AQL Sample Size Calculator
Enter your order quantity and quality limits to get the exact sample size and accept/reject numbers under ISO 2859-1 (ANSI/ASQ Z1.4) — the sampling standard used by inspectors worldwide.
Your inspection
Uses normal inspection severity — the standard for one-off inspections and new suppliers. More on severity ↓
Dimensions, function, safety and destructive checks usually use a Special level on a smaller sub-sample.
Your sampling plan
| Defect class | AQL | Sample | Accept ≤ | Reject ≥ |
|---|
| Check | AQL | Sample | Accept ≤ | Reject ≥ |
|---|
Ac = maximum defects to still pass · Re = defects that fail the lot. Based on ISO 2859-1 / ANSI Z1.4 single-sampling plans, normal inspection. How AQL works →
How the calculation works
Two lookups turn your order quantity into an inspection plan.
The AQL tables
These are the official ISO 2859-1 / ANSI Z1.4 tables. The highlighted cells update live to match the inputs in the calculator above.
Table 1 — Sample-size code letters
Read across from your lot size to your inspection level — the cell where they meet is your code letter.
Table 2 — Single sampling plan (normal inspection)
Each cell shows Ac (accept) and Re (reject). A ↓ means step down to a larger sample; a ↑ means step up to a smaller one (the calculator resolves these for you).
How to use this calculator
Enter the total number of units in your production lot, choose an inspection level, and set an AQL (Acceptable Quality Limit) for each defect class. The calculator applies the ISO 2859-1 (ANSI/ASQ Z1.4) single-sampling tables to return three things: the sample-size code letter, the number of units the inspector should pull and check at random, and the accept/reject numbers for critical, major, and minor defects.
During the inspection, the inspector counts the defects found in the sample and sorts them by class. If the number of defects in any class reaches its reject number, the lot fails that class. If every class stays at or below its accept number, the lot passes.
What AQL means
The Acceptable Quality Limit is the maximum percentage of defective units that is still considered acceptable for a given defect class. A lower AQL is stricter — it tolerates fewer defects and, all else equal, drives a larger sample and a lower accept number. Defects are graded into three classes, and each is usually assigned its own AQL:
- Critical — safety hazards or legal non-compliance. Often set to a very tight AQL, or to zero tolerance (any critical defect fails the lot).
- Major — functional faults a customer would likely return or complain about. Commonly AQL 2.5.
- Minor — cosmetic or workmanship issues that don't affect use. Commonly AQL 4.0.
The 0 / 2.5 / 4.0 combination is the most widely used default for general consumer goods, but you and your supplier should agree AQLs that match your product's risk profile. For what belongs in each class — with worked examples across apparel, electronics, furniture and toys — see our guide to critical, major and minor defects.
General vs special inspection levels
The inspection level sets how large a sample you draw relative to the lot. There are two families, and a typical inspection uses them for different checks:
- General levels (I, II, III) — used for visual and cosmetic inspection, where every sampled unit is examined quickly for appearance, workmanship, colour, and labelling. General II is the default; General I draws a smaller sample (lower cost, less discrimination) and General III a larger one (tighter control).
- Special levels (S-1 to S-4) — used for checks that are slow, costly, or destructive, so a large sample isn't practical: dimensional measurements, function and safety tests, barcode or metal-detection checks, and any test that consumes the unit. They deliberately draw a much smaller sub-sample and accept a higher sampling risk.
This is why a single inspection usually runs more than one sampling plan at once. A common example for a 5,000-unit lot: General II (n=200) for visual defects, plus Level S-2 for dimensional measurements and on-site function tests — a sub-sample of roughly 8–13 units rather than all 200. Two plans, one booking. The calculator's optional "measurement & test plan" reflects exactly this: pick a Special level for the checks you can't run on all 200 pieces.
As a rule of thumb: use a General level (usually II) for the visual AQL, and a Special level (S-1 or S-2 are the most common) for measurements and tests. Your inspection provider will confirm the exact levels on the booking.
How to choose your inspection level
The level is a trade-off: a higher level draws a larger sample, so it catches more defects but takes more time and costs more. Start from the default and move up or down based on risk.
| Level | What it's for | Sample | Choose it when… |
|---|---|---|---|
| General II | Visual & cosmetic — the standard default | Standard | Most inspections — start here unless you have a specific reason not to. |
| General III | Visual & cosmetic, tighter control | Larger | High-value or safety-relevant goods, a new or unproven supplier, tight quality expectations, or a history of defects. |
| General I | Visual & cosmetic, lower cost | Smaller | Low-risk, low-value goods from a trusted supplier, where you'll accept weaker discrimination to save cost. |
| Special S-1 / S-2 | Slow, costly or destructive checks — measurements, function, safety | Very small | On-site tests you can't run on the whole visual sample. S-2 is the most common. |
| Special S-3 / S-4 | The same checks, with more confidence | Small (larger than S-1/2) | Higher-risk tests where you want a bigger test sub-sample. |
In short: default to General II for the visual check; step up to III when a defect slipping through would be expensive — safety, brand reputation, high unit value — or when the supplier is unproven; drop to I only for low-risk goods from a supplier you trust. For measurements, function and safety tests, add a Special level (S-1 or S-2) on a small sub-sample. Try each level in the calculator above and watch how the sample size and accept/reject numbers change — that's the trade-off you're making.
Inspection severity: normal, tightened, reduced
This calculator uses normal inspection severity, which is what nearly all one-off inspections and new-supplier checks use. The standard also defines two other severities, but you don't choose them freely — you move between them based on a supplier's recent track record across a series of lots:
- Normal — the default starting point for any new supplier or one-off inspection.
- Tightened — stricter (lower accept numbers). You switch to it when quality slips — the common rule is 2 rejected lots out of 5 consecutive. It protects you while a supplier is trending badly.
- Reduced — looser, with smaller samples to save cost. A supplier earns it after a consistent run of accepted lots (typically 10 in a row, among other conditions).
Because tightened and reduced only apply once you're running an ongoing AQL programme with switching rules, this tool sticks to normal severity. If you need a tightened or reduced plan for an established supplier, your inspection provider will apply it.
Where this fits in your inspection
AQL sampling is the backbone of a pre-shipment inspection, where an inspector pulls a random sample from finished, packed goods and checks it against your AQLs before you release the balance payment or ship. The same tables are used for in-process inspections and other product checks. For a full walkthrough of how the tables are built and how to read the code letters, see our AQL tables and sample size guide.
This calculator is provided for reference. Your inspection provider will confirm the exact plan on their report; if a value ever differs, follow the plan agreed in your booking.