AQL calculator: sample size and accept/reject numbers
Enter a lot size, inspection level and AQL for critical, major and minor defects. Get the code letter, sample size and accept and reject numbers from the ANSI/ASQ Z1.4 and ISO 2859-1 tables, with the arrow rule applied.
How do you find the AQL sample size?
Find your lot size in Table I of ANSI/ASQ Z1.4 or ISO 2859-1 and read the code letter for your inspection level, usually General II. Then look up that letter and your AQL in Table II-A. A lot of 5,000 at General II and AQL 2.5 gives code L: inspect 200 units, accept 10 defects, reject at 11.
Calculator
Enter your lot and AQL levels
Single sampling, normal inspection, per ANSI/ASQ Z1.4 and ISO 2859-1. The result updates as you type, and nothing you enter leaves your browser.
Lot band: 3,201 to 10,000
General II is the usual choice unless a contract says otherwise.
Your sampling plan
Lot of 5,000 at General II: code letter L. Draw 200 units at random.
- Critical defectsAQL 0Inspect200Accept (Ac)0Reject (Re)1
No AQL column: any critical defect found in the sample rejects the lot.
- Major defectsAQL 2.5Inspect200Accept (Ac)10Reject (Re)11
Plan L, n = 200.
- Minor defectsAQL 4.0Inspect200Accept (Ac)14Reject (Re)15
Plan L, n = 200.
Accept the lot only if every class is at or below its Ac number. Reaching the Re number in any class rejects the lot.
Table I: sample size code letters for your lot
Lot band 3,201 to 10,000. Your inspection level is highlighted.
| S-1 | S-2 | S-3 | S-4 | I | II | III |
|---|---|---|---|---|---|---|
| C (5) | D (8) | F (20) | G (32) | J (80) | L (200) | M (315) |
Table II-A slice: your AQL columns
Each cell is Ac / Re. ↓ means use the first plan below the arrow, ↑ the first plan above it. Your code letter row is outlined and the plans actually used are shaded.
| Code | n | AQL 2.5 | AQL 4.0 |
|---|---|---|---|
| A | 2 | ↓ | ↓ |
| B | 3 | ↓ | 0 / 1 |
| C | 5 | 0 / 1 | ↑ |
| D | 8 | ↑ | ↓ |
| E | 13 | ↓ | 1 / 2 |
| F | 20 | 1 / 2 | 2 / 3 |
| G | 32 | 2 / 3 | 3 / 4 |
| H | 50 | 3 / 4 | 5 / 6 |
| J | 80 | 5 / 6 | 7 / 8 |
| K | 125 | 7 / 8 | 10 / 11 |
| L | 200 | 10 / 11 | 14 / 15 |
| M | 315 | 14 / 15 | 21 / 22 |
| N | 500 | 21 / 22 | ↑ |
| P | 800 | ↑ | ↑ |
| Q | 1,250 | ↑ | ↑ |
| R | 2,000 | ↑ | ↑ |
How does the AQL sampling calculator work?
AQL sampling is a lookup, not a formula. The calculator follows the same two tables an inspector would use from ANSI/ASQ Z1.4 or ISO 2859-1, single sampling under normal inspection.
- Step 1: code letter
- Table I: lot size band × inspection level → letter A to R
- Step 2: plan
- Table II-A: code letter × AQL → sample size n, Ac, Re
- Arrow rule
- ↓ or ↑ in the cell → use the first plan in that direction, with its own n
- 100% rule
- If n ≥ lot size → inspect every unit
- Decision
- Defects ≤ Ac → accept; defects ≥ Re → reject
Each defect class gets its own plan, so critical, major and minor defects can end up with different sample sizes after the arrow rule. Draw the largest sample, then judge each class against its own plan. Critical defects set to 0 use zero acceptance: one critical defect in the sample rejects the lot.
Tables checked on 17 November 2026 against the Z1.4 tables encoded in the open-source AQLSchemes package for R (all 416 cells of Table II-A and all 105 cells of Table I) and against published worked examples.
AQL worked example: a lot of 5,000 units
A shipment of 5,000 units is inspected at General Inspection Level II with the usual consumer goods limits: 0 for critical, 2.5 for major and 4.0 for minor defects.
| Step | Result | |
|---|---|---|
| Lot band5,000 is in 3,201 to 10,000 | 5,000 is in 3,201 to 10,000 | Row 11 |
| Code letterTable I, General II | Table I, General II | L |
| Sample sizeCode L | Code L | 200 units |
| Major, AQL 2.5Table II-A, row L | Table II-A, row L | Ac 10, Re 11 |
| Minor, AQL 4.0Table II-A, row L | Table II-A, row L | Ac 14, Re 15 |
| Critical, 0Zero acceptance on 200 units | Zero acceptance on 200 units | Ac 0, Re 1 |
The inspector finds 0 critical, 8 major and 15 minor defects. Major is within its limit (8 ≤ 10), but minor reaches its rejection number (15 ≥ 15), so the lot is rejected.
