The pandemic highlighted fragilities in food systems and other supply chains. Additional factors, like increased weather variability, drought, or even cyber-attacks can also disrupt supply chains. It is critical to understand the vulnerabilities and identify innovative interventions to maintain food system resiliency in the face of the likelihood of increasing rates of disruption. This is especially true for the U.S. animal agriculture, upon which most Americans rely upon as a major nutritious food source and which also supports the livelihoods of millions of ranchers, farmers, and many others working in agriculture
In response to this challenge, researchers at AgNext are leading a three-year, United States Department of Agriculture and The National Institute of Food and Agriculture (USDA/NIFA) funded study to use cutting edge data science and machine learning approaches to develop new models to identify and evaluate interventions to mitigate disruptions and help to ensure a resilient, affordable, and sustainable supply of these important animal-sourced foods.
This project is a multi-institutional collaborative effort and includes expertise and leadership from:
-
-
- Greg Thoma, Project Lead, AgNext, Colorado State University
-
-
-
- John Sheehan, Soil and Crop Sciences, Colorado State University
-
-
-
- Jennifer Schmitt, University of Minnesota
-
-
-
- David Lobell, Stanford University
-

The team’s long-term goal is to develop an innovative data-based modeling framework for evaluation of interventions intended to mitigate tradeoffs between system resilience and sustainability. Extensive stakeholder engagement is part of this project so researchers can understand and ensure that the final modeling framework considers complexities across the food supply chain.
The project will also advance the field of data science in animal agriculture by virtue of its integrated application of remote sensing, data science, and machine learning to characterize many components of the system.
As a first step in the stakeholder-engagement process, the team hosted a workshop during the 2023 Sustainable Agriculture Summit in Charlotte, North Carolina (Figure 1), entitled “Managing Shocks in the Animal Protein System.” Attendees shared feedback on how the team should best prioritize the model’s potential capabilities and help guide development efforts to those topics with the most opportunity. The team also sought input on the types of regional, national, and global shocks of greatest potential interest: climate/weather-related), animal-disease outbreaks, human-disease and transport and supply chain disruption.
To help drive the conversation, the team presented a prototype system dynamics model exploring tradeoffs in system resilience and sustainability, inviting attendees to directly engage with the tool. This prototype (Figure 2) is now available to the public at the following link.

Workshop attendees proposed a wide range of potential shocks to the system, some of which the team had not previously considered. Some of the shocks suggested included: seed availability, import/export bans, accidents at meat packing plants, labor/immigration-related policy impacts, disruption in supply of key inputs port strikes, and policy changes. One attendee also suggested that the team consider the impact of so-called “negative shocks” or disruptive technologies that boost productivity or eliminate bottlenecks in the supply chain.
The stakeholders also suggested that it would be important for the new models to have a number of key features: decision points related to actions that can build system resilience; including the effect of other dynamics in the supply chain not included in the original simplified model, such as stocker-level supply/demand and auction barn-level dynamics. Attendees also suggested that the model should be expanded to consider more than just price driven decision making.
Additional feedback on the overall modeling approach included recommendations on the most useful geographic scale and appropriate reporting metrics.
The scientific team plans to host a second in-person stakeholder-engagement event at the November 2024 Sustainable Agriculture Summit in Minneapolis.
This research is part of a 3-year, USDA/NIFA-funded project, “Stakeholder-engaged modeling, data science, and machine learning for more resilient and sustainable animal protein food systems,” (NIFA Award #: 2023-67021-39144).