Premium
Assessing non‐normally distributed processes by interval estimation of the incapability index C pp
Author(s) -
Ke J.C.,
Chu Y.K.,
Chung Y.T.,
Lin P. C.
Publication year - 2009
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.979
Subject(s) - index (typography) , estimation , statistical inference , inference , greenwich , interval (graph theory) , computer science , process (computing) , statistics , mathematics , econometrics , engineering , artificial intelligence , combinatorics , environmental science , systems engineering , world wide web , soil science , operating system
Process capability indices have been widely used in the manufacturing industry. While most studies consider estimation of capability indices for normal processes, comparatively little is known about their behavior in non‐normal settings. Greenwich and Jahr‐Schaffrath (Int. J. Qual. Reliab. Manage. 1995; 12:58–71) introduced the incapability index C pp to evaluate processes. In this paper, we explore the interval estimation of the incapability index C pp for non‐normally distributed processes by utilizing seven feasible methods. We further develop an efficient criterion, which is relative coverage, to evaluate the performance of the seven methods. Detailed discussion of simulation results for six non‐normally distributed processes is presented. The results display that the bootstrap pivotal method developed by Wasserman (All of Statistics: A Concise Course in Statistical Inference. Springer Science, Business Media, Inc., 2004) is the best feasible method to estimate C pp . An example is also demonstrated to illustrate how the method may be used in practice. Copyright © 2008 John Wiley & Sons, Ltd.