Examples of Measurement Uncertainty Evaluation This project provides a comprehensive set of worked examples illustrating how the principles of measurement uncertainty evaluation can support and give added value to normative and related practices. Overview This project is providing a comprehensive set of worked examples illustrating how principles of measurement uncertainty evaluation can support documentary standards and guides. It is promoting uncertainty evaluation according to internationally recognized guides across broad disciplines of measurement. The project is delivering new or improved adaptable examples of and templates for uncertainty evaluation to the Joint Committee for Guides in Metrology (JCGM) as publishers of the internationally acknowledged Guide to the expression of uncertainty in measurement (GUM). The project is also providing examples to ten standardization bodies that are specifically related to standards they are developing. Project poster Need Measurement models describe the relationship between quantities we measure or know something about (input quantities) to the quantities of interest (output quantities). In areas such as energy, environment and health care these models are frequently non‑linear and the quantities measured may have substantial uncertainty. In almost all scientific areas, uncertainties associated with the output quantities are to be calculated given the input uncertainties. The traditional approach to uncertainty propagation through a model uses the law of propagation of uncertainty (LPU) of the GUM, but the uncertainty so produced may not always be fit for purpose. Carefully elaborated examples are being prepared in the project that are practical and apply to many domains, capable of delivering reliable results, and as far as possible in a form that can be adapted to actual end-users’ data and knowledge. Since many end-users “learn by example”, a diverse set of practical examples, ranging in complexity from the simple (for example, linear interpolation of pH values) to the sophisticated (for example, impact of voxel size on perceived tumour mass), is being developed. A specific instance is the continued use of illegal substances in sport, which is fuelled by individual, team or club ambitions, or nationalistic desires to provide competitive performances. There is a consequential need in the area of doping control to determine the best approach for the evaluation of uncertainties given an athlete’s sample, and the verification of these uncertainties. In health care, specifically in cancer treatment, where a stringent maximum dose is imposed for reasons of radiation safety, methods beyond traditional LPU may have to be applied so that the computed volumes and flows, and their associated uncertainties, ensure safe dose delivery, especially for paedeatric care, where nano-volumes and micro-flows require particular consideration. The development of examples in these and many other topical and essential areas is well under way in the project. Objectives The overall objective is to provide a comprehensive set of new and improved examples to illustrate uncertainty evaluation methods that are in accordance with recognized international practice, specifically the GUM suite of guidance documents. Some examples concern the traditional metrology areas of calibration, testing, comparison and conformity assessment. Further examples relate to the thematic areas of environment, energy, quality of life, and industry and society. The examples are being offered to the JCGM and its member organisations for use in the examples document JCGM 110 it is developing to illustrate the application of the GUM suite of documents. They are also being provided to standards committees and other organisations that have expressed a need for them. The focus is on supplementing state-of-the-art measurement and research with best practice in measurement uncertainty evaluation. Specific objectives are: To develop examples of measurement uncertainty evaluations capable of acting as template solutions that end users can use for related problems. Examples will include measurement model construction using JCGM 103 (currently being finalized), application of uncertainty evaluation principles for addressing industrial conformity assessments to support JCGM 106, and taking correlations into account as requested by ISO/REMCO, the ISO committee concerned with reference materials. To derive worked examples of uncertainty analyses using the GUM and other methods to assist users to make informed choices on an appropriate uncertainty evaluation method to use. Examples will include an examination of the extent to which the GUM is appropriate for certain applications or whether the Monte Carlo methods of GUM Supplement 1 and GUM Supplement 2, or Bayesian methods, have greater efficacy. To collaborate with JCGM-WG1 (the chief stakeholder), and the standardisation, regulatory and accreditation communities (ISO/REMCO, IEC, CEN, OIML, and ILAC) to ensure that the outputs of the project are aligned with their needs, communicated quickly, and in a form that can readily be incorporated into the JCGM Guides and other documents.
