About
Software and consulting for a clear view of production.
We bring together knowledge of industrial processes and AI-driven software for textile, metal, plastics and other processing companies — and we start by understanding how the plant actually works.
Our origin
Born from a gap we found on the factory floor.
Talking with different industries, we found that something simple to say and hard to do was still missing: a linear, objective view of production — what is being made, where, and what is holding it back.
That is where FabVitals MES came from, designed from the start with artificial intelligence at its core: planning, recording at the workstation and indicators read from the real data of the plant, not from sheets filled in at the end of the day.
Our consultants know the processes of different industries — textile, metal, plastics and other processing. That is why every project starts on the shop floor, not with the software.
We are now choosing the first plants to start FabVitals with us. If yours could be one of them, let us talk. More news soon.
What we do
Understand the flow, write the process, then deliver the software that fits it.
Our purpose is not to sell you a licence. It is to understand your workflow well enough to describe it accurately, and then deliver — together with our software — something that makes the plant measurably better.
We map the workflow
Not the documented one. We follow real orders through the plant and record how information, materials and decisions actually move.
We write the process
The map becomes a written process your team owns. That document is useful on its own, whatever you decide to do next.
We deliver the software
Where the diagnosis calls for a system, we build and integrate it — and we support it afterwards, because we wrote it.
We measure the result
A change that cannot be seen in the numbers is a change nobody will defend six months later.
Where we focus
Industries whose processes we know.
Metal industry
Cutting, forming, machining and assembly, where work orders split and recombine and work in progress is genuinely hard to track.
Textile — apparel and home
Multi-stage processing with subcontracted operations, where a batch leaves the building several times before it ships — and where product traceability is increasingly a regulatory requirement, not a nice-to-have.
Plastics and other processing industries
Transformation operations where yield, scrap and rework decide the margin, and where the data that would explain them is usually recorded too late to act on.
Our consultants know these processes from the inside. That is why we ask the questions we ask — and why we are sceptical of answers that arrive before the questions.
How we think
Four positions we hold, including the inconvenient ones.
Technology is a tool, never the objective
The objective is a better process and better decisions. Software, IoT and artificial intelligence are means. If a changed routine solves the problem, that is the correct answer and we will say so.
You cannot digitalise what you do not understand
A process mapped from a procedure manual describes what should happen. We map what does happen, which is usually different and always more useful.
Most AI projects fail on data, not models
The modelling is rarely the hard part. Consistent, granular, trustworthy records of what happened on the floor are the hard part — and most factories were never built to produce them.
The best result is sometimes no project
We would rather tell you the foundations are missing than sell a system that will not survive contact with your floor. Worse for this quarter, better for the relationship.
How we use AI ourselves
As an analysis instrument, not as the product.
We use AI-assisted methods in our own diagnostic work — to find patterns across production data, stoppage records and process descriptions faster than we could by hand.
This matters for an honest reason: it lets a small team analyse a plant thoroughly in less time, which is what makes a proper diagnosis affordable for a small or medium company. It is a tool that improves our work, in the same way a measuring instrument improves an engineer's work.
It is not a claim that an algorithm decides what your factory should do. Every conclusion we present is one we can explain, defend, and trace back to something we observed on your floor. If we cannot explain why, we do not put it in the report.
Engineering + Industry + Technology
Three disciplines that have to sit at the same table.
Each one alone produces a predictable kind of failure. Holding all three is the whole point of how we work.
Engineering
Rigour about cause and effect. Measuring before concluding, and being specific about what a change is expected to do. Without this, improvement becomes opinion.
Industry
Respect for how a factory actually runs — shift patterns, changeovers, urgent orders, informal knowledge, and the reasons a workaround exists. Without this, a system gets designed for a plant that does not exist.
Technology
Knowing what is genuinely buildable, integrable and maintainable — and what only demos well. Without this, expectations get set that no implementation can meet.
Show us how your factory works.
Let us look at your processes, identify where value is being lost, and determine where technology can actually help.