A very general question that comes up in quality control and is often witnessed or faced by the engineers is to select how many items for a batch for the purpose of inspection in order to be sure that the items that are there in the batch are of a quality that is acceptable. Acceptance Sampling is a procedure that is applicable here and it helps in deciding when there is a need to find out if the lot of items comply with the specifications without actually checking each item individually. This concept of Acceptance Sampling does have advantages over doing up a hundred per cent inspection as it saves on to a lot of time, effort and money. It also so happens with some items, according to their properties, is that a hundred percent testing would destroy the entire batch. Thus when we look at the picture from a managerial perspective, rejection of the complete batch or the shipment on the basis of acceptance sampling than just a certain percentage of defective items which is based on the concept of hundred percent inspections does give a strong incentive to the supplier to be able to adhere to the quality standards.
The approach towards deciding the sample size for acceptance sampling is very straight forward. The basic concepts discuss the concept of sampling distribution. If the researcher were to take repeated samples of a particular size from the population then the distribution of the averages of the specifications of the product would come to a normal distribution with a specific mean and standard distribution. When we talk of sampling distribution, we prefer to use the word sampling error instead. The good part is that, it is not necessary to take repeated samples from the population so as to identify the mean and the variability or the standard error in the chosen population. If the researcher has an idea about the variability in the population, then it is easier to infer the sampling distribution of the mean. In principle, this given information is sufficient in order to estimate the sample size so as to find out a certain change in quality that is needed.
