The Role of Uncertainty Estimation on the Improvement of Measurement Processes and Comparison Between Results

Measurement process is an important part of the routine work in a laboratory. New analytical methods, techniques, instrumentation and software packages for calculation of results and measurement uncertainties make this process to be in constant improvement. The scientists and technicians must receive training to adapt to the changes and to develop a keen sense of awareness of data quality and the uncertainty estimates of the reported results. Correctly evaluating the uncertainties depends on several factors including, but not limited to, experience in the analyses techniques. More important is effectively and correctly communicating the results and the uncertainties in a manner that are understood not only by the originator but also by all other users of those results. International efforts have been made in order to establish standard methods for calculating and expressing measurement uncertainty so that scientists can talk the same language when analysing data and reporting measurement results. Several publications such as the “Guide to Expression of Uncertainty in Measurement” (GUM) [1] and the “Quantifying Uncertainty in Analytical Measurements” (EURACHEM / CITAC Guide CG 4) [2] are excellent source materials for training in the estimation of uncertainties in measurement results according to ISO guidelines. The evaluation of measurement uncertainty has also become a central element in the establishment of quality assurance programs within nuclear analytical laboratories. In this paper, we will discuss uncertainty estimation as a means to evaluate some measurement processes typically used for nuclear material control and accountability. Intra- and inter-laboratory comparisons of measurement results must take into account the uncertainties associated with each result. Examples of such comparisons will be provided and discussed.