Calf Data: Using robotic milk feeders to identify calf training success
Research Poster Health & Life Sciences 2025 Graduate ExhibitionPresentation by Breanna Bone
Exhibition Number 131
Abstract
Robotic milk feeders (RMFs) reduce labor costs in calf rearing but require intensive labor to train calves. The study objective evaluated whether RMF training success was associated with unstable feeding behavior patterns, including milk intake, drinking speed, and rewarded visits. Calves from a single calf raiser were offered near ad libitum milk, with training success defined as consuming >6 L of milk and visiting the feeder >2 times within four days of being on the RMF (n = 463; successful = 149, unsuccessful = 314). Day 4 was set as the baseline for relative changes in behavior. Mixed linear regression models analyzed the day × success interaction using averages from collected variables and relative change, adjusting for day, success status, and pen, with calf as a random effect repeated by day. A significant milk intake × day interaction was observed from day -2 to 10, and day 13 with successful calves having a higher milk intake (P < 0.01). There was no association between training status and drinking speed or rewarded visits (P > 0.62). A significant relative change × day interaction was observed for all variables. On days -2 to 1, successful calves had a greater relative increase in milk intake (P < 0.04). On day -2, successful calves had a greater relative increase in drinking speed (P = 0.0001). By days 10-13, unsuccessful calves had more relative change in rewarded visits (P < 0.01). Calf training success may be identifiable as early as day 2 on the RMF.
Importance
Robotic milk feeders (RMFs) are increasingly used in calf-rearing systems, yet training calves to use them efficiently remains a challenge. Identifying behavioral indicators of training success could optimize calf management leading to improved welfare and performance. This study demonstrates that milk intake and relative change in milk intake, drinking speed, and rewarded visits can predict whether a calf will successfully train on an RMF by day 4 on the feeder, allowing for early intervention. Using RMF data, producers can implement management strategies to support slower-learning calves. These findings contribute to precision livestock farming by enhancing calf management, ultimately promoting better growth and health outcomes in dairy and beef production systems.