AQL - How do I find Acceptable Quality Level?

[qmslady - 2009]

Thanx Jennifer,

I copied this to my computer so I can go over it. We do, however, write which number the part is on our Operator In Process Inspection sheet and on the part itself. So we can justify which part is being written down and specify that "all" are 100% inspected.
QMSLADY

 

[Tim Folkerts]

I think you really should explore SPC and capability measurements.

CAPABILITY MEASUREMENTS like Cpk and Ppk are typical measures of how well you are actually doing. You calculate these by looking at your measured results, finding the mean and standard deviation, and then doing some calcualtions comparing the mean and stadard deviation to the specs. (And as Bob would rightfully point out, for precision machining you might want to do the calculations slightly differently than you would for other processes.)

A typical rule of thumb is to measure at least 30 pieces. More is better, but 30 gives pretty good results. Because capability calculations are based on variable data (ie numbers that can take on many results over the range of interest), a relatively small number of measurements (like 30) will give clear and effective results. These sorts of capability measurements are well known and well accepted in many industries.

This should be done in conjunction with control charts to see that the proceses is behaving consistently.



AQL SAMPLING PLANS, on the other hand, are based on a value you choose. Basically you are say "if the parts meet the chosen AQL (or are better), then the lot will usually be accepted, and if the lots are noticeably worse than the chosen AQL, then the lot will usually be rejected." In your case, an AQL sampling plan at the end (or when they receive the parts) could be effective as insurance on top of your other measurements. For example, if a shipment got mislabeled, a sample of a few parts would quickly spot the error.

Because AQL plans use attribute data (eg pass/fail), they require large samples if you want a high degree of certainty. You might check this thread for info https://elsmar.com/Forums/showthread.php?t=33830. For exampl, if you wanted to be 95% certain that at least 99% of the parts are good, you would have to sample 299 parts! Since you don't make this many parts, then this is not effective. But 30 parts for an estimate of Cpk could give you a similar certainty.


Let me recap my opinions on the matter:

• SPC (control charts) shows that you are being consistent
• Capability (Cpk) shows how well you actually have done relative to specs.
• Sampling (AQL) provides a rough check of how well you have done

Tim F

 

[qmslady - 2009]

Thanks Tim,
I will take this into consideration with all the other good advice I am getting...gotta go to work on it now...
QMSLADY

 

[Jim Wynne]

M'lady,
All of the replies/suggestions you have been getting are serious and express the same doubt as myself, although I do tend to be less serious than most. If I were you, I'd pick one of the "contributors" (cheapest within the USA) and give the person a phonecall. That's probably the quickest way to solve your apparent dilemma. You can then write and let the rest of us know how things progressed.

Give who a phone call?

 

[qmslady - 2009]

Hey Gordan,

That's how we say "HI" in Oklahoma. This is probably Jim's post, but I can't find Gordan's post.

Jennifer gave me a good website to go to, I am starting to understand this a little more. It talks in "plain english" that I can grasp. Still researching the web and coming up with some plans and charts, got "free" 30 day trial on some Sampling Program Software that is helping too. Will let ya'll know what happens and what I decide on.

Thanx...QMSLADYFOR NOW

 

[Gordon Clarke]

Give who a phone call?

Depends on who (if anone) she chooses
I'm not in the USA so I'm out - apart from the fact too, that I'm against AQLs.
I can't get my head around that "defects are acceptable".

 

[Tim Folkerts]

can't get my head around that "defects are acceptable".

I guess this depends on the situation. Consider a box of 100 paper clips. If I have the choice between

1. spending $1 and perhaps having one or two bad paper clips
2. spending $5 for a box certified to have exactly 100 good paper clips

I can tell you I would take #1 just about every time. The "cost" to sort & throw out the 1 or 2 bad clips as I use the paper clips is much less than the cost for "zero defects". Now if the certified box was only $1.02, I might spend the couple extra pennies.

It's always a balance -- the cost of the part, the cost of inspection, the cost when there is a failure ...

For the original post of this thread, it sounds like 100% inspection and 0% failure is the economic decision.



Tim F

 

[Jim Wynne]

Depends on who (if anone) she chooses
I'm not in the USA so I'm out - apart from the fact too, that I'm against AQLs.
I can't get my head around that "defects are acceptable".

It would be better to think of it as acceptable risk level. The term originated a long time ago in US MIL standards when mass inspection was common. Nonetheless, as Tim Folkerts points out, there are times when the economics of the situation does make some defects "acceptable."

 

[Bev D]

OK let's try an alternative approach. (Although Tim's suggestion about SPC has merit for the future)

To summarize, correct me if I've gotten something wrong:
You are doing a 100% inspection, which isn't too difficult given your relatively small batch sizes...
You only write down soem of the data around 10%
you were thinking that you needed some statistical validation for writing down only some of your data.
at this point you know that AQL applies to determining sample size

this is not a typical situation for aerospace and would probably result in many questions.

SO: why don't you simply write down all of the results?

alternatively you can write down only the min and max value

there is no statistical requiremetn behind either of these...make it simple on yourself

 

[Gordon Clarke]

I guess this depends on the situation. Consider a box of 100 paper clips. If I have the choice between

1. spending $1 and perhaps having one or two bad paper clips
2. spending $5 for a box certified to have exactly 100 good paper clips

I can tell you I would take #1 just about every time. The "cost" to sort & throw out the 1 or 2 bad clips as I use the paper clips is much less than the cost for "zero defects". Now if the certified box was only $1.02, I might spend the couple extra pennies.

It's always a balance -- the cost of the part, the cost of inspection, the cost when there is a failure ...

Tim F

I mentioned earlier what I described as critical and non critical dimensions for determining what to concentrate time and effort on. I doubt very much if the products we're discussing are comparable to paper clips.
The "clips" example seems to me a bit far fetched unless we're dealing with people that have been forbidden to use common sense.

I'll give an example along the same lines.

A Russian general and an American general, sitting by a swimming pool, were discussing who had the bravest soldiers.
A Russian soldier walked past, was stopped by the Russian general and ordered to climb to the top of the 30 foot diving board and jump in with a heavy rucksack. He did so and drowned.
The American general stopped an American G.I. and gave him the same order.
The G.I. looked at the general and said, "Are you mad" sir"?
The American general turned and smiled to the Russsian general and said, "Now that is bravery".