Acceptance Sampling and OC Curves: A CQE Study Guide
Acceptance sampling decides whether to accept a lot based on a sample. Here is how sampling plans work and how to read an OC curve for the CQE exam.
Not every lot can or should be inspected 100%. When inspection is destructive, expensive, or simply impractical at volume, quality engineers turn to acceptance sampling—a disciplined way to accept or reject a lot based on a representative sample. It is a staple of the CQE Body of Knowledge and a frequent source of exam questions.
I am John Lee of Alpha Training and Consulting. This guide explains how sampling plans work, how to interpret the operating characteristic curve, and the risk trade-offs the exam expects you to reason about.
How a sampling plan works
A single sampling plan is defined by two numbers: the sample size (n) and the acceptance number (c). You draw n units at random from the lot, inspect them, and count the defectives. If the count is at or below c, you accept the lot; if it exceeds c, you reject it. Double and multiple sampling plans add stages that can reduce average inspection, but the logic is the same.
Reading the operating characteristic (OC) curve
The OC curve is the fingerprint of a sampling plan. The horizontal axis is the lot’s true fraction defective; the vertical axis is the probability that the plan accepts the lot. A perfect plan would accept every good lot and reject every bad one—a vertical line. Real plans produce an S-shaped curve, and the steeper it is, the better the plan discriminates between acceptable and unacceptable quality.
Increasing the sample size steepens the curve and improves discrimination. Changing the acceptance number shifts the curve. Understanding how n and c reshape the OC curve is a core exam skill.
AQL, RQL, and the two risks
| Term | Meaning |
|---|---|
| AQL (Acceptable Quality Level) | The worst quality level still considered acceptable; the plan accepts these lots most of the time. |
| RQL / LTPD | The quality level considered unacceptable; the plan is designed to reject these lots. |
| Producer risk (alpha) | Probability of rejecting a lot at the AQL—a good lot wrongly rejected. |
| Consumer risk (beta) | Probability of accepting a lot at the RQL—a bad lot wrongly accepted. |
Standards you should recognize
The CQE expects familiarity with the major sampling standards: ANSI/ASQ Z1.4 for attributes sampling and Z1.9 for variables sampling, along with the Dodge-Romig tables. You are not expected to memorize the tables, but you should know when each applies and how switching rules (normal, tightened, reduced inspection) respond to a supplier’s demonstrated quality history.
Acceptance sampling pairs naturally with statistical process control: sampling decides what to do with product that already exists, while SPC works to ensure fewer defects are produced in the first place. Study them together and the logic of each becomes clearer.
Frequently asked questions
When is acceptance sampling appropriate?+
When 100% inspection is too costly, destructive, or impractical, and when the risk of accepting some defective units is tolerable. It is a lot-disposition tool, not a process-improvement tool.
What does an OC curve tell you?+
It plots the probability of accepting a lot against the lot’s actual fraction defective. A steep OC curve discriminates well between good and bad lots; a flat curve discriminates poorly.
What is the difference between producer risk and consumer risk?+
Producer risk (alpha) is the probability of rejecting a lot that is actually acceptable. Consumer risk (beta) is the probability of accepting a lot that is actually unacceptable. Sampling plans balance the two.
Written by
John Lee
President, Alpha Training and Consulting
John Lee has more than 25 years of experience in quality engineering and has personally earned every ASQ certification, including the Certified Quality Engineer (CQE). A Shingo Award–winning author, he has helped thousands of engineers pass their ASQ exams on the first attempt with a 94% first-time pass rate.
Certifications: BSME, MBA, CMQ/OE, CQE, CRE, CQA, CQIA, CQI, CCT, CQPA, CQT, CHA, CBA, CSQE, CSQP, CCQM, CSSYB, CSSGB, CSSBB, CPGP, CMBB
Alpha Training and Consulting is an independent training provider and is not affiliated with, endorsed by, or sponsored by ASQ. "ASQ" and "Certified Quality Engineer (CQE)" are trademarks of the American Society for Quality. Exam policies, fees, and Body of Knowledge weightings are set by ASQ and can change—always confirm current details at asq.org.