UNCERTAINTY PROPAGATION IN MULTI-STAGE MEASUREMENTS USING LINEAR REGRESSION ANALYSIS AND MONTE CARLO SIMULATION

A. Silva Ribeiro, J. Alves e Sousa, C. Oliveira Costa, M. Pimenta de Castro

Abstract

Linear Regression Analysis (LRA) is one of the statistical tools most intensively used in all branches of science, with many applications in the study of measurement processes and is therefore important in metrology. The implementation of metrology in quality systems, led to a widespread evaluation of measurement uncertainties based on the GUM uncertainty framework (Guide to the Expression of Uncertainty in Measurement). This methodology, however, has its own restrictions among which one could include the use of LRA in multi-stage measurement. To overcome these restrictions, an alternative approach considers the use of the Monte Carlo method to evaluate LRA uncertainties and, subsequently, to use it further in the evaluation of uncertainties in multi-stage measurement processes.

Keywords

measurement uncertainty, linear regression analysis, Monte Carlo method, multi-stage measurement

Download

PWC-2006-TC21-014u.pdf

DOI

-

IMEKO TC

TC21 - Mathematical Tools for Measurements

Back to the proceedings