Artificial Intelligence Prediction of Heat Stress and Welfare in Dairy Cattle

Abstract

Introduction: Heat stress was becoming a major issue in dairy production systems because of rising environmental temperatures, which were harming the health, welfare, behavior, and milk production in cattle. The conventional methods used to detect heat stress involved manual observation and physiological measurement of parameters in isolation, making it difficult to continuously monitor and detect the stress. This study had developed a prediction model using artificial intelligence to assess heat stress and welfare of dairy cattle considering various parameters such as environment, physiology, behavior, and production. Data had been preprocessed and optimized to find important stress parameters, and machine learning algorithms had been applied for classification of heat stress level and welfare status. Temperature, humidity, temperature humidity index (THI), body temperature, respiratory rate, activity, feeding behavior, and milk yield variation had been identified as important factors for prediction of heat stress using AI-based model. The system had a better ability to monitor continuously compared to traditional methods of evaluation and offered important information on precision dairy farming. It was concluded from this study that an artificial intelligence-based prediction system can improve sustainability in livestock production through heat stress mitigation, monitoring animal welfare and maximizing productivity in varying climates