Transforming dairy farms with data

Animal Science
Statistics and Data Science
Digital Agriculture
Dairy

Martin Perez is helping to develop an ecosystem of networked technologies and management techniques for the 21st century farm.

Amidst farm fields and dairy barns, an agricultural revolution is underway — with algorithms, sensors, and automation changing the face of farming. Martin Perez '25, a doctoral student in animal science at Cornell University, works at the forefront of this change, using the power of data to transform the way dairy farms operate. His research focuses on developing and testing automated monitoring systems for dairy cows, and aims to improve animal well-being, farm efficiency, and the sustainability of the dairy industry.

Perez works in the lab of Dr. Julio Giordano, associate professor of animal science. One of his early projects evaluated commercially available sensor systems that tracked rumination patterns and activity levels in dairy cows. Could these automated monitoring systems effectively detect illnesses like mastitis and metabolic issues, allowing for earlier treatment?

In a randomized clinical trial, one group of cows received intensive manual checks by Perez and his colleagues, while another group was fitted with automated sensors that the researchers then relied on for health alerts. "We found that cow performance didn't differ between the groups," Perez said. "These sensor systems can be used to track cow health without negatively impacting the cows, which validates their potential for health management."

But Perez’s vision extends beyond that initial study. His current research pushes boundaries by combining multiple sensor streams—rumination, activity, body temperature, feeding behavior and more—from a mix of wearable and non-wearable sensors combined with existing data on the cows. Then, using machine learning, he unifies this flood of data into integrated predictive models for cow health and management.

"We used machine learning algorithms to analyze this data and identify patterns that might indicate potential health problems," Perez explained. "We achieved success by combining automated machine learning with more established methods. Our expertise in animal health also played a crucial role.

“This approach allowed us to develop algorithms that perform well in real-world settings, and we're currently testing them on commercial farms,” he added.

schematic showing how data from cow monitors is processed — Perez works with farm data harnessed from cow monitors and sensors, encompassing everything from cows' behavior and physiology to productivity metrics.

Perez’s work is one component of an ambitious project headed by Giordano called the Cornell Agricultural Systems Testbed and Demonstration Site (CAST) for the Farm of the Future. CAST advances data-driven solutions for climate-smart agriculture through developing, testing and demonstrating technologies in real-world farm settings.

"CAST lets us implement new technologies in commercial dairy, crop and livestock operations," said Perez. "We can then rigorously analyze the impacts and share those insights with farmers to facilitate adoption of beneficial innovations."

CAST is made up of three farms in New York state: the Cornell University Ruminant Center in Harford, the Musgrave Research Farm in Aurora and the Cornell Teaching Dairy Barn in Ithaca. Perez and his colleagues tackle agriculture’s digital transformation from multiple angles: cow sensors and predictive analytics; environmental monitoring; and autonomous farm machinery. Data integration and traceability spans the entire food supply chain.

two men stand in a cow barn, looking at sensor data on a phone — Julio Giordano (left) and Martin Perez discuss data from a cow sensor. Photo by Allison Usavage.

Perez sees digital agriculture and data integration as crucial to feeding the world's growing population in a sustainable way. In late 2016, he came to Cornell as an intern in Giordano’s lab to work on the problem. In 2018, he officially began his PhD program under Giordano.

For Perez, CAST represents an opportunity to merge multiple disciplines: animal science, data science, engineering, agronomy and more. "We're bridging that gap between cutting-edge programming skills and deep agricultural knowledge," he said. "It's the only way to develop technologies that truly understand the complexity of farming systems."

As CAST continues growing, Perez hopes to spearhead trials assessing whether early sensor alerts could guide preventative treatments — before cows show clinical symptoms of disease. "We get alerts sometimes and can't find anything obviously wrong with the cow," he explained. "But her rumination, activity and other signals indicate an issue. Can therapies at that point, based solely on sensor data, truly prevent or mitigate illnesses? That's a crucial next frontier."

"We're bridging that gap between cutting-edge programming skills and deep agricultural knowledge. It's the only way to develop technologies that truly understand the complexity of farming systems."

CAST embodies the future of agriculture, Perez said. At the same time, the program's core philosophy centers on a multidisciplinary approach, merging agricultural knowledge and the latest innovations for a better food system.

"We need to keep developing technologies to help farmers, improve cow welfare, and reduce environmental impacts," Perez said. "CAST gives us a unique testbed to pursue those goals through reproducible science and hard data.”

Caroline Stamm ’24 is an animal science major and student writer for the Cornell CALS Department of Animal Science.

Transforming dairy farms with data

Keep Exploring

We openly share valuable knowledge.