Context and purpose of the training
A calibration, test or analysis result must be accompanied by its uncertainty. This is a normative requirement, it can also be an opportunity to take a step back to reflect on the measurement process in order to better control it by identifying and evaluating the different sources of uncertainty.
This training is designed in a structured and progressive way to assess measurement uncertainties according to the reference approach (GUM) and the extension to the Monte-Carlo method.
- Understand the interest of measurement uncertainty.
- Assimilate the basic vocabulary and concepts for the evaluation of uncertainties.
- Know how to evaluate the uncertainty associated with a measurement result according to the GUM method.
- Know and apply the principle of the Monte Carlo method.
- Have the necessary tools for assessing uncertainties, based on the most recent texts.
- Measurement uncertainties - Why?
- Some definitions
- Reminder of statistical notions
- Errors, effects and corrections
- Accuracy/bias, fidelity and exactness
- The GUM methodology
- Expression of the result and Rounding rule
- Examples of uncertainty balance sheet or uncertainty budget
- Examples of calibration certificate
- Monte Carlo method - Supplement 1 of the GUM
Conventional presentation. Exchange and discussion. Presentation of examples from the standards.
Exercises (stated and corrected).
Staff of test or calibration laboratories: metrology manager, technical manager, test engineers, quality manager, technicians, etc.
Laptop with Excel spreadsheet (or equivalent).
Two days (a detailed program can be sent on request).
Pr. Christophe Ley - Professor of Mathematics & Applied Statistics - University of Luxembourg
In partnership with the University of Luxembourg