In order to exploit measurement, test or calibration results, it is essential to know and control the uncertainties associated with them. In response to this challenge, LNE has launched LNE Uncertainty, a software package enabling measurement uncertainty to be assessed by variance propagation in accordance with the Guide to the Expression of Uncertainty in Measurement (Gum) and/or by distribution propagation using Monte Carlo simulations in accordance with Gum S1.
Developed by LNE's experts and entirely free of charge, LNE Uncertainty is a software package for evaluating measurement uncertainty by propagation of variances and/or propagation of distributions, using Monte Carlo simulations.
These methods are described in the Gum and its Supplement 1 (Gum S1) respectively. The Gum (Guide to the expression of uncertainty in measurement) is a guide published by the Bureau International des Poids et Mesures, which sets out the general rules for evaluating and expressing measurement uncertainty.
In addition to the assessment of measurement uncertainty using the Gum and Gum S1 methods, LNE Uncertainty enables you to :
- check the validity of the results obtained with the Gum method according to the criteria described in Gum supplement 1;
- construct an uncertainty budget by performing a sensitivity analysis using two methods: Spearman rank correlation coefficients and Sobol indices;
- process several output quantities simultaneously. The user can refer to Gum supplement 2 for the concept of multiple measurands;
- determine the law of the output variable, using the Kolmogorov-Smirnov test to check that it conforms to various theoretical laws (normal, Student, lognormal, Gamma, Beta, uniform, exponential, etc.).
In addition, to enhance the user experience, LNE Uncertainty integrates the values of fundamental physical constants and their associated uncertainties (Codata 2018).
LNE Uncertainty is the latest addition to the list of software developed at LNE and made available free of charge over the last few years.