简介概要

能力验证结果稳健统计值与有证参考物质认定值的符合性比较研究

来源期刊:冶金分析2010年第11期

论文作者:王海舟 胡洛翡 罗倩华 孔鑫鑫 李小佳

文章页码:1 - 5

关键词:能力验证; 稳健统计; 有证参考物质; 认定值; 差值不确定度; 方法临界差

Key words:proficiency testing; robust statistic; certified reference material; certified value; uncertainty of difference; method critical difference

摘    要:由于能力验证计划系依托"大样本"试验,正常情况下各实验室结果的稳健统计值应更逼近真值,据此可采用有证参考物质作为能力验证计划共同试验样品,依据能力验证结果的稳健统计值,用以比较验证有证参考物质的认定值的符合性。本文分别提出了两种评价方法:其一为能力验证结果稳健统计平均值与有证参考物质认定值间差值的不确定度评价方法。当能力验证结果的稳健统计平均值与有证参考物质认定值之差的绝对值不大于该差值的扩展不确定度时,可以认为该有证参考物质的认定值及其不确定度的评估合理;当能力验证结果的稳健统计平均值与有证参考物质认定值之差的绝对值大于该差值的扩展不确定度时,则应检查该认定值(或能力验证结果稳健统计平均值)可能存在的偏倚或其不确定度的评估是否偏小。其二为有证参考物质认定值的方法临界差评价方法。当有证参考物质的认定值与能力验证结果的稳健统计平均值之差的绝对值不大于具有最小重复性限及再现性限方法的临界差(CD0.95)与能力验证结果稳健统计平均值的方和根时,则表明该有证参考物质相关参数的认定值能满足所有已发布的相关标准方法对临界差的要求,可以用于相关质量控制或方法的评估。

Abstract: Since the proficiency testing schemes depended on "numerous participants" test,the robust statistic value of experimental results from laboratories under normal conditions is more close to the true value.Accordingly,the certified reference materials (CRMs) can be used as common testing samples for proficiency testing schemes,and the robust statistic value of proficiency testing (PT) results can be used for the comparison and verification of the conformity of certified value of CRMs.In this paper,two evaluation methods were proposed.(1) Uncertainty evaluation method of difference between the average robust statistic value of PT results and the certified value of CRM.When the absolute value of difference between the average robust statistic value of PT results and the certified value of CRM was not higher than the expanded uncertainty of the value of difference,it could be considered that the certified value of CRM and uncertainty evaluation were reasonable.When the absolute value of difference between the average robust statistic value of PT results and the certified value of CRM was higher than the expanded uncertainty of the difference,it should be checked whether the evaluated uncertainty was lower or there was possible bias of certified value.(2) Method critical difference evaluation of certified value of CRM.When the absolute value of difference between the certified value of CRM and the average robust statistic value of PT results was not higher than the root of square sum of the method critical difference(CD0.95) with minimum repeatability limit and reproducibility limit and the average robust statistic value of PT results,it indicated that the certified value of related parameters of CRM could meet the requirements of critical difference in issued related standard method,and it could be used for the control of quality or evaluation of methods.

详情信息展示

能力验证结果稳健统计值与有证参考物质认定值的符合性比较研究

王海舟1,胡洛翡1,罗倩华1,孔鑫鑫1,李小佳1

(1.北京市钢铁研究总院)

摘 要:由于能力验证计划系依托"大样本"试验,正常情况下各实验室结果的稳健统计值应更逼近真值,据此可采用有证参考物质作为能力验证计划共同试验样品,依据能力验证结果的稳健统计值,用以比较验证有证参考物质的认定值的符合性。本文分别提出了两种评价方法:其一为能力验证结果稳健统计平均值与有证参考物质认定值间差值的不确定度评价方法。当能力验证结果的稳健统计平均值与有证参考物质认定值之差的绝对值不大于该差值的扩展不确定度时,可以认为该有证参考物质的认定值及其不确定度的评估合理;当能力验证结果的稳健统计平均值与有证参考物质认定值之差的绝对值大于该差值的扩展不确定度时,则应检查该认定值(或能力验证结果稳健统计平均值)可能存在的偏倚或其不确定度的评估是否偏小。其二为有证参考物质认定值的方法临界差评价方法。当有证参考物质的认定值与能力验证结果的稳健统计平均值之差的绝对值不大于具有最小重复性限及再现性限方法的临界差(CD0.95)与能力验证结果稳健统计平均值的方和根时,则表明该有证参考物质相关参数的认定值能满足所有已发布的相关标准方法对临界差的要求,可以用于相关质量控制或方法的评估。

关键词:能力验证; 稳健统计; 有证参考物质; 认定值; 差值不确定度; 方法临界差

<上一页 1 下一页 >

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号