Harmonised Values and Reference Functions in Key Comparisons

摘要

Key comparisons in the field of flow measurement have some peculiarities with impact on data evaluation and performance assessment, in particular on the caluclation of the key comparison reference value (KCRV). The normal KC approach which assumes that all measurements in all laboratories are made at prescribed nominal sampling points faces problems in practical implementation. Furthermore, it is normally assumed that values submitted by participants are independent of each other. It is well-known that this assumption may not hold in specific cases since some of the factors which influence the uncertainty may be the same in all or at least a couple of laboratories. The paper discusses the influence of the above factors. The problems are handled by calculating key comparison reference functions (KCRF) instead of series of isolated KCRV, and by correctly evaluating, and accounting for all correlations present within the data sets.