• 한국전자거래학회지
• /
• 제15권4호
• /
• pp.303-317
• /
• 2010

본 논문에서는 KNN 대체와 LOF 알고리즘의 결합 모델을 확장하여 실시간 비즈니스 프로세스 모니터링을 위한 비정상 종료 예측 방법론을 제안하였다. 기존의 룰 기반 모니터링 방법론은 실시간 프로세스 진행 정도에 따른 비관측 정보에 기인하여 조기 경보 및 실시간 대응이 힘들다는 한계점을 안고 있다. 이를 해결하기 위하여 비관측 정보에 대한 가정 및 진행 중인 프로세스의 향후 경로 예측을 통해 종료 시점에서 예상되는 LOF를 추정하기 위한 알고리즘을 제안하였다. 이 알고리즘을 적용하여 실시간 비즈니스 프로세스 모니터링 과정에서 각 관측 시점마다 종료 시점에서의 결과를 예측함으로써, 전 시점에 걸친 추세를 살펴종료 패턴을 예측할 수 있다. 이를 통해 비즈니스 프로세스의 실시간 진척에 대한 정보를 가시화함으로써 기회 및 위협에 사전에 대응할 수 있게 하여 프로세스 관리 수준의 향상을 기대할 수 있을 것으로 예상된다.

• 한국정밀공학회지
• /
• 제22권6호
• /
• pp.135-143
• /
• 2005

The shoes process management monitoring system for supporting SCM is developed in this study. This system consists of a monitoring program, a conveyer system, a pneumatic unit, a controller, an USB camera, and a server and a client computer. To operate the developed system easily, the monitoring program using LabVIEW in the Windows environment is developed. This program consists of 5 modules: production management, inventory control, media management, defective management, and communication management. The developed system has several advantages: reduced time for managing process work, decreased labor costs, effective operation, and continuous work without an operator. Nowadays advanced manufacturing companies are trying to find a way to check the performance of their production equipments and plants from remote sites. Thus, to manage the developed system from remote sites, communication network is constructed. In order to evaluate the performance of the monitoring system, experiments were performed. The experimental results showed that the developed system provided a reliable performance and a stable communication.

• 인터넷정보학회논문지
• /
• 제21권3호
• /
• pp.83-92
• /
• 2020

최근에 많은 기업 및 조직들이 비즈니스 프로세스 모델의 효율적 운용을 위해 예측적 프로세스 모니터링에 관심이 높아지고 있다. 기존의 프로세스 모니터링은 특정 프로세스 인스턴스의 경과된 실행상태에 초점을 두었다. 반면, 예측적 프로세스 모니터링은 특정 프로세스 인스턴스의 미래의 실행상태에 대한 예측에 초점을 둔다. 본 논문에서는 예측적 프로세스 모니터링 기능 중 하나인 비즈니스 프로세스 인스턴스 실행 잔여시간 예측기능을 구현한다. 잔여시간을 효과적으로 모델링하기 위해 액티비티별 속성에 따른 시간특징 값 분포 차이를 고려하여 액티비티별 특징 정규화를 제안하고 예측모델에 적용한다. 본 논문에서 제안된 모델의 예측성능 우수성을 입증하기 위해서 4TU.Centre for Research Data에서 제공하는 실제 기업의 이벤트 로그 데이터를 통해 선행연구들과 비교평가 한다.

• 산업공학
• /
• 제24권1호
• /
• pp.8-14
• /
• 2011

Principal Component Analysis (PCA) reduces the dimensionality of the process by creating a new set of variables, Principal components (PCs), which attempt to reflect the true underlying process dimension. However, for highly nonlinear processes, this form of monitoring may not be efficient since the process dimensionality can't be represented by a small number of PCs. Examples include the process of semiconductors, pharmaceuticals and chemicals. Nonlinear correlated process variables can be reduced to a set of nonlinear principal components, through the application of Kernel Principal Component Analysis (KPCA). Support Vector Data Description (SVDD) which has roots in a supervised learning theory is a training algorithm based on structural risk minimization. Its control limit does not depend on the distribution, but adapts to the real data. So, in this paper proposes a non-linear process monitoring technique based on supervised learning methods and KPCA. Through simulated examples, it has been shown that the proposed monitoring chart is more effective than $T^2$ chart for nonlinear processes.

• 산업경영시스템학회지
• /
• 제31권1호
• /
• pp.92-100
• /
• 2008

This paper develops a process monitoring system for real-time process control. The practical case is studied on a small and medium marine equipment company. For business process reengineering of the company, we adopt an approach based on information engineering methodology, which consists of four stages : planning, analysis, design, and implementation. The system is developed for Client/Server environment. We discuss the constructing of hardware system for real-time process control at low cost. The developed system is composed of interrelated modules for item master and BOM management, process control, facility management, SQC and work report.

