What is the decision rule?
The decision rule describes how measurement of uncertainty is accounted for, when stating conformity with a specified requirement. Measurement Uncertainty is a parameter, associated with the result of a measurement (e.g. a test) that defines the range of the values that could reasonably be attributed to the measured quantity. The Decision Rule is a system which uses the measurement uncertainty to assist your evaluation of the results.
The decision rule employed by Glass Technology Services
Unless otherwise requested, Glass Technology Services will employ the following decision rule to reduce the risk of false accepts and false rejects.
As a standard approach Glass Technology Services will deem any result which has a greater than 2.5% risk of false acceptance or rejection to be an Indeterminate result and will be highlighted as such.
This diagram shows how the method employed can be used to determine whether the result is an Indeterminate pass or an Indeterminate fail:
Each individual result will have an associated acceptance limit which is based on the uncertainty associated with that individual result. As the uncertainties used are method specific, the uncertainty itself will be fixed. The use of this approach will allow us to calculate the specific risk associated with that individual result. The decision rule employed and agreed by the customer is used to determine whether the result is a Pass, Fail or an Indeterminate result.
What will I see on my reports?
All measurements on reports/certificates issued using accredited test methods will be accompanied by a measurement uncertainty unless specified within the report. Example: Measurement Uncertainty = +/- 5.72%
Where a statement of conformity against a specific or standard is made, a Decision Rule will be applied and the results of this will be included within the report/certificate.
Here is an example report extract which explains the Decision Rule:
Reporting results
Compliant/Pass – The measurement result is within, or below the specification limit when the measurement uncertainty is taken into account. The specific false accept risk is up to 2.5%.
Indeterminate Pass – The measurement result is within, or below the specification limit, however, based on the calculation of conformance probability, the specific risk associated with the result means the PFA is greater than 2.5%. In this case the specific risk will be reported to the customer along with the result.
Indeterminate Fail – The measurement result is outside, or above the specification limit, however, based on the calculation of conformance probability the specific risk associated with the result means the PFR is greater than 2.5%. In this case the specific risk will be reported to the customer along with the result.
Fail – the measurement is outside, or above the specification limit when the measurement uncertainty is taken into account. The specific false reject risk is up to 2.5%.
Any questions?
If you have any questions on the decision rule, your reports or to discuss an individualised approach please contact a.meysner@glass-ts.com