Conformity assessment of a proficiency test using turbine flowmeters based on the concept of tolerance intervals

Conformity assessments can be classified into two categories according to compliance with specifications. The first category involves “simple acceptance” to determine the measurement equivalence. The second category consists of “guarded acceptance” to introduce guard bands, considering the producer...

Full description

Saved in:
Bibliographic Details
Published inMeasurement : journal of the International Measurement Confederation Vol. 238; p. 115287
Main Authors Chun, Sejong, Nam, Hyun-Seung, Yoon, Byung-Ro
Format Journal Article
LanguageEnglish
Published Elsevier Ltd 01.10.2024
Online AccessGet full text

Cover

Loading…
More Information
Summary:Conformity assessments can be classified into two categories according to compliance with specifications. The first category involves “simple acceptance” to determine the measurement equivalence. The second category consists of “guarded acceptance” to introduce guard bands, considering the producer and consumer risks. [Display omitted] •Simple acceptance has an acceptance interval of EN≤1.•Guarded acceptance has an acceptance interval of EN≤0.8 with guard bands of 0.8≤EN≤1.2.•Concerning state is a type of consumer risk which can be summarized by 0.8≤EN≤1.•Warning state is a type of producer risk which can be summarized by 1≤EN≤1.2. A proficiency test determines either conformity or non-conformity states by evaluating the normalized degree of equivalence, EN, among laboratories participating in the test. Concern is paid to the results if EN indicates a boundary between conformity and non-conformity states. Either “concerning” or “warning” can urge the participating laboratories to correct their measurement standards with their measured results. This study investigates how to determine guard bands, i.e., the boundary between conformity and non-conformity, based on the normalized degree of equivalence. Two types of guard bands are demarcated at the center of the boundary of the acceptance interval. 1≤EN≤1.2 is set as a guard band to provide “warning” to a participating laboratory considering producer risks, 0.8≤EN≤1 is set as another guard band to express “concerning” to the participating laboratory considering consumer risks. A proficiency test using turbine flowmeters with temperature correction is conducted to confirm the feasibility of these guard bands.
ISSN:0263-2241
DOI:10.1016/j.measurement.2024.115287