Inspection is a mandatory but non-value activity, so the less we can do is better as long as our inspection plan protects the customer from poor quality. ANSI/ASQ Z1.4, formerly MIL-STD 105E, is a well-established standard for sampling by attributes. A specified sample of n items is inspected, and the lot from which the sample came is accepted if the acceptance number (c) or fewer nonconforming items are found. ANSI/ASQ Z1.4 includes double and multiple sampling plans that reduce the average sample size, but sequential sampling plans are even more efficient.

Attendees will receive a pdf copy of the slides and accompanying notes, and an Excel spreadsheet that converts a traditional sampling plan into a sequential plan.

Learning Objectives

• Use of ANSI/ASQ Z1.4 for sampling by attributes, including selection of the code letter, sample size, and acceptance number
• The operating characteristic (OC) curve that shows the chance of acceptance versus the nonconforming fraction
• The average sample number (ASN) as a function of the nonconforming fraction
• Conversion of any ANSI/ASQ Z1.4 sampling plan into a sequential sampling plan whose protection of the customer from poor quality is essentially identical to that of the original plan, as can be demonstrated with OC curves. This can be done in Microsoft Excel.
• ANSI/ASQ Z1.4 switching rules for normal, tightened, and reduced inspection. Tightened inspection can be deployed as a sequential sampling plan, but reduced inspection creates a more complicated situation in which the traditional plan might be easier to administer.
• Awareness of alternatives such as zero acceptance sampling, which requires even less inspection but will also reject acceptable lots unless their quality is far better than the specified acceptable quality level (AQL). Another alternative is narrow limit gaging that can reduce the sample size enormously but relies on the assumptions that (1) the quality characteristic is normally (bell curve) distributed, (2) it can be measured with go/no-go gages that can be set to specific dimensions, and (3) poor quality results from changes in the process mean and for no other reason.

Who Should Attend 

Quality managers, engineers, and technicians, and others with responsibility for acceptance sampling activities.