Decision rules for workpieces based on total uncertainty

J. X. Wang, X. Jiang, L. M. Ma, Z. G. Xu, Z. Li

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

The decision rules based on measurement uncertainty are given in ISO 14253-1, according to which only measurement uncertainty is considered in the three-dimensional measurement process, and the workpieces are simply accepted or rejected by measurement uncertainty. It can be seen that the decision rules don't accord with the basic requirements of the improved GPS system because correlation uncertainty and specification uncertainty are not taken into account, which are existent actually in a GPS process. In order to solve this problem, this paper presents the decision rules based on total uncertainty, according to which the workpieces should be accepted or rejected by total uncertainty. The biggest difference between the two decision rules is whether correlation uncertainty and specification uncertainty have been taken into account. For a given GPS specification, the key of the decision rules based on total uncertainty is to calculate the compliance uncertainty of the corresponding GPS standard-chain. According to ISO 17450-2, a GPS process should be either in the default state or in the special state. Aiming at the two states, the calculation method for compliance uncertainty of a GPS standard-chain is given respectively. It enables us to generate compliance uncertainty on the verification of a GPS specification, which means the acceptance or refusal of a GPS characteristic can be conducted in a quantitative way. The example of flatness least-square assessment indicates that the veracity of assessment could be improved.

Original languageEnglish
Pages (from-to)1169-1174
Number of pages6
JournalInternational Journal of Advanced Manufacturing Technology
Volume28
Issue number11-12
Early online date29 Jun 2005
DOIs
Publication statusPublished - 1 May 2006

Fingerprint Dive into the research topics of 'Decision rules for workpieces based on total uncertainty'. Together they form a unique fingerprint.

Cite this