The global food system will need to feed 10 billion people by 2050, but currently, more than 10% of American households are food insecure, and global food insecurity has increased by over 30% because of the COVID-19 pandemic (Coleman-Jensen et al., 2020; Baquedano et al., 2021) Rapid population growth, shrinking farmland, dwindling natural resources, erratic climate, and shifting market demand will push the global agricultural production system into a new paradigm (FAO, 2017). This development implies that the food system must become more productive in output, efficient in operation, resilient to climate change, and sustainable for future generations (Vaio et al., 2020).

Artificial intelligence (AI) holds promise in addressing the challenges of this new paradigm by advancing agricultural technologies, improving supply chain management, and generating new knowledge regarding the functioning, interactions, and consequences of the U.S. food system (Liu, 2020). Yet, most AI innovations in agriculture are limited to production data collection and analysis, with some post-farmgate applications targeting product traceability and quality monitoring (Kakani et al., 2020; Misra et al., 2020; Paul et al., 2021). This constraint indicates a clear need for new research that builds on AI and big data beyond the farmgate to answer research questions of national scope related to marketing (including supply chain logistics), food retailing, and consumer behavior.

Interdisciplinary approaches have been recommended for topics including health and agriculture (Waage et al., 2019), food security (Acevedo, 2011; Foran et al., 2014), rural economy and land use (Lowe and Phillipson, 2006), and agri-food science (Horton et al., 2017). The complexity of the social, economic, and environmental interactions in the food system requires problems to be addressed through an interdisciplinary lens (Duffy et al., 1997; Foran et al., 2014). A consequence of not taking a systems approach, and instead pursuing siloed research projects, is that it leads to an incomplete framing of the problem and the development of inadequate research priorities (Lowe and Phillipson, 2006).

Big data, such as the location and description of points of interest (POIs), including grocery stores, restaurants, health facilities and centers, and indoor and outdoor physical activity locations, provides an exciting opportunity to understand and measure the nuances of a ‘healthy’ neighborhood and access to related POIs (Gibson et al., 2014; Wilkins et al., 2019; Shannon et al., 2021). This potential of big data and related data science techniques extends the current work on ‘neighborhood and health’ as neighborhoods can now be measured with rich and detailed multidimensional data. This research will benefit a variety of stakeholders, including policymakers, consumers, and agribusinesses.