P
I recommend performing a Poisson capability study. This is typically used for defects (which are count data), but can be interpreted as you need. It will provide the percentage of a given color with the expected variation about that percentage.
Also here is my data for anyone that wanted to take a look. 48 Bags so far, just kinda threw it all into excel and planned on making it pretty later when I get a chance.
also what program would you all suggest for generating the theoretical data for if the placement of the bars was purely random? that is one of the important things i need to do i think because i want to be able to say whether we can improve the process by mixing the candy better before bagging or if we already mix enough and need to retool to something that will count individual colors in each bag.
Attachments
-
15.3 KB Views: 110
Last edited by a moderator:

