Opens in a new tab

Supplier Receiving Inspection: Definition, Methods, and Best Practices

Reading time
Home / Blog / IQC / Supplier Receiving Inspection: Definition, Methods, and Best Practices
See our solutions

A non-conforming batch that passes through the receiving area undetected costs nothing at the time. It costs money later: machine downtime, parts that need to be sorted urgently, customer complaints, and sometimes product recalls. Supplier receiving inspection is the first line of defense against these costs, yet it is often carried out based on «gut feeling»: we inspect 10 parts out of habit, we inspect everything just to be safe, or we don’t inspect anything at all due to time constraints.

When well-designed, it protects production without unnecessarily tying up the laboratory. This article provides an overview of possible inspection levels, the choice between attribute and measurement-based inspections, sampling plans based on ISO 2859, and the practices that make a difference in the field.

What is a supplier acceptance inspection?

Supplier acceptance inspection (in English incoming inspection) encompasses all inspections performed on raw materials, components, or subassemblies upon delivery, prior to their storage or incorporation into production.

It has two purposes:

  • Decide : Accept or reject a batch based on predefined criteria (drawing, specifications, monitoring plan).
  • Nurturing the Supplier Relationship : Every deviation detected must trigger a request for corrective action and be factored into the supplier's performance evaluation.

ISO 9001 (Section 8.4) requires that products supplied by external service providers be controlled, but does not prescribe a specific method of control. It is up to the company to define a level of verification commensurate with the risk.

100 % inspection, sampling, inspection skip: Which level should you choose?

There are three main options, and a single company generally uses all of them, depending on the product.

Control 100 %

Every part is inspected. This approach is justified for critical characteristics (safety, primary function), for a supplier experiencing a quality crisis, or during the ramp-up phase. Its drawback: it is costly and, contrary to popular belief, it is not foolproof. During visual or repetitive inspections, operator fatigue can cause defects to be overlooked.

Sampling Inspection

A sample is drawn according to a statistical plan, and the decision applies to the entire lot. This is the most common method, provided that the sample size is based on a standard rather than on custom.

Example: For a lot of 1,000 units under General Level II inspection, ISO 2859-1 assigns the code letter J, which corresponds to a sample size of 80 pieces under normal testing conditions. This is a far cry from the «10 samples» often used, which offer no statistical guarantee.

The skip-lot

The shipment is accepted without a physical inspection, based on the supplier’s demonstrated performance. ISO 2859-3 governs this practice: it is reserved for qualified suppliers with a consistent track record and provides for a return to inspection as soon as a deviation occurs.

Attribute-based or measurement-based inspection: ISO 2859 or ISO 3951?

This is a question that is rarely asked, even though it has a direct impact on the audit workload.

Attribute-based inspection (ISO 2859)

Each item in the sample is declared either conforming or nonconforming. The lot is accepted if the number of nonconforming items does not exceed the acceptance criterion. Easy to implement, this method is suitable for appearance defects, «pass/fail» gage inspections, and non-measurable characteristics.

Measurement-Based Control (ISO 3951)

We analyze the measured value of each part: mean, standard deviation, and position relative to the tolerances. A part measured at 9.98 mm with a tolerance of 10 ± 0.05 mm does not provide the same information as a part measured at 10.04 mm, even though attribute-based inspection would classify both as «conforming.».

Because it uses more information, control based on measurements allows, for an equivalent level of protection, significantly smaller samples. In return, it requires a measurable characteristic and a distribution consistent with the standard's assumptions (generally close to the normal distribution).

In precision engineering, where most characteristics are dimensional and have already been measured, ISO 3951 is often underutilized, even though it provides a direct means of reducing inspection time.

Understanding NQA to Set Up Your Plans Properly

The AQL (Acceptable Quality Level) is one of the most misunderstood parameters in acceptance inspection.

An NQA of 1 % does not mean that a lot can contain 1 % of «allowed» defective parts. It is a design parameter of the plan: it corresponds to the average quality level of a supplier that we are willing to accept most of the time across a series of lots. The plan is designed so that a supplier at this level will have the majority of their lots accepted, while a significantly poorer supplier will be identified.

In practice:

  • a tight NQA (0.1 % to 0.65 %) for critical characteristics;
  • an intermediate NQA (1 % to 2.5 %) for major characteristics;
  • a broader NQA for minor defects.

The NQA is discussed and formalized with the supplier, ideally starting with the qualification process.

Streamlining Receiving Inspections: Standard, Enhanced, or Streamlined

ISO 2859-1 is not limited to sample size tables. It provides for switching rules that automatically adjust the strictness of the audit based on the supplier's behavior:

  • Routine checkup : starting point.
  • Entry into an area with enhanced security checks : when two out of five consecutive batches are rejected. The acceptance criteria become stricter.
  • Back to Normal : after five consecutive batches passed the enhanced inspection.
  • Transition to reduced control : when the supplier demonstrates consistently good performance, production is steady, and the quality manager approves the transition. The sample size is then smaller.
  • Termination of the inspection in accordance with the standard : If the supplier remains under enhanced monitoring for too long, the standard calls for suspending the acceptance of shipments until corrective action is taken.

