Table of Contents
A replacement shipment can restore production while leaving the supplier’s original failure mechanism unchanged. Closing a supplier corrective action request (SCAR) at that point confuses commercial recovery with verified prevention.
For electronic components, a useful SCAR connects the affected part and lot to containment records, proven causes, implemented corrections and an effectiveness check. The framework below is an editorial working method informed by 8D problem solving; it is not a claim about a real supplier incident or a certification requirement.
What should a component supplier SCAR contain?
Include a precise nonconformity, the affected material and exposure, action owners, due dates, cause evidence, corrective actions and measurable closure criteria. The supplier should be able to investigate the problem without guessing which requirement was missed.
A problem statement distinguishes requirement from observation: the specified ordering code or condition, what was found, where it was detected and which records support the finding. Avoid substituting “poor quality” for that description.
The ASQ overview of eight disciplines separates containment, verified causes, permanent correction and validation. It also includes investigating why the problem escaped detection. A SCAR can use those distinctions without mechanically requesting an eight-page report for every discrepancy.
Attach the relevant purchase line, lot and packaging identifiers, photographs or measurements, and the affected quantity. Separate confirmed defects from material that is merely suspect. That distinction shapes both the containment boundary and the later denominator used to assess effectiveness.
How do you verify that containment covers the whole suspect lot?
Reconcile the suspect quantity across all locations and process states, and record the actual disposition of each group. A warehouse hold alone does not establish containment of units already in work in progress or shipped onward.
For a hypothetical 1,000-unit lot, the initial ledger might read:
| Material state | Quantity | Containment evidence still needed |
|---|---|---|
| Unissued warehouse stock | 600 | Physical or system hold and reconciled identifiers |
| Work in progress | 250 | Location, affected assemblies and controlled disposition |
| Shipped onward | 150 | Delivery trace and documented downstream action |
| Total reconciled | 1,000 | No unexplained quantity gap |
The arithmetic balances, but that does not mean all 1,000 units are contained. If only the warehouse hold is confirmed, only that 600-unit group has recorded control. Ownership and due dates remain necessary for the other groups.
Expand or narrow the suspect boundary using evidence. Adjacent lots, shared labeling jobs or a common inspection setup may be relevant, but should not be included or excluded solely for convenience. Keep the basis for the boundary visible as the investigation develops.
Why must occurrence and escape causes be investigated separately?
The process that created a defect and the control that failed to detect it are different mechanisms. A corrective action that addresses only one can leave a path for recurrence or continued escape.
Consider a hypothetical wrong-label incident. One occurrence hypothesis is that a print job reused the preceding order’s part number. An escape hypothesis is that the check verified only the product family, not the complete ordering code. Neither becomes a confirmed cause until the workflow and records support it.
A proposed correction might bind the label job to the controlled order record, while a separate verification compares the complete printed code with that record. Challenge the revised process using a deliberately mismatched input in an appropriate controlled test. The test should show whether the correction blocks the defined failure, not merely that the normal workflow still prints a label.
Avoid closing the cause statement at “human error.” It does not explain why the process allowed the error or why the detection step missed it. Document the specific conditions, evidence and competing explanations that were considered.
When is there enough evidence to close a SCAR?
Close it when containment is resolved, the relevant causes are supported, permanent actions are implemented and the predefined effectiveness criteria are met. The observation window must include meaningful opportunities for the failure to recur.
Define that window before reviewing the result: affected part families, shifts or workflows, the number of relevant transactions or lots, the checks performed and the failure criterion. A quiet calendar month with no comparable orders is weak evidence of effectiveness.
Record any remaining limits. Observing no recurrence in the defined cohort does not guarantee zero future defects, but it can support a bounded closure decision. Assign the owner and trigger for reopening the action if the same mechanism reappears.
Where the finding involves a device failure rather than a documentation discrepancy, connect the SCAR to the component failure-analysis evidence package. The laboratory result may support the cause investigation; it does not by itself verify that the supplier’s revised process will prevent another occurrence.
Frequently Asked Questions (FAQ)
Is operator retraining enough to close a supplier corrective action?
Only when verified evidence shows that the action addresses the relevant cause and remains effective. A training attendance record alone does not demonstrate that the failure mechanism or escape path has been removed.
Must every delivery discrepancy use a full 8D report?
No universal requirement applies. Use the contractual process and a response proportionate to the impact, recurrence and uncertainty. Preserve a clear problem statement, action ownership and closure evidence even for a simpler response.
Should a SCAR be closed when replacement material arrives?
No. Replacement material can resolve an immediate supply need while the cause remains unresolved. Keep material recovery, containment and corrective-action closure as separate recorded decisions.
