VARIANCE COMPONENTS OF 'TYPE A' STANDARD UNCERTAINTIES IN A NESTED SEQUENCE OF MEASUREMENTS

摘要

The realization of the International Temperature Scale of 1990 (ITS90) requires that the Laboratories usually have more than one cell for each fixed point. These cells must be compared in a regular basis and the experiments repeated in subsequent days. The measured differences are obtained with two or three standard thermometers and this complete experiment may be repeated some time after. The uncertainty calculation will take into account these time-dependent sources of variability. The "Type A" components of uncertainty can be evaluated by the statistical analysis of the data obtained from designs consisting of nested or hierarquical sequences of measurements. The technique, analysis of variance (ANOVA) and the estimation of the variances components is described. An application example is drafted on the evaluation of these components of standard uncertainty in the comparison of the triple points of two water cells.