• 산업경영시스템학회지
• /
• 제26권1호
• /
• pp.7-16
• /
• 2003

At present, company has to produce a product that consumer like with a competitive price, a good quality, and a fitting time to supply. Process control and quality control are very important to supply with a product uniformly and inexpensively. Process control is given much weight in the quality control in manufacturing system. Statistical process controls(SPC) that are used in process generally have major impact on manufacturing, product design activities, and process development potentially. Control charts in statistical process control method can be interpreted the data from quality characteristics in production process and discriminated between chance variation and assignable variation in process. In addition, control chart can be used to monitor the process output and detect when changes in the inputs are required to bring the process back to an in-control state. The models that relate the influential inputs to process outputs help determine the nature and magnitude of the adjustments required. In this paper, the characteristic of product quality is monitored by control chart during the machining process and construction of quality control cycle is considered to divide into two types in this case that different assignable causes lead to shifts having different magnitudes. Then we are intended to find a process shift magnitude which has economical quality cost policy and are considered to quality cost functions to find a process shift magnitude. Those costs are categorized into the well-known categories of prevention, appraisal, and internal failure and external failure. This paper ends with numerical examples that demonstrate the usefulness of the model.

• Journal of the Korean Statistical Society
• /
• 제23권2호
• /
• pp.403-420
• /
• 1994

An economic design of the $\bar{X}-R$ chart using variable sample size (VSS) scheme is proposed in this paper. In this design the sample size at each sampling time changes according to the values of the previous two sample statistics, sample mean and range. The VSS scheme uses large sample if the sample statistics appear near inside the control limits and smaller sample otherwise. The set of process parameters, such as the sampling interval, control limits and the sample sizes, are chosen to minimize the expected cost per hour. The efficiency of the VSS scheme is compared to the fixed sample size one for cases where there is multiple of assignable causes. Percent reductions of the expected cost in the VSS design are calculated for some given sets of cost parameters. It is shown that the VSS scheme improves the confidence of the procedure and performs statistically better in terms of the number of false alarms and the average time to signal, respectively.

• Journal of the Korean Data and Information Science Society
• /
• 제10권2호
• /
• pp.341-357
• /
• 1999

Greenwich and Jahr-Schaffrath(1995) considered a new process incapability index(PII) $C_{pp}$, which modified the useful index $C^{\ast}_{pm}{$ for detecting assignable causes. The new index $C_{pp}$ provides an uncontaminated separation between information concerning the process accuracy and precision while this kind of information separation is not available with the $C^{\ast}_{pm}$ index. In this paper, we will study about the index $C_{pp}$ based on the bootstrap. First, we will prove the consistency of bootstrap deriving the bootstrap asymptotic distribution for our index $C_{pp}$. Moreover, with the consistency of bootstrap, we will construct six bootstrap confidence intervals and compare their performances. Some simulation results, comparison and analysis are provided. In particular, two STUD and ABC bootstrap methods perform significantly better.

• 대한안전경영과학회지
• /
• 제4권3호
• /
• pp.107-122
• /
• 2002

There are two approaches to process control. The one is engineering process control(EPC) which is one of the techniques very widely used in the process industry and based on control theory which aims at keeping the process on target using manipulating variable. The other is statistical process control(SPC) whose main purpose is to look for assignable causes(variability) in the process. To design an integrated or combined scheme of SPC and EPC is gaining recognition in the process experiences for hybrid industry. This paper aims to investigate recent study concerned on the integration of SPC and EPC. First, we consider the difference between SPC and EPC in simple terms and review various models of EPC for integration including evaluation of previous study. Finally, we suggest some prospective research area concerned on the integration of SPC and EPC.

• 한국통신학회논문지
• /
• 제31권10B호
• /
• pp.936-944
• /
• 2006

인터넷통신 서비스업체는, 최상위 기업경영 성과지표중 하나인 고객만족지수 향상을 위해, 전사적 품질경영을 전개하고 있다. 이와 관련, 고객만족지수와 연관성이 강한 기업내부 주요 인터넷통신 네트워크 품질특성에 대한 관리의 필요성이 제기되고 있다. 본 연구는, 인터넷통신 네트워크 품질특성 모니터 링을 위한 I-MR 관리도 활용체계를 제안한다. 네트워크 품질특성이 관리상태인지 모니터링하고, 네트워크 품질특성이 이상변동을 보이면, 이상원인 조사시점을 결정할 목적으로, 일정간격 연속 측정시점별; 1) 개별측정값 (individual measurements): 및 2) 연속 2개 개별측정값간 이동범위 (moving ranges)를 이용한 통계적 품질관리 체계인 I-MR 관리도를 활용한다. 사례분석에서는, 이상변동 발생시점에서 이상원인이 완전히 제거된다는 가정에 근거하여, 네트워크 품질특성의 실현가능한 치대 기대품질개선율이 계산된다. 인터넷통신 2개 Service A. B. Service별 7개 네트워크 품질특성 대상, 11개 측정지역의 기대품질개선율 평균, 표준편차를 분석, 1-MR 관리도 활용체계의 효과를 검토한다. 사례분석 결과, 기대품질개선율 범위는 다음과 같다. 1) Service A, (1) 평균 (0.31%-11.44%), (2) 표준편차 (9.82%-71.10%); 2) Service B, (1) 평균 (0.29%-17.64%), (2)표준편차 (8.05%-60.59%). 그러므로, 본 연구에서 설계·제안된 I-MR 관리도 활용체계는, 네트워크 품질특성의 평균 산포 개선 및 관리상태 유지에 효과적일 것으로 판단된다.