Measurement Uncertainty Evaluation: GUM Method Online
What is Measurement Uncertainty?
Measurement uncertainty is the parameter that characterizes the dispersion of measurement results and is an essential part of a complete measurement statement: any reported result must include its uncertainty. Per the GUM (JJF 1059.1), Type A evaluation uses statistical methods, typically the standard deviation of repeated observations, while Type B evaluation uses non-statistical information such as calibration certificates, instrument specifications, reference material certificates and experience. Both types are combined in the same way.
When to Use It
Use it when issuing calibration reports, supporting laboratory accreditation under ISO/IEC 17025, stating the uncertainty of a reported measurement, or validating a measurement method. It is also required when a decision depends on whether a result meets a specification, where uncertainty prevents false conformity statements.
How to Use It (Step by Step)
Define the measurement model y = f(x1, x2, ..., xn) and list every uncertainty source: repeatability, instrument indication error, resolution, temperature effects, reference material, reading and sampling. Evaluate the standard uncertainty u(xi) of each source, combine them with the law of propagation of uncertainty, and multiply by a coverage factor to obtain the expanded uncertainty. Report the measurement conditions, data sources and calculation so the result can be reviewed.
Key Formulas / Example
For uncorrelated inputs the combined standard uncertainty is uc = √(Σ(ci·u(xi))²), where ci are sensitivity coefficients, and the expanded uncertainty is U = k·uc. The default coverage factor is k = 2 for approximately 95% confidence, or k = 3 when required. For example, with uc = 0.5 mg and k = 2, the expanded uncertainty is U = 1.0 mg.