Vol. 7, No. 02
Last month, Marty Graw, the new quality manager at Natural Butter Company, tried to demonstrate his own capabilities by pursuing what he thought would be a dazzling capability study. Sketching a chart that he believed would reflect the data and provide visual support for his technicians, he offered the following:
The next step in the process is to determine placement of specification limits. He goes to his SPC reference Practical Tools for Continuous Improvement (Graham and Cleary) and makes an overhead of page 244:
Marty notes that there seem to be virtually no bad parts on the low side, but some appear on the high side. "What percent are actually on the high side?" asks technician Wynfor D'Teem.
Marty has not recorded percentages, nor does he understand how they are calculated, but he recalls that estimating sigma involves dividing R-bar by a weighting factor or . Using lightening-fast calculation, and a finger to the breeze, Marty states firmly that the way to estimate the percent out of spec is to divide by the upper spec limit:
Wynfor scrutinizes the white board and decides that 16.6 percent looks about right for the shaded area. Has Marty inadvertently managed to hit the nail on the head?
Click here for relevant sections of Practical Tools for Continuous Improvement.
Please direct questions or problems regarding this web site to the Webmaster.