Optimised measurement uncertainty and decision-making when sampling by variables or by attribute

Wherever there are uncertainties, there are risks of incorrect decision-making in conformity assessment associated both with sampling and measurement. The present work reviews how an earlier criterion for ‘fitness for purpose’ in analytical measurement can be extended to more general measurements in conformity assessment, in terms of a decision theory approach where the costs of analysis are balanced against the costs associated with the consequences of incorrect decision-making. This is considered to be an important step towards establishing clearer procedures for setting and specifying tolerances and associated uncertainties, and in facilitating acceptance of conformity in general by both customer and supplier. Examples are taken from legal metrology as one of the first applications of a costs-based, ‘fitness for purpose’ approach in this field. An original discussion in economic terms is given of common rules in conformity assessment, including the ‘shared risk’ principle and maximum permissible uncertainties. This work also includes a new optimised uncertainty methodology to testing and sampling by attribute. This goes beyond traditional statistical sampling plans in that it assigns costs to both testing and consequences of incorrect decision-making.