A gradient approach to efficient design and analysis of multivariate EWMA control charts

This paper deals with the statistical design of control charts and describes a methodology to identify control schemes with action and warning limits that, for a fixed in-control average run length (ARL), minimize the ARL at a given out-of-control situation as measured by a shift in the process mean. This methodology allows the design of control charts with improved ARL performance for moderate shifts without degrading the in-control ARL as typically happens when run rules are applied to a conventional control chart. These optimal control schemes significantly reduce the out-of-control average run length compared with nonoptimal schemes.

The multivariate exponentially weighted moving average (MEWMA) control chart, in which the smoothing parameter is a principal diagonal matrix, is established based on the assumption that the samples follow the standard normal distribution to monitor the multivariate quality characteristics. However, monitoring quality characteristics always exists in the whole process. To make the MEWMA control chart more effective, a modified multivariate exponentially weighted moving average control chart, namely practical MEWMA (PEWMA) control chart, is proposed, in which the smoothing coefficient matrix is extended to a full-rank matrix which the smoothing parameter are all not zero. Analysis of both the in-control and out-of-control statistical performance of PEWMA control chart illustrate that the average run length of PEMWA control chart is more superior to the MEWMA. Finally combined with a factory example, a minimisation cost model is constructed based on the Lorenzen-Vance economic model to optimise the parameter combination.

In order to improve the monitoring efficiency of the multivariable autocorrelation process and reduce the cost of process control, an economic design method of residual Multivariate Exponentially Weighted Moving Average (MEWMA) control charts with variable sample size and sampling intervals (VSSIs) was proposed. First, residual MEWMA control charts with VSSI were designed for monitoring multivariable autocorrelation processes. Second, the economic model of the VSSI residual MEWMA control chart was established, and a data example was given. The optimal design parameters of the control chart were obtained by using a genetic algorithm to minimize the expected cost per unit time. Third, using orthogonal experimental design and regression analysis, a sensitivity analysis of the economic model was carried out to get the relationship between the model parameters and the design parameters of the control chart. Finally, through the optimality analysis, it was verified that the proposed economic model was superior to the models established by the variable sampling intervals strategy and the variable sample size strategy and had less expected cost per unit time.

Since multi-attribute control charts have received little attention compared with multivariate variable control charts, this research is concerned with developing a new methodology to employ the multivariate exponentially weighted moving average (MEWMA) charts for m-attribute binomial processes; the attributes being the number of nonconforming items. Moreover, since the variable sample size and sampling interval (VSSI) MEWMA charts detect small process mean shifts faster than the traditional MEWMA, an economic design of the VSSI MEWMA chart is proposed to obtain the optimum design parameters of the chart. The sample size, the sampling interval, and the warning/action limit coefficients are obtained using a genetic algorithm such that the expected total cost per hour is minimized. At the end, a sensitivity analysis has been carried out to investigate the effects of the cost and the model parameters on the solution of the economic design of the VSSI MEWMA chart.

This article is concerned with the effect of the methods for handling missing values in multivariate control charts. We discuss the complete case, mean substitution, regression, stochastic regression, and the expectation–maximization algorithm methods for handling missing values. Estimates of mean vector and variance–covariance matrix from the treated data set are used to build the multivariate exponentially weighted moving average (MEWMA) control chart. Based on a Monte Carlo simulation study, the performance of each of the five methods is investigated in terms of its ability to obtain the nominal in-control and out-of-control average run length (ARL). We consider three sample sizes, five levels of the percentage of missing values, and three types of variable numbers. Our simulation results show that imputation methods produce better performance than case deletion methods. The regression-based imputation methods have the best overall performance among all the competing methods.

