Using R for Assessing Measurement Uncertainty

• 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): Hung-Kung Liu*+ and William F. Guthrie and Antonio Possolo and Stephen L.R. Ellison
• Companies: National Institute of Standards and Technology and National Institute of Standards and Technology and National Institute of Standards and Technology and Laboratory of the Government Chemist

Abstract

In recent years, the trend toward laboratory accreditation based on standards like ISO 17025 has greatly accelerated the need to include sound uncertainty statements in measurement results. Thus, a goal of statistical metrology has become providing software tools to properly assess measurement uncertainties. This talk describes the development and use of R functions for assessing measurement uncertainties using the methods outlined in the Guide to the Expression of Uncertainty in Measurement (also commonly referred to as the "GUM"). These flexible and easy to use R functions can handle a variety of typical correlation structures, and perform a validity check on the results by simulating the coverage probabilities actually attained by the uncertainty intervals.