Hurricane Sandy: Using Technology For Effective Response
As Hurricane Sandy barrels up the eastern coastline of the United States, aiming for landfall late Monday or early Tuesday morning somewhere between northern Virginia and southern New Jersey, Direct Relief is hard at work with our technology partners at Palantir to assess needs, determine likely emergency scenarios and mount an effective, targeted response for our clinical partners in the path of the storm.
Palantir is a set of advanced data integration, visualization, and analysis tools which allows Direct Relief to pull together all of the information sources needed into a common framework to better understand and manage complex problems in near real time.
Are Direct Relief’s clinical partners situated near socially vulnerable populations, flood risk zones, and probable coastal flooding areas? How does physical risk and vulnerability relate to social vulnerability? Can we anticipate needs for essential medicines in emergencies based upon analysis of what we have shipped to various clinical partners in the past? What are the likely scenarios for population movement which may stretch the resources of inland primary care health centers in the event of evacuation?
Rather than treating these dimensions of Direct Relief’s disaster response as separate and distinct, Palantir pulls them all for us into a continuous braid of analytic workflows to improve the overall intelligence and efficiency of our response.
Starting, for example, with a geographic layer showing county-level values of the social vulnerability index and flood-related damage estimates from the University of South Carolina’s Hazards and Vulnerability Research Institute, we build statistical correlations with data from the Centers for Disease Control and Prevention (CDC) on disease prevalence rates to score counties in terms of their health risks, population needs, and disaster impacts.
Pulling that analysis directly into a map view to see where the most at-risk counties are located, we can cross reference with clinical addresses and storm scenarios to prioritize problem areas and response requirements. Rapid, highly-targeted analysis of historical product flows for health centers in these risk zones focuses attention on specific material needs.
In combination with partner communications our analysis enables gains in speed, scope, and precision to build requests for donations from our corporate supporters, allowing Direct Relief to deliver better aid in time to improve health and save lives.
Our hope for the future is that intuitive information tools like Palantir will allow the aid workers at Direct Relief to collaborate more effectively with computers to amplify humanitarian aid and make truly significant gains in our ability to meet the needs of the millions of people who depend upon our nation’s network of nonprofit safety-net health centers for essential care during and after crises like Hurricane Sandy.