Cpk definition for EBM and SBM bottles,

[mcpe72]

Our company desires to have a reference to show costumers what we offer, in terms of capability and don't ask cpk for certain features and recommend perhaps a cpk of 1 or only not out of spec, the nature of the process doesn't allow high precision, SBM maybe not the same for EBM (necks could when are calibrated).
I would consider critical dimensions, Volumen, top load and the neck so we won't have leaks (T an ID, and H to apply the caps)

 

[Miner]

What is your question?

 

[mcpe72]

my boss is asking me to define values for CPK, such T dimension 1.33, weight 1. etc, i think EBM is not as accurate as bottles made with SBM, i have to do a plant for both kinds of processes so when our customers ask, we have a reason behind why we choose 1 of 1.33 or only no out of spec ( i though this allow me more spread, because i wont have to comply with a cpk value, as long I'm inside spec) when we do qualifications, we are R&D so its not ongoing production, so the explanation would be, because of the nature of the process the values have high variance and its ok to have 1 for cpk. is there any standard that says that? or should i use AQL instead of cpks?

 

[ChrisM]

You don't "ask" or "define" Cpk - Cpk is derived by measuring how capable your process is.
Have a look here for some basics:

A Guide to Process Capability (Cp, Cpk) and Process Performance (Pp, Ppk)

A guide to Process Capability (Cp, Cpk) and Process Performance (Pp, Ppk) Analysis.

www.1factory.com

If you have a highly capable process, your Cpk figure, when you calculate it, will be well in excess of 1.5, it could even be 2 or more

 

[mcpe72]

I know there are processes that are very precise and require high cpks I understand that when lifes are involved such in an airplane, one would "need" it, our product, bottles will be use once (soap, shampoo, things of that kind), I think that its a waste requiring that in a plastic bottle, i meant that in those cases when one can be lax, its ok for a cpk that is low? would you "fixed" it with a broad tolerance, still we don't want those boundaries that far only to satisfy a number.

 

[Bev D]

I know there are processes that are very precise and require high cpks I understand that when lifes are involved such in an airplane, one would "need" it, our product, bottles will be use once (soap, shampoo, things of that kind), I think that its a waste requiring that in a plastic bottle, i meant that in those cases when one can be lax, its ok for a cpk that is low? would you "fixed" it with a broad tolerance, still we don't want those boundaries that far only to satisfy a number.

I think you - and/or your boss - are confusing specification limits with Cpk/process capability. You have said it yourself - “to satisfy a number”. It is a game don’t play it.
A process spread that is much tighter than the specification limits would only really help when there is a tolerance stackup problem or wear condition such as would be seen in aerospace or automotive. This is not the case in plastic bottles.
If you engineer the specifications properly then you only need to meet the specificaitons. The key here is engineering the specifications. Put your mental, monetary and physical effort into that.

 

[mcpe72]

we are OK with the specification and being not out of spec, we are conscious that when the process is out of control we would know it and need to be fixed, or we will have lots of rejects, this is for the final customer that requires this number, its kind of an explanation to appease the final customer, they are not happy with our qualification giving a cpk lower than 1.33 in one dimension, we used PPK and they want cpk with this kind of mintabs "analysis"

 

[mcpe72]

in the old times people was happy with having all in spec, now they want to apply more rules, than in my opinion, don't help much and generate scrap when things are still ok, i would check that the bottles don't leak, don't break and have the needed content, but they are asking us cpks for lots of dimension and don't agree with that but customers are requiring it, although the nature of the process inherit lots of variability, such top load ( depending how the material moves we can have different thickness and have big differences between bottles)

 

[ChrisM]

"but they are asking us cpks for lots of dimension and don't agree with that"
- Simple, just politely refuse to supply the information on the basis that it is your "Intellectual property". Is the bottle to your design or theirs?
If this is causing you more problems than it is worth, just decline to supply what they are asking for, and if necessary turn the work down.

 

[ScottK]

Wow this is blast from the past. I worked in Stretch Blow and Extrusion Blow (as well as Injection Blow and Injection molding) in the late 90s as the one and only corporate quality engineer for 5 plants. That company was absorbed long ago by Berry Plastics. At that time all Cpk meant to us was that customers required critical dimensions to show a Cpk of 1.33 or better on data taken over the course of an 8 hour run from samples of every cavity. The critical dimensions really depended on the bottle and it's use. Neck dimensions were always the top priority whether threaded or not. Also wall thickness. Usually, overall height and width as well.

I'm sure a lot of improvements have been made since late 2000 but during my time in the industry EBM was the hardest to control for body dimensions and wall thickness.

All that said - in the end Cpk was meaningless because rather than adjusting the process we always adjusted the tolerances.