A recent invention on the University of Maryland campus has taken food safety inspection to a whole new level. Using this invention, chain restaurants can easily monitor food safety practices in their outlets across the nation. Restaurants and cafes can also stay ahead by comparing their performance with their competitors.
All this and more will be ensured by a national database recently developed by Associate Provost of Learning Initiatives, Executive Director of the Teaching and Learning Transformation Center, and Professor of Computer Science Ben Bederson, and his team comprised of UMD Professor of Economics Ginger Jin, UCLA Associate Professor of Business Management Philip Leslie, UMD Computer Science Ph.D. graduate Alexander Quinn, and UMD Economics Ph.D. graduate Ben Zou.
The database has been chosen as one of the nine nominees for Inventions of the Year as part of this year’s Celebration of Innovation and Partnerships at the University of Maryland.
The database represents a huge leap from those built using manually-collected data that may easily miss the big picture and have little impact on compliance actions, said Bederson.
He uses data robots to collect data automatically from local government websites. “Our data robots cover a large number of local jurisdictions over the U.S., continuously detecting new data posted by each jurisdiction, and integrating them into a single, standardized, and cumulative database,” he says.
The approach is cost-effective, robust and scalable compared to manual alternatives.
The researchers have also developed analytical tools that can be used to compare food-quality and safety across chains, regions, food outlets like restaurants and cafes, convenience stores, and grocery markets. This will improve inspection efficiency and promote retailer compliance, resulting in a decrease in food-borne illnesses.
The technology enables the analysis of millions of food safety inspection reports across the country. It will be useful for restaurants that want to compare their outlets located in other jurisdictions and find out which ones are problematic as well as compare their performance against competitors.
The team has also started Hazel Analytics, a regulatory data analytics company,which Bederson said is direct outgrowth of their academic collaboration around food safety inspection data.
“As we shared our work with industry players, government agencies such as the FDA and CDC, and other academics, our intuition was confirmed that there was commercial value in our database and analytical approach,” he said.
Hazel now produces a commercial grade restaurant inspection database and analytical services for the food service industry.
“We are currently in close talks with McDonald’s and Jack in the Box, and have had introductory discussions with several other major national chains. We expect to have our first paying customers this year,” Bederson said.
“Building our system to reliably collect information from so many different jurisdictions was a formidable engineering challenge,” he said. Another difficulty the team faced was developing normalization algorithms to compare data across jurisdictions where the data is very different. For some web pages, they had to write custom ‘scrapers’ to get the data and for others they had to interpret the already available databases.
The database is publicly available at InspectionRepo.com at no cost for non-commercial use.
The Office of Technology Commercialization (OTC) helped Bederson and his group to develop, license and commercialize the technology. “With OTC's help, we have worked closely with the Maryland Technology Development Cooperation [TEDCO] and received Phase I $100K funding from the Maryland Innovation Initiative [MII] program,” Bederson said.
Bederson plans to apply the web “scraping” technology to other inspection programs implemented by the local governments. He said that it can also be used in federal inspection programs like meat inspections conducted by the U.S. Department of Agriculture.
March 31, 2015