Data is the crucial piece that shapes the course of actions in disaster response situations. It empowers decision makers to plan responses for a multitude of worst and best case scenarios.
It is the basic building block of a robust early warning system.
However, this rich and nebulous data comes engulfed in the trio of Vs - Volume, Velocity and Variety.
Just 1 sq. mile area of flood disaster study generates around 12 TB of data. The tremendous volume of data that piles up from flood, hurricane, earthquake and other natural/manmade disasters puts the best data storage solution to test.
The rate of production of data can be attributed to climate changes, advances in technology and socio economic developments. Climate changes cause physical changes to the geography of a region, which calls for updated mappings. Advancements in technology enable frequent measurements and transmission of map data. Socio economic changes have a direct impact on the demographics, which is a key variable in risk assessment systems.
A myriad of weather sensors, social media feeds, structured and unstructured data sources spew out data in multiple formats. This vibrant cocktail of information needs to be stored and analyzed to put to meaningful use.
The trio of Vs can be harnessed by applying best practices in:
Data management - that cover timely refresh and archive cycles
Cleansing- involves validation and verification
Security - that guards from undesirable, harmful access that may risk authenticity of data
The Analytics Layer
Once the 3 Vs of data have been conquered, the stage is set for powerful analytics to flex its muscles.
Descriptive analytics gives a hindsight of what happened.
Diagnostic analytics explain why it happened.
Predictive analytics put forth a foresight of what will happen.
Prescriptive analytics shows the way to make it happen.
The Visualization Layer
Easy interpretation of results from data analysis is as important as the analysis itself. Millions of rows of data are a challenge to the common visualization formats, i.e. dashboards, reports, graphs and charts. It takes extensive domain understanding to get data in the right shape to be consumed by human interfaces on the web, mobile devices, GIS apps etc.
IT Begins With Data...And Ends with Disaster Resilience
The Risk assessment system that results as a combination of the 3 layers, i.e. Data, Analytics and Visualization is well equipped for mitigation and emergency planning. First response teams can be most effective by capitalizing on forewarnings, by stocking up on supplies and working out logistics. Regulatory agencies can enforce policies that minimize losses.
Hazus-MH is a tool developed for FEMA Risk mitigation that assess population needs related to emergency management. It enables comparisons between various scenarios and the respective mitigation actions.
It enables identification of vulnerable areas that may require special planning considerations. By estimating potential losses that may occur when disaster hits, planning agencies can make informed decisions on allocation of resources for most effective and efficient response and recovery.
For more information on Hazus- MH and to test drive visit http://www.niyamit.com/industries/#riskanalysis or contact email@example.com
(This is an excerpt from my presentation at AIDF 6th Annual Disaster Relief Summit on 19th November at Washington D.C. It was an interesting experience to join a team of recognized experts from aid and government agencies, United Nations, NGOs, donors, academia and private sector as we discussed innovations and best practices for disaster response. We presented NiyamIT's risk assessment/planning tool, and led the round table discussions on Early Warning Systems.)