measurement accuracy and uncertainty when making statements of conformity to specified requirements. It focuses on the simple acceptance and rejection decision rule, which is the most commonly used decision rule in measurement practice. The rule is based on the American national standard (ANSI) ASME B89.7.3.1-2001 Guidelines for Decision Rules and other historical documents. The rule is used in almost all situations, with measured values inside tolerance limits considered acceptable and those outside considered unacceptable.
Measurement professionals are aware that there is always error in measurement, and the use of the simple acceptance and rejection decision rule has always been accompanied by some stipulation that the measurement quality is sufficiently good. The U.S. Military Standard, MIL-STD-120 Gage Inspection, was released in 1950, which stated that the accuracy tolerances of the measuring equipment should not exceed 10% of the tolerances of the parts being checked. This rule has transformed over the years into the Test Accuracy Ratio (TAR), or test accuracy ratio.
Decision rules and associated ratios like the TAR are all related to risk in measurement systems. Organizations must determine what decision rule meets their unique needs and which ratios, if any, are considered acceptable. The purpose of this technical bulletin is to add clarity to the simple acceptance and rejection decision rule regarding how to calculate the ratios often part of the decision rule.
The evaluation of measurement uncertainty became more prevalent in commercial calibration practice in the late 1990s. As calibration laboratories began calculating and documenting uncertainty, the practice of using Test Uncertainty Ratio (TAR) calculations began to be replaced with the test uncertainty ratio (TUR). TUR requirements are now found in many national and international standards for calibrating measuring equipment. The TAR is generally larger than the TUR, as measurement uncertainty includes all sources of variation and not just the specified accuracy of the measuring equipment. Both TAR and TUR can be useful in measurement practice, with TUR being most useful in selecting calibration providers and TAR useful in selecting new measuring equipment.