metRology: A New R Package for Measurement Science

• Activity Number: 67
• Type: Topic Contributed
• Date/Time: Sunday, August 1, 2010 : 4:00 PM to 5:50 PM
• Sponsor: Section on Quality and Productivity
• Author(s): Stephen L.R. Ellison*+
• Companies: Laboratory of the Government Chemist
• Keywords: metrology ; software ; open source ; measurement uncertainty

Abstract

This talk presents the elements, features and architecture of a new open source R package for metrology applications; that is, for statistics applied to measurement science. The implementation uses S3 classes to provide a simple and extensible framework for exploring statistical procedures in metrology. Features currently include plots for structured inter-laboratory comparisons, implementation of current methods of measurement uncertainty evaluation and location estimates for metrology comparisons. Uncertainty evaluation methods include symbolic or numerical derivatives and Monte Carlo evaluation, including support for correlation. Classes and methods are defined for uncertainty evaluations, inter-laboratory study data, and location estimates. The principal features will be demonstrated, with examples. The rationale for the choice of architecture will be presented and discussed.