MiniTab 17 Question-Standard Deviation Default

[Coury Ferguson]

Our company uses Minitab 17 to provide our Automotive Customers with Cpk/Ppk data. I have noticed that Minitab 17 defaults to 6 Standard Deviations. Is there a way to change that default to 3 Standard Deviations or is it going to have to be changed every time when developing Ppk/Cpk data to be submitted to our customer?

I am not an expert in Minitab, so maybe there is a simple way to change the default, that I am not aware of.

 

[jfries2]

Using Minitab 18 here, but per the help screen on Capability Analysis:

"By default, the tolerance is 6 standard deviations wide (3 standard deviations on each side of the process mean). "

So, 6 should be what you want. If you do need to change it, go to "Stat"<"Quality Tools"<"Capability Analysis"<"Normal"<Click on "Options", then enter a different K value.

 

[Miner]

jfries2 response is correct for both versions of Minitab.

 

[Coury Ferguson]

Thanks for your information. We use 3 SD not 6 SD. I guess I will have to change it every time.

 

[Miner]

By changing it to 3, Minitab will interpret this as +/- 1.5. Are you certain that is what you intend?

 

[Coury Ferguson]

By changing it to 3, Minitab will interpret this as +/- 1.5. Are you certain that is what you intend?

Miner,

Yes.

Our data is within LCL and UCL. When using it at 6 SD our data becomes skewed. Our Ppk/Cpk when using the 6 SD does not meet the 1.66/1.33 that the customer wants, but at 3 SD it does, if not better.

Here is an example: We have a TP UCL .5 mm and LCL .0 when we made the parts the TP (don't ask me why the customer wanted the Ppk/Cpk data for a TP) was reading at .0 mm-.25 mm the data skewed far left. But the process was in control and producing the requirements of the drawing.

Another example would be DIA UCL 12.9 mm LCL 12.6 mm nominal 12.75 mm.

We tried everything to get the data to meet the Ppk/Cpk requirements and using the 3 SD instead of 6 SD bought the process in within the customer's requirements.

It is a shame in my opinion, that we have to manipulate data on good parts.