scholarly journals Application of a six sigma model to evaluate the analytical performance of urinary biochemical analytes and design a risk‐based statistical quality control strategy for these assays: A multicenter study

2021 ◽  
Author(s):  
Qian Liu ◽  
Guangrong Bian ◽  
Xinkuan Chen ◽  
Jingjing Han ◽  
Ying Chen ◽  
...  
Keyword(s):  
Quality Control ◽  
Multicenter Study ◽  
Control Strategy ◽  
Six Sigma ◽  
Sigma Model ◽  
Analytical Performance ◽  
Statistical Quality Control ◽  
Statistical Quality ◽  
Quality Control Strategy

Application of Six Sigma Statistical Quality Control to Monitoring the Deposition of Contaminating Particles

1992 ◽  
Vol 35 (5) ◽  
pp. 27-32
Author(s):  
Douglas Cooper ◽  
Jan Babcock ◽  
Frances DiPietro
Keyword(s):  
Quality Control ◽  
Normal Distribution ◽  
Six Sigma ◽  
Particle Deposition ◽  
Statistical Quality Control ◽  
Frequency Distributions ◽  
Control Approach ◽  
Statistical Quality ◽  
Class 1 ◽  
Number Of Particles

The goal of this work was to help in the control of particle deposition in manufacturing cleanrooms by the application of statistical quality control methods, specifically for the Class 1 cleanroom at the Semiconductor Research and Development Center at IBM East Fishkill. Each week a set of silicon wafers was inspected and the number of particles present were recorded ("precount"). The wafers were then put in various locations throughout the cleanroom. The wafers were left exposed to cleanroom air for one week, and then the number of particles were determined again. The quality control approach was the "Six Sigma" system developed originally by Motorola, Inc. In another paper, we described how these measurements were put into a Six Sigma framework: upper specification limit, target level, and estimates of the standard deviations.1 Here, we present results from 30 weeks of measurements and discuss the choice of assumed measurement distribution function (for example, the normal versus the lognormal distribution). Some locations had lognormal count frequency distributions. Others had count distributions with normal distribution components or with unclassified distribution components. Values of skewness and kurtosis were almost all greater than what would be expected from a normal distribution. In particular, this means that larger percentages were outside the limits μ±κs than would be true for a normal distribution, so that predictions of percentages of nonconforming instances would be erroneously low if the normal distribution were assumed.

Statistical Quality Control of HIV-1 ELISA Test Performance

1994 ◽  
Vol 89 (428) ◽  
pp. 1200-1208 ◽  
Author(s):  
R. C. Gentleman ◽  
M. S. Hamada ◽  
D. E. Matthews ◽  
A. R. Wilson
Keyword(s):  
Quality Control ◽  
Test Performance ◽  
Elisa Test ◽  
Statistical Quality Control ◽  
Statistical Quality ◽  
Hiv 1
Sign in / Sign up

Export Citation Format

Share Document