Research harnesses AI decision modeling to improve livestock sustainability
AgriLife Research’s Karun Kaniyamattam targets agriculture industry challenges
The difference between a profitable season and a devastating loss for a livestock operation often hinges on a producer’s ability to predict the future.
Karun Kaniyamattam, Ph.D., a Texas A&M AgriLife Research artificial intelligence, AI, modeling scientist, Bryan-College Station, aims to provide the industry with a “digital crystal ball” to help livestock producers make smarter, faster decisions about animal health, production efficiency and profits.
Since joining the agency in 2023 as an assistant professor in the Texas A&M Department of Animal Science, Kaniyamattam has emerged as a leading voice in livestock data analytics and AI modeling.
Key research initiatives and the future
The tools Kaniyamattam and his team have created position Texas A&M AgriLife as a leader in research for AI-driven sustainability in livestock production.
His research is designed to address three of the most urgent challenges facing livestock production: producer profitability, reducing antimicrobial use and quantifying environmental impact.
“Livestock systems generate enormous amounts of data, but the challenge is turning that information into practical insights,” Kaniyamattam said. “AI allows us to analyze relationships among animal health and environmental and economic factors across livestock systems in ways that were not previously possible.”
They have developed 16 web-based tools so far, including BovineTwin, a real-time 3D tool that creates a “digital twin” to visualize and predict individual animal emissions versus productivity tradeoffs.
Kaniyamattam’s research impact is underscored by the scope of grants and projects he leads or has spearheaded, including:
- Data-driven efforts in bovine respiratory disease.
- Using system dynamics to evaluate sustainable livestock management practices and diversify income for beef producers.
- Developing AI agent-based models to simulate beef supply chains and optimize resources.
- Advancing early disease diagnosis through computation and improved livestock productivity.
- Forecasting tools for Texas market profitability.
- Qualitative biological and economic modeling to strengthen U.S. herd resilience.
“The possibilities for AI applications in livestock sustainability are vast,” Kaniyamattam said. “The only limitation to the ways we can develop this technology is that it must address the challenges most important to producers.”
Leading expertise in AI for livestock production
A major focus of Kaniyamattam’s research is bovine respiratory disease, the leading cause of morbidity and antimicrobial use in the U.S. beef industry. The disease costs U.S. and global beef cattle systems more than $3 billion annually.
Conventional management relies on subjective visual diagnosis that is often delayed. But Kaniyamattam’s team is developing a digital solution — “predictive disease intelligence.”
By integrating computer vision and sensor-derived behavioral data, his computer models can identify at-risk animals before they show obvious clinical symptoms. This precision allows for more judicious use of antibiotics, supporting animal and environmental welfare, and saving producers money in Texas and beyond.
Internationally, Kaniyamattam’s expertise in AI modeling for livestock production has brought him to lead a cross-sectoral team from the U.S. and South Africa. The team focuses on machine learning and decision modeling in beef production. The project aims to build an open-source machine learning model that producers worldwide can use to evaluate livestock management strategies.
The next generation and AI: It starts in the classroom
Kaniyamattam’s expertise also extends into the classroom, where he trains the next generation of research leaders. He has mentored over 120 students in AI decision-modeling activities between 2024 and 2026.
He designed a specialized course, Decision Science for Sustainable Livestock Systems. It aims to fill a critical curricular gap in using “low-code” learning, and systems thinking-backed, team-based solutions for real-world challenges.
“We want to develop an ecosystem where Aggies become AI-driven solutions architects,” he said.
AI will be a critical part of future livestock sustainability, Kaniyamattam said, and it is important that students are prepared to advance the industry.
“As livestock systems face increasing pressure from environmental and economic uncertainty, our work offers a proactive path forward,” he said. “By translating complex biological data into actionable intelligence, we want to ensure sustainable and profitable livestock production for generations to come.”