In many statistical process control (SPC) applications, the quality of a process or product is represented by a relationship or the so-called quality profile, between a quality characteristic and one or more explanatory variables. The process quality characteristics are sometimes measured with a reasonable degree of approximation. Often, especially when qualitative assessments arise, evaluation of quality characteristics is carried out ambiguously or using linguistic qualifiers. The fuzzy sets theory has proved to be a well-established approach for dealing with uncertainty due to the approximate measurement, ambiguity in subjective evaluations, or vagueness in linguistic variables. Our main purpose was to present and compare four methods of monitoring nonlinear fuzzy profiles, for which different nonlinear fuzzy regression modeling approaches are considered. The first two methods are “a data-driven fuzzy rule-based” and “an extended least square support vector machine (LS-SVM),” for which the profile is characterized without considering a predefined mathematical relationship. However, for the other two methods, a specific form of the profile was needed. The third method, namely “a modified fuzzy regression model,” was initially invented for linear models. Besides, the fourth method employs “the fuzzy least square method” based on linearizing transformation. The exponentially weighted moving average (EWMA) control statistic was used to derive the control statistic to be plotted on the univariate as well as multivariate control charts. An extensive simulation study was conducted to compare the performance of the methods, and the average run length (ARL) criterion was considered to assess the detect-ability of control charts against various out-of-control conditions. Our comparison results indicated that the multivariate EWMA (MEWMA) chart based on the LS-SVM method outperforms the rest in detecting process shifts with smaller values of ARL when the process undergoes out-of-control conditions.

In this research, the main parameters of the multivariate cumulative sum (CUSUM) control chart (the reference value k, the control limit H, the sample size n and the sampling interval h) are determined by minimising the Lorenzen–Vance cost function [Lorenzen, T.J., and Vance, L.C. (1986), ‘The Economic Design of Control Charts: A Unified Approach’, Technometrics, 28, 3–10], in which the external costs of employing the chart are added. In addition, the model is statistically constrained to achieve desired in-control and out-of-control average run lengths. The Taguchi loss approach is used to model the problem and a genetic algorithm, for which its main parameters are tuned using the response surface methodology (RSM), is proposed to solve it. At the end, sensitivity analyses on the main parameters of the cost function are presented and their practical conclusions are drawn. The results show that RSM significantly improves the performance of the proposed algorithm and the external costs of applying the chart, which are due to real-world constraints, do not increase the average total loss very much.

Water treatment is intended to change the original water quality that does not fulfill the health requirements become a water for human consumption and must comply with the levels of certain parameters. Quality control can be done by forming a Multivariate Exponentially Weighted Moving Average (MEWMA) control chart. In the Multivariate Exponentially Weighted Moving Average (MEWMA) control charts with λ = 0.25 and UCL = 13.92658 seen that process controlled statistically. Once the process is under control, it can be done analysis of the ability of the process to determine whether the process fulfill the specifications or not. In the calculation process capability univariate each characteristics and multivariate process capability index values obtained more than 1 means that the process is going well. Keywords : water quality, Multivariate Exponentially Weighted Moving Average (MEWMA), process capability.

It is an increasingly common practice to monitor several related quality characteristics of a product or process using a multivariate control chart procedure. Several types of multivariate control charts, including Hotelling's χ 2 and T 2 control charts, have been developed in attempts to improve monitoring by using the correlation structure that exists between quality characteristics. The purpose of this paper is to summarize the assumptions made regarding the out-of-control process shift in the economic design of multivariate control charts and to address their consequences. We study the average run length (ARL) properties of the χ 2 control chart using a numerical example and show that this chart can perform ineffectively under the assumed out-of-control conditions when designed using the economic approach. Following Healy,[1] we offer an alternative procedure that has improved ARL properties and overall performance. These results can be important to researchers and practitioners who are interested in using the economic design of multivariate control procedures.

