Acceptance sampling can determine whether you should accept or reject an entire lot of a product. With variables, a sampling plan compares a single measured characteristic, such as thickness, concentration, or surface roughness against compared to at least one specification.

Using Minitab, you can enter data from particular lots and determine whether an entire lot should be accepted or rejected.

A camera assembly manufacturer receives camera lenses in shipments of 3,600 lenses several times a week. The quality team has implemented a sampling plan to inspect the lenses to determine whether they are acceptable for use. A defective lens is one that is thicker than 0.415 inch (USL) or thinner than 0.395 inch (LSL).

From the lot of 3,600 lenses, the manufacturer and its supplier agree to set the acceptable quality level (AQL) to 100 defectives per million and the rejectable quality level (RQL) to 600 defectives per million. Using a producer's risk (alpha) of 0.05 and a consumer's risk of 0.10, Minitab determines that an appropriate sampling plan is to randomly select and inspect 259 of the 3,600 lenses. From these data, Minitab calculates the mean and the standard deviation. The manufacturer will accept the lot if it meets the following three conditions:

·    Standard deviation is less than the maximum standard deviation; in this case, 0.0027533.

·    (Mean – lower spec) / standard deviation is greater than or equal to the critical value; in this case, 3.44914. The lower specification for the camera lens is 0.395.

·    (Upper spec – mean) / standard deviation is greater than or equal to the critical value, in this case, 3.44914. The upper specification for the camera lens is 0.415.

You randomly select 259 lenses from a batch and record lens thickness measurements.

Data: Camera.MTW (available in the Sample Data folder).