Which Sampling Tables for AQL?

P

pammesue

I swear I have searched the interwebs and found a ton of different tables to use for sampling. I haven't actually found the official ones because there are so many!

I see that the MIL STD is cancelled?

There are several ANSI ones to choose from. Which one should I use?

Level II and 2.5 4.0 major and minor

No Critical c=0 defects. There are some c=0 charts that have a very high number of samples taken. The other charts I have found ...which didn't specify the c=0 which would be understood i would think....have much smaller samples.

please help
 

Scott Catron

True Artisan
Super Moderator
I believe the "official" replacement for MIL-STD-105 is ANSI/ASQ Z1.4, "Sampling Procedures and Tables for Inspection by Attributes".

Edit: Steve beat me to it.

Also, Wikipedia lists as a replacement MIL-STD-1916, "DoD Preferred Methods for Acceptance of Product"
 

Statistical Steven

Statistician
Leader
Super Moderator
I believe the "official" replacement for MIL-STD-105 is ANSI/ASQ Z1.4, "Sampling Procedures and Tables for Inspection by Attributes".

ANSI/ASQ Z1.4 is the equivalent to MIL-STD 105E. As far as where to start, you can start with General Inspection Level II and Normal. Using the switching rules you can move to Tightened or Reduced.
 

David-D

Involved In Discussions
MIL-STD-105 (attribute based AQL acceptance sampling) was canceled by the Department of Defense as it's use of AQLs and accepting of known levels of defective product was antithetical to the zero defects direction undertaken in the 1990s. ANSI/ASQ has rebadged it as their own and renamed it as Z1.4; they also created an analogous photocopy of the DoD's MIL-STD-414 for AQL based variable sampling called Z1.9.

Since the mid 1990s, the DoD has moved to MIL-STD-1916 as the preferred method for acceptance for non-destructive testingwhich is an Accept on Zero methodology rather than being based upon AQLs. It incorporates provisions for attribute, variable, continuous, and contractor proposed sampling methods. There is also an ISO copy of 1916 but I don't remember the specific number.

David
 

David-D

Involved In Discussions
As others said, if AQL sampling is what you really want (especially if the AQLs are already being specified by others) you want Z1.4 for attribute or Z1.9 for variable sampling. If AQLs aren't actually specified then there is a multitude of other sampling options out there to consider, each with their own advantages and disadvantages

David
 
Top Bottom