• 품질경영학회지
• /
• 제44권4호
• /
• pp.751-770
• /
• 2016

Purpose: Two-stage ${\bar{X}}$ chart is a useful tool for process control when a surrogate variable may be used together with a performance variable. This paper extends the two-stage ${\bar{X}}$ chart to a three stage version by decomposing the first stage into the preliminary stage and the main stage. Methods: The expected cost function is derived using Markov-chain approach. The optimal designs are found for numerical examples using a genetic algorithm combined with a pattern search algorithm and compared to those of the two-stage ${\bar{X}}$ chart. Sensitivity analysis is performed to see the parameter effects. Results: The proposed design outperforms the optimal design of the two-stage ${\bar{X}}$ chart in terms of the expected cost per unit time unless the correlation between the performance and surrogate variables is modest and the shift in process mean is smallish. Conclusion: Three-stage ${\bar{X}}$ chart may be a useful alternative to the two-stage ${\bar{X}}$ chart especially when the correlation between the performance and surrogate variables is relatively high and the shift in process mean is on the small side.

• Management Science and Financial Engineering
• /
• 제6권1호
• /
• pp.1-19
• /
• 2000

This papeer proposes a general approach of the multivariate expontially weighted moving average(MEWMA) chart, in which the smoothing matrix has full elements instead of only diagonal elements. The average run length (ARL) properties of this scheme are examined for a diverse set of quality control environments and the information to design the chhart is provied. Performance of the scheme is measured by estmating ARL and compared to those of two group cumulative sum (CUSUM) chats. The comparison resullts show that the MEWMA chart can improve its ARL performance in detecting a small shifts out-of-control in the start-up stage, the general MEWMA chart of a full smoothing matrix appears to offer an exceptional protection aginst departures from control in the process mean.

• 품질경영학회지
• /
• 제28권2호
• /
• pp.70-88
• /
• 2000

Short runs where it is neither possible nor practical to obtain sufficient subgroups to estimate accurately the control limit are common in modem business environments. In this study, the standardized control chart, Hillier's exact method, Q chart, EWMA(Exponentially Weighted Moving Average) chart for Q statistics and EWMA chart for mean and absolute deviation among many SPC(Statistical Process Control) techniques for short runs have been reviewed and advantages and disadvantages of these techniques are discussed. The simulation experiments to compare performances of these variable charts for process mean and variations are conducted for combination of subgroup size, scale and timing of shifts of process mean an/or standard deviation. Based upon simulation results, some guidelines for practitioners to choose short run SPC techniques are recommended.

• 품질경영학회지
• /
• 제40권2호
• /
• pp.176-185
• /
• 2012

After a work of Derman and Ross(1997) that considered simple main runs rules and derived ARL (Average Run Length) using Markov chain modeling, $\bar{X}$ control chart based on diverse alternative main and supplementary runs rules that is the most popular control chart for monitoring the mean of a process are proposed. This paper reviews and discusses the-state-of-art researches for these runs rules and classifies according to several properties of runs rules. ARL derivation for a proposed runs rule is also illustrated.

• 응용통계연구
• /
• 제20권3호
• /
• pp.597-604
• /
• 2007

현대 사회에서 건강관리와 공중보건 분야의 이상원인 탐지와 감시를 위하여 관리도 기법을 많이 사용하고 있다 예를 들어, 외과의사의 수술 수행성과의 변화나 만성병과 전염병 등의 비율 변화를 탐지하는데 관리도를 사용할 수 있다. 이 논문에서는 건강관리와 공중보건 분야에서 많이 사용하는 VLAD (variable life-adjusted display) 관리도를 소개하고, 이 관리도에서 관리한계선을 설정하는 방법을 제안하였다.

• 품질경영학회지
• /
• 제26권1호
• /
• pp.126-142
• /
• 1998

Traditional statistical process control(SPC) assumes that consective observations from a process are independent. In industrial practice, however, observations are ofter serially correlated. A common a, pp.oach to building control charts for autocorrelatd data is to a, pp.y classical SPC to the residuals from a time series model fitted. Unfortunately, one cannot completely escape the effects of autocorrelation by using charts based on residuals of time series model. For the detection of variance change in the process we propose a CUSUM of squares control chart which does not require the model identification. The proposed CUSUM of squares chart and the conventional control charts are compared by a Monte Carlo simulation. It is shown that the CUSUM of squares chart is more effective in the presence of dependency in the processes.

