Determining the Correct Sample Size when AQL points to two Sample Sizes

[Hiccup]

I need to know if the method I am using to select the sample size is correct. Is there a standardized method for these kind of situations?

Example:

I have a lot size of 100 units inspected at GII so I pull 20 samples (F). The AQL levels are 1.5 for major and 2.5 for minor. Following the arrow in the 1.5 column, the chart says the sample size size is actually 32 units. So what I do is pull 32 samples. I inspect 20 units for critical, major and minor defects. The last 12 units I only check for critical and major.

Is this the correct way? What if I run into two arrows pointing in opposite directions (sample size 50 at 0.5ma, 0.65mi), two arrows pointing down, two arrows pointing up, etc.

Your expertise is greatly appreciated.

 

[Edward Huang]

I need to know if the method I am using to select the sample size is correct. Is there a standardized method for these kind of situations?

Example:

I have a lot size of 100 units inspected at GII so I pull 20 samples (F). The AQL levels are 1.5 for major and 2.5 for minor. Following the arrow in the 1.5 column, the chart says the sample size size is actually 32 units. So what I do is pull 32 samples. I inspect 20 units for critical, major and minor defects. The last 12 units I only check for critical and major.

Is this the correct way? What if I run into two arrows pointing in opposite directions (sample size 50 at 0.5ma, 0.65mi), two arrows pointing down, two arrows pointing up, etc.

Your expertise is greatly appreciated.

I don't know what you meant by "the chart says the sample size size is actually 32 unit"
100 units sample 20-----right!
1.5 for major------defection exceed 1 unit, reject
2.5 for minor------defection exceed 2 unit, reject

 

[Hiccup]

On a sample size of 20 at 1.5, you hit an arrow pointing down. According to the AQL chart "Use first sampling plan below the arrow". From my interpretation, this means you use the next sample size down which would be 32. So now I have the AQL chart telling me to pick two different sample sizes.

 

[Edward Huang]

Did you use MIL STD 105E AQL?
The arrow just tell you can use "0 1" as above for 1.5 major and "1 2" as below for 2.5 minor

 

[Hiccup]

We are using ANSI/ASQ Z1.4-2003 (pretty much the same as MIL STD 105E, yeah?)

Maybe I am understanding this wrong but in your suggestion, that would mean for major defects, the an AQL level of 1.5 has the same ac/re as an AQL level of 0.65 on a sample size of 20 units. That doesn't make sense to me.

 

[Edward Huang]

That actually mean it should be no major defects between 1.5 and 0.65 because batch sizes are too small.

 

[acceptancenewbie]

I think you are doing right

Major issue sampling: 32 (1,2)
Minor issue sampling: 20 (1.2)

You may decide to be conservative and do 32 (1,2) for all. Let me know what you decided?

 

[Mike S.]

I need to know if the method I am using to select the sample size is correct. Is there a standardized method for these kind of situations?

Example:

I have a lot size of 100 units inspected at GII so I pull 20 samples (F). The AQL levels are 1.5 for major and 2.5 for minor. Following the arrow in the 1.5 column, the chart says the sample size size is actually 32 units. So what I do is pull 32 samples. I inspect 20 units for critical, major and minor defects. The last 12 units I only check for critical and major.

Is this the correct way? What if I run into two arrows pointing in opposite directions (sample size 50 at 0.5ma, 0.65mi), two arrows pointing down, two arrows pointing up, etc.

Your expertise is greatly appreciated.

100 units, 1.5% AQL, sample size = 32 pieces
100 units, 2.5% AQL, sample size = 20 pieces
Follow the arrow. In any given row/column combination, there is only one arrow, but the table is small and sometimes you need a straightedge to align things.

There are also AQL "slide rules" to make it easier.