What is acceptance sampling?

Acceptance sampling is a major component of quality control and is useful when the cost of testing is high compared to the cost of passing a defective item or when testing is destructive. It is a compromise between doing 100% inspection and no inspection at all. In particular, in cases when the quality of a supplier's process is unknown, acceptance sampling may be a good option as opposed to 100% inspection. Acceptance sampling can be done on attributes or measurements of the product.

You can use acceptance sampling to develop inspection plans that enable you to accept or reject a particular lot of incoming material based on the data from a representative sample.

Example of an attribute acceptance sampling plan

For example, you receive a shipment of 10,000 microchips. You either cannot or do not want to inspect the entire shipment. An attribute sampling plan can help you determine how many microchips you need to examine (sample size) and how many defects are allowed in that sample (acceptance number).

In this case, suppose your acceptable quality level (AQL) is 1.5% and the rejectable quality level (RQL) is 5.0%, and you assume alpha = 0.05 and beta = 0.1. Minitab generates a sampling plan that indicates that you need to inspect 209 chips. If 6 or less of the 206 inspected microchips are defective, you can accept the entire shipment. If 7 or more chips are defective, you must reject the entire shipment.

Example of a variables acceptance sampling plan

For example, you receive shipments of 2500 plastic pipe segments each week and you need to verify that the wall thickness measurements meet specifications. You either cannot or do not want to inspect the entire shipment. A variables sampling plan can help you determine how many pipes you need to measure (sample size) and the criteria for accepting or rejecting an entire lot (critical distance).

In this case, the lower specification for the wall thickness of the piping is 0.09". You and the supplier agree that the acceptable quality level (AQL) is 100 defectives per million and the rejectable quality level (RQL) is 300 defectives per million, and you assume alpha = 0.05 and beta = 0.1. Minitab generates a sampling plan that indicates that you need to measure 760 pipes and indicates that the critical distance is 3.55748. You then measure a sample of 760 pipes and use the accept/reject tool in Minitab to indicate whether a shipment should be accepted or rejected.