Data Analysis for Proficiency Testing

TL;DR

This paper introduces a multivariate statistical model for assessing laboratory measurement equivalence in proficiency testing, accounting for source variability and data complexity, with application to engine power measurements across multiple labs.

Contribution

It proposes a novel nonergodic multivariate model that incorporates type B variation and handles multilevel data in proficiency testing.

Findings

Model effectively assesses laboratory equivalence.

Application demonstrates practical utility in engine power testing.

Provides a detailed asymptotic analysis for nonergodic models.

Abstract

Proficiency Testing (PT) determines the performance of individual laboratories for specific tests or measurements and it is used to monitor the reliability of laboratories measurements. PT plays a highly valuable role as it provides an objective evidence of the competence of the participant laboratories. In this paper, we propose a multivariate model to assess equivalence among laboratories measurements in proficiency testing. Our method allow to include type B source of variation and to deal with multivariate data, where the item under test is measured at different levels. Although intuitive, the proposed model is nonergodic, which means that the asymptotic Fisher information matrix is random. As a consequence, a detailed asymptotic analysis was carried out to establish the strategy for comparing the results of the participating laboratories. To illustrate, we apply our method to analyze the data from the Brazilian Engine test group, PT program, where the power of an engine was measured by 8 laboratories at several levels of rotation.