Calculating the process sigma for call rate
February 29, 2004 at 4:31 pm #34762
NwajeiParticipant@Frank Include @Frank in your post and this person will
be notified via email.
I would appreciate some expert guidance on how I could develop a process sigma on a metric that has been used in our company to evaluate the intensity of customer response as a function of the install base. It’s more a business metric as it also measures our cost of poor quality against EMA revenue.
Call rate is an annualized number that sums the calls (monthly) divided by the average installed population (latest 3 months) *12. For example a annualized call rate 6 for January would suggest that the average unit install calls in on average 6 times a year for service.
I have charted it on an XmR control chart since Jan 2003 but I am not sure if I’m using the standard deviation component correctly (in this case the mean R).
If my product spec dictates that no more than 5 calls/yr is allowed (let’s assume its the CCR), the mean of 5.567 and the mean R for the data equal to .3727, then is the s estimator equal to .33 (MR/d2 or .3727/1.128) where D2 is the sample size two factor for control charts? This would then yield a process sigma of 1.7.
Here’s the data…
date callsJan 5.0Feb 5.5Mar 6.0Apr 5.6May 5.1Jun 5.5Jul 5.1Aug 5.4Sep 5.8Oct 6.1Nov 5.8Dec 5.9
Another option would be to estimate s by multiplying MR by 1.047 ( a correction factor).
For those statistically inclined, I would appreciate your insights.
Thanks in advance.
The forum ‘General’ is closed to new topics and replies.