ISO 9001 LA study materials
ISO 9001 LA study resources

Cascade Fabrications: an exclusion that does not hold

ISO 9001 LA · Put your knowledge to work in context.

All case studies

Cascade Fabrications: an exclusion that does not hold

Cascade Fabrications supplies welded frames. Its quality manual states clause 8.3 (design and development) is excluded because "design is provided by the customer". In practice, for 6 of its 11 customers, Cascade's engineering cell converts customer drawings into weld programs, selects filler material, and specifies dimensional tolerances that do not appear on the customer drawing; for two jobs it proposed an alternative steel grade that the customer accepted. Separately, the auditor finds 4 of 25 gauges used on a live customer job carry calibration labels expired by 5 to 9 months, against a 12-month interval in the calibration procedure. The quality manager describes both issues as "documentation only".

1. Is the exclusion of clause 8.3 justified? Set out your reasoning and the evidence you need before writing a finding.

Read the worked answer

Test. Clause 8.3 may be excluded only where the requirements cannot be applied (4.3), and the justification must be maintained.

Analysis. Converting a customer drawing into a weld program is not automatically design. It becomes design when the organization adds or determines requirements beyond the customer's: specifying tolerances not on the drawing, selecting the filler material, and especially proposing an alternative steel grade that changes product characteristics and which the customer then accepts. Those are design and development outputs (8.3.3) and design changes (8.3.6), with the customer acting as a reviewer.

Evidence to collect: compare one customer drawing against the released weld program line by line; obtain the correspondence for the two alternative-grade jobs; check whether any verification/validation or review records exist for them.

Likely finding. Nonconformity against 4.3 and 8.3.1 - the exclusion is not justified for those 6 customers. A weaker case exists where the customer approves every output and Cascade adds nothing; then it is process planning under 8.5.1. Be prepared also to raise a second, separate finding that where 8.3 does apply, its records (inputs, review, verification, changes) do not exist - but only after the applicability question is settled.

2. How do you decide whether the expired-calibration condition is a minor or a major nonconformity?

Read the worked answer

Severity follows effect on the system's ability to meet requirements and the risk of nonconforming output - not the auditee's label and not the number of gauges alone.

Questions that decide it:

  1. Was product released (8.6) using measurements taken with the four expired gauges?
  2. Can conformity be re-established - re-measurement with calibrated equipment, or customer concession?
  3. Is this isolated or a system failure? 4 of 25 (16%) expired by up to 9 months on a live job points at the recall/label monitoring process failing, not one person's lapse.

Reasoning to a grade. A documented procedure exists but is not effective across the gauge population: that is a breakdown of a whole control and is likely to recur - major if release decisions depend on those measurements. It is minor if the affected characteristics are independently verified (e.g. CMM check at final) and no product impact can be shown. Write the finding on the condition with dates and gauge IDs, and let severity follow the verification result.

3. What further evidence would you gather on the calibration issue before leaving the site, and why?

Read the worked answer

The finding must show consequence, not just a label date.

  • Full gauge list with IDs, last/next calibration dates, so the population is quantified rather than sampled.
  • Recall: all product measured with the four gauges since expiry, and whether it is still in the warehouse, in transit, or at the customer.
  • Re-verification results using a calibrated instrument - this converts an administrative finding into a product-conformity finding or closes it.
  • Calibration interval rationale (7.1.5.2 a) and whether the register flags overdue items, i.e. whether the process monitored itself.
  • Operator and inspector interviews on how gauge status is checked before use (8.5.1).
  • Customer notification requirements and 8.7 handling for any suspect product.
  • Earlier audits: did the internal audit programme ever cover calibration (9.2), and was this an input to management review (9.3.2)?

This evidence also protects you if the organization contests the severity grade at the closing meeting.