multivariate control
Recently Published Documents

PurposeThis study aims to design a multivariate control chart that improves the applicability of the traditional Hotelling T2 chart. This new type of multivariate control chart displays sufficient information about the states and relationships of the variables in the production process. It is used to make better quality control decisions during the production process.Design/methodology/approachMultivariate data are collected at an equal time interval and are represented by nodes of the graph. The edges connecting the nodes represent the sequence of operation. Each node is plotted on the control chart based on their Hotelling T2 statistical distance. The changing behavior of each pair of input and output nodes is studied by the neural network. A case study from the cement industry is conducted to validate the control chart.FindingsThe finding of this paper is that the points and lines in the classic Hotelling T2 chart are effectively substituted by nodes and edges of the graph respectively. Nodes and edges have dimension and color and represent several attributes. As a result, this control chart displays much more information than the traditional Hotelling T2 control chart. The pattern of the plot represents whether the process is normal or not. The effect of the sequence of operation is visible in the control chart. The frequency of the happening of nodes is recognized by the size of nodes. The decision to change the product feature is assisted by finding the shortest path between nodes. Moreover, consecutive nodes have different behaviors, and that behavior change is recognized by neural network.Originality/valueModifying the classical Hotelling T2 control chart by integrating with the concept of graph theory and neural network is new of its kind.

A potential eco-friendly method without organic solvents is presented by integrating a chromatographic fingerprint and multivariate control chart based on Q residuals to differentiate grape juices from different farming practices. The sample preparation was only water dilution, and the mobile phase was water acidified with sulfuric acid, which can be readily neutralized before its disposal. The proposed method is shown to be a simple way to distinguish between organic and non-organic grape juices in a non-target way, successfully evaluating an external validation data set, where organic and non-organic samples were correctly assigned. Through the chromatographic profile, it is possible to suggest that one of the species responsible for this distinction may be from the anthocyanins class.

Abstract In recent years, the manufacturing industry has tended to reduce mass production and produce in small quantities, which is called “Short Run Production”. In such a situation, the course of the production process is short, usually, the number of productions is less than 50. Therefore, a control chart for the short run production process is required. This paper discusses the comparison between multivariate control chart for short run production (V control chart) and T2 Hotelling control chart applied to sunergy glass data. Furthermore, a simulation of Average Run Length (ARL) was carried out to determine the performance of the two control charts. The results obtained are that the production process has not been statistically controlled using either the V control chart or the T2 Hotelling control chart. The number of out-of-control on the control chart V using the the EWMA test is more than the T2 Hotelling control chart. Based on the ARL value, it shows that the V control chart is more sensitive than the T2 Hotelling control chart.

Abstract Water that is used as the basic human need, requires a processing process to get it. Water quality control in Tirtanadi Water Treatment Plant is still univariate, while theoretically the quality characteristics of water quality are correlated and there is also an autocorrelation due to the continuous process. In this study, quality control is performed on three main variables of water quality characteristics, namely acidity (pH), chlorine residual (ppm), and turbidity (NTU) using several multivariate control charts based on Multioutput Least Square Support Vector Regression (MLS-SVR) residuals. MLS-SVR modelling is used to overcome and get rid of autocorrelation. The input results of the MLS-SVR model are specified from the significant lag of the Partial Autocorrelation Function (PACF), which in this study, is the first lag. The results of the MLS-SVR input model and the optimal combination of hyper-parameters produce residual values that have no autocorrelation anymore. The residuals are used to develop the Hotelling’s T 2, Multivariate Exponentially Weighted Moving Average (MEWMA), and Multivariate Cumulative Sum (MCUSUM) control charts. In phase I, we found that the processes are statically controlled. Meanwhile, in phase II, the monitoring results show that there are several out-of-control observations.

Abstract Monitoring the quality of drinking water needs to be conducted considering the important role of water in human life. Mixed Multivariate EWMA-CUSUM (MEC) chart is a multivariate control chart developed for observing the mean process. Based on the previous studies, this chart has better performance in detecting a shift in the process mean. In this research, the MEC is applied to observe the grade of drinking water. However, there is autocorrelation in drinking water data which lead to more false alarm occurred. Therefore, the Multioutput Least Square Support Vector Regression (MLS-SVR) model is employed to reduce or even remove the autocorrelation in the data. Using the optimal hyperparameter, the MLS-SVR algorithm produces the residuals of phase I with no autocorrelation. Those residuals are then used to form the MEC control charts. When the MEC is used to monitor the residual in phase I, there is no signal of out-of-control found. Further, in phase II, out-of-control observations are detected. The MEC chart can detect more signals out of control compared to the conventional Hotelling’s T 2 and Multivariate Exponentially Moving Average (MEWMA) charts.

While researchers and practitioners may seamlessly develop methods of detecting outliers in control charts under a univariate setup, detecting and screening outliers in multivariate control charts pose serious challenges. In this study, we propose a robust multivariate control chart based on the Stahel-Donoho robust estimator (SDRE), whilst the process parameters are estimated from phase-I. Through intensive Monte-Carlo simulation, the study presents how the estimation of parameters and presence of outliers affect the efficacy of the Hotelling T2 chart, and then how the proposed outlier detector brings the chart back to normalcy by restoring its efficacy and sensitivity. Run-length properties are used as the performance measures. The run length properties establish the superiority of the proposed scheme over the default multivariate Shewhart control charting scheme. The applicability of the study includes but is not limited to manufacturing and health industries. The study concludes with a real-life application of the proposed chart on a dataset extracted from the manufacturing process of carbon fiber tubes.