INITIAL POSTING: This is with reference to Minitab 14 attribute analysis using p-chart. The results window shows a Summary Stats box. This gives a " PPM Def " and "Process Z ". This PPM number corresponds to a Process Z-value which is calculated by finding the Z-value with shift for the PPM and then adjusting this Z-value for the shift by subtracting 1.5. Table-S, Appendix-E from "Implementing Six Sigma" shows that a Z-value without the shift corresponds to a PPM of 1374 approx. What am I missing here ? How can we convert/adjust the Z-value without converting/adjusting the PPM for the shift ?
RESPONSE: I will be making references to my book, "Implementing Six Sigma", since you referenced it in your email. First, when sample sizes are large p-charts can indicate out-of-control signals that can be questionable (see section 10.22, 2nd edition, page 237). Secondly, when a process is in control it is said to be predictable. I prefer to express process capability/performance metrics as a long term defect/defective rate(see section 11.23, 2nd edition page 297). This type of process capability/performance statement makes sense and is easy to understand by all. The defective rate for an in control failure rate control chart is simply the center line (see section 11.16 page 283, 2nd edition). This approach is much simpler and less confusing than dealing with sigma-shifts, sigma quality level, Cp/Cpk, and any other process capability indices.
Hope this helps. Contact me if you would like to discuss further.
forrestbreyfogle 8/26/2004
INITIAL POSTING: This is with reference to Minitab 14 attribute analysis using p-chart. The results window shows a Summary Stats box. This gives a " PPM Def " and "Process Z ". This PPM number corresponds to a Process Z-value which is calculated by finding the Z-value with shift for the PPM and then adjusting this Z-value for the shift by subtracting 1.5. Table-S, Appendix-E from "Implementing Six Sigma" shows that a Z-value without the shift corresponds to a PPM of 1374 approx. What am I missing here ? How can we convert/adjust the Z-value without converting/adjusting the PPM for the shift ?
RESPONSE: I will be making references to my book, "Implementing Six Sigma", since you referenced it in your email. First, when sample sizes are large p-charts can indicate out-of-control signals that can be questionable (see section 10.22, 2nd edition, page 237). Secondly, when a process is in control it is said to be predictable. I prefer to express process capability/performance metrics as a long term defect/defective rate(see section 11.23, 2nd edition page 297). This type of process capability/performance statement makes sense and is easy to understand by all. The defective rate for an in control failure rate control chart is simply the center line (see section 11.16 page 283, 2nd edition). This approach is much simpler and less confusing than dealing with sigma-shifts, sigma quality level, Cp/Cpk, and any other process capability indices.
Hope this helps. Contact me if you would like to discuss further.
Forrest Breyfogle
forrest@smartersolutions.com
512-918-0280
www.smartersolutions.com