Field Introduction
Mimorian is an industrial intelligence platform that models equipment, structures fault diagnosis and captures the know-how of maintenance teams through a multi-agent AI architecture. Guided diagnosis, a structured decision tree enriched by every field intervention, is its most concrete application.
It is 2:30 pm. Your compressor's variable speed drive shuts down abruptly. No clear error code. Production halted. Loss: 2k€ per hour.
Scenario 1 (without guided diagnosis): Your technician calls the veteran, who is unavailable. He dismantles the drive "to take a look". Two hours later, he has identified a potentially faulty component. He orders a part. A 48-hour lead time. Total loss: 100k€.
Scenario 2 (with Mimorian guided diagnosis): He opens the "Variable speed drive: complete shutdown" checklist. Twenty minutes later, he has diagnosed a failure of the smoothing capacitor (C1). The part arrives the next day. Loss: 24k€.
The difference? 76k€. And a guided diagnosis.
What is a Mimorian guided diagnosis?
A Mimorian guided diagnosis is a structured decision tree (an intelligent checklist, not a simple tick-box list): symptoms → probable causes (probabilities) → diagnostic tests → remedies.
Every element is documented with its origin: "Cause #1 probability 65%: confirmed on 45 identical drive failures. Tests validated by the manufacturer Schneider."
It is transparent. No black box.
For the full methodological framework (differences with the CMMS attached procedure, FMEA, the 5 evaluation criteria before adoption), read our complete guide to AI-guided diagnosis in industrial maintenance.
A concrete example: Schneider ATV930 drive, complete shutdown
Step 1: Observation and context (5 minutes)
The technician selects "Schneider ATV930 drive" and "Symptom: complete shutdown". Mimorian displays the signs to observe: 400V input voltage, 24V DC internal voltage, fan running. He ticks: 24V LED off, fan inert, machine was running normally.
Step 2: Structured diagnosis (5 minutes)
Mimorian displays the probable causes based on the history of 67 identical failures:
- Cause #1: Internal power supply failure (72%). Tests: measure the voltage at the terminals of capacitor C1
- Cause #2: 24V auxiliary power supply failure (18%). Tests: check the 2A fuse
- Cause #3: Transformer failure (10%). Tests: winding resistance
Step 3: Tests in sequence (10 minutes)
The technician runs the tests in order. Result: C1 voltage = 0V, black mark on the capacitor. Confirmation: 100%.
Step 4: Remedy decided, execution (30 minutes)
Mimorian displays the remedy with the exact part (Schneider 9700-4055, 240€), the intervention time (45 minutes), the safety procedure and the detailed steps. The technician carries it out. The drive restarts.
Step 5: Knowledge capture (5 minutes)
The technician records the field feedback. Mimorian enriches the guided diagnosis for next time.
Why it is faster than trial and error
Before (without guided diagnosis): Calling support (30 minutes on hold), exploring at random (2h), dismantling to take a look (1h), calling support again... Total: 6 to 8 hours of haphazard diagnosis plus 48h of waiting = 56h.
With guided diagnosis: Opening the guided diagnosis (1 minute), symptoms plus context (5 minutes), probable causes (1 minute), tests in order (10 minutes), decision (2 minutes). Total: 50 minutes plus part lead time = 24 to 48h.
Gain: a 90% reduction in diagnosis time. A 70% reduction in downtime. Across a plant with 20 failures a year, that is 400 technician hours saved and 50k€ of downtime avoided.
Continuous enrichment
After a year, you have 50 field feedback entries for the Schneider drive. Your local guided diagnosis has been refined: the probabilities shift, rare cases emerge, local tips are added, lead times become more precise.
After 3 years, you have an embodied collective intelligence. A new technician? They diagnose like a veteran from week 1.
Conclusion: From intuition to process
A Mimorian guided diagnosis does not replace the technician's intuition. It structures it, shares it and enriches it continuously. The result: diagnoses 10x faster, expertise that is never lost, and augmented technicians.
Want to see a guided diagnosis in action? Request a personalised demo on your own equipment.
For the theoretical and comparative framework (attached procedure vs FMEA vs AI guided diagnosis), read our complete guide to AI-guided diagnosis in industrial maintenance. For the know-how capture dimension, see also our complete guide to know-how capture in industrial maintenance.
📚 Sources :
- ISO 17359:2018 : Condition Monitoring and Diagnostics of Machines
- Panopto, 2018 : Workplace Knowledge and Productivity Report
- Gawande, A. (2010) : The Checklist Manifesto. Référence fondatrice sur l'utilisation de check-lists structurées.
- Schneider Electric, 2023 : Documentation technique ATV930.