Inadequacy for mathematical models of measurement objects and risk calculations according to ISO/IEC 17025-2019

Abstract

The problem of inadequacy of mathematical models of measurement objects is considered in connection with the problem of “definitional uncertainty of measurement” and the need for risk management in accordance with GOST ISO/IEC 17025-2019 “General requirements for the competence of testing and calibration laboratories”. The prehistory of the problem is described: from the formation of moment and compositional approaches to estimating accuracy to the introduction of a special term “inadequacy error for mathematical model of measurement object”. The negative impact of hopelessness of conceptual and terminological transformations in metrology and critical contradiction of the applicability estimates for “statistical methods” of GOST R ISO/IEC 31010-2011 “Risk management – Risk assessment techniques” and “Guide to the expression of uncertainty in measurement” is noted. It is shown that taking into account the inadequacy of probabilistic models in risk calculations is a necessary condition for results reliability.