![](https://ghaaemi.ir/wp-content/uploads/2023/04/distributions.gif)
The Type B evaluation of uncertainty involves estimating the quantity u^2j, which may be taken as an approximation to the corresponding variance, the existence of which is assumed. The quantity u^2j is treated like variance and the quantity uj like standard deviation,and where appropriate the covariances should be treated in a similar way.
In case the probability distribution of xp is subjective on a predetermined distribution, a quantity of the nature of variance is determined from the knowledge of its distribution. Hence Type B evaluations are based on the predetermined distributions as well
It may be emphasized that in both cases the distributions are models that are used to represent the state of our knowledge. Uncertainties due to any input quantity Xp should be evaluated by both Type A and Type B evaluation methods
Prior to BIPM directive, in 1980s, uncertainty consisted of two components: one used to come from random errors and the other due to systematic errors. The two components of uncertainty used to be named as random uncertainty and systematic uncertainty. There is not always a simple correspondence between the classification into Types “A” and “B” and the previously used classification into “random” and “systematic” uncertainties. The term “systematic uncertainty” can be misleading and should be avoided.
In case the measurand – the output quantity – is a function of several input quantities, all variances and covariance should be combined by quadrature method. Variances obtained by Type A evaluation method are not distinguished from those obtained by Type B evaluations. So all variances and covariances should, therefore, be treated as variances or covariances in strict statistical sense.
As mentioned earlier, for those uncertainties, whose estimates have not been obtained by independent repeated observations, Type “B” evaluation method is used. This type of uncertainty is calculated by judgement using all relevant information on the variability of the uncertainty.
For example:
Previous measurement data
Experience and general knowledge of the behaviour and properties of relevant materials and instruments
Manufacturer’s specification
Data provided in the calibration and other certificates
Uncertainty assigned to reference data taken from handbooks
Uncertainties may creep in a measurement process due to the use of:
Standards
Measuring instruments
Inherent characteristic of the instrument under calibration
Various physical constants
Values of physical properties of the material used in standards and measuring instruments
Operating conditions
Common Uncertainties Evaluated by Type B
Uncertainty as reported in the calibration certificates of the standard or the instrument used.
Uncertainty due to interpolation between the calibration points of the standard used in the measurement.
Uncertainty due to the change in environmental conditions, such as temperature, pressure and relative humidity of air.
Uncertainty due to ability to reset, repeatability and threshold discrimination of the instruments used.
Uncertainty due to the value taken for some physical constants, or properties of the materials used in the process of measurement, such as values of density of water, acceleration due to gravity and expansion coefficients.
Uncertainty in applied corrections based on measurements or the data obtained from standard handbooks
Source : measurement uncertainty