Nutritional management of dairy cattle is of importance to the industry due to its influence on production performance and association with large expenses for producers. Current ration formulation may be improved by predicting feeding recommendations for individual animals, rather than groups of animals, through precision feeding. Automated feeding systems (AFS) designed to deliver individual rations must include response-based models that utilize individual cow production data to make feed recommendations. These models require large data sets of individual cow responses to a variety of nutritional interventions. As a result, an experiment was designed to collect individual response data from 24 Holstein cows fed supplemental top dresses. After analyses, dry matter intake (DMI), milk yield (MY), milk fat yield, milk protein yield, feed efficiency, and activity were significantly affected by top dress (P < 0.001). These results suggest opportunity to use precision feeding to implement economically optimal ration recommendations designed to increase dairy cow production. Therefore, a second experiment was conducted in order to develop and test two algorithms that targeted individualized feeding to increase feed efficiency. Milk protein percentage (P = 0.008) and feed efficiency (P < 0.001) were significantly affected by a 3-way interaction between top dress, algorithm, and week. These results highlight the opportunity for precision feeding to increase the efficiency of individual dairy cows. Although the control group resulted in greater income over feed costs than either of the developed algorithm feeding strategies, algorithm refinement and modification may result in more efficient feeding recommendations that are economically viable.

Nutritional management of cattle is crucial to the dairy industry. The feeding of dairy cattle is the largest expense for producers and directly influences cow production. In particular, precision feeding of dairy cattle may have the ability to lower costs for farmers and increase the productivity of dairy cows. Currently, cattle are fed in group configurations, where cows with similar nutrient requirements are offered the same diet. However, individually feeding dairy cows utilizing precision technologies may have the ability to increase the production performance of cattle. Utilizing precision feeding to individually feed dairy cattle requires automated feeding systems (AFS) designed to decrease the additional labor associated with feeding animals as individuals. However, algorithms designed to predict individual animal nutrient requirements are lacking for use in AFS. As a result, large data sets of individual cow responses to varying diets are necessary to train algorithms designed to predict unique ration formulations for individual animals. Two experiments were developed to collect individual animal production responses that were used to develop two response-based algorithms capable of influencing feed efficiency of individual cows. The results from these experiments highlight the potential for precision feeding of dairy cattle to influence individual animal feed efficiencies and milk production. Future improvements in algorithm development and training are necessary in order for these feeding strategies to be economically worth the investment of AFS on commercial dairy farms.

Master of Science