Making predictive maintenance usable for your business

Many manufacturers started with predictive maintenance in recent years. Sensors on critical machines, a Sensorfact or Festo dashboard, a subscription per machine. The technology works, the alerts come, and yet little changes on the shop floor.

The problem is not the sensors or the models. It sits in between. Between the alert and the action. Between the dashboard and the ERP. Between the maintenance team that sees it and the planner who has to schedule it.

• The PdM dashboard sits on a separate tab nobody opens after the first months
• Alerts arrive via email or app but disappear among the regular messages
• Work orders are still created manually based on the alert
• Time tracking and cost accounting of a PdM alert is booked separately
• Nobody knows anymore what an alert cost or delivered
• After a breakdown it is often found that there was a warning, but it was not followed up

This is where our added value sits. Not in more sensors or better models. In the working software flow that turns an alert into an executed action.

Connect to your ERP and CMMS

An alert from Sensorfact, Festo or ifm automatically becomes a work order in your existing maintenance system (Exact, AFAS, Ultimo, Planon or whatever you use). No manual handover, no double entry. The technician gets the alert where he already works.

AI agents that interpret and route alerts

Not every alert is equally urgent. An AI agent reads the alert, assesses the context (which machine, how critical, what is the history), and decides what should happen. Urgent to the maintenance team. Routine to planning. Unimportant to the log. Works alongside your existing PdM tool, reads the alerts via API or email.

Dashboards that actually get used

A dashboard that fits how your company works. PdM alerts together with production KPIs, planning, cost accounting. Not a separate tab nobody visits, but a work screen shared by the shop floor and the office. On tablet, on PC, on a large screen in the workshop.

Make cost accounting and return visible

What did an alert cost? What did it prevent? How many maintenance hours against how many hours of production loss? We build the reporting your PdM system does not deliver, and link it to your time tracking and invoicing. Only then do you see whether the predictive maintenance investment pays off.

An honest story: the AI in our solution is not in the sensor analysis. Your PdM supplier already does that well. Our AI is in the workflow around it.

AI agent for alert triage

An agent that reads, classifies and routes every alert to the right person. Based on the machine, the type of alert, the planning and the history, it decides whether this is an emergency, a planned maintenance moment, or something that can wait until next quarter.

Document AI for maintenance history

Service tickets, technician reports, historical failures. An AI that reads these documents and makes them searchable. When an alert now comes in for bearing X on pump Y, the system knows that the same pump had three similar alerts last year.

AI assistant for the maintenance team

A chat interface where technicians can ask questions of your own knowledge base. "What was the procedure for bearing Y last time?" "Which parts do I need for this alert?" Works on your own documentation and history.

No sensor analysis, no FFT on vibration data. We leave that to the specialists. But AI where we are strong: the workflow, the documentation, the integration.

Patrick, our founder, worked eighteen years in industrial roles, much of it in maintenance, wear and rotating equipment. At Stork Gears & Services, at Lagersmit on sea seals, at Castolin Eutectic on wear problems. Clients such as Shell, Wärtsilä and IHC called on him for maintenance strategies. He knows what a maintenance team needs from an alert, because he stood on the other side of that alert for years.

In recent years we build AI systems ourselves around maintenance administration and document processing for heavy equipment. Working software, in production at clients, not a proof of concept. The combination of shop-floor engineering and software building is rare in the Netherlands, and crucial for PdM integration: you get someone who understands the technician, the planner and the system.

We start with a conversation about your existing situation. Which PdM system do you have, how do alerts come in, what happens to them now, where is the pain? Then we scope a defined topic. We build it on your own infrastructure. One-off build cost, optional maintenance contract for further development. You own the solution.

Defined integration project (e.g. Sensorfact to ERP with automatic work orders): from around 8,000 euro. Full PdM software package (dashboards, AI agents, document AI, cost reporting): from 25,000 euro. Our rates and approach are transparent on the approach page.

View the demo →