The Elsmar Cove Wiki More Free Files The Elsmar Cove Forums Discussion Thread Index Post Attachments Listing Failure Modes Services and Solutions to Problems Elsmar cove Forums Main Page Elsmar Cove Home Page
Google
  Web Elsmar.com
*Please be aware that SOME RECENT forum threads may not yet be indexed by Google.

View Full Version : Sampling Plan - AQL effect on sample size


BigEd09
11th June 2009, 02:37 PM
Hello, I am trying to understand why someone would reference an AQL level with the ANSI Z1.4-2003 sampling plans as this has no bearing on the sample size. The requirement on the drawing is as follows:

"INSPECTION REQUIREMENTS: ALL NOTES AND CIRCLED DIMENSIONS TO
BE INSPECTED USING SAMPLING PLAN - ANSI/ASQ Z1.4-2003, LEVEL II,
AQL = 2.5 AND C = 0 UNLESS OTHERWISE NOTED"

Am I correct in my "assumption" that if the AQL were 4.0, that the sample size would not change in the above requirement?

I also have a C=0 sampling plan that is not from the ANSI standard and the lot size is determined by the size of the lot and the AQL. Unfortunately the sheet does not reference its origin.

Thanks to all that reply.
Ed

Tim Folkerts
11th June 2009, 03:16 PM
This seems to be a self-contradictory requirement.

If you use the Z1.4 tables, then the acceptance number ("AC" in Z1.4 or "C" in other contexts) is determined by the table. It cannot be arbitrarily set to zero. So either you use C=0 or you use the numbers from Z1.4


On the positive side, for small lots (up to 50 items), the Level II, Normal inspection with AQL happens to be AC = 0. So for up to 50 items in the lot you would be fine.

The tables tell you:


up to 50 items: inspect 5, accept with no defects found
51-150 items: inspect 20, accept 1 defect
151-280 items: inspect 32, accept 2 defects.

If you inspect 150 items and use AC=0, then you re actually going to AQL = 0.65, rather than 2.5.

Something in the requirement needs to be changed - in my humble opinion.


Also, going to AQL = 4 would indeed change the sample size, at least for small lots. Up to 25 items you would inspect 3 items (AC=0). 26- 90 items you would inspect 13 items (AC=1).



Tim

BigEd09
11th June 2009, 05:13 PM
Tim,

Thank you for your feedback. If I could give you an example to make sure I am clear.

Based on the ANSI standard for a lot of 1300 pcs - Inspection level 2, the sample size would be 125 pcs.

If I follow the c=0, AQL 2.5 part of the requirement, using the C=0 plan that I have, the sample size would be 23.

Major difference for inspection requirement. I need to plead my case with our customer and I just want to be as clear as possible. Thank you fro takign the time to reply.

Have a great afternoon.
//Ed

Tim Folkerts
11th June 2009, 07:14 PM
Tim,
Based on the ANSI standard for a lot of 1300 pcs - Inspection level 2, the sample size would be 125 pcs.

That sounds right.

If I follow the c=0, AQL 2.5 part of the requirement, using the C=0 plan that I have, the sample size would be 23.If your are going to have C=0 for AQL = 2.5, then i see a sample size of 5 from the tables.



I think you can keep any two of the qualifiers "Level II", "AQL=2.5" and "C=0". Once you choose two of these, then the other one will be set. As I read the tables for a lot of 1300:


"Level II", "AQL=2.5" --> C=7 (and a sample of 125)
"Level II", "C=0" --> AQL = 0.1 (and a sample size of 125)
"AQL=2.5" and "C=0" --> Level S-1 (and a sample size of 5)


Tim

aproddutoor
15th June 2009, 06:02 PM
What is the sampling plan used to collect data?

What is the sampling plan used for Inspection?

Can they be the same?

Lyregi
22nd June 2009, 05:16 AM
I also have another question on AQL which I hope you folks will not mind me posting in this thread.

Supposedly:

Lot Size: 12,000 units
AQL: 0.65 at General Inspection Level II
Sample size stated: 315
Accept/Reject stated: 5/6

My company practises as follows:
Samples inspected: 800 units
Accept/Reject: 5/6 for minor defects, 3/4 for major defects

What happens if 2 units inspected fall under minor defects and 1 unit inspected fall under major defects? Should I reject the entire lot for such incident?

:cool:

reynald
22nd June 2009, 05:31 AM
My company practises as follows:
Samples inspected: 800 units
Accept/Reject: 5/6 for minor defects, 3/4 for major defects

What happens if 2 units inspected fall under minor defects and 1 unit inspected fall under major defects? Should I reject the entire lot for such incident?

:cool:

When I was still on the outgoing inspection function, our judgement was that the batch would pass. That is, we do not add the count of nonconformities for the major and the minor. Judgment is done indepently per category.

This set-up was communicated and agreed with both internal and external customers though.

Lyregi
22nd June 2009, 05:47 AM
When I was still on the outgoing inspection function, our judgement was that the batch would pass. That is, we do not add the count of nonconformities for the major and the minor. Judgment is done indepently per category.

This set-up was communicated and agreed with both internal and external customers though.

Thanks, pardon me if my interpretation of your judgement is wrong. Am I right to say:

If lot consists of 2 minor defects and 3 major defects, I will still accept this lot?:thanks: