{
  "schema": "vedokrok.public-item.v1",
  "release_id": "MHC-RPUB-20260920-75ad787a",
  "url": "/knowledge/run-failure-mode-analysis-before-the-consequential-change",
  "id": "MHC-D-RESEARCH-0732",
  "version": "0.1.0",
  "title": "Run failure-mode analysis before the consequential change",
  "summary": "Ask how the process can fail before production teaches you.",
  "kind": "protocol",
  "body": "Walk each important step and list plausible failure modes, their effects, existing controls and detectability. Use rough severity/likelihood/detectability ratings only to prioritize attention. Spend the review on weak controls and hard-to-detect high-impact modes, not on debating a precise risk score.",
  "limits": [
    "FMEA prioritization numbers are not calibrated probabilities. Use them to structure attention, not simulate certainty."
  ],
  "topics": [
    "union-learning-from-decisions-and-errors"
  ],
  "intents": [],
  "source_ids": [
    "RS-D8547C3B32436378"
  ],
  "evidence": [
    {
      "claim": "FMEA is a prospective method for identifying possible failure modes and considering their effects before or during process improvement.",
      "source_id": "RS-D8547C3B32436378",
      "role": "supports",
      "note": "Priority scores are heuristics; they should not be mistaken for calibrated failure probabilities.",
      "locator": "Description and uses"
    }
  ],
  "use_when": [
    "You are about to change a process, migration, automation or workflow where one missed failure mode could have a large blast radius."
  ],
  "avoid_when": [
    "FMEA prioritization numbers are not calibrated probabilities. Use them to structure attention, not simulate certainty."
  ],
  "example": "Bulk update: wrong selection filter → thousands of unintended records → current control: manual review → detection: after activation → safeguard: pre-run count and sampled IDs.",
  "check": "At least one high-impact or hard-to-detect failure mode has a stronger prevention, detection or recovery control.",
  "steps": [
    "At least one high-impact or hard-to-detect failure mode has a stronger prevention, detection or recovery control."
  ],
  "sources": [
    {
      "id": "RS-D8547C3B32436378",
      "title": "Failure Mode and Effects Analysis",
      "url": "https://digital.ahrq.gov/health-it-tools-and-resources/evaluation-resources/workflow-assessment-health-it-toolkit/all-workflow-tools/fmea-analysis"
    }
  ],
  "relations": [
    {
      "from": "MHC-D-RESEARCH-0732",
      "to": "MHC-D-RESEARCH-0733",
      "type": "useful_with",
      "url": "/knowledge/prefer-a-system-strengthening-action-over-another-reminder"
    }
  ],
  "collections": [
    {
      "id": "RC-D475A7021FD981F0",
      "title": "Turn mistakes and outcomes into changes you can verify",
      "url": "/collections/turn-mistakes-and-outcomes-into-changes-you-can-verify"
    }
  ]
}