MULTIVARIATE CONTROL CHARTS BASED ON DATA DEPTH FOR SUBGROUP LOCATION AND SCALE - CASE STUDY

Production Practice ◽

Classic Approach ◽

Multivariate Control ◽

T2 Control Chart

The purpose of the article is to present a method for determining control charts, which allow to control few interrelated quality characteristics. Often, in production practice, there is a need to simultaneously control several interrelated quality characteristics. The use of univariate control charts, separately for each quality characteristics, may lead to inadequate corrective actions of a production process. In such situations, it is recommended to use multivariate control charts, for example, the T2 control chart. However, the use of this classic approach involves making complicated calculations. Therefore, the author of this paper suggests using multivariate control charts based on data depth proposed by Liu. In this paper, the author presented the idea and principles of the multivariate control chart based on data depth and then, using it to assess the statistical stability of the process in a manufacturing company, engaged in the production of window fittings.

On multivariate control charts

Production ◽

10.1590/s0103-65132011005000010 ◽

2011 ◽

Vol 21 (2) ◽

pp. 235-241 ◽

Cited By ~ 8

Author(s):

Marianne Frisén

Keyword(s):

Control Chart ◽

Industrial Production ◽

Control Charts ◽

Multivariate Control Chart ◽

Industrial production requires multivariate control charts to enable monitoring of several components. Recently there has been an increased interest also in other areas such as detection of bioterrorism, spatial surveillance and transaction strategies in finance. In the literature, several types of multivariate counterparts to the univariate Shewhart, EWMA and CUSUM methods have been proposed. We review general approaches to multivariate control chart. Suggestions are made on the special challenges of evaluating multivariate surveillance methods.

Comparing the Performance of Several Multivariate Control Charts Based on Residual of Multioutput Least Square SVR (MLS-SVR) Model in Monitoring Water Production Process

Journal of Physics Conference Series ◽

10.1088/1742-6596/2123/1/012018 ◽

2021 ◽

Vol 2123 (1) ◽

pp. 012018

Author(s):

M Ahsan ◽

T R Aulia

Keyword(s):

Water Quality ◽

Quality Control ◽

Control Charts ◽

Continuous Process ◽

Quality Characteristics ◽

Least Square ◽

Support Vector ◽

Water Quality Control ◽

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.

DIAGRAM KONTROL MULTIVARIAT np DAN DIAGRAM KONTROL JARAK CHI-SQUARE DALAM PENGENDALIAN KUALITAS PRODUK KAIN DENIM (Studi Kasus di PT Apac Inti Corpora)

Jurnal Gaussian ◽

10.14710/j.gauss.v7i4.28866 ◽

2018 ◽

Vol 7 (4) ◽

pp. 385-396

Author(s):

Dwi Harti Pujiana ◽

Mustafid Mustafid ◽

Di Asih I Maruddani

Keyword(s):

Production Process ◽

Control Chart ◽

Control Charts ◽

Process Capability ◽

Chi Square ◽

Distance Control ◽

Denim Fabric ◽

Denim fabric sort number 78032 is one type of fabric in the last 4 years almost every month produced by PT Apac Inti Corpora. In the continuity of denim fabric production process, there are data defects (non-conformity) that causes the quality of denim fabric decreases. To maintain the consistency of the quality of products produced in accordance with the specified specifications, it is necessary to control the quality of the production process that has been running for this. Multivariate control charts attributes used are multivariate control charts np using the number of samples and the proportion of disability data with correlation between variables while the chi-square distance control charts use squared distances with uncorrelated data between variables. The results showed that in the multivariate control chart np there were 2 out-of-control observations in the phase II data using control limits from phase I data already controlled by the value of BKA of 636321.4. While in the chi-square distance control chart showed all observations are in in-control condition with BKA value of 0.06536. Controlled production process obtained multivariate process capability value for multivariate control np diagram of 0.625142 <1 which means the process is not capable, while the value of process capability in the chi-square distance control chart is 1.1329> 1 which means the process is capable. Keywords: denim fabric, multivariate np control chart, chi-square distance control chart, multivariate process capability

On Nonparametric Multivariate Control Charts Based on Data Depth

Frontiers in Statistical Quality Control 6 ◽

10.1007/978-3-642-57590-7_13 ◽

2001 ◽

pp. 207-227 ◽

Cited By ~ 8

Author(s):

Zachary G. Stoumbos ◽

L. Allison Jones ◽

William H. Woodall ◽

Marion R. Reynolds

Keyword(s):

Control Charts ◽

Data Depth ◽

Propuesta de Carta de Control Multivariada utilizando la Media Winsorizada basada en la Carta |S| / Proposed Multivariate Control Chart using Winsorized Mean based on the |S| Chart

Revista Ingeniería ◽

10.15517/ri.v29i1.34379 ◽

2018 ◽

Vol 29 (1) ◽

pp. 65-79

Author(s):

