A supplier’s report says a casting is acceptable, while receiving inspection reports a different result. The fastest useful response is to preserve the part and compare the evidence. Repeatedly measuring until one side obtains a favorable number rarely resolves the underlying issue.

The investigation should establish whether both parties evaluated the same characteristic, under the same revision and a compatible method. It should also separate the immediate lot disposition from the longer-term corrective action.

Confirm identity and revision first

Match the part, drawing revision and inspection report. Check whether the supplier’s result relates to the actual delivered component or to a different sample. A traceability mismatch can create an apparent dimensional dispute before measurement technique is even considered.

The acceptance requirement should be the controlled contractual one. A receiving inspector should not silently apply a later drawing revision, and a supplier should not rely on an obsolete tolerance. Resolve the document baseline before debating the readings.

Compare the definitions behind the numbers

Two methods may use different datum references, fitting algorithms or contact locations. They can therefore report different values for a nonideal feature. The investigation should compare the measurand and setup rather than only the final decimal number.

This side-by-side review often reveals a concrete difference that can be tested directly.

Preserve the component during the investigation

Avoid rework, forced assembly or aggressive cleaning that could change the disputed feature before the comparison is complete. If the part must be handled or moved, document its condition. A changed component cannot provide a clean comparison with the original inspection result.

Where transport damage is possible, retain packaging evidence as well. A damaged edge or shifted component may explain a localized difference, but that conclusion should be supported by observations rather than assumed to be the carrier’s fault.

Use a controlled joint measurement plan

Agree how repeat measurements will be performed and how the result will be interpreted. A reference laboratory can help when needed, but its task must be defined. Sending the part away with the instruction to tell us who is right is less useful than specifying the characteristic and decision rule.

Consider measurement uncertainty where relevant, especially near a tolerance limit. Preserve all valid results and explain any excluded measurement. The final conclusion should be understandable to a reviewer who was not present during the discussion.

Separate containment from corrective action

The customer may need an immediate decision about using, sorting or holding the lot. That decision can proceed under the approved quality process while the root cause is still being investigated. Do not present a temporary concession as a permanent change to the drawing.

For TripleAce Castings orders, a clear discrepancy report should include the part identity, drawing characteristic, measured values, method and photographs where useful. Once the cause is established, update the relevant inspection or manufacturing instruction and verify the change on appropriate evidence.

Measurement and conformity resources

NIST: Measurement Process Characterization supports controlled comparison of measurements. NIST: Conformity Assessment Basics explains the relationship between requirements and evidence. The dispute process should follow the parties’ actual contract and quality procedures.