Data science consultancy in Rotterdam

It builds models that extract from images or measurements what you currently check by eye or record by hand. For images, that means classification (what is in the image), detection (where each object is) or segmentation (which pixels belong to it). For sensors, it means identifying anomalies and trends in time series of vibration, sound or temperature.

These models are part of the AI applications we build and then manage.

According to its website, the Floating Farm has been operating in Rotterdam since May 2019 as the world’s first floating farm. The cows’ feed consists largely of by-products from the city, such as spent grain from Rotterdam breweries.

For the Floating Farm, we installed the cameras ourselves and implemented a computer vision model using image classification in Python to make the feeding process easier to monitor and reduce feed waste.

Not automatically. According to the Autoriteit Persoonsgegevens, camera surveillance needs a legal basis, must be necessary, and any analysis functions not needed for that purpose must be switched off. Counting or inspection is a new purpose. If a camera is aimed at employees, the works council must give its prior consent.

At GVB, footage from the vehicles’ security cameras is not viewed or shared; it is only analysed by the model. Your privacy officer or legal adviser must determine whether your proposed use is permitted.

According to Havenbedrijf Rotterdam, drones are already widely used in the port: for surveillance, measuring dry-bulk stockpiles and traffic patterns at terminals, and inspecting cranes and wind turbines. You apply to Havenbedrijf Rotterdam for a flight area.

A model can extract from the footage what an inspector currently has to look for, such as a volume or an anomaly at a fixed point. Drones can easily track people, the AP notes; take that into account when planning the flight route.

The Rotterdam Makers District comprises RDM Rotterdam and Merwe-Vierhavens (M4H) and focuses on new technology based on digitalisation, robotics and additive manufacturing. RDM is being developed by Havenbedrijf Rotterdam, Hogeschool Rotterdam and Techniek College Rotterdam.

For a manufacturing company there, a camera above the production line or a sensor on a machine is the usual starting point. We do not have a case study of our own in the Makers District; the Floating Farm is our Rotterdam example.

Either is possible. On a device near the camera, known as edge computing, the footage stays on site and you send on only counts or alerts. That works well on a quay or in a building with a weak connection, or when a quick response is needed.

In the cloud, you have more computing power and can deploy and monitor models centrally, for example on Databricks or Microsoft Fabric. Often, the solution combines both.

Sometimes. Annex III covers AI used as a safety component in the management and operation of critical digital infrastructure, road traffic, and the supply of water, gas, heating or electricity. According to the European Commission, those rules apply from 2 December 2027. Emotion recognition in the workplace is prohibited, except for medical or safety reasons.

Your legal adviser or compliance team must assess whether your application falls within its scope; we can help you take stock. See AI Act.

When a simple measurement or better record-keeping in your existing system already solves the problem, or when there is no decision you need to improve yet. If you first want to find out where AI could benefit your organisation, start with AI consultant Rotterdam.

If you need an AI application beyond images and sensors that we can build and manage, see AI consultancy Rotterdam.