How do nutritionists turn model outputs into better rations on the farm? Recorded live at the American Dairy Science Association (ADSA) 2026 Annual Meeting, this episode explores how nutritional models for dairy cattle are developed, applied and refined to support dairy ration formulation. Researchers and consulting nutritionists cover the history of the Cornell Net Carbohydrate and Protein System (CNCPS). They also compare mechanistic and empirical models and explain how to pair model outputs with on-farm observation, sensor data and experience.
Dr. Andrew Lapierre shares the nearly 50-year history of CNCPS, and Dr. Jonas de Souza previews his work on fatty acid modeling. Dr. Veridiana Daley reviews milk protein modeling, amino acid nutrition and new technologies that could sharpen decision-making. (00:00 – 10:45)
The consulting nutritionists explain how they use models' day to day. They discuss how ration software has evolved and why model outputs still need practical farm management and observation behind them. (10:46 – 15:12)
The panel compares empirical and mechanistic models. Models are the best interpretation of available data, not perfect pictures of biology. That's why it's critical to understand each model's assumptions before acting on its recommendations. (15:13 – 19:25)
NASEM and CNCPS often reach different answers because they rely on different assumptions, microbial compositions and prediction methods. Neither is perfect, but both are useful for evaluating amino acid supply and nutrient responses. (19:26 – 24:08)
Genetic progress has raised production potential, yet many post-absorptive efficiency assumptions still come from older datasets. The panel names nutrient use efficiency and variable animal responses as key areas for future research. (24:09 – 29:46)
Nutritionists use models to guide gradual ration changes within real-world constraints. They also point out that management decisions can affect performance as much as the diet itself. (29:47 – 33:32)
Rumination, rumen health and other precision dairy sensor data could improve intake predictions, flag health events and support nutritional decisions. (33:33 – 37:43)
The panel looks at dry matter intake optimization and forage quality challenges. They also ask whether maximizing intake is always the most profitable strategy. (37:44 – 43:29)
The group sees room to expand response modeling and improve feed libraries. They also want to better characterize ingredients and add amino acid and fatty acid data, all while validating models against research and field datasets. (43:30 – 58:41)
Looking ahead, the panel discusses better fatty acid and amino acid prediction systems, sensor integration, artificial intelligence and stochastic modeling. They also call for stronger collaboration between researchers and field nutritionists, with better data and ongoing validation driving progress. (58:42 – End)
Please subscribe and share with your industry friends. Invite more people to join us at the Real Science Exchange virtual pub table. In addition, please be sure to register for our upcoming Real Science Lecture Series webinars.
Finally, if you want one of our Real Science Exchange t-shirts, screenshot your rating, review, or subscription, and email a picture to anh.marketing@balchem.com. Include your size and mailing address, and we’ll mail you a shirt.