Editorial Review:
A solid, rigorous, yet comprehensible analysis of process capability indices, this work bridges the gap between theoretical statisticians and quality control practitioners, showing how an understanding of these indices can lead to process improvement.
Cached date: AWS Called=true
You may also be interested in these products:
These categories may also be of interest to you:
Customer Reviews
Average Customer Rating: 
first good statistical treatment for general distributions 2008-01-24
The quality movement in the US in the 1980s and 1990s has led to a great deal of quality efforts that stress statistical measurements of process capability. These indices are intended to answer the question "How often will a manufactured part fall outsided specification limits?" Certain standards and tests based on capability indices have validity when the process variation has a normal distribution.
However, it has been my experience in the medical device industry that many processes are non-normal and that the application of the normal theory in these cases can lead one astray. Others have found this to be the case in the automobile industry as well as in other industries.
Also some people treat these indices as though they are known constants when in practice we almost always use sample estimates of means and standard deviations in our calculation of the index. This means that the "index" is itself an estimate of the capability parameter.
These issues are recognized and emphasized by Kotz and Johnson in this wonderful little monograph. It was the first book to address many of these issues and to summarize what it known based on the scattered literature. They treat all the major indices and present normal theory and bootstrap alternatives among others. It is very authoritative and is an important reference for anyone dealing with these quality control issues.
first good statistical treatment for general distributions 2001-05-24
The quality movement in the US in the 1980s and 1990s has led to a great deal of quality efforts that stress statistical measurements of process capability. These indices are intended to answer the question "How often will a manufactured part fall outsided specification limits?" Certain standards and tests based on capability indices have validity when the process variation has a normal distribution.However, it has been my experience in the medical device industry that many processes are non-normal and that the application of the normal theory in these cases can lead one astray. Others have found this to be the case in the automobile industry as well as in other industries. Also some people treat these indices as though they are known constants when in practice we almost always use sample estimates of means and standard deviations in our calculation of the index. This means that the "index" is itself an estimate of the capability parameter. These issues are recognized and emphasized by Kotz and Johnson in this wonderful little monograph. It was the first book to address many of these issues and to summarize what it known based on the scattered literature. They treat all the major indices and present normal theory and bootstrap alternatives among others. It is very authoritative and is an important reference for anyone dealing with these quality control issues.
|
|