A control chart is one of the statistical process techniques that is used to monitor different processes. Some processes are characterized by functions or profiles, and a profile is a functional relationship between the dependent and independent variable(s) used to monitor the quality of the process. Several research studies were conducted on linear profiling where only fixed effects are considered. However, in this research, we focus on random effects as they represent the differences between profiles and thus are more proper for interpretation. Two approaches are proposed in this study for Phase II profile monitoring; the first approach is the nonparametric via residuals and the second is the semiparametric approach, where this technique combines the parametric estimates with a portion of the nonparametric estimates to the residuals. Usually, parametric estimations lead to biased estimates when the model is misspecified, whereas nonparametric estimates may give high variances, and thus semiparametric estimates are preferred. New nonparametric and semiparametric multivariate exponential weighted moving average (MEWMA) control charts are introduced and their performances compared to the parametric approach for different samples and shift sizes, and the correlation between and within profiles was considered. The average run length (ARL) and average time to signal (ATS) criteria are used for choosing the best approach. Simulation studies and real datasets were utilized for comparing the performance of the proposed MEWMA charts.

The ability of a control chart to detect a shift in the process can be determined by the average run length (ARL), a widely used performance indicator for single sampling interval (SSI) control charts. When it comes to variable sampling interval (VSI) schemes, average time to signal (ATS) is utilized instead of ARL. Although analysis of ARL has been extensively studied for SSI schemes, calculation of the ATS for VSI control charts is still in its infancy due to several theoretical challenges. In this context, the paper derives closed-form expressions for time to signal distribution and the ATS for a spatial design of VSI multivariate Bayesian control chart referred to as the dual sampling scheme (DSS). In a DSS strategy, the process is being monitored initially with a longer sampling interval, which is changed to the shorter sampling interval at a designed switching point. Unlike previous approximate developments where the switching point is ignored, the paper derives exact expressions for computation of time to signal distribution and ATS within such a Bayesian framework. Derivations of the ATS are based on an intuitively pleasing definition of an artificial absorbing state and coded values of the posterior probability using Markov chain theory.

Design of X-bar control chart based on Inverse Rayleigh Distribution under repetitive group sampling

The multivariate exponentially weighted moving average (MEWMA) control chart is a control charting scheme that uses weighted averages of previously observed random vectors to monitor the mean vector of a process. The in-control average run length (ARL) for the MEWMA control chart can be expressed as the solution of an integral equation. We give a FORTRAN program that uses the quadrature method to approximate the solution of this integral equation. The program can be used to obtain either the in-control ARL for a given value of the upper control limit or the appropriate upper control limit for a desired in-control ARL.

Autocorrelation leads to a bias estimator of standard control charts. It is important to develop control chart that allows autocorrelation and to evaluate its performance. The objective of this paper is to evaluate the performance of multioutput least square support vector regression (MLS-SVR)-based multivariate exponentially weighted moving average (MEWMA) control chart for monitoring multivariate autocorrelated data. For first order of vector autoregressive (VAR) and first order of vector moving average data, the proposed control chart tends to yield stable in-control average run length at about 200. The proposed control chart becomes more insensitive due to the increase of MEWMA smoothing parameter. In the real application, the proposed method is successfully applied to monitor water turbidity and chlorine residual data in the drinking water manufacturing process.

In constructing any control chart, the sample size and the number of samples taken are important issues. When destructive tests are applied to collect data, the sample size and the number of subsamples are usually limited. Lack of sufficient observations affects parameters estimates. Re‐sampling techniques are always suggested to increase the accuracy of parameters estimates. Also, the presence of outlying observations severely affect re‐sampling and the estimates of the underlying distribution parameters. The limited sample size, limited number of subsamples and presence of outliers are more important in multivariate process analysis. In this study, bootstrap re‐sampling technique and robust estimators of mean vector and variance‐covariance matrix are simultaneously utilized to design a multivariate robust T2 control chart to address the three problems in hand. The performance of the proposed robust T2 control chart is evaluated and compared to the Hotelling T2 control chart by means of average run length (ARL). Simulation results indicate that the suggested robust T2 control chart outperforms the Hotelling T2 in presence of outliers and similarly when there is no outlier. Design of robust control chart for processes generating auto‐correlated data applying bootstrap re‐sampling technique is an area for further research.