Logo of the Physikalisch-Technische Bundesanstalt
DKD

13. Measurement Uncertainty

Philip M. Fleischmann

 

„Sicher is’, dass nix sicher is, drum bin ich vorsichtshalber misstrauisch.“ (Karl Valentin)
One thing’s certain and that is that nothing is certain, so I’d rather be suspicious.

To think that measurements might be uncertain is a rather horrible idea for the user. Not to mention measurements carried out by calibration laboratories! But even when using the most accurate methods, the measurement uncertainty cannot be zero.

And this is exactly what the Technical Committee Measurement Uncertainty deals with.

The exchange of information plays an important role. This also implies the balancing act between practical handling and mathematically correct formulation: "rooted sum of squares", "correlation coefficients" and "Monte Carlo simulation" - these are the tools to be used.

Given that the technical committee consists of participants from all technical committees, it is sometimes difficult to find a common procedure for the best solution to a problem.

But maybe that' s the nature of it, to be able to give only an "approximately accurate" indication?

The Technical Committee is currently preparing a guide on the application of decision rules and risk assessment.

This is supposed to provide users in calibration laboratories with a practicable tool to fulfil the essential points of sections 7.8.6.1 and 7.8.6.2 of DIN EN ISO/IEC 17025:2018 in a practice-oriented manner.

Furthermore, the Technical Committee is working on a guideline for the determination of calibration intervals in relation to the measurement uncertainty calculation.

The Technical Committee has published the following guides:

  • DKD-L 13-1: Praxisgerechte Ermittlung der Messunsicherheit
    (Practice-oriented determination of the measurement uncertainty – so far only in German)
  • DKD-L 13-2: Validation of measurement uncertainty budgets
  • DKD-L 13-3: Rounding of Results and Measurement Uncertainties in Calibration Certificates