AI-Optimized Controlled Environment Agriculture

Controlled environment agriculture — indoor farming systems that regulate temperature, light, water, and nutrients to grow food year-round — offers a promising pathway for food system resilience in regions with short growing seasons and precarious supply chains, including much of Wyoming. But CEA has a significant sustainability liability: its energy intensity. Without careful management, the electricity demands of indoor farming can undermine the very environmental goals it is meant to serve.

This project addresses that tension directly, using optimization modeling to manage how CEA systems consume energy. The core technical challenge is synchronizing the internal environment of a CEA facility — with its crop growth requirements, temperature thresholds, and lighting schedules — with an external grid that varies continuously in both cost and carbon intensity. The approach identifies sequences of operational decisions (when to run heaters, adjust lighting, cycle water systems) that minimize electricity cost and carbon footprint across a full growing season while maintaining crop yield.

The work is conducted at the University of Wyoming, developing and validating the optimization framework against a detailed CEA facility model and historical energy grid data. The spatial dimension of the work extends this analysis across the Mountain West — including the Jackson Hole, Denver, and Salt Lake City regions — coupling the optimized CEA model with scenarios for future renewable energy expansion to identify where low-carbon, low-cost indoor agriculture is most viable as the grid continues to decarbonize.

The long-term vision is a publicly available optimization platform that CEA operators, planners, and policymakers can use to evaluate the energy and climate compatibility of indoor agriculture investments — and to identify the places and grid conditions where CEA can function not as an energy burden but as a flexible, productive load that supports grid resilience.