MATHMET

The European Centre for
Mathematics and Statistics in Metrology

Guideline documents from joint research projects

 

Uncertainty evaluation for regression problems: best practice guide

This guide contains the case studies used in the NEW04 project "Novel mathematical and statistical approaches to uncertainty evaluation" as particular examples, but it expresses the problem in a general way so that it can be applied to many other regression problems in metrology. In particular, the guide is applicable to the regression problems resulting from calibration problems in metrology. The guide addresses statistical models that have been classified with respect to (i) linearity or nonlinearity (in relation to the parameters to be estimated), (ii) the distribution of measurement errors being normal or non-normal, (iii) the covariates being assumed to be known exactly or with an associated uncertainty, and (iv) whether or not prior knowledge is available. In addition the guide explains how to determine a prior distribution required for the Bayesian analysis. The two cases of relevant prior knowledge and no prior knowledge are considered. The former case includes incorporation of physical constraints, such as the non-negativity of mass concentrations, as well as eliciting and pooling expert knowledge.
related project:
EMRP NEW04
Document:
download
Contact person:
EmailElster
 

Uncertainty evaluation for computationally expensive problems: best practice guide

Based on the data sets and case studies considered in the EMRP NEW04 project "Novel mathematical and statistical approaches to uncertainty evaluation", this document provides guidance on the evaluation of uncertainties for computationally expensive problems. The aim of the guide is to advice metrologists who encounter mathematical models such as partial differential equations, which make a naive Monte Carlo uncertainty propagation impractical.
The guide contains useful literature references to start from, and it addresses statistical models and numerical implementations issues. The methods considered are polynomial chaos, surrogate Gaussian process models and smart sampling strategies.
related project:
EMRP NEW04
Document:
download
Contact person:
EmailWright
 

Decision-making and conformity assessment in multivariate cases: best practice guide

Based on the data sets and case studies considered in the EMRP NEW04 project "Novel mathematical and statistical approaches to uncertainty evaluation", this document provides guidance on decision-making and conformity assessment in multivariate cases. The aim of the guide is to advice metrologists, technicians and engineers in industry who are faced with decision-making tasks in cases where the criterion to assess is multivariate. The guide contains literature references on the state-of-the-art in the field and provides insights into the case studies from the EMRP project. New approaches to decision-making taking into account multivariate uncertain quantities are presented which have been applied successfully in the project.
related project:
EMRP NEW04
Document:
download
Contact person:
EmailEmardson
 

Industrial dynamic measurements: best practice guide

This guide has been produced by the EMRP IND09 project "Traceable dynamic measurement of mechanical quantities". The aim of the guide is to provide practical information and advice to engineers and technicians in industry who carry out dynamic measurements in the course of their work. It contains insights and experience gained during the IND09 project for dynamic calibration and measurement of force, torque and pressure. The guide addresses technical issues for carrying out dynamic calibration and measurement and it provides comprehensiv advice on the mathematical and statistical analysis of dynamic measurements.
related project:
EMRP IND09
Document:
webpage
Contact person:
EmailEsward
x
This website uses cookies occasionally to provide you with the best web browsing experience. However, no web-analytics tracking based on cookies is employed here.