Problem Management
Find it: Management → Problem Management and Problem-Asset Mapping (Operations group).
Sound familiar?
- "We've fixed this exact failure five times — why is it still a surprise?"
- "When this symptom shows up, what's usually the cause?"
- "Tribal knowledge leaves when the senior tech retires."
Every repeat failure is a problem you've already solved — if you wrote it down.
The Problem
The same failure gets diagnosed from scratch every time. Known causes live in heads, not the system. New technicians can't lean on past findings, and symptom-to-cause reasoning is reinvented per ticket.
What OpexMX Problem Management Does
Builds a library of known problems, groups them, maps them to assets, and links symptoms to likely causes — with AI that proposes new entries by reading the manuals.
Known Problems, Classified
Each problem has a type (BREAKDOWN, MALFUNCTION, QUALITY, SAFETY), a group, a priority, and a status — so the catalog is structured, not a dumping ground.
Symptom → Cause Links
Link symptoms to problems with RCA weights (calculated and expert likelihood). When a symptom reappears, the likely cause is already ranked — diagnosis stops being guesswork.
Asset Mapping & Sub-Actions
Map a problem to the assets it affects, with sub-actions for the fix. The known resolution travels with the asset.
AI Capture from Manuals
Describe a symptom; the AI embeds it, searches the asset's manual chunks, and proposes a structured problem — {name, type, description, actions, confidence} — as a preview. Promote it in one transactional write (problem + symptom links + sub-actions).
How It Works
- Capture problems as you solve them (or let AI draft from manuals).
- Group and classify them.
- Link symptoms to likely causes with weights.
- Map to assets with fix sub-actions.
- Next time the symptom appears, the cause and fix are already known.
Who Uses This?
| Role | What They Care About |
|---|---|
| Reliability Engineers | A growing, searchable known-problem library |
| Technicians | The likely cause, fast |
| Planners | Repeat-failure patterns to feed PM/RCM |
| New hires | Hard-won knowledge, written down |
Pro Tips
- Capture at resolution time — it's cheapest then.
- Let AI draft, then have a senior validate before promoting.
- Weight symptoms by experience; revisit the weights over time.
- Feed repeat problems into RCM and PM reviews.
Integrations
- Tickets: symptoms on tickets surface known causes
- Assets: problems map to the equipment they affect
- Knowledge Center: AI capture reads the asset's manuals (RAG)
- RCM / Reliability: repeat problems drive analysis