When the arrow changes the sample
Take a lot of 800 at General II. Table I gives code J, which is 80 units. At AQL 2.5 the plan is 80 units, Ac 5, Re 6. At AQL 0.065 the J cell holds a down arrow: the first plan below is code L, so you inspect 200 units and reject on one defect. At the other end, a lot of 100 is code F (20 units), and at AQL 1.0 the F cell points up to code E: 13 units, Ac 0, Re 1.
AQL table: code letters and single sampling plans
Table I gives the sample size code letter for each lot size and inspection level. The number in brackets is the sample size for that letter under normal inspection.
| Lot size | S-1 | S-2 | S-3 | S-4 | I | II | III |
|---|---|---|---|---|---|---|---|
| 2 to 8 | A (2) | A (2) | A (2) | A (2) | A (2) | A (2) | B (3) |
| 9 to 15 | A (2) | A (2) | A (2) | A (2) | A (2) | B (3) | C (5) |
| 16 to 25 | A (2) | A (2) | B (3) | B (3) | B (3) | C (5) | D (8) |
| 26 to 50 | A (2) | B (3) | B (3) | C (5) | C (5) | D (8) | E (13) |
| 51 to 90 | B (3) | B (3) | C (5) | C (5) | C (5) | E (13) | F (20) |
| 91 to 150 | B (3) | B (3) | C (5) | D (8) | D (8) | F (20) | G (32) |
| 151 to 280 | B (3) | C (5) | D (8) | E (13) | E (13) | G (32) | H (50) |
| 281 to 500 | B (3) | C (5) | D (8) | E (13) | F (20) | H (50) | J (80) |
| 501 to 1,200 | C (5) | C (5) | E (13) | F (20) | G (32) | J (80) | K (125) |
| 1,201 to 3,200 | C (5) | D (8) | E (13) | G (32) | H (50) | K (125) | L (200) |
| 3,201 to 10,000 | C (5) | D (8) | F (20) | G (32) | J (80) | L (200) | M (315) |
| 10,001 to 35,000 | C (5) | D (8) | F (20) | H (50) | K (125) | M (315) | N (500) |
| 35,001 to 150,000 | D (8) | E (13) | G (32) | J (80) | L (200) | N (500) | P (800) |
| 150,001 to 500,000 | D (8) | E (13) | G (32) | J (80) | M (315) | P (800) | Q (1250) |
| 500,001 and over | D (8) | E (13) | H (50) | K (125) | N (500) | Q (1250) | R (2000) |
Table II-A below shows Ac / Re for the AQL values used most often. Arrows mean use the first plan in the arrow's direction. The calculator above covers every AQL from 0.010 to 10.
| Code | n | 0.65 | 1.0 | 1.5 | 2.5 | 4.0 | 6.5 |
|---|---|---|---|---|---|---|---|
| A | 2 | ↓ | ↓ | ↓ | ↓ | ↓ | 0 / 1 |
| B | 3 | ↓ | ↓ | ↓ | ↓ | 0 / 1 | ↑ |
| C | 5 | ↓ | ↓ | ↓ | 0 / 1 | ↑ | ↓ |
| D | 8 | ↓ | ↓ | 0 / 1 | ↑ | ↓ | 1 / 2 |
| E | 13 | ↓ | 0 / 1 | ↑ | ↓ | 1 / 2 | 2 / 3 |
| F | 20 | 0 / 1 | ↑ | ↓ | 1 / 2 | 2 / 3 | 3 / 4 |
| G | 32 | ↑ | ↓ | 1 / 2 | 2 / 3 | 3 / 4 | 5 / 6 |
| H | 50 | ↓ | 1 / 2 | 2 / 3 | 3 / 4 | 5 / 6 | 7 / 8 |
| J | 80 | 1 / 2 | 2 / 3 | 3 / 4 | 5 / 6 | 7 / 8 | 10 / 11 |
| K | 125 | 2 / 3 | 3 / 4 | 5 / 6 | 7 / 8 | 10 / 11 | 14 / 15 |
| L | 200 | 3 / 4 | 5 / 6 | 7 / 8 | 10 / 11 | 14 / 15 | 21 / 22 |
| M | 315 | 5 / 6 | 7 / 8 | 10 / 11 | 14 / 15 | 21 / 22 | ↑ |
| N | 500 | 7 / 8 | 10 / 11 | 14 / 15 | 21 / 22 | ↑ | ↑ |
| P | 800 | 10 / 11 | 14 / 15 | 21 / 22 | ↑ | ↑ | ↑ |
| Q | 1,250 | 14 / 15 | 21 / 22 | ↑ | ↑ | ↑ | ↑ |
| R | 2,000 | 21 / 22 | ↑ | ↑ | ↑ | ↑ | ↑ |
Normal, tightened and reduced inspection
Inspection starts at normal severity. The standard then switches between normal, tightened and reduced inspection based on recent lot results. This calculator gives normal plans only, because tightened plans (Table II-B) and reduced plans (Table II-C) apply only once those rules trigger.
