Skip to content
Get On My Space Get On My Space Get My Coverage Plan

AQL sampling plan for during production inspection (dpi): what importers book before shipment

The short answer

An AQL sampling plan turns a DPI from a general look-around into a measurable test: this many units pulled, these defect limits applied, this verdict recorded with evidence. Importers who book that plan explicitly get reports they can act on and suppliers who know exactly what is expected. This guide explains the scope to book, the numbers behind it, and the regulated checkpoints that deserve special handling before shipment.

Key takeaways

  1. AQL sampling translates the completed lot size into a defined sample size plus acceptance numbers for major and minor defects, agreed before the visit.
  2. DPI applies the plan to the quantity actually complete on visit day, so timing the visit at 20 to 30 percent completion determines the statistical value.
  3. Critical defects sit outside AQL tolerance logic and trigger containment on any occurrence, regardless of the sample counts.
  4. Marking consistency, including CE marking where applicable, belongs in the DPI scope because re-labeling finished goods is slow and expensive [1].
  5. Orders serving several destinations need variant-aware sampling, since technical requirements legitimately differ between markets [2].

Importers often book a during production inspection with a vague instruction to check quality, then wonder why the report feels thin. The missing piece is usually the sampling plan: how many units get pulled, from which cartons and line output, against which defect limits, and with what happens next. Without those answers agreed in advance, the visit becomes a matter of opinion rather than evidence.

AQL sampling gives both sides a shared language for those answers. It translates lot size into sample size and defect tolerance into acceptance numbers, so the factory knows the test before it runs and the buyer can defend the verdict afterward. Applied mid-production, the same plan also tells you whether correction is still economical, because the defect rate is measured while most of the order is still unbuilt.

This article explains what importers should specify when they book AQL sampling inside a DPI: the scope choices, the numbers behind the tables, and the special handling for regulated checkpoints such as marking and market variants. It assumes no statistical background and focuses on the booking decisions that determine whether the visit protects the shipment.

What the AQL plan decides before anyone visits

Every sampling plan answers three questions in writing: how large the lot is, how many units get inspected, and how many defects of each severity the lot can contain while still passing. The lot for a DPI is the quantity completed and available on visit day, counted from finished cartons plus identified current output. The sample size grows with the lot up to defined breakpoints, then levels off, which is why inspecting at 25 percent completion yields a smaller but still meaningful sample while leaving room to correct.

Illustration: What the AQL plan decides before Decorative illustration for the section "What the AQL plan decides before"; visual only, carries no data.

Defect classification drives the limits. Critical defects, meaning unsafe or noncompliant units, are normally excluded from tolerance entirely: one occurrence fails the affected scope. Major defects, which the customer will notice or return, get a tight limit such as 1.0 or 1.5. Minor defects, small blemishes within tolerance, get a looser limit such as 2.5 or 4.0. These numbers are commercial agreements written into the purchase order, not laws of nature, and both sides should sign them before production starts.

The inspection level sets how much scrutiny applies. The general level used for most consumer goods balances confidence against visit cost, while special levels with smaller samples suit checks that are destructive or time-consuming. Your booking instruction should state the level, the AQL values per severity, and whether the sample spreads across all colors and sizes proportionally. With those three lines written down, the inspector can work without phone calls and the factory cannot dispute the method afterward.

How the DPI sample gets pulled on the day

Carton selection and unit pulls

Inspectors select cartons systematically across the finished stock, opening a spread from different pallets and production dates rather than the top layer of the nearest stack. From each selected carton they pull units for workmanship review, then set aside a subset for functional and packaging tests. Current line output is sampled separately and identified as such, so drift between yesterday's cartons and today's line becomes visible in the data.

Recording the evidence

Each sampled unit contributes a row of evidence: measurements where tolerances apply, function results where a script exists, and photos for every defect with the classification noted. The report states the lot size found, the sample size taken, the defects counted by severity, and the resulting verdict against the agreed limits. That record is what lets a buyer in another country make the same decision the inspector made on the line. Confirm these recording points when booking so nothing is left to assumption. See the inspection scope.