• 응용통계연구
• /
• 제26권6호
• /
• pp.903-913
• /
• 2013

최근 소셜 네크워크 서비스의 발달로 인해 개인의 감정이나 의견을 표현하는 소셜 데이터들이 하루에도 수백만 건씩 생산되고 있다. 또한 소셜 데이터는 개인의 의견에 또 다른 생각을 더하는 등 정보의 생산과 소비가 누구나 가능해짐으로써 사회현상을 잘 반영해주는 도구로 성장하고 있다. 본 연구에서는 블로그에 올라온 부정적인 감성어들을 분석하여 블로거의 감성변화를 탐지하기 위해 다변량 관리도를 이용하고자 한다. 이를 위해 2008년 1월 1일부터 2009년 12월 31일 사이에 생성되었던 모든 블로그를 사용하였다. 품질 특성치가 다변량으로 주어지는 경우 호텔링의 $T^2$ 관리도가 널리 사용된다. 그러나 이 관리도는 품질 특성치들의 분포가 다변량 정규분포라는 가정을 하고 있어, 비정규 다변량 자료에 대한 관리도의 성능은 좋지 않다. 이에 본 논문에서는 Sun과 Tsung (2003)이 제안한 써포트 벡터머신에서 단일 집합 분류 기법 중 하나인 SVDD(support vector data description) 알고리즘과 이를 확장한 K-관리도를 소개하고, 실제 데이터 분석에 적용해 보았다.

• 품질경영학회지
• /
• 제30권4호
• /
• pp.1-14
• /
• 2002

Traditional SPC techniques are looking out variation of process by fixed sampling interval and fixed sample size about every hour, the process of in-control or out-of-control couldn't be detected actually when the sample points are plotted near control limits, and it takes no notice of expense concerned with such sample points. In this paper, to overcome that, consider VSI(variable sampling interval) EWMA control charts which VSI method is applied. The VSI control charts use a short sampling internal if previous sample points are plotted near control limits, then the process has high probability of out-of-control. But it uses a long sampling interval if they are plotted near centerline of the control chart, since process has high possibility of in-control. And then a comparison and analysis between FSI(fixed sampling interval) and VSI EWMA in the statistical aspect and economic aspect is studied. Finally, we show that VSI EWMA control chart is more efficient than FSI EWMA control chart in the both aspects.

• 응용통계연구
• /
• 제19권1호
• /
• pp.81-96
• /
• 2006

본 연구에서는 분포를 알 수 없고 비대칭인 공정자료에 실용적으로 적용할 수 있는 간단한 관리도법을 제안하였다. 비대칭 분포를 따르는 공정자료에 정규성 가정에 기초한 슈하르트 관리도를 그대로 적용하면 비대칭성이 증가할수록 제 1종 오류를 범할 확률이 증가할 가능성이 높아지며 변동을 관리하는데 효율성이 떨어지게 된다. 이러한 문제를 해결하기 위해 본 연구에서 제시한 관리도는 관리한계선을 사분위수에 기초하여 정하는 방안을 제시하고 있다. 이러한 방법으로 관리한계선을 그릴 경우 제 1종 오류도 감소하게 되고, 비대칭분포를 하는 공정자료에 대하여 매우 실용적이라고 하겠다.

• 한국의류학회지
• /
• 제24권7호
• /
• pp.1044-1055
• /
• 2000

The purpose of this study is to suggest sizing chart for a clean room wear. 3 control dimensions(Stature, Bust girth, B.N.P.∼Wrist point length) were chosen as 3 axes of clean room wear size chart. A loss function was used to determined intervals of stature, Bust girth and B.N.P.∼Wrist point length of size chart, because the loss function introduces the concept of frequency to size chart for better customer's size satisfaction. From the size table whose intervals had been determined by a loss function. The 4 sizes individually were suggested for clean room wear size chart by sex. The 3 sizes individually were suggested for clean room head cover size chart by sex too. The suggested size chart would be considered more feasible than present size chart. Also they are suggested supply reference measurement chart relevant to clean room wear manufacturing for 13 most frequent sizes.