NC Agricultural Analytics Platform Supports Data-Driven Farming

By Amy Burtch

When Brad Lewis landed at NC State University, he saw something within the N.C. Plant Sciences Initiative that he did not expect to see. 

“I saw different groups working together collaboratively, and it was a huge breath of fresh air,” says Lewis, the program manager for the N.C. Agricultural Analytics Platform. This makes sense as the N.C. Plant Sciences Initiative (N.C. PSI) is one of the university’s leading interdisciplinary research units. 

Agriculture and agribusiness is a $100+ billion industry in the state of North Carolina. Therefore, it’s important to develop solutions for storing, analyzing and maintaining large-scale datasets to support stakeholders and improve agricultural systems in the state. 

Enter the N.C. Ag Analytics Platform. 

With funds appropriated by the North Carolina General Assembly, the platform is a N.C. PSI and N.C. Food Animal Initiative project operated in partnership with North Carolina A&T University and data and AI company, SAS. 

The platform aims to support the management and utilization of data to sustain growth, innovation and data-driven decision-making within the agricultural industry in North Carolina.

Supporting Agricultural Research

Ag Analytics Platform Program Manager Brad Lewis was brought on in the fall of 2024 to look holistically across the platform’s projects and find synergies in the data and outcomes being created by the platform. This position unites his experience in both education and technology, following 20 years at IBM. 

The platform is in its third budget year, with NC State and NC A&T granted $1 million each. It started with simple touchpoints, moved into joint workshops and training sessions, and has since morphed into consultation and steady collaboration. 

NC State faculty are invited to submit project proposals to the platform. Currently, there are seven active projects, with four projects in either the completed or maintenance phase. The platform supports projects using a combination of on-premises and cloud-based analytics. It can leverage the AI and analytics platform, SAS Viya, when appropriate, working closely with SAS experts to accelerate the development of data analytics solutions.

“The Ag Analytics team supports and supplements,” Lewis says. “We come alongside the research teams to help in any way we can.”

Power of Data

Machine learning is one of the benefits of working with SAS Viya, Lewis explains. The software can build predictive models, which help foresee a multitude of agricultural situations. Beyond support from SAS, the platform provides automated data collection from sensors, metadata tracking and deployment of web-enabled dashboards.

Examples of platform projects include:

• BeanPACK is an agronomic soybean decision support tool that assists farmers with planting and harvesting dates. 
• Moth trap sensors, used to track corn earworm moths, were improved using cellphone technology; data is now captured in batches to avoid data loss. 
• Nema-AI is a project in collaboration with the N.C. Department of Agriculture and Consumer Services that will deploy machine-learning models to identify nematodes (worm pests) through a mechanized microscope, improving a previously laborious process.

Impact of Weather

In February 2025, Center for Environmental Farming Systems Director Michelle Schroeder-Moreno initiated her platform research project, “Understanding the Long-Term Impacts of Weather and Sustainable Agricultural Systems.” She wants to answer the question: How do farming practices and climate impact crop yields and soil health? 

“The overall goal is to understand, over the long-term, how farming systems — whether organic, conventional, forestry, crop animal rotation or land released from agriculture — impact crop yields, as well as the soil fertility, structure and biological systems over time,” Schroeder-Moreno says. 

Schroeder-Moreno and the CEFS interdisciplinary research team are using a 25-year dataset from the Cherry Research Station’s Farming Systems Research Unit. As she notes, there are few long-term farming systems research studies in the United States due to their expense and long-term investment. 

Her research team and SAS initially worked together to organize data in a specific way to conduct analysis and then transfer all meaningful notes from the dataset, originally collected on paper. They are now creating crop-specific models and seeking to understand what is biologically meaningful. 

“SAS centered us on how to organize the data, analysis and explanation,” Shroeder-Moreno says. “We would not have been able to do this sophisticated analysis without them.”

Focus on Lagoon Management

Associate Professor of Industrial and Systems Engineering and Operations Research Sara Shashaani and Associate Professor of Biological and Agricultural Engineering and Extension Specialist Mahmoud Sharara have partnered with the Ag Analytics Platform for their project, “Uncertainty Quantification for Robust Irrigation and Lagoon Management in Farms.” The research is part of a larger externally funded project investigating better management practices for hog lagoons. 

With 1,500 storage lagoons in eastern North Carolina, an area boasting the state’s highest pig production, this project is essential to supporting new solutions for a critical economic enterprise for the state. 

Shashaani and Sharara want to create a model that can help predict the nitrogen concentration in the soil of storage lagoons, which can be impacted by weather and lagoon management and irrigation practices. According to Sharara, understanding the nitrogen concentration of the lagoons determines how much manure a farmer can apply and when to apply it. 

“With the assistance of SAS Viya and Brad’s Ag Analytics Platform team, we are using data sets and machine learning tools to figure out how to simulate that concentration over time,” Sharara says. 

SAS has helped by organizing and analyzing the data, and then arriving at an AI-based model, allowing the research team to predict nitrogen concentration at different times of the year. This AI-based model can be integrated into the team’s larger model, enabling them to run “what if” scenarios to analyze farms over extended periods of time. 

The team has also used the AI-based model to see how changes in the data affect its predictions, helping make the model — and its recommendations — more reliable in an uncertain, changing environment.

The ability to update AI-based models in an automated manner with new collections of weather and farm data is an aspiration that the team has pursued in this collaboration with SAS and the Ag Analytics Platform.

“Partnering with SAS has been an interesting and rewarding process of mutual learning because we are teaching one another about the system,” Shashaani says. 

With each project it supports — from improving sustainability and profitability to engaging farmers in best management practices — the N.C. Agricultural Analytics Platform is shaping the future of agriculture.