Acceptance Sampling is a statistical measure used in quality control. A company cannot test all of its products because of ruining the products, or the volume of products being very large. Acceptance sampling resolves this by checking a sample of product for defects. The process includes sample size, batch size and the number of defects suitable in the batch. This process permits a company to determine the quality of a batch with a particular degree of statistical certainty without having to check every unit of product. The statistical reliability of a sample is usually measured through a t-statistic.
For Example - If a company manufactures millions of product and tests 10 units with one default then an hypothesis would be made on probability that 10,000 of the 1,00,000 are defective. Though, this could be a grossly inaccurate representation. More dependable conclusions can be made by raising the batch size higher than 10, and rising the sample size by doing more than merely one test and averaging the results. When done properly, acceptance sampling is a highly efficient tool in quality control.