Capability Studies - 5 different dimensions - How many parts?

[Dawn]

Recently we did a capability study on a press with 5 different dimensions. They took parts that were ran for 1 hour. This took approximately 3 full working days with 3 quality techs to complete. Is it a standard when doing a capability study to take parts for 1 full consecutive hour?

 

[Spaceman Spiff]

... not according to the PPAP manual, 3rd Edition. The manual states that an 8 hrs. run is required to do capability study. However, what you did is often what injection molders do to tweek in the tool (correlating all the cavitites). I guess the bottomline is what is it that you want out of the capability study.

 

[Marc]

I think you're saying it took 3 techs 3 days (9 man days) to measure the 5 dimensions on each part...

Naw - we're missing something here. I know there is an 8 hour run requirement, but some suppliers produce 1 item every 2 hours and some produce 2,000 per hour. Logic tells us there is a limit.

That having been said, I don't have the answer - we need some to hear from someone who has done it.

Calvin to Spiff - you're high volume?

 

[Spaceman Spiff]

I'm not sure if we are high volume, but we produce on an average 1000 assemblies per day per line (my last job we produced 30,000 assemblies per day per line).

 

[billsfan]

Hi all,
I wouldn't consider myself an expert, but if I had a dollar for every PPAP I've done I could retire.
Let's see if I can shed some light....
Remember the 7 P's - "Proper Prior Planning Prevents Piss Poor Performance." Have a plan before you start making parts. (I also prefer having my customers' buy in on the plan.)
First, page 2 of the PPAP manual (third edition) talks about a significant production run (1.2.1). As has been mentioned previously, some companies could manufacture thousands of parts in 8 hours, some just a handful. You must do what makes sense to you and your customer. After all, you are trying to prove that you can produce product consistently meeting the design record and specifications. Some or perhaps all of the parts that you run will be used for the process study.
Next, page 6 describes the initial process studies (1.2.2.9) By now you should have determined what if any "critical or key characteristics" are on this part, either specified by your customer or determined internally. This or these characteristics are what you will do you capability study on.
You must perform a Gage R&R on the equipment used to measure this or these characteristics. Then you start collecting data. As stated in NOTE 4 PG. 6) "a short term study should be based on a minimum of 25 subgroups containing at least 100 readings..." Now if you make 1000 parts per hour obviously you're not going to measure every part. I would take samples throughout the run until I had 100 readings. If you make 100 parts a day I would either measure all of those parts or get the customer to agree to something less.
You don't want to mislead your customer or your company. If you have a process that produces MANY parts per hour and you only produce 100 consecutive parts for the capability study, what does that really tell you when you're done? That you can run 100 parts successfully? What about the next 1000, 10000 or million? If it only took you 15 mins. to run these parts, now you know that you can run successfully for 15 mins. but what about the next few hours, days, weeks, years? Wouldn't you like to know that you have a stable and capable process prior to shipping parts off to the customer? So would they! PPAP is how you prove it.
Sorry, I seem to have rambled...
Hope this helped.

 

[Marc]

Do ramble on. This is the kind of detailed, reasoned info we need in response to questions here. Any more advice and details will be much appreciated!

 

[billsfan]

Marc & "the gang",
If it's details you want, then let's try an example. First, something simple. You die stamp metal parts, coil fed, single die. Which meens you produce one part every time the machine cycles. The only "critical or key characteristic" on the customer drawing is length. This dimension is 1" +/- .01. First you need a piece of measuring equipment capable of measuring out to 3 decimal places (10x rule). Perform a Gage R&R. If aceptable results, perform initial process capability study.
This is where the rubber meets the road. To keep the example simple, let's say that this is the first time making these parts, but that you make very similar parts of different lengths. After set-up approval, you start collecting samples over time for your capablity study. Again to keep this simple, let's say that you make 50 parts per hour. According to your previously agreed upon plan, you'll be taking 5 parts per 1/2hour, measuring the length of each and charting on an x bar & r chart. You run for 10 hours, produce 500 parts, & collect data from 100 of the 500 parts. This data can then be analyzed to determine Cpk/Ppk.
Things start getting complicated when you're talking about multi-cavity molds or dies, multiple stations or fixtures and numerous "critical or key characteristics".
To get more specific, I would need more information. If I haven't shed any light let's at least hope I didn't muddy the waters!
Later dudes......