In recent years, Accept-on-Zero (AOZ) plans have become popular. These plans, also known as zero-defective plans, are acceptance sampling plans with acceptance threshold of c=0. In other words, they dictate batch acceptance only if no non-conforming items are found in the sample.
The intention behind AOZ plans is to “protect the consumer” by creating the impression that batches with non-conforming items are unacceptable, thereby putting pressure on the producer to produce high quality items. This was the intention of the U.S. Department of Defense (DOD) when it issued MIL-STD-1916 on April 1996 which contains AOZ plans for attributes, variables, and continuous sampling. Unfortunately, a different intention has led to the popularity of AOZ plans: companies who are sensitive to legal litigation by their customers (such as automotive and pharmaceutical companies) perceive AOZ plans as legal protection due to avoiding evidence of passing non-conforming items to the consumer. However, it is clear that technically AOZ plans do not offer better consumer protection compared to c>0 plans that satisfy the same LTPD or AQL criterion. The main criticism of AOZ plans is the misconception that they create that a sample with zero non-conforming items implies a perfect batch (which has no non-conforming items). Obviously, this is not the case.
While AOZ plans are popular, they are quite controversial. From a technical point of view, they present an advantage in terms of requiring smaller samples compared to corresponding c>0 plans. However, the price of a smaller sample is the lowered discrimination rate between good and bad lots, which in this case translates into a higher producer risk. Graphically, OC curves associated with c=0 plans have a maximum at p=0 from which they drop faster than c>0 plans. Hence, the c=0 batch acceptance probability drops fastest at the best quality levels.
In practice, there exist several AOZ standards and tables. SQC Online provides calculators for obtaining AOZ plans using the popular Mil-Std-1916 standard as well as AQL-based plans using Squeglia’s tables. For detailed information on the different AOZ tables and schemes see Chapter 17 in Schilling & Neubauer’s book Acceptance Sampling in Quality Control, 2nd Edition, CRC Press, 2009.