Asset downtime risk and predictive maintenance action

Elite LLM Handoff

Explain the reduce_asset_downtime recommendation using the governed Fortress output and caveats.

governed_handoff:
  - Goal: reduce_asset_downtime
  - Primary path: downtime_prevented, score 0.564, high confidence
  - Supporting signals: planned_intervention, spare_part_shortage, technician_gap, temperature_rise, vibration_threshold_breach
  - Runtime relationship sets returned: 8
  - Domain journeys: 120; events: 520; relationship sets: 120

Outcome Prompt JSON

{
  "schema": "kynticai.elite.outcome-prompts.v1",
  "caseStudyId": "manufacturing-field-asset-risk",
  "domain": "Manufacturing / field operations",
  "stakeholder": "Reliability lead",
  "runtimeQuery": "A production asset has a vibration threshold breach, rising temperature, spare-part shortage, and technician availability gap. What maintenance action should reliability take next to prevent downtime?",
  "promptContract": "You are Elite, the KynticAI recommendation-writing layer.\n\nYou receive governed Fortress output, not raw customer data. Use only the supplied relationship-set facts, scores, confidence labels, alternatives, caveats, and outcome goal.\n\nWrite a decisive operational recommendation for the accountable team. Your output must include:\n1. Recommended action.\n2. Why this action is supported by the comparable relationship sets.\n3. What to do now.\n4. When to choose one of the alternative actions.\n5. What outcome to capture so the system improves.\n\nRules:\n- Do not invent facts outside the governed handoff.\n- Do not promise revenue, retention, clinical, legal, financial, safety, or compliance outcomes.\n- Do not apologise for using synthetic demo data.\n- Be commercially useful, specific, and direct.\n- Keep human review and claim boundaries visible where the domain requires them.",
  "governedFacts": [
    "Goal: reduce_asset_downtime",
    "Primary path: downtime_prevented, score 0.564, high confidence",
    "Supporting signals: planned_intervention, spare_part_shortage, technician_gap, temperature_rise, vibration_threshold_breach",
    "Runtime relationship sets returned: 8",
    "Domain journeys: 120; events: 520; relationship sets: 120"
  ],
  "outcomePrompts": [
    {
      "rank": 1,
      "outcome": "downtime_prevented",
      "action": "Reserve the critical spare, schedule a planned intervention window, and dispatch the qualified technician before automatic shutdown.",
      "confidence": "High",
      "score": "0.564",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Asset downtime risk and predictive maintenance action\nDomain: Manufacturing / field operations\nStakeholder: Reliability lead\nRuntime query: A production asset has a vibration threshold breach, rising temperature, spare-part shortage, and technician availability gap. What maintenance action should reliability take next to prevent downtime?\n\nGoverned facts:\n- Goal: reduce_asset_downtime\n- Primary path: downtime_prevented, score 0.564, high confidence\n- Supporting signals: planned_intervention, spare_part_shortage, technician_gap, temperature_rise, vibration_threshold_breach\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 520; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 1\n- Outcome: downtime_prevented\n- Action: Reserve the critical spare, schedule a planned intervention window, and dispatch the qualified technician before automatic shutdown.\n- Confidence: High\n- Score: 0.564\n- Evidence summary: Comparable relationship sets ended in downtime_prevented and share signals such as planned_intervention, spare_part_shortage, technician_gap, temperature_rise, vibration_threshold_breach.\n- Supporting relationship sets: set-b9eebc3b76ab, set-30d87d158f4c, set-b7fa01947f62, set-dbd7e7183031, set-35bfbd1cb6b2\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Recommended action: Reserve the critical spare, schedule a planned intervention window, and dispatch the qualified technician before automatic shutdown.\n\nWhy this is supported: Fortress ranked downtime_prevented first with score 0.564 and high confidence. Comparable relationship sets ended in downtime_prevented and share signals such as planned_intervention, spare_part_shortage, technician_gap, temperature_rise, vibration_threshold_breach. The supporting relationship sets are set-b9eebc3b76ab, set-30d87d158f4c, set-b7fa01947f62, set-dbd7e7183031, set-35bfbd1cb6b2.\n\nWhat to do now: brief the reliability lead with the runtime query, the shared signals (planned_intervention, spare_part_shortage, technician_gap, temperature_rise, vibration_threshold_breach), and the exact action. Keep the action specific to this scenario rather than turning it into a generic campaign or workflow rule.\n\nWhen to choose an alternative: switch if the live evidence is closer to spare_part_reserved (Reserve the critical spare and expedite replenishment before assigning the work order.); technician_dispatched (Dispatch the correctly certified technician with the diagnostic checklist before closing the alarm.).\n\nOutcome to capture: record the action taken, the reviewed outcome, and any contradictory signal so the next Scout/Fortress comparison has stronger evidence."
