From: David Chan, Jenny Chan

> ...can anyone tell me that the relationship between AQL and PPM..
> somebody told me that 0.065% AQL = 650PPM, is it true ???
>
> i personally think that it isn't , but cannot get proven evident.. can
> u help me ??
>
> tx..
>
> Jenny Chan...

I can't help you with the actual numbers off hand, but AQL is Acceptable Quality Level as a per cent. Used in inspection to determine sample size, you decide the AQL level you want and a chart gives you a sample size. Your knowns are the lot size which you look up on the table, and the AQL level you want.

PPM is Parts Per Million which can be converted to a %.

Maybe one of the others can help you out with details and numbers.

Acceptable Quality Level (AQL) is defined by ANSI/ISO/ASQC A3534 as follows: "When a continuing series of lots is considered, a quality level which for purposes of sampling inspection is the limit of satisfactory process average."

The associated AQL and Lot Tolerance Percent Defective (LTPD) are summarized graphically by the operating characteristic curve.

<font COLOR="#DF0218"><BLOCKQUOTE>…somebody told me that 0.065% AQL = 650PPM, is it true ???</BLOCKQUOTE></font>

Strictly speaking, no. But, a case could be made that if the AQL was, in fact, 0.065%, then the PPM might be 650. The problem is, AQL is an Acceptable Quality level, not the Actual Quality Level. AQL is the quality level that would be detectable under a certain sampling plan.

Also strictly speaking, the LTPD * 10,000 = PPM nonconforming.

Regards,
Don

I got this e-mail today. I know this is an old thread, but it's worth opening back up for a bit.

On 6/18/02 6:29 PM, arnulfom wrote:

> Hi Marc,
>
> Sorry for bother you, look, I saw a thread at the cove forum where you talk
> about the PPM vs AQL.
>
> I have a doubt and could not get in touch with any one in the forum due to
> failure comunication problems, my question as follows:
>
> If I have an AQL=0.1, which is the number of defective parts per million I
> will give my customer?<hr>Also see AQL (Acceptable Quality Level) vs. LTPD (Lot Tolerance Percent Defective) (http://Elsmar.com/Forums/showthread.php?s=&threadid=1818)

and

Information about PPM since giving up AQL - Historical description of requirements (http://elsmar.com/Forums/showthread.php?s=&threadid=1900)

You can't get to PPM from AQL. There are many OC curves that will pass through a single AQL. If you're interested in the conversion, you'll need to specify both ends of the curve and then look at the associated AOQL levels. Typically, you'll find the worst case AOQL level at Pa = .50.

I hate the word "Acceptable" in AQL, it suggests we are doomed to failure before we start.:mad: :frust:

Geoff

Is it a realistic understanding of processes and variation ?

Cp and Pp do that for me.

And also define ppm

avigeva

ppm is parts per million, whereas PPM is Parts Per Million.

M Greenaway said:
ppm is parts per million, whereas PPM is Parts Per Million.
Usually non-conforming

Marc said:

> If I have an AQL=0.1, which is the number of defective parts per million I
> will give my customer?

Check this:

Multi Cavity Sampling Plan - MSPs (Multi-Stream Processes) (http://Elsmar.com/Forums/showthread.php?s=&threadid=4352)

Good reference threads! Thanks!

About Capability index and Acceptance methods, I found this article, I tink could be of any help.

*** DEAD LINK REMOVED ***

:bigwave:

FYI,
A quick review of the mil-std's found that all relating to product sampling have been canceled,, i.e.,1235 105,106,107,414
The replacement is Mil-std-1916 which is combination of a sampling plan with zero defect acceptance and Cpk values. Interesting.

Interesting info, Sam. Thanx.

But most of there standard were changed to private domain

like

MIL-STD-105E (1989)
Sampling Procedures and Tables for Inspection by Attributes
Cancelled by NOT 1 dated 27 FEB 95. Future acquisitions should refer to acceptable NGS on sampling procedures and tables for inspection by attributes, such as ANSI/ASQC Z1.4-1993.

MIL-STD-414 (1968)
Sampling Procedures and Tables for Inspection by Variables for Percent Nonconforming
Cancelled by NOT 2 dated 2 FEB 99. Future acquisitions should refer to acceptable non-Government standard on sampling procedures and tables for inspection by variables for percent defective such as ANSI/ASQC Z1.9-1993.

But they are still the same (with some minor changes) :frust: but if you want it, you have to pay.

ref:

http://www.variation.com/techlib/standard.html