Process Capability
Six Sigma – iSixSigma › Forums › Old Forums › General › Process Capability
- This topic has 5 replies, 5 voices, and was last updated 14 years, 7 months ago by
RGP.
-
AuthorPosts
-
October 12, 2007 at 8:15 am #48406
We would like to calculate process capability of a pH control process,where the USL and LSL are 9.8 and 9.2 respectively. Process data reported by QC is in increments of 0.1 (e.g. 9.4 or 9.5). Can we consider this data as continous and calculate the Cpk or this data has to be taken as discrete and transformed before it can be used for estimation of process capability.
What should be the right way of estimating process capability under such circumstances.
Thanking you in anticipation please,
0October 12, 2007 at 12:00 pm #163024You need not to convert the data to discrete. Better to calculate process capability with continuous data. Use Cp and Cpk for process capability calculation.
One suggestion: In this exampe process tolerance (USL – LSL) is only 0.6. but least count in scale is 0.1, if you can change your least count to0 .01, it will give you better result. If you can collect that in this decimal it is better otherwise you can go ahead with the existing scale.0October 12, 2007 at 1:18 pm #163027It appears that you are part of the quality team at your facility. Just some additional thoughts: (1) pH meters have electrodes that typically require more maintenance than say a thermal couple. Therefore, out-of-control condition detected by the SPC chart may or may not be a process shift. It could be an instrument shift. (2) Understand the pH meters calibration procedure and frequencies. Is it being calibrated frequently enough? (3) How often are the electrodes replaced? The probe response time to a change in pH depends upon the age of the electrodes. (4) Do you have a 2nd or 3rd pH meter that you can use to verify the results. Have the 2nd and 3rd meters been correlated to the primary pH meter?
All of these issues are important considerations in establishing a stable process. The last thing you want is a statistically stable process in which the pH is outside the specification limits. Finally, all of these considerations must be balanced against the specifics of the process. A world-class process involves SPC-plus. The plus involves all the specifics needed to produce the best control possible.
Here is a question that I would like an answer to: What is the impact on measurement accuracy when the pH meter is used continuously for 24 hours ..?0October 13, 2007 at 2:59 am #163052
Deming FanParticipant@Deming-FanInclude @Deming-Fan in your post and this person will
be notified via email.This paper by the guru, Dr Don Wheeler on measurement errors will help you :
http://www.spcpress.com/pdf/Relative_Probable_Error.pdf0October 22, 2007 at 6:08 am #163453
Andrew KennettParticipant@Andrew-KennettInclude @Andrew-Kennett in your post and this person will
be notified via email.RGP asks some very good questions. In my food plant we really struggle with pH measurement because the food grade non-glass pH probes are accurate but inprecise and there is the problem of drift. We chart pH data but it is ignored (I know why chart when you are going to ignore?). Any body had success with these twin SPC challanges of a lack or precision and insttrument drift?
Andrew0October 25, 2007 at 11:05 am #163820Posting this again, hopefully as a reply
I have a couple of questions:
1. What is the root cause of the drift?
2. Doesn’t the pH of a food product determine, in part, how good the food taste? For example, if the pH of Coke or Pepsi were significantly different from its target value (when shipped), would it taste significantly different than when the pH is on target? If so, pH is a very important factor in product quality.
3. I assume that the addition of phosphoric acid is what Coke and Pepsi use to control pH. Am I on target with this comment?0 -
AuthorPosts
The forum ‘General’ is closed to new topics and replies.