ISO-IEC

Uncertainties



Taranath Code


Like there is nothing perfect in the world we live in, measurements too - be it of any kind - are not perfect and are influenced by various factors resulting in uncertainties in measurement. For this reason, any measurement result needs to be accompanied by an uncertainty estimate to make it more meaningful. The evaluation of uncertainties is quite complex and the capability and competence of Analytical and Precision Balance Co. to have a sound management system, carry out calibrations, evaluate uncertainties, etc. are what gives Analytical and Precision Balance Co. the recognition and accreditation by IAS.

Uncertainty statements are of immense importance, more particularly when high levels of calibrations are involved. The uncertainty values help the user in evaluating his/her further processes, fine tuning them and have better control on product quality.

Written by B.N.Taranath,Staff Metrologist, International Accreditation Service.

Type A & B Uncertainties



Analytical & Precision Balance has developed a format and procedure for the estimation of uncertainty of all calibrations that it does. While estimating uncertainty of calibrations all factors which effect the calibration are considered and reasonably estimated.

This document satisfies the requirements of ISO 17025: Section 5.4.6. 5.4.6.1 "A testing laboratory performing its own calibrations shall have and shall apply a procedure to estimate the uncertainty of a measurement for all calibrations and types of calibrations". 5.4.6.2 "Testing laboratories shall have and shall apply procedures for estimating uncertainty of measurement".

This procedure applies to all quantitative tests and calibrations performed under the scope of accreditation. Typically only methods and standards classified as category IV methods as defined by this procedure are required to provide uncertainty estimates.

Definitions:

Uncertainty: The unknown difference between a measured result and the actual value.
Calibration Uncertainty: The uncertainty associated with the calibration standard or master used while performing a calibration. Generally, this is the expanded uncertainty estimate at 95% confidence. Confidence may also be stated in terms of the coverage factor, k=2.
Resolution: The finest increment of change that may be observed on an instruments scale.
Repeatability: A statistical value (typically 1 standard deviation) calculated based on the results of repeated measurement.
Combined Uncertainty: A calculated statistical value using the root sum squared technique, RSS, (the square root of the sum of the squares) that includes all identified sources of uncertainty
Expanded Uncertainty: Calculated uncertainty at the prescribed confidence interval. This lab assumes ~95% confidence and identifies the expanded uncertainty as U95%.