Tropical Storm Andrea made landfall on June 6 near Steinhatchee, Fla., a Gulf Coast town south of the Florida panhandle. It traveled up the East Coast for three days on a rain-drenched path that caused flooding in spots from Miami Beach to Massachusetts.
For Aon, the world’s largest insurance broker, storm damage claims are a common type of report for the insurers, who are its customers, to process. But it was only this year that Aon’s software-as-a-service unit, Aon eSolutions, started incorporating mapping software so its customers could analyze risks involved in events like Andrea on a BI dashboard.
Using data from the National Oceanographic and Atmospheric Administration, Aon created a map layer showing the storm’s path and embedded that into a map showing the locations of its customers’ facilities (see image at the top of this article). Mark G. LeVeque, an Aon eSolutions product consultant, said this view enables an analyst at one of his client firms to examine the relative magnitude of risk at each facility by viewing what is near the storm’s path. A warehouse, a manufacturing plant, a retail store or an office building—each location would have its own risk profile based on its employees and assets.
LeVeque’s presentation was one of several highlighting the intersection of geographic data and risk management use cases at the recent Esri Business Summit. The summit, held in July in San Diego, coincided with the Esri annual user conference that assembles geographic information system professionals and technologists to discuss applications related to the company’s mapping software.
Insurers and other risk managers pointed out that Web-based mapping software, and the ability to mesh different kinds of datasets with mapping visualizations, opened up their field to a new way of thinking and different ways to do business. What traditionally has been a tabular way of thinking in tables, dominated by business intelligence (BI) tools with text and numbers, has broadened to include more visual displays represented by maps.
“There’s a growing interest by the insurance industry and risk managers to use geospatial technology,” said Russ Johnson, public safety director at Esri. “They were interested before, but now they see an opportunity that is more visible and sharable.”
Seeing Risk in Red Dots on a Map
Aon eSolutions uses IBM’s Cognos BI system as a key component for its risk management SaaS offering, and the company started working with Esri mapping software in 2012 after a version became available for Cognos, LeVeque said.
Now the company’s RiskConsole, a dashboard-like risk management information system offering, is on the market with maps. Various views enable an analyst to click on a map position and see data about properties, histories of claims and risk profiles. For example, an assisted living care center could be watching the track of a storm and assess whether it needed to evacuate residents. A manufacturer could compare and contrast worker compensation data by location, to see if certain factories were more or less prone to injury reports. A natural gas company could juxtapose map layers of its underground pipelines with areas at risk for flooding or other hazards.
Report lag—the number of days after an event for a business to file a loss report with an insurance claims administrator—is another use for GIS. The insurance industry has found the longer the lag time, the higher the claim costs, LeVeque said. So mapping time lag in different regions where a storm or other incident has occurred can help a risk manager identify where the lags are. He demonstrated this by showing a map of the New York City metropolitan area, where red, yellow and green dots indicated various states of time lags for the clients of a hypothetical risk manager. Red dots demand attention right away.
“There’s something about the map that people find compelling,” LeVeque said. “To be a red mark on the map, that has a greater impact.”
Other examples are likely to emerge, LeVeque said. He estimates that 90 percent of the data insurers collect for use in Aon’s services include a geographic component. Now the company is researching whether it makes sense to add crime statistics as a mapping layer.
When Weather Turns into Advantage
Weather data was the subject of several risk management use cases at the Esri conference. One presenter argued that weather projections can also open business opportunities by helping businesses anticipate rising or falling demand for products and services.
Analysis of historical, real-time and forward-looking weather data leads to location-based opportunities, what Steven A. Root, a meteorologist and CEO of WeatherBank Inc., called “weather triggered marketing and demand planning” and enabling business analysts to model the effects of weather on sales.
North America has an infrastructure ready to support these analytical models, including a robust foundation of data sources—from the federal government, to academic researchers to private services like AccuWeather. The data provides both historical views from 1950, up-to-date, real-time, location-specific information on 200 different weather metrics and extended forecasts for the next two years.
For example, the winter of 2013 persisted late into spring around much of the country, compared to a record warm spell in March 2012. Understanding forecasts ahead of time could lead a retailer to stock different goods for sale, whether it’s clothing or gardening supplies.
“Weather lifts or compresses sales based on consumers’ perceptions,” Root said. “Understanding weather means a competitive advantage.”
Correction, August 12, 2013: The original version of this article has been updated to correct the reference to Aon eSolutions risk management information system. It is RiskConsole, not risk console. Also, a risk term of art is “report lag,” not “time lag” as appeared originally. A clarification: RiskConsole’s primary users are risk managers and risk management departments that manage properties and purchase insurance, not specifically insurance firms. That means an analyst clicking on a map would view data about a property, not a policy holder.