• 품질경영학회지
• /
• 제26권1호
• /
• pp.108-125
• /
• 1998

The EWMA(Exponentially Weighted Moving Average) has recently received a great deal of attention in the quality control literature as a process monitoring tool on the shop floor of manufacturing industires, since it is easy to plot, to interpret, and its control limits are easy to obtain. Most a, pp.ications of the EWMA for process monitoring have concentrated on the problem of detecting shifts of a process mean and a process standard deviation with ARL(Average Run Length) properties. But there may be the necessity of controlling linearity on product quality such as the correlation coefficient to the process operator. Control managers may want to protect the increase of a process correlation coefficient value, such as 0, between two variables of interest. However, there are few studies concerned on this part. Therefore, we propose EWMA models for a process correlation coefficient using two transformed statistics, T-statistic and (Fisher's) Z-statistic. We also present some results of simulation by SAS/IML and compare two models.

• 약학회지
• /
• 제60권2호
• /
• pp.85-91
• /
• 2016

Drying of granules for tablet formulation is one of the important unit operations. The loss on drying method is traditionally used for this purpose. However, it is a time-consuming method, requiring at least 1 h. Moreover, it is ineffective in monitoring the moisture content of granules during the drying process. In this study, online real-time monitoring of moisture content during the drying process was successfully performed using near-infrared (NIR) spectroscopy. NIR spectra were collected during 15 different drying batches for developing a reliable NIR spectroscopic method. Such a large number of batches were used to develop a more robust partial least squares (PLS) model. NIR spectra collected from 12 batches were used for developing the model that was validated by predicting the moisture content of the samples in the remaining 3 batches. The standard errors of predictions (SEPs) in the measurement of batch 1, batch 2, and batch 3 were 0.52%, 0.57%, and 0.56%, respectively. The online NIR spectroscopic method developed in this study was reliable and accurate in monitoring the moisture content during the drying process.

• 한국전자거래학회지
• /
• 제12권2호
• /
• pp.219-231
• /
• 2007

본 논문에서는 복합 이벤트 처리를 적용하여 비즈니스 활동 분석(BAM)을 위한 이벤트 분석 방안을 제안한다. 이는 프로세스 관리자가 프로세스가 종료되기 이전에 발생 가능한 위험을 감지하고 모니터링하기 위하여 실시간에 진행되는 이벤트에 대하여 조기 경보를 제공하기 위하여 개발되었다. 본 연구에서는 의미 있는 위험을 가지는 이벤트를 추출하는, 프로세스 기반의 이벤트 모니터링 과정을 제시하였다. 복합 이벤트 패턴은 과거 누적된 이벤트 로그를 바탕으로 정의되며, 이벤트의 위험도는 그 패턴들에 기반하여 평가된다. 제안된 방법론은 홈쇼핑업체의 서비스 프로세스의 예를 이용하여 설명한다.

• 한국기계가공학회지
• /
• 제21권6호
• /
• pp.89-97
• /
• 2022

Ultrasonic metal welding has been widely used for joining lithium-ion battery tabs. Weld quality monitoring has been an important issue in lithium-ion battery manufacturing. This study focuses on the weld quality monitoring in ultrasonic metal welding with the longitudinal-torsional vibration mode horn developed newly. As the quality of ultrasonic welding depends on welding parameters like pressure, time, and amplitude, the suitable values of these parameters were selected for experimentation. The welds were tested via tensile testing machine and weld strengths were investigated. The dataset collected for performance test was used to train the multi-layer perceptron neural network. The three layer neural network was used for the study and the optimum number of neurons in the first and second hidden layers were selected based on performances of each models. The best models were selected for the horn and then tested to see their performances on an unseen dataset. The neural network models for the longitudinal-torsional mode horn attained test accuracy of 90%. This result implies that proposed models has potential for the weld quality monitoring.

• 한국정밀공학회지
• /
• 제13권12호
• /
• pp.30-37
• /
• 1996

On-line process control equipment for CO$_{2}$ laser cutting is not available for industrial applications. The major part of the industrial laser cutting machines are adjusted off-line by highly educated engineers. The quality inspection of the sample is visual and referred to different quality scales. Due to the lack of automation the potential laser users hesitate to implement the cutting method. The first step toward an automation of the process is the development of a process monitoring system and the research is cincentrated on the area of on-line quality monitoring during CO$_{2}$ laser cutting. The method is based on the detection of the emitted light from the cutting front by photo diode. The developed monitoring system consists of the OP Amplifier, A/D convertor, power supply and PC. The signal from the photo diode has been undertaken from Fourier analysis and statistical analysis with real time. The photograph of striation pattern was taken by metallurgical microscope. As a result, it is possible to predict the striation pattern according to the beam traveling speed.