linedown-warroom

Diagnose production line stoppages and price restart paths across MES, CMMS, and ERP systems.

Updated Aug 17, 2026
One-click install
npx skills add https://github.com/MetaFloor-AI/metafloor-scm-plugin-openai --skill linedown-warroom-metafloor-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: linedown-warroom
Source: https://github.com/MetaFloor-AI/metafloor-scm-plugin-openai/tree/main/skills/business-workflows/linedown-warroom
Command: npx skills add https://github.com/MetaFloor-AI/metafloor-scm-plugin-openai --skill linedown-warroom-metafloor-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When a critical production line trips at 02:00, every minute of delay burns scrap, idle labor, and missed customer commitments. This Skill turns a raw MES downtime code into a priced, ranked restart decision—repair in place, reroute to an alternate line, or hold and absorb—while gating every committing write behind human approval. ## Core Features & Use Cases - Constraint Diagnosis: Maps downtime codes (e.g., E-45 hydraulic fault) to constraint classes and verifies the true cause with a second independent read before spending on any lever. - Restart Path Pricing: Computes expected total cost per path using burn rate, this asset's actual CMMS repair-time spread, changeover cost, and commitment-miss exposure, then ranks options. - War-Room Orchestration: Opens an incident bridge, fans tasks out to maintenance, materials, and planning in parallel, and fires committing writes (MES resequence, SAP reschedule, CMMS work order, material move) in dependency order with check-and-abort guards. - Use Case: A stamping press trips on the night shift with three orders due in 36 hours. The Skill confirms the fault, prices repair vs. reroute vs. hold ($101K vs. $70K vs. $80K), recommends the reroute, and holds the customer notification until the plant manager approves a confirmed path. ## Quick Start Ask the agent to diagnose a line-down event from an MES downtime code and recommend the lowest-cost restart path with approval gates on all committing actions.

Frequently Asked Questions about linedown-warroom

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I respond to a production line down event?

Start by classifying the MES downtime code into a constraint class (equipment fault, component starve, quality hold) and confirm it with a second independent read. Then price each restart path—repair in place, reroute, or hold—by total cost including burn rate and missed commitments, and present the ranked options for plant manager approval.

How do I decide between repairing a line versus rerouting production?

Reroute pays when the changeover cost is less than the burn rate multiplied by the repair-minus-changeover time plus the missed commitments it avoids. Price both paths using the asset's actual CMMS repair-time spread rather than the book mean, since a defined changeover cost can buy down a fat repair-overrun tail.

Which systems does this workflow integrate with?

It reads line status from MES systems like Siemens Opcenter or Rockwell FactoryTalk, repair history from CMMS like IBM Maximo or SAP PM, orders and inventory from SAP MM, and alternate-line qualification from Kinaxis. Committing writes go back to MES, SAP, and CMMS in dependency order after approval.

Why should repair estimates not use the CMMS book mean?

The book mean is generic and often optimistic for a specific asset. Build the spread from the asset's own recent repair actuals for that code, set the overrun probability from how often past repairs exceeded the median, and flag low-confidence estimates when fewer than five actuals exist.

When is a customer notification sent during a line-down event?

The customer notification is treated as irreversible egress and fires only after the restart path is committed and explicitly approved by the plant manager. Sending a premature heads-up on an unconfirmed path can trigger the very escalation the response is trying to prevent.

What happens if a committing write fails mid-recovery?

The MES resequence is the anchor; if a downstream write like the SAP reschedule fails, halt the sequence, flag the mismatch on the incident bridge, and reconcile MES and SAP before notifying the customer. Never leave production moved with commitments unrescheduled.