There is a methodology called "Acceptable Quality Level" (AQL) that can be easily applied to determine product sample size to determine quality. This method defines the number of samples that must be taken from a larger population to determine quality. The American Society for Quality (ASQ) publishes a book that defines and explains this sampling method. It is called "Zero Acceptance Number Sampling Plans" by Nicholas Squeglia. This book explains how the c=0 sampling method is an upgrade over the MIL-STD-105E (military standard) that has been used by some government contractors and the military. The c=0 sampling plan methodology can lead to higher levels of productivity in your inspection and quality assurance departments by significantly reducing sampling quantities. The c=0 designation is significant since you want zero defects to be found in a sample of products that are chosen randomly from a larger population. There is an easy-to-use c=0 chart that defines the number of samples that should be taken to statistically determine if a specific population size meets a predefined quality level. In this case the number of defects has to be zero for a population or lot to be accepted as a quality lot.

Things You Will Need
  • AQL c=0 chart

  • Total number of products to be sampled

Step 1.

Determine or count the total number of products in a certain lot or population.

Step 2.

Use the c=0 chart to determine the number of samples that must be taken from the overall population.

Step 3.

Randomly choose the correct number of samples from the entire population. If there are multiple boxes of the same product you should open several boxes to get a true random sample of the correct quantity.

Step 4.

Inspect the samples according to the accepted predetermined specifications or criteria.

Step 5.

Accept the entire population lot size if all criteria are deemed acceptable.

Step 6.

Reject the entire population lot size if one or more defects are found in the random sample.


To get a true random sample, it's best to have just one person choose the sample.