Setting an AQL inspection standard for stamping 1 parts from Vietnam sounds simple until a rejected shipment lands. Our Vietnam team has seen vague standards destroy entire orders and buyer trust.
To set an AQL inspection standard for stamping parts from Vietnam, classify defects as critical, major, and minor, apply ISO 2859-1 General Inspection Level II, assign AQL 0 for critical, 1.0–2.5 for major, and 4.0 for minor defects, then write everything into the purchase contract.
That is the short answer. But stamping parts carry unique risks — burrs, cracked bends, hole positions — so let me walk you through each decision step by step.
How do I choose the right AQL level for stamping parts sourced from Vietnam?
A few years back, one of our U.S. clients rejected a full lot of brackets over burrs. The AQL was fine on paper. The problem was that nobody had defined which AQL applied to what.
Choose your AQL level based on defect risk, not habit. Use General Inspection Level II under ISO 2859-1 as the default, then assign AQL 0 for critical defects, AQL 1.0 or 2.5 for major functional defects, and AQL 4.0 for minor cosmetic issues on stamped parts.
/ICHI-%20Photo/150-ton%20Stamping%20Press%20Machine%201.jpg)
First, understand what AQL actually is. AQL is not a quality target. It is a sampling and acceptance rule acceptance sampling 2. It tells an inspector how many parts to pull from a lot, and how many defects trigger a rejection. It does not guarantee a defect-free shipment. Many buyers miss this point, and it causes disputes later.
For stamping parts, the risk profile is different from consumer goods. A scratch on a hidden bracket flange rarely matters. A hole punched 0.3 mm out of position can stop an entire assembly line. So the AQL you pick must match the consequence of failure, not just industry habit.
Match AQL to how the part is used
Here is the framework we use when we help U.S. buyers source stamped parts through our Vietnam office:
| Part usage scenario | Suggested major AQL | Suggested minor AQL | Notes |
|---|---|---|---|
| Structural or load-bearing bracket | 1.0 | 2.5 | Tighten sampling; dimensions are king |
| Standard mounting clip, hidden in assembly | 2.5 | 4.0 | Function matters, cosmetics less so |
| Visible cover or cosmetic panel | 1.5 | 2.5 | Appearance moves up in priority |
| Plated or coated component | 1.0–1.5 | 4.0 | Coating failures are functional, not cosmetic |
Also consider inspection level. Level II is the standard starting point. If your supplier is new, or a prior lot failed, move to tightened inspection. If the part carries a safety-critical dimension, consider 100% inspection for that single characteristic. Sampling saves cost, but it should never gamble with a feature that can hurt someone or halt a production line.
Finally, remember the trade-off. A tighter AQL catches more defects but raises inspection time, rejection rates, and cost. Vietnamese factories often push for looser AQLs to protect ship dates. Your job is to hold the line only where risk truly justifies it.
What defect classifications should I use when inspecting stamped parts from Vietnamese suppliers?
Our engineers review stamping drawings every week, and the most common gap we find is not tolerances — it is that nobody has agreed on what counts as a "major" defect.
Classify stamped-part defects into three groups: critical defects that threaten safety or legal compliance with zero tolerance; major defects that cause fit, assembly, or function failures; and minor defects that are cosmetic deviations with no functional impact. Document each class with photos.

Defect classification is where most buyer-factory disputes actually start. A factory in Bac Ninh may call a burr "minor." Your assembly team in Ohio calls it "major" because it cuts an operator's glove. Neither side is lying. They are just using different definitions. So write the definitions down before production starts.
The three defect classes, applied to real stamped parts
Take a stamped steel bracket like the ones we regularly source — a zigzag profile with punched slots for fasteners. Here is how classification works in practice:
| Defect class | Bracket example | Cosmetic panel example | Plated part example |
|---|---|---|---|
| Critical (AQL 0) | Crack at a bend affecting load-bearing | Sharp edge or burr creating injury risk | Exposed base metal with corrosion risk |
| Major (AQL 1.0–2.5) | Slot position out of tolerance, blocks assembly | Visible dent or warped surface | Plating peel or blistering in functional zone |
| Minor (AQL 4.0) | Light scratch on a hidden flange | Faint surface mark below agreed threshold | Slight color variation outside main view area |
Notice something important. The same defect type can shift class depending on location. A burr on a hidden edge may be minor. The same burr on a grip surface becomes critical. This is why generic AQL checklists fail for stamping.
Stamping-specific defects you must define
Your defect list should always cover: burr height limits, dimensional tolerances on holes and bends, flatness, edge and hole cracks, deformation from handling, coating or plating thickness, and surface scratches by zone. For each one, state a measurable limit. "No excessive burrs" is useless. "Burr height max 0.1 mm on edges A and B" is enforceable. In our experience, adding a one-page photo standard — good part, borderline part, reject part — cuts inspection disputes dramatically.
How do I set sample sizes and inspection criteria for my Vietnam stamping orders?
When we run final inspections for clients at factories near Hai Phong, the sampling math takes five minutes. Agreeing on the criteria beforehand is what saves the shipment.
Use your lot quantity and General Inspection Level II to find the code letter in the ISO 2859-1 table, then read the sample size and accept/reject numbers for each AQL. For example, a 3,000-piece lot yields a sample of 125 pieces.

