Image recognition & process monitoring
In most production and logistics environments, cameras are already in place, but the footage is rarely used to learn anything from. We turn that into a system: one platform that watches along on the cameras you already have — and where there are none yet, adding them is rarely the bottleneck. Not a device on the line that checks one thing, but one system that can handle dozens of different checks at once.
This lets you safeguard your quality system in the broadest sense — from product inspection to hygiene, progress and preventive maintenance — and get new employees up to speed faster. Where you now see afterwards that something went wrong, you will soon see why: the same day, with evidence that stands up to an audit.
The partner behind this pillar
VAIBS from Harderwijk makes existing cameras smart. The company is a European Platinum Partner of the Vaidio Video AI Analytics platform and adds its own, self-trained concepts to it. Vaidio is camera-independent and connects to virtually any ONVIF IP camera and existing VMS or NVR systems; all it needs is a video stream — no extra hardware, no replacement of your camera network. The platform now offers more than forty production and logistics analytics — from product inspection to behaviour and safety. It watches along and looks back, but does not replace your video management system: it does not control cameras. What VAIBS can do, and what is virtually unique in this market: demonstrate per AI model where the training data comes from, whether it was obtained with the client’s consent and how the model was trained — including that it contains no bias and does not profile on ethnicity.
Use cases in practice
The platform is set up to operate within the GDPR and the EU AI Act. VAIBS supplies the platform; Supply Today takes the lead — from quality-system requirement (IFS/HACCP/GMP) and MES integration to a working operation, step by step through a staged approach.
01
Quality assurance in production and warehouse (IFS / HACCP / GMP)
“Our quality requirements are becoming ever more stringent, but visual checks are snapshots and hard to sustain with staff turnover. A seal that doesn’t close, a wrong label, a forgotten hairnet, or an order that ends up with the wrong customer — we often only see it afterwards, and the cause is anecdotal.”
Read how we solve this →
02
Accelerated onboarding: new employees up to speed faster
“Turnover in our production and logistics is high and our people come from very different backgrounds and languages. Getting new employees up to standard takes weeks, and in those weeks we see it in productivity and in quality.”
Read how we solve this →
03
Making waste and raw-material overuse visible
“At the end of the week we see that a line uses more raw material than the standard, but not where it leaks away. Product on the floor by the slicer, overfilled bags, wrong mixing ratios — the figures show that it happens, not where and when.”
Read how we solve this →
04
Efficiency: explaining downtime, disruptions and slowdowns (OEE)
“MES shows that a line stops or runs slowly, but not exactly why. Wrongly stacked crates, pallets in the way, an operator bypassing a fault, manual rework slowing the line — that stays out of sight, so we steer on assumptions.”
Read how we solve this →
Does my IT department have to free up time for this first? Not for a pilot: it can run fully isolated, outside your company network. For roll-out, IT is needed — access to the camera streams, and a server in the network or a cloud connection. That is a limited, one-off effort, and only comes up once you have seen that it works. A link with MES, WMS or ERP comes later still, only when you want figures per batch, supplier or shift.
Do I need to buy new cameras? Usually not: we start with what is already there. Where coverage is missing, buying more is rarely the bottleneck — a suitable camera costs a fraction of the installation and embedding. The bulk of the investment lies in getting it to work, not in the hardware.
One point of contact: we start with a low threshold (observing, stage 1) and build up in a controlled way to MES integration and predictive AI — you learn at low risk. Privacy is designed in, not bolted on afterwards: the platform is set up to operate within the GDPR and the EU AI Act — with audit trails, logging, blurring and ethically sourced datasets, offline or in a European cloud. Supply Today takes the lead and ensures embedding in the quality and management system; VAIBS supplies the platform. Not the technology but the working outcome is what counts.
