146
8 Measurement Uncertainty
8.20 Advantages and Limitations of the Modeling
Procedure for Evaluation of Uncertainty
To summarize, it is worth emphasizing both the advantages and the limitations of
using modeling in the evaluation of the uncertainty of chemical measurements.
Advantages:
– Uniform and commonly accepted method of evaluation of uncertainty;
– Analytical procedure can be described in the form of a mathematical equation;
– Enables the optimization of the analytical procedure;
– Accepted by international organizations.
Limitations:
– Analytical procedure is often not easily modeled;
– Not all uncertainty contributions are easily quantified;
– Procedure requires a good knowledge of mathematical statistics;
– Procedure requires a detailed analysis of the measuring procedure;
– Often leads to the underestimation of the uncertainty.
8.21 Experimental Methods of Evaluation of Uncertainty
The modeling method requires the identification and clear determination of all constituents of the measuring procedur—those that influence the measuring result and its
uncertainty. It means that all stages of the procedure are well defined, their influence
on the result is known and it is possible to estimate the value of uncertainty for all
constituents.
In chemical measurements, where the measuring procedure comprises many
stages, it is not always possible to display a full mathematical model. The measurement result often depends on parameters that are hard to predict. The laboratory
practice shows that some values determined during the measurements reflect the
influence of a few sources of uncertainty at the same time.
Evaluation of uncertainty: useful data
– Data collected during the validation of the measuring procedure in the laboratory;
– Data collected during quality control (i.e. control charts) in the laboratory;
– Data from inter-laboratory comparisons (inter-laboratory validation);
– Data from proficiency testing.
Validation of measuring procedure in the laboratory
8 Measurement Uncertainty
8.20 Advantages and Limitations of the Modeling
Procedure for Evaluation of Uncertainty
To summarize, it is worth emphasizing both the advantages and the limitations of
using modeling in the evaluation of the uncertainty of chemical measurements.
Advantages:
– Uniform and commonly accepted method of evaluation of uncertainty;
– Analytical procedure can be described in the form of a mathematical equation;
– Enables the optimization of the analytical procedure;
– Accepted by international organizations.
Limitations:
– Analytical procedure is often not easily modeled;
– Not all uncertainty contributions are easily quantified;
– Procedure requires a good knowledge of mathematical statistics;
– Procedure requires a detailed analysis of the measuring procedure;
– Often leads to the underestimation of the uncertainty.
8.21 Experimental Methods of Evaluation of Uncertainty
The modeling method requires the identification and clear determination of all constituents of the measuring procedur—those that influence the measuring result and its
uncertainty. It means that all stages of the procedure are well defined, their influence
on the result is known and it is possible to estimate the value of uncertainty for all
constituents.
In chemical measurements, where the measuring procedure comprises many
stages, it is not always possible to display a full mathematical model. The measurement result often depends on parameters that are hard to predict. The laboratory
practice shows that some values determined during the measurements reflect the
influence of a few sources of uncertainty at the same time.
Evaluation of uncertainty: useful data
– Data collected during the validation of the measuring procedure in the laboratory;
– Data collected during quality control (i.e. control charts) in the laboratory;
– Data from inter-laboratory comparisons (inter-laboratory validation);
– Data from proficiency testing.
Validation of measuring procedure in the laboratory