Collar-based prediction of individual dry matter intake in dairy cows: validation across international research farms and evidence of familial variation in commercial herds

This project develops and tests a collar‑based sensor system that predicts each dairy cow’s dry matter intake (DMI). Measuring individual DMI is a major bottleneck for evaluating feed efficiency (FE) on commercial farms because traditional methods require a lot of labor and don’t scale well. The new system continuously records behavior with wearable collars and builds a model that links activity, rumination, and actual feed consumption. We validate the model using historic data from research farms around the world that have automated electronic feeding systems and manual tie‑stall measurements. The project shows how to set up the data‑processing pipeline, turn accelerometer and rumination logs into useful features, and apply regression

modeling to estimate daily DMI. The estimated DMI is then used to calculate each cow’s energy‑corrected milk (ECM) to feed‑efficiency ratio. This work supports precision dairy tools that are scalable and non‑invasive, offering an alternative to physical feeding gates. The framework helps students learn about sensor integration, predictive analytics, and livestock biometrics in modern dairy technology.

Message: Hi FE, I need help with “Collar-based prediction of individual dry matter intake in dairy cows: validation across international research farms and evidence of familial variation in commercial herds” in “Food Technology”