The mechanics are straightforward once you see them laid out. Here is a simplified reference for common stamping lot sizes at Level II:
| Lot size | Sample size | Accept / Reject at AQL 1.0 | Accept / Reject at AQL 2.5 | Accept / Reject at AQL 4.0 |
|---|---|---|---|---|
| 501–1,200 | 80 | 2 / 3 | 5 / 6 | 7 / 8 |
| 1,201–3,200 | 125 | 3 / 4 | 7 / 8 | 10 / 11 |
| 3,201–10,000 | 200 | 5 / 6 | 10 / 11 | 14 / 15 |
| 10,001–35,000 | 315 | 7 / 8 | 14 / 15 | 21 / 22 |
Read it like this. For a 3,000-piece lot at AQL 2.5 for majors: inspect 125 random pieces. If you find 7 or fewer major defects, the lot passes. Find 8, and it fails. Critical defects at AQL 0 mean any single critical defect rejects the lot.
What sampling alone will miss
Here is the hard truth from our project work: AQL sampling cannot see process drift. Progressive dies wear over long runs. A hole that measured perfectly at piece 1,000 may drift out of tolerance by piece 80,000. A 125-piece sample pulled at random may miss this entirely. So for high-precision stamped parts, we pair final AQL inspection with two extra layers. First, a first article inspection 4 to approve the initial output against the drawing. Second, in-process checks tied to die maintenance cycles, sometimes backed by Cpk 5 data on critical dimensions. This is how you catch tool wear before it becomes a rejected container.
Also define the practical details in writing: which measurement tools are used, which drawing revision governs, and how random selection is done across cartons. Ambiguity here is where inspections fall apart.
What steps can I take to ensure my Vietnamese supplier consistently meets my AQL standards?
A purchasing manager once told me his real fear was not one bad lot — it was never knowing which lot would be the bad one. Consistency is the whole game.
Ensure consistency by approving a golden sample before mass production, writing AQL levels and defect definitions into the contract, requiring first article and in-process inspections, using third-party or local inspection teams in Vietnam, and escalating to tightened inspection after any failed lot.

One inspection at the end of production is a snapshot. Consistency comes from a system. Having run our own office in Vietnam with a local team, we have learned that Vietnamese stamping suppliers respond very well to clear, documented expectations — and drift quickly when expectations stay verbal.
Build the system in five steps
- Approve a golden sample. Before mass production, sign off a master sample and keep sealed duplicates on both sides. Every future dispute gets resolved against this physical reference, not against opinions.
- Put the standard in the contract. State the AQL per defect class, the inspection level, the drawing revision, critical dimensions, measurement methods, and cosmetic zones. Vietnam's own national standards reference ISO 2859-1 and expect AQL values to be contractually defined, so factories understand this language.
- Inspect through the process, not just at the end. Cover incoming raw material, first article approval, in-process checks during the run, and final pre-shipment AQL inspection. Early checks catch material and die issues before they multiply across thousands of parts.
- Use eyes on the ground. If you cannot visit factories in Bac Ninh, Hai Phong, or Ho Chi Minh City yourself, use a third-party inspector or a sourcing partner with local staff. Remote quality management for stamping simply does not work through email alone.
- Escalate and de-escalate by history. After a failed lot, switch to tightened inspection until the supplier proves recovery. Reward strong history with normal inspection. This dynamic approach beats a single fixed AQL forever.
One more practical point. Many Vietnamese stampers serving automotive and electronics buyers now hold ISO 9001 6 or IATF 16949 7 certifications, and some Tier-1 shops are adopting AI-based optical inspection. Aligning your AQL requirements with these existing systems — rather than imposing a parallel one — gets faster buy-in and better results.
Conclusion
Vague standards create rejected lots, delays, and disputes. A written AQL plan — clear defect classes, correct sampling, golden samples, and staged inspections — turns Vietnam stamping sourcing into a predictable, low-risk operation.
Frequently Asked Questions
How do I choose the right AQL level for stamping parts sourced from Vietnam?
A few years back, one of our U.S. clients rejected a full lot of brackets over burrs. The AQL was fine on paper. The problem was that nobody had defined which AQL applied to what.
What defect classifications should I use when inspecting stamped parts from Vietnamese suppliers?
Our engineers review stamping drawings every week, and the most common gap we find is not tolerances — it is that nobody has agreed on what counts as a "major" defect.
How do I set sample sizes and inspection criteria for my Vietnam stamping orders?
When we run final inspections for clients at factories near Hai Phong, the sampling math takes five minutes. Agreeing on the criteria beforehand is what saves the shipment.
Footnotes
1. Background explanation of the metalworking process central to the entire article. ↩︎
2. Background concept explaining the statistical basis of AQL sampling methodology. ↩︎
3. Primary international standard governing AQL sampling procedures referenced throughout the article. ↩︎
4. Explains the quality control process used to validate initial production output. ↩︎
5. Statistical concept for process capability referenced when discussing tool wear monitoring. ↩︎
6. Official ISO page for the quality management certification referenced for suppliers. ↩︎
7. Official body for automotive quality standard mentioned as supplier certification. ↩︎
