an orchard. The detection of papaya on natural background is difficult because colors of fruits and background such as leaves are similarly green. We cannot separate it from leaves by color information. Therefore, this research will use shape information instead. First, we detect an interested object by detecting its boundary using edge detection technique. However, the edge detection will detect every objects boundary in the image. Therefore, shape description technique will be used to describe which one is the interested object boundary. The good shape description should be invariant in scaling, rotating, and translating. The successful concept is to use Fourier series, which is called "Fourier Descriptors". Elliptic Fourier Descriptors can completely represent any shape, which is selected to describe the shape of papaya. From the edge detection image, it takes a long time to match every boundary directly. The pre-processing task will reduce non-papaya edge to speed up matching time. The deformable template is used to optimize the matching. Then, clustering the similar shapes by the distance between each centroid, papaya can be completely detected from the background.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.4
• /
• pp.506-516
• /
• 1998

In this paper, an algorithm is proposed to detect the malfunction of plasma-etching characteristics using EPD signal trajectories. EPD signal trajectories offer many information on plasma-etching process state, so they must be considered as the most important data sets to predict the wafer states in plasma-etching process. A recent work has shown that EPD signal trajectories were successfully incorporated into process modeling through critical parameter extraction, but this method consumes much effort and time. So Principal component analysis(PCA) can be applied. PCA is the linear transformation algorithm which converts correlated high-dimensional data sets to uncorrelated low-dimensional data sets. Based on this reason neural network model can improve its performance and convergence speed when it uses the features which are extracted from raw EPD signals by PCA. Wafer-state variables, Critical Dimension(CD) and uniformity can be estimated by simulation using neural network model into which EPD signals are incorporated. After CD and uniformity values are predicted, proposed algorithm determines whether malfunction values are produced or not. If malfunction values arise, the etching process is stopped immediately. As a result, through simulation, we can keep the abnormal state of etching process from propagating into the next run. All the procedures of this algorithm can be performed on-line, i.e. wafer to wafer.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.18 no.12
• /
• pp.38-44
• /
• 2019

Various automation studies have been conducted to detect defective products based on product images. In the case of machine vision-based studies, size and color error are detected through a preprocessing process. A situation may arise in which the main features are removed during the preprocessing process, thereby decreasing the accuracy. In addition, complex systems are required to detect various kinds of defects. In this study, we designed and developed a system to detect errors by analyzing various conditions of defective products. We designed the deep learning algorithm to detect the defective features from the product images during the automation process using a convolution neural network (CNN) and verified the performance by applying the algorithm to the checker-switch failure detection system. It was confirmed that all seven error characteristics were detected accurately, and it is expected that it will show excellent performance when applied to automation systems for error detection.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.28 no.1
• /
• pp.100-107
• /
• 2018

Objectives: This study examines exposure to hazardous substances in the working environment caused by exposure to toxic substances produced in the aluminum die casting process in the automobile manufacturing industry. Materials and Methods: The exposure concentration levels, detection rates and time-trend of 15 hazardous factors in the aluminum die casting process over 10 years(from 2006 to 2016) were used as a database. Results: The study found that hazardous factors in the aluminum die casting process were mostly metals. The rate for detected samples was 70.6%(405 samples), and that for not detected samples was 29.4%. The noise for an eight-hour work shift showed a 49.7% exceedance rate for TLV-TWA. Average noise exposure was 89.0 dB. The maximum exposure level was 105.1 dB. Conclusion: The high numbers of no-detection rates for hazardous substance exposure shows that there is no need to do a work environment measurement. Therefore, alternatives are necessary for improving the efficiency and reliability of the work environment measurement. Moreover, to prevent noise damage, reducing noise sources from automation, shielding, or sound absorbents are necessary.