25th FVI Expert Breakfast

Topic: "Netflix for Maintenance" – How René Menard brings technical knowledge into minds with AI and micro-learning.

In this session, René Menard (Menard Lab) reported together with experts in digital learning from the practice of special machine construction. The problem: onboarding is too expensive and too slow.

• The 120-day pain: René Menard laid the numbers on the table: In complex special machine construction, it takes up to 120 days for a new service technician to be fully operational. This causes costs of about €40,000 per head before the first euro is earned.
• Away from "supervised reading": No one reads 500-page operating manuals ("primers") anymore. The solution lies in micro-learning: AI automatically breaks down complex manuals into small "knowledge nuggets" (3-5 minutes) that can be consumed interactively on a smartphone – similar to language learning apps.
• Skill check instead of certificate belief: Menard Lab has radicalized the application process. Instead of reading resumes, applicants must take a digital skill check. Those who answer 70% of the questions (e.g., on sensors) correctly get the job. This makes the process objective and fast.
• Learning on demand (gap-closing): Those who do not pass the check do not fail, but automatically receive only the learning units they are missing. Instead of sending everyone to an expensive "basic training," only the delta is trained.
• Agile knowledge update: An important feature in the discussion: When technical parameters change (e.g., a new lubricant is prescribed), the AI scans the documents and pushes only this change as an update to the team. The knowledge remains current without having to reprint manuals.

Classification: ADAM turns "training knowledge" into real "application knowledge"

Here we sharply differentiate: While other tools offer "dry runs" (quizzes), ADAM delivers the knowledge directly in the "real case."

• Learning by Doing (ADAM) vs. Learning by App: Separate learning apps have the problem that the knowledge is often forgotten when standing in front of the machine. Our approach: ADAM is the "mentor in the pocket." The technician does not learn in advance but receives the "knowledge nugget" (e.g., the video on filter replacement) exactly at the moment he opens the work order in ADAM. This is just-in-time learning.

• The dynamic qualification matrix: René Menard spoke of control by skills. ADAM operationalizes this. Our approach: ADAM uses the skill profiles of employees for smart dispatching. The system knows: "This order requires 'High Voltage Certificate.' Employee A does not have the certificate, so Employee B gets the order." We actively prevent untrained personnel from accessing critical systems.
• Knowledge transfer without hurdles: Creating learning content is often laborious. Our approach: With ADAM's GenAI functions, the experienced master can simply speak a photo or voice memo into the system history ("Attention, always vent here first!"). ADAM automatically turns this into a work instruction for the next colleague. Thus, every employee becomes a trainer without having to build a "training."

Conclusion: Training is good, assistance is better. ADAM ensures that what is learned does not stay in the head but lands on the machine. We close the gap between "theory" (training) and "practice" (repair).