Walgreens’ Use of AI for Flu Tracking is Nothing to Sneeze At

Combining geographical location data with demographic data and powerful artificial intelligence, the pharmaceutical retailer relies on GSI mapping to shed light on the spread of influenza and help direct medication inventory needed in its stores. The technology even plays a role in national COVID-19 infection tracking.


Article Key

Image credit: Walgreens

Cold and flu season is upon us, and Walgreens once again will be tracking it for us. For the past several years, the Walgreens Flu Index monitors flu cases across the U.S. and displays the results in a dynamic online map. Walgreens tracks the sale of antiviral medications used to treat influenza across its network of 9,000 pharmacies. The collected data then feeds into a geographic information system from Esri, and the resulting map, updated weekly, reveals where the highest number of flu cases occur.

The Walgreens Flu Index is a nice service for customers, who can check influenza instances in their state, county, or zip code—and perhaps get a flu shot—but Walgreens uses the data to inform inventory decisions. In stores with increasing infections, Walgreens can ship out more flu-related products to accommodate the surge. Using predictive analytics, the company can forecast potential hot spots and prepare store inventory accordingly.  

In addition to tracking flu cases, Walgreens uses the mapping tool more broadly internally. Dubbed WalMap, the tool synchronizes Walgreens data and allows employees to add and subtract layers of information via a dashboard to address unique needs. For example, Walgreens execs use spatial analytics to determine where to open stores and what merchandise to stock.

Using WalMap, the company can view geographical and demographic information to ascertain store viability and customer retention in a new area. A mobile version allows employees to go in the field to collect additional data or add notes to an aerial view of a location.

A Granular Map of the World

The WalMap tool is built on Esri’s ArcGIS. Esri combines spatial algorithms with machine learning and data analytics, and presents the data through maps, graphs, and other visuals on a dashboard. The company has more than 1,000 analytical tools to give location data additional heft. The Esri Geospatial Cloud allows customers to identify patterns and use predictive models to improve business operations. The cloud-based GIS mapping solution provides users with 24/7 access from anywhere, so companies can share data and collaborate easily.

Screen grab of US map tracking flu vaccine distribution

Image credit: Walgreens

With the ArcGIS Velocity add-on, users can drop in spreadsheets or feed data from Internet of Things platforms, message brokers, or third-party APIs. Users can overlay that data on Esri’s location and demographic data to glean insights that might otherwise escape notice.

ArcGIS Velocity also allows users to visualize, process, and analyze data feeds in real time, store them as big data, and perform queries and analysis, so customers can offer software-as-a-service (SaaS) IoT applications. Every feed can be its own layer on the map, and multiple layers can provide granular insights. 

AI Brings COVID-19 to Light

The Esri technology is also behind the Johns Hopkins University COVID-19 Dashboard. The map was developed by Ensheng Dong, a JHU doctoral student, who previously had completed an internship at Esri and was very familiar with the capabilities of ArcGIS. Dong created the map in January, shortly after hearing about COVID-19, to monitor cases near his family in China. As the pandemic grew, he worked with Dr. Lauren Gardner, professor of engineering at JHU, to expand the mapping tool. Dong initially had to gather, classify, and synchronize the COVID data himself—often manually. In February Esri’s Living Atlas of the World team helped Dong and the Johns Hopkins team automate data scraping.

Since going live in April, the JHU COVID-19 dashboard has expanded from displaying simply active COVID-19 cases to incorporating other factors: deaths, recovering cases, testing rates, case-fatality ratio, hospital capacity, insurance availability, and demographics, such as age, ethnicity, and employment. According to Gardner, the JHU map gets between 3 and 4.5 billion hits a day, and other organizations, including policy makers, are pulling the data to their own internal dashboards for use.

The ability to collect data and overlay that information against a plethora of other geographic information allows users to discover hidden patterns and trends that can impact business, society and the environment, while providing transparency to the public.

  • Learn more about Esri GIS mapping and its ArcGIS platform. 


Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries.

Find Intel® IoT Market Ready Solutions

Search the Intel® IoT partner ecosystem and discover 130+ end-to-end IoT solutions.

Become a Member of the Intel® Partner Alliance

If you are delivering repeatable and scalable IoT offerings using Intel technology, the Intel® Partner Alliance program gives you more opportunities to build your business. Get started.