Technical reference
asks for where possible, so that a deviation is less likely to recur. It sets no required method. What matters is that the cause explains the event, the action addresses the cause, and a later check shows the failure has stopped.
Why failures repeat: the fix addressed the symptom; the cause list makes 'other' the easy answer; retraining is the default action; each repeat is logged as a new minor event, so no one sees a pattern; or no one ever checks that the action worked.
Repeats are also evidence about the plan. If a is hard to meet day after day, then the equipment, capacity or method may not fit the operation. The and the may need another look, as well as the people.
Parameters
| When to do a full analysis | Your call, by consequence. Serious events, failures and repeats are the usual triggers |
|---|---|
| Method | None prescribed by . Choose one that fits, from asking why repeatedly to a full cause map |
| Who | Someone who can read the evidence, plus the people who do the work. Bring in outside help when needed |
| A cause is good enough when | It explains the reading, it explains why good cases stayed good, and changing it would prevent the failure |
| Effectiveness check | A date and a measure set when the action is written |
| Trend review | Periodic in §19.10. Set your own interval, and read repeats by unit, product and shift |
Records that prove it
- One-sentence problem statement
- Evidence read: records, observation, conversations
- The chain of reasons to the root cause
- Effect on the scope of the hold
- Action, owner and date
- Effectiveness check and result
- Change made to plan, method or training
CCP or prerequisite?
That depends on the for your process. The number, if it has one, belongs to your plan.