ISO 28592:2017 PDF

Overview

ISO 28592:2017 (ISO 28592:2017) specifies double sampling plans by attributes with minimal sample sizes, indexed by producer’s risk quality (PRQ) and consumer’s risk quality (CRQ). It provides attribute-based acceptance inspection procedures for lots of discrete items, covering inspection by percent nonconforming and by nonconformities per 100 items. Nominal risk combinations supported are (5%, 5%), (5%, 10%) and (10%, 10%). The standard’s statistical theory and supporting tables/figures are given in Annex A.

Key topics and technical requirements

• Decision rule (attributes):
  • If the first random sample contains 0 nonconforming items → accept the lot.
  • If the first sample contains ≥ 2 nonconforming items → reject the lot.
  • If the first sample contains exactly 1 nonconforming item → draw a second random sample; accept if the second sample has 0 nonconforming items, reject otherwise.
• Plan indexing: Plans are indexed by PRQ (acceptable quality level from producer view) and CRQ (unacceptable quality level from consumer view).
• Objectives: Choose first and second sample sizes to minimize the maximum expected sample size with respect to incoming quality, subject to the nominal producer’s and consumer’s risks not being exceeded.
• Use cases within the standard: Acceptance sampling and hypothesis testing (test that process quality equals PRQ vs. CRQ).
• Outputs: Tables and figures of sample sizes, operating characteristics, average sample sizes, actual producer’s/consumer’s risks, and average outgoing quality (AOQ).

Practical applications and who should use it

• Quality engineers and inspectors choosing efficient acceptance sampling procedures when inspection cost is high but occasional second samples are acceptable.
• Procurement and supplier quality management for lot disposition of isolated lots or short production runs where the sum of both sample sizes is typically ≤ ~10% of lot size.
• Manufacturers operating short series or pilot runs where AQL-indexed switching rules (e.g., ISO 2859-1) are unsuitable.
• Organizations running continuing series may use these plans when failed lots are 100% inspected and nonconforming items replaced, though ISO 2859-1 should also be considered for continuing series.
• Statisticians applying hypothesis testing of lot/process quality using PRQ vs CRQ comparisons.

Related standards

• ISO 2859-1 (double sampling and AQL-indexed plans) - consider when using switching rules for continuing series.
• ISO 28590–28599 series - related acceptance sampling procedures and sequential sampling standards (renumbered series referenced in ISO foreword).

Keywords: ISO 28592, double sampling plans, attributes, minimal sample sizes, producer’s risk quality (PRQ), consumer’s risk quality (CRQ), acceptance sampling, percent nonconforming, nonconformities per 100 items, AOQ, inspection plans.