Sample review · anonymised · no model call on this page
What an 8D Gate REJECT looks like
This is a stored result from a real 8D Gate run on a fictional spot-weld 8D (company, part and people names replaced). The report found the cause of undersize weld nuggets, but its permanent actions only add inspection and training, and it never explains which control should have caught the escape.
Why it was rejected, in short
- Corrective actions treat the symptom (re-inspect, retrain, extra peel tests), not the uncontrolled weld-schedule change.
- The escape cause is “inspection frequency not followed”, a compliance lapse with no named failed control.
- D6 has valid in-plant peel results, but no results with the extra 100% peel checks removed, so it does not show the fix works on its own.
Decision
REJECT
Score 48/100 · REVISE THEN RESUBMIT
Executive summary
Reject. The root cause method establishes occurrence (uncontrolled weld-schedule change), but the escape cause, an inspection-frequency lapse, names no failed control. Occurrence actions are incomplete: no change control on the weld schedule restore and no dual-approval fix; the permanent actions address only symptoms. D6 does not show the process works without the extra inspection. The escape analysis is not closed.
Rationale: Overall 48, capped at 54 by ca_unrelated_to_rc (permanent actions are extra inspection and retraining); the escape gap adds critical_evidence_missing and D6 adds effectiveness_not_demonstrated (both cap 59). The D4 method chain is sound for occurrence, but the escape cause, "inspection frequency not followed", is a compliance lapse: no control (sampling plan, MSA, poka-yoke) designed to catch undersize nuggets is named. The 40 in-plant peel coupons (5.6–6.2 mm) on the restored schedule are valid evidence, but the effectiveness claim rests on the extra inspection itself and there are no results after containment is removed. The schedule restore has no recorded approval. Gate caps: unrelated CA 54, escape evidence 59, effectiveness 59 → final cap 54; score 48 → REJECT.
Gate failures
- Escape point not identified · critical_evidence_missing · cap 59 · CONDITIONAL
'Inspection frequency not followed' names a lapse, not the failed control. Name the sampling plan / check that should have caught undersize nuggets and why it did not.
- Permanent CA Symptom-Only · ca_unrelated_to_rc · cap 54 · REJECT
CA1–CA4: visual re-inspect, retraining, extra peel. No fix to uncontrolled schedule change or absent password/dual-approval control.
- D6 effectiveness not demonstrated · effectiveness_not_demonstrated · cap 59 · CONDITIONAL
The effectiveness claim is 'extra inspection will prevent recurrence'. The in-plant peel coupons are valid, but there are no results on the restored schedule with the extra 100% peel checks removed.
How to read this: each gate sets a maximum score (its cap). The lowest cap wins, here 54, and the overall score (48) sits under it, so the result is REJECT. Caps are not section scores: D4 can be rated ACCEPTABLE for its occurrence analysis while its escape gap still triggers a gate.
Section ratings
- D1 · GOOD
- D2 · GOOD
- D3 · ACCEPTABLE
- D4 · ACCEPTABLE
- D5 · WEAK
- D6 · WEAK
- D7 · WEAK
- D8 · CRITICAL
Why D8 is CRITICAL: D8 itself carries no weight in the score. It is flagged because the report asks to close the 8D while the engineering change, dual-approval fix, FMEA update, schedule-restore approval and customer sort are still open. Closing with open actions is the problem, not the team recognition.
Critical gaps
- Escape evidence missing: no named inspection control (sampling plan, MSA, limit, poka-yoke) designed to detect undersize. Frequency lapse is process compliance; root cause of the inspection failure must be addressed.
- Occurrence CA incomplete: the weld schedule change has no engineering change (ECN) or change control. The HMI restore (9.2 kA on 2026-02-05) has no record of who approved it.
- Permanent CA symptom-only: visual re-inspect, retraining, extra peel do not fix uncontrolled parameter change or absent dual-approval. Schedule lock and password policy TBD.
Fix list
- Identify the inspection control (sampling plan, MSA, SPC) that should have detected undersize nuggets. Audit 2026-01-12 weekly peel skip and provide data showing why it failed. — closes Escape evidence missing. Gate: critical_evidence_missing.
- Issue an engineering change (ECN) to lock weld schedule WS-J3-05 under change control; add a dual-approval step for HMI parameter edits; document who approved the 2026-02-05 schedule restore. — closes Occurrence CA incomplete. Prevent future uncontrolled changes.
- Prove the fix without containment: run 30 pcs on the restored weld schedule with 100% peel checks removed and report nugget diameters, or state that containment stays until this is done. — closes D6 effectiveness not demonstrated. Gate: effectiveness_not_demonstrated.
Advisory only — not a customer or auditor decision.
Example
From this sample review, 8D Gate drafts a reply an SQE could send back to the supplier. Copy it and edit it in your email before sending. Nothing is sent or stored.
Draft reply to the supplier
Drafted in your browser from this review. Nothing is sent or stored. Copy it, then edit it in your email before you send.
Get the free 8D / D4 checklist + updates