• 산업경영시스템학회지
• /
• 제21권45호
• /
• pp.237-246
• /
• 1998

In most statistical process control(SPC), control charts are used in which samples are taken and a suitable statistic is determined and plotted. In these control charts, control limits, ${\mu}{\pm}textsc{k{\sigma}}$, from which a decision is made are mostly ${\mu}{\pm}3{\sigma}$ and current literature in control charts are mainly concerned with detecting a shift in the mean. Therefore, when $\sigma$ is increased considerably after a long time, using control limits set at the first time causes a great deal of economic loss. In this paper the solutions to determine new control limits which maximizes the profit per unit produced and reduce $\sigma$ to economically optimal level for a certain cost when $\sigma$ is increased due to process deterioration are proposed. By applying new control limits, $\alpha$ error decreases considerably compared to apply initial control limits when $\sigma$ is increased due to process deterioration. Therefore, false alarm investigation cost drops down to the level of initial a error. And also this solution provides useful information regarding replacement of a process when the process is reviewed regularly.

• Journal of the Korean Data and Information Science Society
• /
• 제23권5호
• /
• pp.1037-1044
• /
• 2012

Multivariate control charts for effectively monitoring every component in the dispersion matrix of multivariate normal process are considered. Through the numerical results, we noticed that the multivariate control charts based on sample statistic $V_i$ by Hotelling or $W_i$ by Alt do not work effectively when the correlation coefficient components in dispersion matrix are increased. We propose a combined procedure monitoring every component of dispersion matrix, which operates simultaneously both control charts, a chart controlling variance components and a chart controlling correlation coefficients. Our numerical results show that the proposed combined procedure is efficient for detecting changes in both variances and correlation coefficients of dispersion matrix.

• Journal of the Korean Statistical Society
• /
• 제29권3호
• /
• pp.319-336
• /
• 2000

A mixed model with a white noise process and an IMA(0,1,1) process is considered as a process model. It is assumed that the process is a white noise in the absence of a special cause and the process changes to an IMA(0,1,1) due to a special cause. One useful scheme in measuring the process level is to use the variable measurement interval (VMI) between measurement times according to the value of the previous chart statistic. The advantage of the VMI scheme is to measure the process level infrequently when in control to save the measurement cost and to measure frequently when out of control to save the off-target cost. This paper considers the VMI scheme in order to detect changes in the process model from a white noise to an IMA(0,1,1). The VMI scheme is shown to be effective compared to the standard fixed measurement interval (FMI) scheme in both statistical and economic contexts.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2006년도 춘계공동학술대회 논문집
• /
• pp.92-100
• /
• 2006

In many common control charting situations, the statistic to be charted can be viewed as the output of a linear filter applied to the sequence of process measurement data. In recent work that has generalized this concept, the charted statistic is the output of a general linear filter in impulse response form, and the filter is designed by selecting its impulse response coefficients in order to optimize its average run length performance. In this work, we restrict attention to the class of all second-order linear filters applied to the residuals of a time series model of the process data. We present an algorithm for optimizing the design of the second-order filter that is more computationally efficient and robust than the algorithm for optimizing the general linear filter. We demonstrate that the optimal second-order filter performs almost as well as the optimal general linear filter in many situations. Both methods share a number of interesting characteristics and are tuned to detect any distinct features of the process mean shift, as it manifests itself in the residuals.

• Industrial Engineering and Management Systems
• /
• 제11권4호
• /
• pp.385-389
• /
• 2012

With the development of computer storage and the rapidly growing ability to process large amounts of data, the multivariate control charts have received an increasing attention. The existing univariate and multivariate control charts are a single hypothesis testing approach to process mean or variance by using a single statistic plot. This paper proposes a multiple hypothesis approach to developing a new multivariate control scheme. Plotted Hotelling's $T^2$ statistics are used for computing the corresponding p-values and the procedure for controlling the false discovery rate in multiple hypothesis testing is applied to the proposed control scheme. Some numerical simulations were carried out to compare the performance of the proposed control scheme with the ordinary multivariate Shewhart chart in terms of the average run length. The results show that the proposed control scheme outperforms the existing multivariate Shewhart chart for all mean shifts.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1993년도 Fifth International Fuzzy Systems Association World Congress 93
• /
• pp.1098-1101
• /
• 1993

The purpose of this work is to build a fuzzy model of a batch culture for a process control. The process is highly nonlinear system with large delay. This paper presents two methods of modeling the process behavior. One is a method of recognizing them by fuzzy rules that are contracted by the pattern analysis in consideration of skilled operators' way. The other is a method of predicting them by approximate linear models and fuzzy rules by statistic analysis.

• Journal of the Korean Data and Information Science Society
• /
• 제26권4호
• /
• pp.1027-1034
• /
• 2015

A control chart is very useful in monitoring various production process. There are many situations in which the simultaneous control of two or more related quality variables is necessary. When the joint distribution of the process variables is multivariate normal, multivariate Shewhart control charts using the function of the maximum likelihood estimator for monitoring the dispersion matrix are considered for the simultaneous monitoring of the dispersion matrix. The performances of the multivariate Shewhart control charts based on the proposed control statistic are evaluated in term of average run length (ARL). The performance is investigated in three cases, where the variances, covariances, and variances and covariances are changed respectively. The numerical results show that the performances of the proposed multivariate Shewhart control charts are not better than the control charts using the trace of the covariance matrix in the Jeong and Cho (2012) in terms of the ARLs.

• 산업경영시스템학회지
• /
• 제18권34호
• /
• pp.35-42
• /
• 1995

In the plastic extrusion process, product quality is influenced by work condition for temperature of cylinders and dies. The work conditions are various, so it is difficult to standardization of the work conditions. Therefore, the work conditions are depended on the workers of experience and skill. In the plastic extrusion process, it has five control heating points on the cylinder and three control heating points on the die. In addition, there is one control point on the extrusion process. It is extrusion speed. In this case, we don't know how these affect product quality. We structure the multiple linear regression equation with the temperature of cylinders and dies as independent variables and the product weight as dependent variable. We solve this equation using statistic computer package named Juse-Qcas.

• 통합자연과학논문집
• /
• 제12권3호
• /
• pp.78-84
• /
• 2019

In many cases, the quality of a product is determined by several correlated quality variables. Control charts have been used for a long time widely to control the production process and to quickly detect the assignable causes that may produce any deterioration in the quality of a product. Numerical switching performances of multivariate cumulative sum control chart for simultaneous monitoring all components in the dispersion matrix ${\Sigma}$ under multivariate normal process $N_p({\underline{\mu}},{\Sigma})$ are considered. Numerical performances were evaluated for various shifts of the values of variances and/or correlation coefficients in ${\Sigma}$. Our computational results show that if one wants to quick detect the small shifts in a process, CUSUM control chart with small reference value k is more efficient than large k in terms of average run length (ARL), average time to signal (ATS), average number of switches (ANSW).

• 품질경영학회지
• /
• 제25권1호
• /
• pp.100-115
• /
• 1997

Shewhart control chart is a basic technique to monitor the state of a process. We observe samples of size four or five and plot some statistic(e.g., mean or range) of each sample on the chart. When setting up the chart, we need to obtain u, pp.r and lower control limits. It is common practice that those limits are calculated from the preliminary 20-40 samples presumed to be homogeneous. However, it may ha, pp.n in practice that the samples are contaminated by outlying observations caused by various reasons. The presence of outlying observations make the control limits wider and hence decrease the sensitivity of the charts. In this paper, we introduce robust control charts with tighter control limits when outlying observations are present in the preliminary samples. Examples will be given via simulation study.