ASTM Offers Guide for Measurement Systems Analysis
ASTM E2782 was developed by Subcommittee E11.20 on Test Method Evaluation and Quality Control, under the jurisdiction of ASTM International Committee E11 on Quality and Statistics.
According to Stephen Luko, a statistician with Hamilton Sundstrand and a longtime committee member and former chairman of E11, when an object is measured and a result presented, typically some decision will be made or action taken as a result of the measurement result. In such cases, Luko notes, it is always reasonable to ask the following questions: “What is the error in the measurement result?” and “What is the error in the measurement process?”
“From a practical point of view, the operator of a measured quantity wants to know how different might the measured quantity be from the true value of what has been measured,” says Luko. “This is the essence of MSA.”
As an example, Luko says that if the measurement error is too large, this result may render the measurement results either useless or of limited value. There are several components to measurement error as well, and the statistical practice of measurement systems analysis shows you how to separate these components and improve measurement performance for the various components.
The new standard will be useful anywhere measurement systems are used or where measurement error is important to assess. This includes work by many ASTM committees, along with user communities found in manufacturing and engineering firms, and in more general business practices.
As currently approved, ASTM E2782 applies to variable measurements and nondestructive applications. Subcommittee E11.20 plans to incorporate destructive measurement applications as well as applications involving attribute-type measurements in future revisions of the standard and invites all interested parties to contribute to this work.