Models to estimate overall analytical measurements uncertainty: Assumptions, comparisons and applications

Evaluation of analytical results reliability is of core importance as crucial decisions are taken with them. From the various methodologies to evaluate the fitness of purpose of analytical methods, overall measurement uncertainty estimation is more and more applied. Overall measurement uncertainty allows to combine simultaneously the remaining systematic influences to the random sources of uncertainty and allows assessing the reliability of results generated by analytical methods. However there are various interpretations on how to estimate overall measurement uncertainty, and thus various models for estimating it. Each model together with its assumptions has great impacts on the risks to abusively declare that analytical methods are suitable for their intended purpose. This review paper aims at (i) summarizing the various models used to estimate overall measurement uncertainty, (ii) provide their pros and cons, (iii) review the main areas of application and (iv) as a conclusion provide some recommendations when evaluating overall measurement uncertainty.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► Review of the models used to combine systematic and random influences. ► Highlight of the main pros and cons of these overall uncertainty models. ► Performance evaluation of these models with respect to compliance assessment. ► Review of the main area of applications of these overall uncertainty models.