Introduction

  • Overview of ANSI/ASQ Z1.4 including definition of the sample size n and acceptance number c based on (1) the lot size, (2) the acceptable quality level, and (3) the inspection level.
  • Average sample number ASN (how many items we expect to inspect as a function of the quality in terms of nonconforming fraction) and operating characteristic curve (likelihood of acceptance as a function of the quality of the lot)
  • Conversion of any ANSI/ASQ Z1.4 plan into a zero acceptance plan with an identical chance of rejecting a lot at the rejectable quality level (RQL). ANSI/ASQ Z1.4 plans do not have official RQLs but, for analytical purposes, we treat the nonconforming fraction at which there is a 90% chance of rejection as the RQL. The principle is the same as for basic discovery sampling.
  • Do not, however, use zero acceptance plans when quality does not exceed the AQL because of the much higher likelihood of rejecting acceptable lots. Nothing is free in industrial statistics, and the tradeoff for the much smaller sample size is the higher risk of rejection of lots we should accept. This is especially true when conformance with ANSI/ASQ Z1.4's switching rules (which will be covered in the webinar) is required.
  • When relatively mediocre quality precludes zero acceptance plans, alternatives are available with smaller ASNs than single-sample ANSI/ASQ Z1.4 plans. These include double and multiple sampling, and also sequential sampling plans.
  • Narrow limit gaging plans offer even smaller sample sizes but rely on the assumption that (1) the critical to quality characteristic is normally distributed and (2) the go/no-go gage can be set to specific dimensions, e.g. with gage blocks.
  • Deployment of reduced inspection plans as zero acceptance plans is complicated by the fact that two decisions must be made for each lot; whether to accept the lot, and whether to invoke the switching rules. When this is the case, it may be simpler to just use the reduced inspection plan.

Why should you attend?

Inspection is a required but non-value-adding activity, so the less we do the better—as long as we meet customer requirements in terms of protection from poor quality. Zero acceptance plans (c=0 plans) offer smaller sample sizes than ANSI/ASQ Z1.4's (formerly MIL-STD 105's) single, double, and multiple sampling plans, and even sequential sampling plans. The tradeoff consists of a higher likelihood of rejection at the acceptable quality level (AQL) so this plan is recommended only when quality far exceeds the AQL.

Who Will Benefit

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

ENROLLMENT OPTIONS

On Demand
ENROLL NOW

Bill Levinson, P.E., FASQ, is the principal of Levinson Productivity Systems PC which specializes in industrial statistics, quality management, and lean manufacturing. He is a leading authority on He Know More

William Levinson