The winners of this year’s Rabobank-MIT Food and Agribusiness Innovation Prize were a team from Harvard University that’s bringing healthier meal options to low-income communities and an MIT team that’s crowdsourcing and crunching crop-price data for smarter investing.
EatWell, a team of students from Harvard T.H. Chan School of Public Health and outside experts, won the first-place prize of $15,000 for an idea to make and sell simple, low-cost meal kits in “food deserts,” which are urban areas where it’s difficult to buy affordable, nutritious food.
Context Insights — based on an MIT Sloan School of Management class project — won a second-place $10,000 prize for its plan to crowdsource crop-price predictions from smallholder farmers in Africa and aggregate that data for governments and microfinance institutions to increase crop investments that alleviate market volatility.
The five other competing teams were:
- Proventus: Analyzing agronomic data to provide farmers better access to affordable loans and help investors make profitable returns;
- AgroRacks: Constructing greenhouses near massive data center farms that produce waste heat, so the waste heat will warm the greenhouses;
- Cloud Agronomics: Making a drone-based imaging system for farms that precisely maps damage, so farmers know which crops to remove;
- Strella Biotech: Developing sensitive, durable biosensors that can detect when produce is ripe or rotten during the storing and shipping processes; and
- PreserveAir: Building cooling trailers for shipping produce that enable farmers to minimize dehydration losses and improve fruit quality.
The competition was sponsored by Rabobank — one of the largest banks in the world that caters specifically to food and agribusiness clients — and supported by the Abdul Latif Jameel World Water and Food Security Lab (J-WAFS) and the MIT Food and Agriculture Club.
EatWell, a nonprofit, plans to sell meal kits at kiosks in areas with food deserts, such as in the Mattapan neighborhood of Boston — the team’s first market. In Mattapan, 60 percent of the neighborhood’s 6,000 families are considered low-income, public transportation is sparse, and there’s only one grocery store, team member Dan Wexler, a Harvard School of Public Health student, said in the team’s winning pitch. “As a result, a third of residents are food insecure, having limited access to affordable, nutritious food,” he said.
More than 40 million Americans are food insecure, living in 6,500 food deserts nationwide, with significantly higher rates of obesity, diabetes, and chronic heart disease, he added, “and residents need affordable, nutritious meal options if we want health outcomes to improve.”
EatWell first surveys a community’s residents about meal preferences. Then, a professional chef and dietitian create simple, 30-minute, healthy recipes. Based on the Mattapan surveys, for instance, the team created a chicken pot pie pasta dish with loads of vegetables and a Haitian-based pork griot dish.
“We really plan to tailor the recipes to each community we work with. We find that being part of the community is really important and what matters to us. We want to grow with and out of the community,” Wexler told MIT News.
At $15, the kits are less expensive per meal than competitors — such as Blue Apron — because they aren’t shipped and don’t have in-box refrigeration, Wexler said.
The competition’s prize money will help EatWell buy ingredients and build kits for a pilot in Mattapan in August. After that, the team plans to expand to other neighborhoods in Boston, before moving to Providence, Rhode Island, and beyond.
The team also plans to hire residents to run its kiosks, Wexler added. “There’s a large population of seniors in Mattapan. Not all of them are working, and [they] would love to have part-time jobs where they can make a little bit of money,” he said.
Context Insights is very much MIT-grown: The idea spawned from 15.027 (Opportunities in Developing Economies), taught by Tavneet Suri, an associate professor of applied economics who is also scientific director of Abdul Latif Jameel Poverty Action Lab (J-PAL) Africa and the co-chair of J-PAL’s agriculture sector.
In the class, students were tasked with developing a solution for any global market issue. Team members Tarit Rao-Chakravorti and Tim Reed, two MIT Sloan School of Management students, chose the issue of commodity volatility — such as fluctuating corn prices — in Africa.
In developed countries like the United States, data and financial tools exist to forecast crop prices. This helps farmers secure crop investments that ensure prices never dip below a certain point. In areas like sub-Saharan Africa, however, no such data or tools exist, Rao-Chakravorti told MIT News.
Branching off from research Suri had conducted using mobile-based surveys in Africa, the two developed a platform for crowdsourcing crop-price predictions from knowledgeable local farmers. “We use the wisdom of the crowds to forecast the price of basic commodities like maize that are critical to life in sub-Saharan Africa,” Rao-Chakravorti said.
The team shoots text messages to the farmers, asking them to predict the price of corn at any given time. They then treat the farmers like “hedge-fund analysts,” Rao-Chakravorti said, “paying them according to the accuracy of their forecasts.” Payments come in the form of mobile money, but the team hasn’t yet determined the exact payment amounts.
The team crunches the raw data and provides aggregated price predictions to governments to form purchasing policies, as well as to commodity exchange firms. A hope is those firms will use the information to start futures markets, where participants buy and sell commodity contracts for delivery on specific dates, helping crop prices stay steady and competitive. Those markets “are critical for the development of a sustainable financial system,” Rao-Chakravorti said.
The team is currently piloting the platform with the East Africa Exchange, the largest regional commodity exchange in East Africa, which aims to link smallholder farmers to agricultural and financial markets.
The contest’s prize money will be used to provide incentive payments to farmers to start providing predictions. “Our platform relies on predictions from our forecasters, and this money will go directly to them,” Reed told MIT News. “This also helps prove the science, because crowdsourcing is an emerging field.”
Rao-Chakravorti and Reed are currently in a class with Thomas Malone, the Patrick J. McGovern Professor of Management at MIT, founding director of the MIT Center for Collective Intelligence, and a crowdsourcing expert, to refine its platform and its crowdsourcing strategies.
This year, there were nearly 30 submissions for the competition, which started last November. In January, judges selected the seven finalist teams, which were paired with mentors that helped develop business plans and pitches. Teams are not limited to MIT affiliates.