Re: stable system?

[Kromkowski@aol.com]

Here are my thoughts.

1.  Stop using the Standard Deviation (likely calculated by least squares on 
a computer or calculator).  Use an Ind.X/Mov. Range chart.  This preserves 
information found in data order.  

2.  Using an IndX/mR,  data manipulation (a "special" kind of special cause 
(rules of course to be determined in advance) one of manner run rules) is not 
the first thing that comes to mind. It doesn't come to mind at all, 
especially with so few samples.

3.  The range chart suggests it might be worth it looking for reason for 
difference between point 3 and point 4.  (Personally, I'd take the time to 
get 30 data points and redo the chart.)

4.  Precision.  Your measurements are only to the nearest gram so I 
personally tend to think that your limits should be rounded to the same 
measure (or at least no more precise than 1/2 the smallest unit used -- .5 
grams since one can say that the measuring device (at least back in the 
analog days) can come up with information in between the "ticks".  (Do they 
even make digital scales with this little precision?).  Although, in 
rereading your post, I am now confused as to whether you are posting the 
values for "ten individual samples" or "ten averages of ten samples". 

5.  No way current system can meet goal.  Who is getting the big bucks for 
"setting" that "stretch" goal.  

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