Blogs

Let predictive analytics be your guide through stormy weather

Government Market Segment Manager, IBM

As a matter of fact, you DO need a weatherman to see which way the wind blows. Bob Dylan was wrong on this one. Certainly weather is all around us, which is why we naturally think of the weather in personal terms: Will I need a jacket tonight? Should I bring an umbrella? Should I set my alarm ahead so I can shovel my driveway tomorrow morning? Or—if you live in California—will it ever rain again?

But as it turns out, weather has an integral role in everything from retail sales to hedge funds to your utility bill; indeed, weather plays a very big part in our economy. And nowhere is weather more important than in the area of emergency management. Having a firm grasp on current and future weather conditions helps emergency preparedness organizations manage and ready themselves for emergencies.

http://www.ibmbigdatahub.com/sites/default/files/pinwheel_blog.jpgSome emergencies, of course, are directly attributable to weather—floods, hurricanes, tornadoes. But weather is a critical factor even in emergencies that are not caused by weather. For example, when planning to shelter thousands of people after a human-caused disaster, it’s sure good to know whether waterproof tents will be needed. And those who are fighting a forest fire (though many forest fires are in fact started by lightning strikes) need to know in which direction the wind is blowing, as well as at what speed. Similarly, planning for moving large groups of people out of a disaster zone isn’t complete without an idea of what weather conditions the people will be moving in.

Kaitlin Noe, a colleague of mine, wrote a blog post discussing the past, present and future of emergency management in which she highlighted how the practice of emergency management is improving thanks to technology use. Even more particularly, I’ll note, emergency management is improving thanks to real-time access to weather data.

IBM has announced an emergency management platform that benefits from a partnership with The Weather Company, as Time Magazine noted when describing how IBM is helping towns predict disasters: “The new platform marries live weather forecasts with a hyper-local map of a city’s infrastructure, painting in yellow and red the predicted damage to come.” Here, then, is a whole new element in emergency management: Not only do live weather feeds provide invaluable data that can help in planning for and managing disasters, but also, in combination with analytics, the IBM solution adds prediction to the calculus. Using historic data from natural disasters, for example, predictive models can be developed to provide instant assessments of infrastructure failure within a disaster zone under various weather conditions—just consider, for instance, wind’s effects on power lines.

Or look at another segment of public safety. Some police departments have built predictive models to identify “hot” zones where specific types of crime are likely on certain days and at particular times. Such models help guide force deployment. Many factors play into such models, but can you guess what one factor is? That’s right—weather. Car burglaries, for example, are particularly likely on rainy nights.

We’ve all heard about the Internet of Things: how sensors and other technology are rapidly connecting everything else. The possibilities are very exciting (though I’m not sure I want my toaster to text me when my bagel is ready), especially because of what the Internet of Things brings to emergency management. The Weather Company, for example, operates a network of 125,000 weather sensors, some of which are even attached to jet airliners. Surprised? Don’t be. Real-time access to weather information is just a connection to another node in the Internet of Things, this time involving the atmosphere of the earth itself.

Capitalize on weather data insights

Join the ongoing conversation using #SaferPlanet.