Using Data Science to Optimally Allocate COVID-19 Tests to Protect Travelers
LOS ANGELES – July 27, 2020 – A multi-disciplinary team of researchers, entrepreneurs, and policy makers announced an AI-based project, nicknamed “Eva,” that uses data to support decision making by the Greek government as it reopens the tourist industry vital to its economy amid the worldwide COVID-19 pandemic.
Kimon Drakopoulos and Vishal Gupta from the Marshall School of Business at the University of Southern California; Hamsa Bastani from the Wharton School at the University of Pennsylvania; Jon Vlachogiannis, founder of AgentRisk; and the Greek government came together earlier this summer to build the machine-learning platform. Greece is home to approximately 11 million people, but has welcomed more than 33 million visitors annually in recent years, with tourism accounting for close to 20 percent of the country’s employment.
“The AI system developed by Bastani, Drakopoulos, Gupta, and Vlachogiannis has been an asset both for preparing the opening of the country to visitors from all over the world, as well as for allowing flexibility in decision making regarding our COVID-19 strategy,” said Nikos Hardalias, Greece’s Civil Protection and Deputy Minister for Crisis Management, who heads the COVID-19 Response Taskforce for the country.
“Tourism is vital to the Greek economy and in times of a pandemic controlling the flow of visitors is extremely delicate both operationally and from a public health point of view,” continued Hardalias. “The developed solution has allowed the Greek Government to make crucial decisions with confidence due to the ability to continuously monitor the epidemiological characteristics of all countries that we accept visitors from. It is great to see how science can complement our national response to this challenge in keeping the local population and our visitors safe.”
Eva combines real-time testing data with information from a simple form that visitors complete 24 hours before arrival to build a risk profile for each visitor. Based on that profile, Eva suggests which visitors should be tested for COVID-19 on arrival and which can safely be admitted without testing. Crucially, Eva uses past data and optimization to simultaneously improve its own risk predictions while also identifying sick visitors before they enter the country, all subject to Greece’s current COVID-19 testing capacity.
The system provides several benefits for travelers and decision makers by leveraging data to enhance public safety:
- Efficiency: With as many as 40,000 people per day arriving at points of entry around the country, Greece cannot test everyone who might bring coronavirus into the country. Using data to assess risk factors focuses testing on the riskiest travelers, enhancing public health and safety while responsibly allocating valuable testing resources.
- Convenience: A streamlined operations including the pre-arrival form, expedited testing, and seamlessly connected databases minimizes disruptions to travelers. Most travelers are not subject to additional screening, and, those who are, are usually on their way to enjoying Greece within 24 hours without wasting valuable vacation time.
- Responsiveness: By leveraging real-time data to allocate resources, Eva’s analytics support rapid decision-making, allowing policy-makers to quickly respond to unexpected super-spreader events or flare-ups.
“For me, this is about not only applying my work in data science to help the people of Greece,” said Drakopoulos, “but also the people of the world who love to travel and worry about the safety of doing so.”
Bastani, Gupta, and Drakopoulos collaboratively developed Eva’s underlying algorithms, emphasizing learning from real-time data, and wrote its implementation. Vlachogiannis is the software architect of the machine learning pipeline, which allows seamless and secure access to anonymized data from disparate Greek government databases in near-real-time. Recently, Drakopoulos has been embedded with Greek public health and policy leaders, overseeing Eva’s deployment and liaising with the rest of the team as they continue to tailor Eva to Greece’s unique circumstance.
“One of the most exciting elements of Eva is its ability to learn, improve, and evolve. Adapting in real-time is crucial in this pandemic, where the situation on the ground can change dramatically in a day or two,” said Gupta.
Bastani adds, “New testing results are continuously incorporated into the dynamic learning algorithm, giving Eva a distinct advantage over static COVID-19 screening policies. This is an exciting step forward in evidence-based policy-making.”
No screening procedure can possibly find every infected visitor. Eva dovetails with Greece’s existing contact-tracing system to catch anyone who slips through the cracks. Overall, Eva’s risk-profiles, test allocations and other data analytics form a real-time dashboard, visually representing the latest information to the Greek Government to inform decision-making.
“In addition to a day of hope for those who love travel and long for a path out of the pandemic, this is also a huge day for data science, machine learning and algorithmic support for good governance,” said Vlachogiannis.
About Kimon Drakopoulos, Vishal Gupta, and the Marshall School of Business
Kimon Drakopoulos (https://www.kimondrakopoulos.com/) is an Assistant Professor of Data Sciences and Operations, whose research focuses on epidemics modeling, social networks, and information economics.
Vishal Gupta (http://faculty.marshall.usc.edu/Vishal-Gupta/) is an Assistant Professor of Data Sciences and Operations, developing new algorithmic approaches to data-driven decision-making in settings where data and/or resources are scarce, with applications in healthcare, revenue management and business analytics.
Consistently ranked among the nation’s premier schools, USC Marshall is internationally recognized for its emphasis on entrepreneurship and innovation, social responsibility and path-breaking research. Located in the heart of Los Angeles, one of the world’s leading business centers and the U.S. gateway to the Pacific Rim, Marshall offers its 6,000-plus undergraduate and graduate students a unique world view and impressive global experiential opportunities. For more information, visit www.marshall.usc.edu.
About Hamsa Bastani and the Wharton School
Hamsa Bastani (https://oid.wharton.upenn.edu/profile/hamsab/) is an Assistant Professor of Operations, Information, and Decisions, whose research focuses on developing novel machine learning algorithms for data-driven decision-making, with applications to healthcare operations, revenue management, and social good.
Founded in 1881 as the world’s first collegiate business school, the Wharton School of the University of Pennsylvania is shaping the future of business by incubating ideas, driving insights, and creating leaders who change the world. With a faculty of more than 235 renowned professors, Wharton has 5,000 undergraduate, MBA, executive MBA and doctoral students. Each year 13,000 professionals from around the world advance their careers through Wharton Executive Education’s individual, company-customized, and online programs. More than 99,000 Wharton alumni form a powerful global network of leaders who transform business every day. For more information, visit www.wharton.upenn.edu.
About Jon Vlachogiannis and AgentRisk
Jon Vlachogiannis is an inventor, data scientist and serial entrepreneur, with numerous patents in analytics and highly scalable systems, and a successful exit in the big data and analytics space.
Outside Project Eva, Jon is the founder and CEO of AgentRisk.
AgentRisk is an Algorithmic Wealth Management Platform with origins in Silicon Valley founded in 2015 by a team of successful tech entrepreneurs. It creates and utilizes proprietary algorithms to manage wealth in an automated, data-driven fashion eliminating human biases. AgentRisk’s customers are successful Entrepreneurs, Celebrities and Athletes who trust data, transparency and value being a part of a network of similar minded people. For more information, visit https://agentrisk.com/
###