turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- PROC SHEWHART example S Chart with Varying Sample ...

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

05-20-2016 11:56 AM

Hi,

I'm looking for a published example (online or in the docs) of data and results from running an S chart (i.e. control chart for within group standard deviations), where the group size varies.

I see an example of this for an XRChart:

http://support.sas.com/documentation/onlinedoc/qc/ex_code/141/shwmns.html

But I want an SChart (or XSCHART), using the default method used by SAS (ASTM 1976).

Goal is to find some published data to use in a validation report. Would appreciate any suggestions. (I'm going to try and buy the ASTM book, but hoping to find something online faster than snail mail).

Thanks,

--Q.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

05-20-2016 01:34 PM

Not sure what "ASTM 1976" means, but have you seen the chapter "Analyzing Quality Data with SAS", which includes an XSCHART example on p 9?

The excerpt is from Rutledge *Just enough SAS: A Quick-Start Guide to SAS for Engineers.* The data are available from the book web page.. Just click "Example Code and Data."

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

05-20-2016 02:18 PM

Thanks Rick, I hadn't seen that book.

Unfortunately, I'm looking for an example of an S chart (or XSCHART) where the subgroup size varies, so the process limits on the S chart will also vary accordingly.

For example, this:

data have; do lot=1 to 20; do n=1 to floor(ranuni(3)*20); *random number of values in each group; y=100+sqrt(10)*rannor(3); output; end; end; run; proc shewhart data=have; xschart y*lot; run;

Produces a chart like:

By "ASTM 1976" was referring to the reference for the calculation in th docs:

American Society for Testing and Materials (1976), ASTM Manual on Presentation of Data and

Control Chart Analysis, 1916 Race Street, Philadelphia, PA 19103.

Amazon is selling it for $5, so hopefully I'll be lucky when it arrives. : )

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

05-20-2016 03:07 PM - edited 05-20-2016 03:16 PM

If you go to Google and search in "Images" for

shewhart x chart unequal sample size

you will see many examples. I'm not sure whether these qualify as "published examples."

For example, one instance is a set of course notes:

http://www.math.montana.edu/jobo/st528/documents/chap6b.pdf (p 68-69)

If you need a published example of high importance, you can find example data and chart on pp 244-248 of

*Introduction to Statistical Quality Control* by Douglas C. Montgomery, 4th Ed.

Do the results have to be created by SAS?

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

05-20-2016 03:16 PM

Thanks again Rick. I've got Montgomery 2nd Ed open on my desk, but it looks to me like the example has a couple errors, and I don't think it's using the same method as SAS. I should probably get the latest edition. Seems like Montgomery is a standard text for this stuff. No, I don't need it to be created by SAS. And in a validation sense, arguably better if the truth I use is not from SAS.