Attribute vs. variable data

[Kromkowski@aol.com]

Although you addressed this to statisticians, I respond below:  

"how many bags they should have produced according to the machine 
specifications"

How was it that the machine specifications where made?  (This is a rhetorical 
question initially) I think that use should use an X chart (I'd begin with 
IndX/Mr and later _if rational_ subgrouping makes sense).  There is clearly 
no independence of data.  [You also said np chart, but if you have 
percentages then wouldn't you actually call it a p-chart? P-charts are best 
used for charts for proportion defective.  So you could use this kind of 
chart for taking a look at the proportion of bags that are broken or 
otherwise defective.  Not a lot of sense in having a machine being really 
efficient at creating waste.]

I note once again that this as well as other similar post could serve as 
wonderful case studies.  

To that end, post some actual data!

We can then collectively (you can do the initial calculations and we can 
check) find out if the process is stable.

If signals for special cause investigation, you can share with us your 
investigation.

You can also share with us the investigation for common causes of variation, 
theories for change and plans to test, etc.

I note the machine specifications may well come into play, once the X chart 
shows statistical control and you begin attempts to reduce common cause 
variation and optimally center the process in relation to downstream 
customers (nb., centering may not mean running the machine at maximum output 
if messes up things down and up stream in light of overall aim of enterprise. 
 Consider selling too big or too small a copier example or the long term 
overall benefit of running the machine like crazy for a week and then waiting 
10 weeks to repair it.

The sharing of the data, etc., (a summary of your PDSA cycle) as proposed 
seems a fair price to pay to the DEN for providing you information.

"Good management and good supervision require knowledge of the calculations 
that will separate the two kinds of cause."  OOTC at 316.
--
John David Kromkowski
Attorney at Law
6600 York Road - Suite 108
Baltimore, Maryland   21212-2028
Kromkowski@aol.comTelephone:  (410) 377-6248Facsimile:  (410) 372-0624
=====================================================================