The Role of Mechanistic Models in Precision Dairy Nutrition and On-Farm Decision Support

Models have a strong history of application in dairy production, where their objectives have revolved around optimally feeding and growing livestock. These predominantly nutritional (and typically quasi-mechanistic) models have evolved to mathematically express our cumulative biological knowledge, developed in order to understand and manipulate nutrient dynamics in the animal. In the field, these models serve as ‘decision support tools’ and are often linked to optimization algorithms to determine optimal ration composition. Academically, mechanistic models (MM) also serve to extract meaningful information from data and increase our biological understanding of complex systems. Current limitations of this modelling approach, in terms of application on-farm, revolve around their manual nature and lengthy input requirements. Meanwhile, the emergence of big data and associated machine learning (ML) approaches are taking the world (and some agriculture sectors) by storm, propelled by their quick development time and precision on-farm. No doubt they will have a major role to play in the future of on-farm decision support systems for dairy production. However, their slow adoption rate in animal production thus far may be due to the current degree of digitalization, the utility offered, return-on-investment (ROI), and/or the challenge of maintaining sensitive technology in corrosive, dusty and dirty environments. More than likely, the future of on-farm decision support will be a hybridization of these two approaches (MM and ML), utilizing the respective strengths and augmenting weaknesses of each approach. This talk will examine the use of MMs on-farm and in precision dairy, and contemplate future directions of precision dairy.

This abstract will be presented at the 2021 Animal Nutrition Conference of Canada. Check out all the lectures and speakers here.

More information in https://animalnutritionconference.ca/.