In the early weeks of the COVID-19 pandemic, as governments issued lockdown orders and stay-at-home advisories, grocery store shelves went bare. Shoppers scooped up flour and yeast, canned goods, frozen vegetables, meats and any other staples that they were concerned could run out before they got the chance to return to the stores.
With few exceptions, there were no real concerns about food shortages in the United States, said Jayson Lusk, distinguished professor and head of Purdue’s Department of Agricultural Economics. However, there was no easy way for consumers or policymakers to know that since data that could be used to calculate the risk to the country’s food supply is spread piecemeal across multiple government agencies like the U.S. Department of Agriculture and the Bureau of Labor Statistics.
This experience motivated Lusk to lead a partnership among Purdue University, the Foundation for Food & Agriculture Research and Microsoft to create open-access online dashboards that can track and report the factors that could lead to food supply disruptions during national and global emergencies.
“The data are not always user-friendly for people unless you’re a specialist and know where to go to find it all,” Lusk said. “There were disruptions for consumers that didn’t have to occur. In the early aftermath of that, I was part of a group brainstorming the types of information that needs to be accessible and what we can do to help the food and agriculture sectors, government and consumers understand what’s going on during an emergency. We believe these dashboards can help.”
FFAR is providing more than $220,000 to Purdue, with Microsoft matching those funds and supplying cloud technology, business intelligence, artificial intelligence and machine learning tools for the project. The total investment in the project is more than $500,000.
“The coronavirus pandemic has highlighted the need for easily accessible data on potential vulnerabilities in the food supply chain,” said Sally Rockey, executive director of FFAR. “This research is exposing those vulnerabilities in real time, providing policymakers and industry with the information needed to prevent bottlenecks and ensure food security.”
While the dashboards are being developed with COVID-19 in mind, they will be adaptable to any other situation that could cause a food supply disruption. The work will build off the Purdue Food and Agriculture Vulnerability Index, which quantifies the potential risk to the supply of agricultural products because of farm and agricultural worker illnesses from COVID-19.
“There are certainly still concerns about COVID, but the idea is to develop multiple dashboards on different topics so that we’re prepared to respond to a variety of issues that might come up during a large-scale emergency,” Lusk said. “The dashboard we have now is based on publicly available data and only tells you so much. With these new dashboards, we’ll integrate machine learning and make extrapolations to better estimate and predict disruptions in the food supply sector.”
One area that has been a concern during COVID-19 is the meat processing sector, where high rates of COVID among workers can disrupt operations. Lusk said dashboards will zoom in on the county level to show the percentage of residents who are sick and use machine learning to estimate the number of workers in a plant that may be affected or could be in the near future. That information can be useful to health officials and government agencies, who could shift resources to those areas or enact rules to counteract the problems.
The work also offers researchers data to identify the impacts of policy decisions during a pandemic or other type of large-scale emergency.
“These dashboards may give us the opportunity to understand how regulations impact food supply during a pandemic,” Lusk said. “That can help inform future decisions so that we can keep food flowing from areas of low demand to high and getting to the people who most need it in difficult times.”