文章基本信息

标题：Robust methods in acceptance sampling
本地全文：下载
作者：Elisabete Carolino ; Isabel Barão.
期刊名称：RevStat : Statistical Journal
印刷版ISSN：1645-6726
出版年度：2013
卷号：11
期号：1
页码：67-82
出版社：Instituto Nacional de Estatística
摘要：In the quality control of a production process (of goods or services), from a statisticalpoint of view, the fo cus is either on the process itself with application of StatisticalProcess Control or on its frontiers, with application of Acceptance Sampling (AS) andExperimental Design. AS is used to inspect either the process output (final pro duct)or the process input (raw material). The purpose of the design of a sampling plan isto determine a course of action that, if applied to a series of lots of a given quality,and based on sampling information, leads to a specified risk of accepting/rejectingthem. Thus AS yields quality assurance. The classic AS by variables is based on thehypothesis that the observed quality characteristics follow the Gaussian distribution(treated in classical standards). This is sometimes an abusive assumption that leadsto wrong decisions. AS for non-Gaussian variables, mainly for variables with asym-metric and/or heavy tailed distributions, is a relevant topic. When we have a knownnon-Gaussian distribution we can build specific AS plans asso ciated with that distri-bution. Alternatively, we can use the Gaussian classical plans with robust estimatorsof lo cation and scale — for example, the total median and the sample median aslocation estimates, and the full range, the sample range and the interquartile range,as scale estimates. In this work we will address the problem of determining AS plansby variables for Extreme Value distributions (Weibull and Fr′echet) with known shapeparameter. Classical plans, specific plans and plans using the robust estimates forlocation are determined and compared.
关键词：quality control; acceptance sampling; acceptance sampling by variables; robust methods