The purpose of this Technical Specification is to specify a validation procedure to show if an Alternative Method (AM) can be used as an alternative to the Standard Reference Method (SRM), both implemented to determine the same measurand. This document has been drawn up for laboratories working in air pollution measurements (and consequently examples taken from this sector are included in the appendices).
In particular, this Technical Specification provides the statistical tools and different criteria to evaluate the alternative method; this does not release the person responsible for this validation from bearing technical and analytical judgement on the evaluation of the different criteria.
Three steps are described in the validation procedure:
⎯ description of the AM and setting of the field of equivalence (range and type of gas matrix);
⎯ determination of the performance characteristics of the AM and calculation of the overall uncertainty where appropriate and check of compliance of the maximum overall uncertainty allowed for the SRM;
⎯ check of repeatability and lack of systematic deviation of the AM in the field in comparison with the SRM for the type of matrix defined in the field of equivalence.
NOTE Some parts of the second step of the validation of the alternative method should be performed by a recognised test-house.
If the AM fulfils the requirement of the procedure, then the laboratory that carried out the whole validation process is allowed to use it as a SRM in the field application where the equivalence has been demonstrated.
However, if the validation process involves at least 4 different accredited laboratories performing simultaneously parallel measurements in the field, and if the AM passes with success all the tests of the procedure, then this method could be proposed to CEN, who can decide to consider this AM as a new reference method (ARM).
The use of this procedure implies that a reference method has been defined by the regulator or in a contract and has been validated.
This Technical Specification only considers the case of linear quantitative methods.