{"title":"Conformance Decision Rules to Support ISO/IEC CD2 17025 Under Revision","authors":"R. Stern","doi":"10.51843/wsproceedings.2016.32","DOIUrl":null,"url":null,"abstract":"The current draft of ISO/IEC CD 17025 contains a clause to document the decision rule for conformity along with the risk and statistical assumptions in the test method or procedure (7.1.1d). So how shall we do this? A common strategy for managing measurement decision risk is to choose a guard-band that results in a desired false-accept risk given a tolerance limit, the calibration process uncertainty, and the a priori probability. JCGM 106:2012 provides guidance for the implementation of prior knowledge in conformance decisions, however a good estimate of the a priori probability may be difficult to obtain. Historical device population information for estimating a priori probability may not be readily available such as \"testing\" applications where the device is a prototype. This paper presents the false-accept risk and corresponding false-reject risk for four decision rules. We provide simple equations to limit risk to pre-determined amounts. Several case studies illustrate when to use \"Specific\" or \"Global\" risk. When \"Global\" risk is appropriate, the decision rules can be applied with limited knowledge of a priori probability.","PeriodicalId":162467,"journal":{"name":"NCSL International Workshop & Symposium Conference Proceedings 2016","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"NCSL International Workshop & Symposium Conference Proceedings 2016","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.51843/wsproceedings.2016.32","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The current draft of ISO/IEC CD 17025 contains a clause to document the decision rule for conformity along with the risk and statistical assumptions in the test method or procedure (7.1.1d). So how shall we do this? A common strategy for managing measurement decision risk is to choose a guard-band that results in a desired false-accept risk given a tolerance limit, the calibration process uncertainty, and the a priori probability. JCGM 106:2012 provides guidance for the implementation of prior knowledge in conformance decisions, however a good estimate of the a priori probability may be difficult to obtain. Historical device population information for estimating a priori probability may not be readily available such as "testing" applications where the device is a prototype. This paper presents the false-accept risk and corresponding false-reject risk for four decision rules. We provide simple equations to limit risk to pre-determined amounts. Several case studies illustrate when to use "Specific" or "Global" risk. When "Global" risk is appropriate, the decision rules can be applied with limited knowledge of a priori probability.
ISO/IEC CD 17025的当前草案包含一个条款,用于记录符合性决策规则以及测试方法或程序中的风险和统计假设(7.1.1d)。那么我们该怎么做呢?管理测量决策风险的一个常用策略是选择一个保护带,该保护带在给定容限、校准过程不确定性和先验概率的情况下产生期望的误接受风险。JCGM 106:2012为一致性决策中先验知识的实施提供了指导,然而,对先验概率的良好估计可能很难获得。用于估计先验概率的历史设备人口信息可能不容易获得,例如设备是原型的“测试”应用程序。本文给出了四种决策规则的误接受风险和相应的误拒绝风险。我们提供简单的公式,将风险限制在预先确定的数额。几个案例研究说明了何时使用“特定”或“全局”风险。当“全局”风险合适时,决策规则可以在有限的先验概率知识下应用。