Researchers have developed a novel digital twin framework for precision livestock farming, focusing on predicting core body temperature in dairy cattle to detect heat stress. This framework integrates physics-based simulations with machine learning models, including a Gaussian process, Kalman filter, and a Markov chain, to model thermoregulation, individual deviations, real-time data alignment, and activity states. The system uses a three-stage stacked ensemble of LightGBM models, achieving a cross-validated R2 of 0.783 and a 92.38% prediction interval coverage probability for 2-hour ahead forecasts. AI
IMPACT This framework could improve animal welfare and farm management by enabling early detection of heat stress in livestock.
RANK_REASON The cluster contains a research paper detailing a new methodology and framework for a specific application. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