- Normal to tightened: when 2 of 5 or fewer consecutive lots are not accepted on original inspection.
- Tightened to normal: when 5 consecutive lots are accepted on original inspection.
- Normal to reduced:only when production is steady, recent results meet the standard's criteria, and the responsible authority approves.
- Reduced to normal: as soon as a lot is not accepted, production becomes irregular or other conditions warrant it.
If lots keep failing on tightened inspection, the standard stops sampling acceptance until the supplier has improved the process. Check the exact criteria in the edition of ISO 2859-1 or ANSI/ASQ Z1.4 your contract names.
What are the common AQL sampling mistakes?
- Reading AQL as the allowed defect rate. AQL 2.5 does not mean 2.5% defects are fine in your lot. It describes the process average the plan is designed to accept.
- Ignoring the arrow.Using the code letter's own sample size with the Ac/Re from another row gives a plan that does not exist in the standard.
- Sampling from the top of the pallet.Units must be drawn at random from the whole lot, or the plan's protection does not hold.
- Mixing lots. Each lot should come from one production run under the same conditions.
- Unclear defect classes. Agree in writing which defects are critical, major and minor before inspection starts. An FMEA worksheet is a good place to rank which failure modes matter most.
Sampling estimates quality from a fraction of the lot. Intellyx Automated Vision QC inspects every unit on the line with cameras, and Perimattic's computer vision for manufacturing team builds custom inspection stations. For the wider system of inspection plans, records and corrective actions, see quality management system development. If lot sizes swing with orders, Perimattic Demand OS forecasts demand so production lots, and the inspection work that comes with them, can be planned ahead.
Once you know how many units fail, put numbers on it with the defect rate and cost of quality calculator, see how scrap pulls down overall equipment effectiveness, or price a rejected lot with the production cost calculator. The DPMO & Six Sigma Calculator and OEE Calculator on this site cover the same ground for Perimattic Intellyx users.
Which calculator should you use next?
Each of these answers a question this result raises. All 11 free manufacturing tools show their formula or method.
Frequently Asked Questions
What does AQL 2.5 mean?
AQL 2.5 means the worst process average you are prepared to tolerate is 2.5 defective units per 100. The sampling plan is built so lots from a process running at or better than 2.5% are accepted most of the time. It does not promise that any single lot has 2.5% defects or fewer, and it is not a target defect rate.
How do I calculate the AQL sample size?
You do not calculate it with a formula; you read it from the standard tables. Table I gives a code letter from the lot size and inspection level, and Table II-A gives the sample size and accept and reject numbers for that letter and your AQL. If the cell holds an arrow, follow it to the first plan it points to and use that plan’s sample size.
What is the difference between inspection levels I, II and III?
They set how much you inspect relative to lot size. General II is the default. Level I takes smaller samples and gives less discrimination between good and bad lots, while Level III takes larger samples and gives more. Special levels S-1 to S-4 use very small samples for destructive or costly tests where large samples are not practical.
What happens when the AQL table shows an arrow?
An arrow means no plan exists for that code letter and AQL. Move in the arrow’s direction to the first cell with numbers and use that row’s sample size along with its accept and reject numbers. For example, a lot of 800 at General II is code J (80 units), but at AQL 0.065 the arrow points down to code L: inspect 200 units, accept 0, reject 1.
What AQL should I use for critical, major and minor defects?
A common choice in consumer goods inspection is 0 for critical defects, 2.5 for major and 4.0 for minor, which is what this calculator loads by default. The standard does not prescribe these values. Your customer contract or quality agreement should name the AQL for each defect class, and safety-related defects usually get zero acceptance.
What if the AQL sample size is larger than my lot?
Inspect every unit in the lot. This happens with small lots, particularly when an arrow sends you to a larger plan. The accept and reject numbers still come from the plan you landed on, but with 100% inspection you are counting actual defects in the lot rather than estimating them from a sample.
Are ANSI/ASQ Z1.4 and ISO 2859-1 tables the same?
For single sampling under normal inspection, the code letters, sample sizes and accept and reject numbers match, and both descend from MIL-STD-105E. The documents differ in wording and some procedural details, so name the standard your customer specifies on your inspection report.
Want these numbers live, not once a month?
- Free consultation
- No commitment
- Tailored to your team