This standard supersedes and replaces the old AMT (NMTBA) "Definition and Evaluation of Accuracy and Repeatability of Numerically Controlled Machine Tools," as well as the German VDI/DGQ 3441 on the same subject.

The "Test code for machine tools-Part 2: Determination of accuracy and repeatability of numerically controlled axes" ISO 230-2: 2006 can be purchased from:


The revised standard reflects new thinking behind the statistical approach to accuracy and repeatability; the previous assumption of random (Gaussian, bell-shaped or normal) distribution of programmed target hits has been abandoned. The new thinking reflects newly-acquired knowledge about factors affecting accuracy (e.g., friction, thermal effects, etc.), some of which could be controlled. Consequently, one does not deal with "standard deviation" (sigma) but rather with "standard uncertainty" (term used from the theory of uncertainty of measurement).

Another major change in the new standard is that instead of using 3 "standard deviations" (old standard), 2 "standard uncertainties" are used in line with a host of other characteristics when reporting the resulting accuracy and repeatability. The calculation for "standard uncertainty" is the same as for "standard deviation."

For additional information contact Paul Warndorf, AMT’s Vice President–Technology, at 703-827-5291 or e-mail: