it

Innovation for animal welfare

November 13, 2025

Reading time: 3 minutes

If we asked people today to imagine Artificial Intelligence, many would picture invisible algorithms, neural networks, and data in the cloud. Others might think of supercomputers or autonomous robots that move and make decisions on their own.
But what if we asked them to imagine AI in a dairy farm? It’s probably not the first scenario that comes to mind. Yet, even in livestock farming, AI can make a tangible difference, supporting farmers in their daily decisions.

 

Because animal welfare can be measured and improved and that’s exactly what we’re working on.

What our system does in practice

Our framework is designed to simplify daily farm operations by automating the observation and analysis of key animal welfare parameters. The collected data is processed in real time, providing farmers with objective indicators and timely alerts.

 

Automatic, markerless identification
Each animal is uniquely recognized through the natural patterns of its coat—without the need for ear tags or RFID devices.

Body Condition Score (BCS) monitoring
By automatically analyzing images, the system estimates the physical condition of each animal. This allows farmers to intervene promptly, preventing metabolic or reproductive issues.

Locomotion Score (LS) analysis
Through posture estimation and motion analysis models, the system evaluates the animal’s gait, detecting early signs of lameness or pain. Timely action helps reduce veterinary costs while improving animal well-being.

Behavioral analysis
By continuously monitoring video footage, the platform measures the time animals spend eating, resting, and interacting. These data help identify stress conditions, abnormal behaviors, or episodes of aggression.

Animal welfare in dairy cattle

We have focused specifically on the application of AI in dairy farming, automating key analyses such as weight estimation, Body Condition Score, and Locomotion Score monitoring, crucial parameters especially in the pre and post-calving periods. Abnormal variations in weight or body condition are often the first indicators of metabolic disorders.

 

Using AI systems based on image analysis from cameras installed in pens, calving areas, and corridors allows for continuous assessment of animal conditions, generating automatic alerts that enable farmers to intervene precisely and promptly.
At the same time, behavioral analysis evaluates parameters such as feeding and drinking time, standing vs. lying time, posture, and gait symmetry.
Together, these insights create a dynamic, objective picture of animal welfare, translating raw data into actionable decisions.

Why this is a “concrete” innovation

AI for livestock farming isn’t (at least for now) about science fiction or robots.
It’s about cameras, local servers, meters of Ethernet cable, software, and data that every day help farmers reduce waste, improve herd health, and optimize production.
Each algorithm is a building block contributing to a more sustainable livestock industry, where animal welfare and productivity are not in conflict, but reinforce one another.