Rister Junior Barreto Pombo ◽

Angellys Paola Ariza Guerrero ◽

Roberto José Herrera Acosta

Keyword(s):

Covariance Matrix ◽

Control Chart ◽

Control Charts ◽

Quality Monitoring ◽

Global Quality ◽

Simultaneous Evaluation ◽

Multivariate Control Chart ◽

Implementation Methodology ◽

Resumen— El monitoreo global de la calidad de un producto está sujeto a la evaluación simultánea de varias de sus características; es necesario bajo estas condiciones la implementación de las cartas de control tipo multivariadas. La variabilidad, en este caso la matriz de varianza covarianza, es sin duda el más importante de los estadísticos desde la perspectiva multivariada, que puede ser monitoreada con distintas cartas. Entre éstas se encuentran: las cartas Shewhart, CUSUM y EWMA. En este artículo se desarrolla una metodología de implementación de la Media Winsorizada en la carta de control multivariada de varianza efectiva |S|, encontrando una gran utilidad en procesos con valores extremos. El estudio muestra además una comparación entre la carta de control tradicional multivariante y la carta propuesta, que muestra mayor sensibilidad.Abstract— The global quality monitoring of a product is often subject to the simultaneous evaluation of several of its features; under these circumstances it is necessary to implement multivariate control charts. Variability, in this particular case, the variance-covariance matrix is indisputably the most important of the statistics from the multivariate perspective and it can be monitored with different charts, among these: Shewhart, CUSUM and EWMA. This article develops the Winsorized Mean in the effective variance multivariate control |S|-chart implementation methodology and it was demonstrated that the modification was more efficient when the sample hat outliers. This study shows a comparison between the traditional multivariate control chart and a proposed chart which was found to have more sensitivity.

Ch. 14. Multivariate control charts: Hotelling T2, data depth and beyond

Handbook of Statistics - Statistics in Industry ◽

10.1016/s0169-7161(03)22016-2 ◽

2003 ◽

pp. 573-593

Author(s):

Regina Y. Liu

Keyword(s):

Control Charts ◽

Data Depth ◽

An Application of Univariate and Multivariate Control Charts in Monitoring Water Quality

ASM Science Journal ◽

10.32802/asmscj.2020.sm26(1.3) ◽

2020 ◽

pp. 1-7

Author(s):

Siti Rahayu Mohd Hashim ◽

Azwaan Andrew ◽

Wilter Azwal Malandi

Keyword(s):

Water Quality ◽

Control Chart ◽

Control Charts ◽

Ph Value ◽

Statistical Process ◽

Monitoring Process ◽

Process Dispersion ◽

Water Treatments ◽

Control chart is a tool for detecting an out-of-control signal in statistical process control (SPC). It is widely used in process monitoring in order to detect changes in process mean or process dispersion. This study aims to illustrate the application of multivariate control charts in monitoring water quality at one of the water treatments plants in Kota Kinabalu, Sabah. The tested water quality variables in this study are turbidity, pH value, dissolved oxygen (DO) and concentration of ferum. Two multivariate control charts, Hotelling’sT2 and MCUSUM control charts are constructed under the violation of the multivariate normality assumption. The purpose is to study the effect of non-normal data upon the monitoring process using the selected multivariate control charts. By comparing the monitoring process between the two types of control charts, the consistency of the results is studied. All the univariate and multivariate control charts produced out-of-control signals from different points, hence inconclusive results obtained. Keywords: Water quality; multivariate control chart; univariate control chart; Hotelling’s T2; MCUSUM

Multivariate Control Chart and Lee–Carter Models to Study Mortality Changes

Mathematics ◽

10.3390/math8112093 ◽

2020 ◽

Vol 8 (11) ◽

pp. 2093

Author(s):

Gisou Díaz-Rojo ◽

Ana Debón ◽

Jaime Mosquera

Keyword(s):

Control Chart ◽

Control Charts ◽

Causes Of Death ◽

Mortality Model ◽

Probability Of Death ◽

Multivariate Control Chart ◽

Model Residuals ◽

Multivariate Control ◽

Carter Model

The mortality structure of a population usually reflects the economic and social development of the country. The purpose of this study was to identify moments in time and age intervals at which the observed probability of death is substantially different from the pattern of mortality for a studied period. Therefore, a mortality model was fitted to decompose the historical pattern of mortality. The model residuals were monitored by the T2 multivariate control chart to detect substantial changes in mortality that were not identified by the model. The abridged life tables for Colombia in the period 1973–2005 were used as a case study. The Lee–Carter model collects information regarding violence in Colombia. Therefore, the years identified as out-of-control in the charts are associated with very early or quite advanced ages of death and are inversely related to the violence that did not claim as many victims at those ages. The mortality changes identified in the control charts pertain to changes in the population’s health conditions or new causes of death such as COVID-19 in the coming years. The proposed methodology is generalizable to other countries, especially developing countries.