Quality and Statistics
One of the purposes of quality assurance systems, according to standard ISO 9000:
“… Primarily to prevent deviations from product requirements …”
Quality control of process capability and product characteristics can be expressed in quantitative or numerical terms by applying statistics [Chatfield 1983].
The accuracy of a measurement method can be used for statistical process control in two different situations:
- To verify that requirements for product characteristics are met within the production process, assuming that the measurement conditions do not change
- To verify that requirements on measurement quality are met for actual the measurement process, assuming that the production conditions are stable
Numerical quality management techniques
Some important applications of statistical methods in management of measurement quality are:
Acceptance Inspection of test results
- under repeatability conditions
- For example, a costly chemical analysis
_______________________________________________________________________________
Determination of limits [ISO 5725-6]
A repeatability limit, r, is the value of the absolute difference between two single test results that can be expected obtained under repeatability conditions to be less than or equal to with a certain probability (usually 95%):
r = 2.8 sr
Reproducibility, R, is similarly defined for test results obtained under reproducibility conditions:
R = 2.8 sR
_______________________________________________________________________________________
________________________________________________________________________________________
ISO 5725-6: 1994 Use in practice of accuracy values
Applications of statistics in Measurement & Testing 
From “Determination of limits”: r = 2.8 sr
What is sr, please?
Thank you
sr is the precision standard deviation under repeatability conditions