Digital transformation of veterinary control as a factor in reducing the energy intensity of milk production

The paper substantiates the introduction of robotic contactless cattle health monitoring to enhance milk production energy efficiency. Research methods included mathematical modeling of energy losses and thermogram analysis. An automated system architecture based on infrared thermography and convolutional neural networks integrated into herd management is proposed. It was found that subclinical mastitis, typically detected on days 5–7, increases specific energy intensity due to lower yields with constant costs. Transitioning to digital monitoring enables early diagnosis (days 2–3) via temperature gradients (ΔT > 1.0°C). This supports a «management by deviation» strategy, preventing milk rejection and keeping energy efficiency within normal ranges by minimizing resource waste.