Scope choiceWhat to specify when bookingWhy it mattersCommon mistake
Lot definitionCompleted quantity on visit daySets the correct sample sizeQuoting the full order as the lot
Sample spreadAcross cartons, dates, colors, sizesCatches clustered defectsPulling from one pallet only
AQL valuesMajor 1.0-1.5, minor 2.5-4.0 typicalDefines pass or fail objectivelyLeaving limits to the inspector
Critical ruleZero tolerance with containmentProtects safety and complianceAveraging criticals into majors
Variant coverageEach market variant sampledPrevents wrong-market shipmentsTesting only the dominant variant
Illustration: Scope choice What to specify when Decorative illustration for the section "Scope choice What to specify when"; visual only, carries no data.

Regulated checkpoints inside the sampling scope

Two checkpoints deserve explicit scope lines whenever they apply. The first is conformity marking for destination markets. Where the CE marking is required, the mark, the model designation, the ratings plate, and the accompanying warnings must match the approved artwork from the first units off the line [1]. Inspectors verify this on early output because correcting artwork and plates mid-production costs little, while re-labeling thousands of finished cartons costs dearly and delays the vessel.

The second is market-variant control for orders split across destinations. Plug types, voltage ratings, manual languages, and warning symbols legitimately differ between markets, and international trade rules recognize that technical regulations may vary while disciplining how they are applied [2]. The DPI scope should therefore require the inspector to confirm segregation: the right variant in the right carton with the right documents. Sampling only the dominant variant leaves the smaller destination exposed to a full-lot error.

Write both checkpoints into the booking as named lines with their own verdicts rather than burying them under general workmanship. A report that states marking passed and variant segregation passed, each with its own evidence, lets the shipping department release with confidence. A report that never mentions them leaves the buyer guessing at the exact point where guessing is most expensive.

Before your next order reaches 25 percent completion, write the sampling plan into the booking: lot definition, sample spread, AQL values, the critical rule, and variant coverage. A one-paragraph scope agreed in advance produces a report that reads like evidence instead of opinion, and evidence is what gets defects fixed while there is still time.

Sources

  1. Wikipedia: CE marking —— Overview of CE marking as conformity marking for products sold in the European Economic Area.
  2. WTO: Agreement on Technical Barriers to Trade —— WTO portal for the TBT Agreement covering technical regulations, standards and conformity assessment.

Frequently asked questions

How is the AQL sample size actually determined for a DPI visit?

AQL tables set the sample size from the lot size and the inspection level, then set the acceptance and rejection numbers from the chosen AQL values for major and minor defects. Tighter AQL values mean fewer defects tolerated. For DPI the lot is the quantity completed on visit day, and the inspector records that quantity so the sample size is traceable. The key point is that the plan is agreed before the visit, not improvised on the line.

Does passing AQL during DPI mean the whole order is defect-free?

No, and it should not be used that way. AQL sampling gives statistical confidence that the defect rate sits below an agreed threshold; it does not promise zero defects in the uninspected units. Critical defects are normally handled outside AQL logic with a zero-tolerance rule: any critical finding triggers containment regardless of the count. Read the AQL result as a process signal, then verify correction with a focused re-check.

What do inspectors check about CE marking during production?

The CE marking indicates the manufacturer's declaration that the product meets applicable European requirements, and the marking, instructions, and model identification must be consistent across the goods. During DPI the inspector checks that early output already carries the correct mark and that variant-specific plates are not mixed. Finding a marking error mid-production is far cheaper than re-labeling finished cartons.

Why does the DPI scope split samples by destination market?

Technical regulations and conformity assessment procedures differ legitimately across destination markets, and the WTO TBT Agreement disciplines how those measures are prepared and applied so they do not create unnecessary trade obstacles. In a DPI this means verifying that each market variant in the order, such as plug types, label languages, and manual versions, is correctly segregated on the line. The AQL sample should cover each variant rather than only the dominant one.

admin

Writer at Get On My Space