Counting defects for inspections of a gift box assortment. Help needed.

Hi,
I've been going in circles recently for a customer order I have to have inspected....
I have an 50 piece assortment of 5 different components that are all packaged into 1 carton. There are 500 cartons of 50. Each component has it's own quantity, dimensions, weight, and cosmetics.

As a simple (made up) example:
I have an order for 500 gift boxes of golf balls. Each gift box is an assortment of 5 different golf balls:
Ball type A Qty = 29
Ball type B Qty = 9
Ball type C Qty = 6
Ball type D Qty = 3
Ball type E Qty = 3
Total balls/gift box = 50

I know how to inspect each ball, with major and minor defects.
Let's say for major defects I select an AQL level of 2.5%.

I am planning to use Inspection Level 2, General Inspection, with a sample size of 50 gift boxes. Accept is then 3 major defects. 4 defects or more is reject the lot.
So my questions:
A. Do I count as 1 Defect if in an entire gift box I find 1 major defect on a golf ball? That means if 4 balls defective in 4 different gift boxes (total 2,500 balls), the entire lot is reject!
B. Should I allow 3 major defects in1 gift box, and not count it as a defect? But if there are 4 major defects then that box is 1 defect in the total lot of 50 inspected gift boxes?
C. When I open a gift box, should I inspect all 50 golf balls? 50 x 50 = 2,500 balls to inspect. Too many!

Golf balls are used independently, and not parts that must fit together to create something, you know, like a SpaceX rocket.

Your advice and comments appreciated!
Roger

 

Hello Roger,
I will try to help you by using the Military Standard Sampling Procedure.

For an Inspection level 2 of General Inspection and a lot size between 1201 and 3200 (on your example 2500 balls), according to table I - Sample Size Code Letters of the Military Standard, corresponds to the letter K.

This letter must then be crossed with the table of AQL's according to the type of inspection. For a normal type of inspection (Table II-A) the letter K obtained from the previous table gives a Sample Size of 125. Crossing this value with the 2.5% AQL would give you 8 rejects out of the total number of samples taken (in this case 125).

Saying this, you should sample the entire batch randomly. Using your example again, if you have 500 boxes and 50 balls inside each one, you should randomly inspect 125 balls spread throughout the entire lot and, if those 125 balls you detect 8 balls with major defects in it, yes you should reject the entire lot.

I am saying this assuming you cannot inspect every item separately. For me, the correct method is to inspect every different type of ball individually (1 inspection for Ball Type A, 1 inspection for Ball Type B, etc...) as you say they have different characteristics.

Trying my best to explain it in a clear way. Hope this helps.

 

Carlos,
Many thanks for your answer and clear explanations. It gives me the exact answer that I can apply to my actual inspection.
I had looked for a couple of hours online and never found any help.
Roger