• Journal of the Korean Data and Information Science Society
• /
• 제19권2호
• /
• pp.609-620
• /
• 2008

There may be several parallel equipments having the same function in a multi-stage manufacturing process, which affect the product quality differently and have significant differences in defect rate. The product quality may depend on what equipments it has been processed as well as what process variable values it has. Applying one model ignoring the presence of different equipments may distort the prediction of defect rate and the identification of important quality variables affecting the defect rate. We propose a procedure for data segmentation when constructing models for predicting the defect rate or for identifying major process variables influencing product quality. The proposed procedure is based on the principal component analysis and the analysis of variance, which demonstrates a better performance in predicting defect rate through a case study with a PDP manufacturing process.

• 품질경영학회지
• /
• 제31권4호
• /
• pp.219-225
• /
• 2003

Two­dimensional warranty data can be modelled using two different approaches： two­dimensional point process and one­dimensional point process with usage as a function of age. The first approach has three different models. First of all, bivariate model is appealing but is not appropriate for explaining warranty claims. Next, the rest of the two models (marked point process, and counting and matching on both directions independently) are more appropriate for explaining warranty claims. However, the second one (counting and matching on both directions independently) assumes that the two variables (variables representing the two­dimensions) are independent. Last of all, one­dimensional point process with usage as a function of age is also promising to explain the two­dimensional warranty claims. But the models or variations of them need more investigation to be applicable to real warranty claim data.

• 인터넷정보학회논문지
• /
• 제19권6호
• /
• pp.83-89
• /
• 2018

워크플로우 프로세스 인텔리전스와 시스템에서 워크플로우 프로세스 마이닝 및 분석 문제가 중요해지고 있다. 워크플로우 프로세스 인텔리전스의 품질을 향상시키기 위해서는 워크플로우 프로세스 마이닝 및 분석을 수행할 때, 워크플로우 실행 이벤트 로그를 저장하는 효율적이고 효과적인 데이터 센터가 필수적이다. 본 논문에서는 워크플로우 이벤트 로그 데이터 센터를 효율적으로 구성하고 XES 형식으로 워크플로우 프로세스 실행 이벤트 로그를 효과적으로 저장하기 위한 3차원 프로세스 기반 데이터 큐브를 제안한다. 이의 검증 단계로서, 프로세스 기반 데이터 큐브가 워크플로우 프로세스 패턴과 해당 워크플로우 프로세스 실행 이벤트 내역에서 실행 비율 및 업무전달관계와 같은 분석적 지식을 발견하는데 얼마나 적합한지를 보여주기 위해 프로세스 마이닝 실행 예제를 제시한다. 결과적으로, 프로세스 기반 데이터 큐브와 이를 활용한 프로세스 마이닝 시스템의 구현을 통해, 워크플로우 프로세스의 기본적 제어흐름 패턴을 성공적으로 발견할 수 있음을 확인했다.

• 대한산업공학회지
• /
• 제41권6호
• /
• 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.

• 산업경영시스템학회지
• /
• 제33권1호
• /
• pp.33-40
• /
• 2010

Because sampling interval for data collection tends to be short compared with the overall processing time, in chemical process, instrumental process related tanks or furnace collected data have a significant autocorrelation. Insufficient control technique and frequent control actions cause unstable condition of the process. Traditional control charts which were developed based on iid (independently and identically distributed) among data cannot be applied on the existence of autocorrelation. Also unstable process is difficult to identity or diagnose. Because large-scale process has a lot of measurable variables and multi-step-structures among data, it is difficult to find relation between measurable variables and nonconformity. In this paper, we suggested an appicable model to diagnose the process and to find relation between measurable variables (CTQ) and nonconformity in the process having autocorrelation, unstable condition frequently, a lot of measurable variables, and multi-step-structure. And we applied this model to real process, to verify that the process engineers could easily and effectively diagnose the process and control the nonconformity.

• 융복합기술연구소 논문집
• /
• 제3권1호
• /
• pp.41-44
• /
• 2013

Process and manufacturing data are numerously accumulated to the enterprise database in industries but little of those data are utilized. Data mining can support a decision to manager in process from the data. However, it is not easy to field managers because a proper adoption of various schemes is very difficult. In this paper, six scenarios are conducted using data mining schemes for the various situations of field claims such as yield problem, trend analysis and prediction of yield according to changes of operating conditions, etc. Scenarios, like templates, of various analysis situations are helpful to users.

• 산업경영시스템학회지
• /
• 제25권6호
• /
• pp.1-6
• /
• 2002

Most of process data control have been designed under the assumption that there are independence between observed data. However, it has been difficult to apply the traditional method to realtime data because they are autocorrelated, and they are not normally distributed. And the more, they have fluctuating means. Already the control method for these data was proposed by Markov Modulated Gaussian Process. Therefore, this study take into account MMGP's traits especially for the MMGP's autocorrelation.

• Journal of the Korean Data and Information Science Society
• /
• 제16권2호
• /
• pp.401-410
• /
• 2005

In this paper, we calculate the premium rate of reliability insurance policy for brake pads for automobiles using real failure data obtained from use-condition. We try process capability analysis for the manufacturing process of brake-system. We describe the performance factors which have an effect on failure characteristics of brake pads. We also obtain the maximum likelihood estimates of shape and scale parameters of the fitted Weibull distribution for brake pads.

• 한국컴퓨터정보학회논문지
• /
• 제11권6호
• /
• pp.59-67
• /
• 2006

비즈니스 프로세스 관리도구인 BPM을 SCM 부문에 적용하면 효율적인 프로세스 관리 및 제어가 가능하다. 또한 BPM은 SCM을 구성하는 프로세스를 효과적으로 통합하여 실행시킬 수 있다. 이러한 접근 방법은 SCM 프로세스의 진행과정을 보다 효율적으로 관리하고 모니터링 할 수 있도록 한다. 또한 프로세스 수행결과를 분석하여 프로세스의 개선에 대한 방안을 수립할 수 있다. 이에 본 논문에서는 이러한 BPM을 SCM 환경에 도입한다 아울러 SCM 프로세스를 효과적으로 통합 실행하고 업무 프로세스를 개선하는 방안을 데이터마이닝 기법을 적용하여 제시한다.

• 한국빅데이터학회지
• /
• 제3권1호
• /
• pp.47-54
• /
• 2018

비즈니스 프로세스 분석의 기존 연구들은 비즈니스 프로세스에 포함된 업무, 고객 서비스, 작업자 편의, 수행시간 예측 등 다양한 요소를 분석하였다. 이러한 요소를 정확히 분석하기 위해서는 정보시스템에 기록된 실제 이력 데이터를 활용하는 것이 효과적이다. 프로세스 마이닝은 이벤트 로그 데이터로부터 비즈니스 프로세스의 여러 가지 요소를 분석하는 기법이다. 본 사례 연구는 구매 대행 업체의 업무 수행 데이터에 프로세스 마이닝를 적용하여 구매 대행 프로세스의 업무 흐름, 수행 시간, 담당자 등의 프로세스 운영 분석을 수행하였다.