Planning: SCC-PG DFARM: A Smart, Connected and Distributed Community Hydroponics System

Everyone has a right to affordable and nutritious food. Unfortunately, over 40 million Americans faced food insecurity in 2022. In Knoxville, Tennessee, the community where this project is located, a 2023 targeted survey of 771 residents of economically disadvantaged and diverse communities funded by the United Way of Greater Knoxville (UWGK) found that 74% of respondents met the U.S. Department of Agriculture’s definition for very low or low food security. The purpose of this project is to pilot and evaluate a distributed system of hydroponics to empower food insecure households to produce their own food year-round to increase food security. The project is entitled DFARM, short for Distributed Farming. DFARM will have three smart and connected components. One will allocate food production targets to each participating household and other participating sites in the DFARM system. The second component of DFARM will support the distribution of harvested food to participants. The third component of DFARM will focus on day-to-day management of the hydroponic systems. This planning grant will support an authentic and intensive community engagement effort. Among the many letters of collaboration submitted to support this proposal are those from the Knox County Health Department, Knox Pride, Rooted East, Women with Vision, and the Knoxville Food Council. These stakeholders and others will be asked to consider numerous important questions about DFARM, including: how should Knoxville’s food community organize to administer DFARM; and what type of decision support does the community wish to have to allocate harvested food? DFARM provides significant social and technical research challenges. Major social science research questions include: how should a collaborative hydroponic food production system be structured so that collaborators’ preferences and goals of equity and reduction of food insecurity are satisfactorily co-addressed; and can the collaborative approach for DFARM yield synergistic benefits that are sufficiently greater than a simple collection of individual hydroponic systems? Important computer science research questions include: what algorithm or model should be used to allocate food production targets given an objective function that may have multiple goals subject to a complex set of constraints, including constraints that may frequently change over short periods of time (e.g., the number and capacity of DFARM’s hydroponic systems, households’ food needs and preferences); and what approach should be used to build an AI assistant to support the food distribution decision process? In answering these questions, this project will pioneer advanced AI-driven solutions, including constraint optimization and machine learning techniques, for real-time management and optimization of hydroponic systems, setting a new standard for urban agriculture. The project's technological advancements will be closely interwoven with social impact goals, ensuring that the innovations alleviate food insecurity and enhance community well-being. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.