Reference and additional methods for measurement uncertainty evaluation

In this paper, the authors introduce additional methods to the mainstream GUM Guide to the expression of uncertainty in measurement for testing laboratories, namely methods based on repeatability or reproductibility and trueness approaches. These methods comply with the concepts and recommendations of the GUM. The additional methods are introduced to solve the case where a measurement model could not be established. In some circumstances, a measurement model is not available either due to technical difficulties or cost constraints. In this situation a " global statistical model" is used. In fact the "statistical model" represents a simplification of the measurement model where some input quantities have been merged. The most important application area of these approaches certainly concerns testing activities, analytical chemistry and medical laboratories. Medical laboratories are now encouraged to evaluate their uncertainty of analysis, either by regulation or on a voluntary basis in an accreditation process. In this paper, the authors present the GUM approach and additional methods, their coherence, and their common requirements: a clear definition of the measurand and a technical analysis of the measuring process to identify the uncertainty factors. (C) 2016 Elsevier Ltd. All rights reserved.