It is this mechanism that makes the acceptance inspection both protective and cost-effective: we focus our efforts on suppliers who are falling short, and we ease up on those who have proven themselves. However, it must actually be implemented, which requires tracking the history of each supplier/product pair. This is rarely maintained over the long term in a spreadsheet.

Best Operational Practices

Segment suppliers and product lines by risk

Consider two factors: the criticality of the characteristic (safety, regulatory, functional) and the supplier’s historical performance (PPM, audit results, complaints). A critical component supplied by an unreliable supplier warrants enhanced inspection; a standard consumable from a qualified supplier may be subject to reduced inspection or a skip-lot procedure.

Establish acceptance criteria before the first delivery

Tolerances, critical defects, characteristics to be measured, quality assurance (QA): everything must be agreed upon with the supplier in advance. A criterion that is disputed at the time of a rejection is one that was not discussed thoroughly enough beforehand.

Check the documents, not just the paperwork

Incoming inspection often begins with the documentation: material certificates (such as Type 3.1 certificates in accordance with EN 10204), certificates of conformity, supplier inspection reports, and lot identification. A compliant part accompanied by an incorrect material certificate remains a problem.

Test using calibrated instruments

An acceptance decision is valid only if the instrument used is calibrated and suitable. A caliper that is out of calibration, or a measuring tool with insufficient repeatability relative to the tolerance, can result in the acceptance of a defective lot or the rejection of a good lot. There is a direct link between acceptance inspection and instrument fleet management.

Document each inspection in a traceable manner

Batch, quantity delivered, sample size, results, decision, inspector, instrument used. These records are used for supplier reviews, audits, and potential disputes.

Resolve nonconformities until they are closed

A rejected lot must be physically identified, placed in quarantine, recorded, and reported to the supplier along with a request for corrective action. Following through on the resolution is just as important as initiating the process: without it, the same defect will recur.

Using Receiving Data to Manage Suppliers

Each inspection yields actionable data: rejection rates by supplier, recurring defects, and gradual drift of a dimension toward a tolerance limit. Measurement-based inspection is invaluable here, as it allows us to detect when a supplier’s production is drifting out of tolerance before they deliver parts that are out of tolerance.

When analyzed regularly, this data makes it possible to:

  • objectively compare suppliers based on the same product code;
  • detect a deviation before it turns into a refusal;
  • support supplier reviews with facts rather than impressions;
  • justify the switch to reduced inspection or skip-lot.

Equipping Your Supplier Acceptance Inspection Process

Strictly applying ISO 2859 and ISO 3951 manually quickly becomes unmanageable once you’re dealing with dozens of suppliers and hundreds of SKUs. Dedicated incoming inspection software handles the tedious tasks: generating sampling plans, calculating acceptance criteria, automatically applying switching rules, and maintaining history by supplier and by part number.

This is the approach taken by Ellistat’s IQC module, which incorporates progressive inspection and dynamic adjustment in accordance with ISO 2859 and ISO 3951. By combining measurement-based inspection and automatic switching, it reduces the number of inspections required to achieve the same quality level by half. The results are based on instruments whose calibration status is verified, and every decision remains traceable.

Conclusion: Monitor Less, but More Effectively

An effective supplier acceptance inspection is neither systematic nor improvised. It is based on standardized sampling plans, a thoughtful selection of attributes and measurements, pass/fail rules that are actually enforced, and data used to help suppliers improve. The goal is not to increase the number of inspections, but to focus efforts where risk warrants it. Ultimately, a well-designed acceptance inspection process reduces the need for itself.

FAQ

What is the difference between ISO 2859 and ISO 3951?

ISO 2859 defines attribute inspection plans: each part is judged to be either conforming or nonconforming. ISO 3951 defines measurement-based inspection plans, which use measured values. For the same level of assurance, measurement-based inspection generally requires a smaller sample size.

How to Choose the Right NQA?

The NQA depends on the severity of the defect. A tight NQA is used for critical characteristics, an intermediate NQA for major characteristics, and a more lenient NQA for minor defects. It must be agreed upon with the supplier before the first deliveries.

When should you switch to reduced monitoring?

According to ISO 2859-1, a switch to reduced inspection is possible when the supplier demonstrates consistently good performance, maintains a steady production rate, and receives authorization from the quality manager. Any rejected lot results in a return to normal inspection.

Is an acceptance inspection required under ISO 9001?

ISO 9001 requires control over the products and services provided by external suppliers, but does not specify a particular method of control. The company selects the level of verification appropriate to the risk: inspection upon receipt, supplier audits, certificates, or a combination of these methods.

How many parts should be inspected in a batch?

It depends on the batch size, the inspection level, and the QAA. For example, for a batch of 1,000 pieces at General Level II, ISO 2859-1 specifies a sample size of 80 pieces for a standard attribute inspection. An inspection based on measurements in accordance with ISO 3951 generally allows this sample size to be reduced.