A Nonparametric Multivariate Statistical Process Control Chart Based on Change Point Model

Author(s):  
Yafei Xu ◽  
Ostap Okhrin
Author(s):  
Umar Abubakar Adamu ◽  
Gulumbe Shehu Usman ◽  
Dikko Hussaini Garba

Recently, much attention has been raised on effects of high increase in drugs counterfeiting and sub-standard quality which leads to many casualties in Nigeria. The Multivariate Statistical Process Control Charts approach was employed to examine such defects especially in assessing the official physico-chemical quality of chloroquine phosphate tablet (BP250mg) which claimed to contain the required quality properties. The Multivariate Exponentially Weighted Moving Average (MEWMA) Control Chart gives a powerful and reliable control chart than the widely used Hotelling’s T2˗Control Chart, which detects the smallest shift in the product process means and have minimum process variability. Also, the Matrix of scatter plots indicated the existence of relationship among the process variables and the Principal Component Analysis (PCA) minimized the rate of dimensionality of the process variability, which captured most of the variables outliers and retained the first Principal Components (PC) that explained over 99% variability of the product. To this end, the study results shows that the product quality characteristics (process variables) is under control (stable) and conform to international standard as specified by BP 2002.


2014 ◽  
Vol 971-973 ◽  
pp. 1435-1439
Author(s):  
Mei Hua Duan ◽  
Xue Min Zi

It is common to monitor several quality characteristics of a process simultaneously in modern quality control, and it is called multivariate statistical process control (MSPC) in the literature. A change-point control chart for detecting shifts in the mean of a multivariate statistical process is developed for the case where the nominal value of the mean is unknown but some historical samples are available. This control chart is called distribution-free multivariate control chart based on change-point model. Its distribution robustness is a significant advantage where we usually know nothing about the underlying distribution. And the simulated results show that this approach has a good performance across the range of possible shifts.


2000 ◽  
Vol 24 (2-7) ◽  
pp. 291-296 ◽  
Author(s):  
B. Lennox ◽  
H.G. Hiden ◽  
G.A. Montague ◽  
G. Kornfeld ◽  
P.R. Goulding

AIChE Journal ◽  
2010 ◽  
Vol 57 (9) ◽  
pp. 2360-2368 ◽  
Author(s):  
Bundit Boonkhao ◽  
Rui F. Li ◽  
Xue Z. Wang ◽  
Richard J. Tweedie ◽  
Ken Primrose

Sign in / Sign up

Export Citation Format

Share Document