Association of AI-Predicted Milk Yield Residuals to Behavioral Patterns and Transition Success in Multiparous Dairy Cows
The transition period in dairy cows, spanning six weeks around calving, is a critical phase in which cows are more susceptible to infectious and metabolic diseases, impacting general health and productivity. An artificial intelligence (AI) model was used to predict the expected milk yields, to be subtracted from the actual yields to obtain the milk yield residuals. We found that cows with more negative milk yield residuals exhibit a behavioral pattern associated with failing transition. This outcome strengthens the potential application of the AI-predicted milk yield residuals as a transition indicator, paving the way for its implementation as an automated health monitoring tool.