    },
    {
      "rank": 2,
      "outcome": "spare_part_reserved",
      "action": "Reserve the critical spare and expedite replenishment before assigning the work order.",
      "confidence": "Low",
      "score": "0.1721",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Asset downtime risk and predictive maintenance action\nDomain: Manufacturing / field operations\nStakeholder: Reliability lead\nRuntime query: A production asset has a vibration threshold breach, rising temperature, spare-part shortage, and technician availability gap. What maintenance action should reliability take next to prevent downtime?\n\nGoverned facts:\n- Goal: reduce_asset_downtime\n- Primary path: downtime_prevented, score 0.564, high confidence\n- Supporting signals: planned_intervention, spare_part_shortage, technician_gap, temperature_rise, vibration_threshold_breach\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 520; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 2\n- Outcome: spare_part_reserved\n- Action: Reserve the critical spare and expedite replenishment before assigning the work order.\n- Confidence: Low\n- Score: 0.1721\n- Evidence summary: Comparable relationship sets ended in spare_part_reserved and share signals such as purchase_order, replenishment_expedited, spare_reserved, stock_low.\n- Supporting relationship sets: set-f7ca03201aa7\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Alternative action: Reserve the critical spare and expedite replenishment before assigning the work order.\n\nWhy this route exists: Fortress returned spare_part_reserved at rank 2 with score 0.1721 and low confidence. Comparable relationship sets ended in spare_part_reserved and share signals such as purchase_order, replenishment_expedited, spare_reserved, stock_low. The supporting relationship sets are set-f7ca03201aa7.\n\nWhen to use it: choose this instead of the primary route when the live evidence matches purchase_order, replenishment_expedited, spare_reserved, stock_low more closely than the current top-ranked outcome.\n\nWhat to do now: make the action explicit for the reliability lead, preserve the caveat for this domain, and avoid inventing extra facts beyond the governed handoff.\n\nOutcome to capture: record whether spare_part_reserved happened and feed that reviewed result back into the relationship-set evidence."
    },
    {
      "rank": 3,
      "outcome": "technician_dispatched",
      "action": "Dispatch the correctly certified technician with the diagnostic checklist before closing the alarm.",
      "confidence": "Low",
      "score": "0.1557",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Asset downtime risk and predictive maintenance action\nDomain: Manufacturing / field operations\nStakeholder: Reliability lead\nRuntime query: A production asset has a vibration threshold breach, rising temperature, spare-part shortage, and technician availability gap. What maintenance action should reliability take next to prevent downtime?\n\nGoverned facts:\n- Goal: reduce_asset_downtime\n- Primary path: downtime_prevented, score 0.564, high confidence\n- Supporting signals: planned_intervention, spare_part_shortage, technician_gap, temperature_rise, vibration_threshold_breach\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 520; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 3\n- Outcome: technician_dispatched\n- Action: Dispatch the correctly certified technician with the diagnostic checklist before closing the alarm.\n- Confidence: Low\n- Score: 0.1557\n- Evidence summary: Comparable relationship sets ended in technician_dispatched and share signals such as certified_technician, diagnostic_checklist, dispatch_completed.\n- Supporting relationship sets: set-baf64826ed77\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Alternative action: Dispatch the correctly certified technician with the diagnostic checklist before closing the alarm.\n\nWhy this route exists: Fortress returned technician_dispatched at rank 3 with score 0.1557 and low confidence. Comparable relationship sets ended in technician_dispatched and share signals such as certified_technician, diagnostic_checklist, dispatch_completed. The supporting relationship sets are set-baf64826ed77.\n\nWhen to use it: choose this instead of the primary route when the live evidence matches certified_technician, diagnostic_checklist, dispatch_completed more closely than the current top-ranked outcome.\n\nWhat to do now: make the action explicit for the reliability lead, preserve the caveat for this domain, and avoid inventing extra facts beyond the governed handoff.\n\nOutcome to capture: record whether technician_dispatched happened and feed that reviewed result back into the relationship-set evidence."
    }
  ]
}