IBM's acquisition of the The Weather Company's B2B properties back in October drew plenty of attention, but since then the company has been fairly quiet about the deal.
No more, however. The Weather Company has now launched a commercial offering called Deep Thunder, which merges hyper-local forecasting models developed since the mid-1990s by IBM Research under that name with The Weather Company's own RPM (rapid precision mesoscale) forecasting model. Deep Thunder will also "also use historical weather data to train machine learning models that will help businesses predict the actual impact of weather," the company said in a statement.
CIO describes Deep Thunder's capabilities in a detailed feature article:
Deep Thunder is tuned for forecasts at a 0.2 to 1.2 mile resolution, allowing businesses to understand the exact weather conditions in their location. For example, in aviation, local inclement weather contributes in a direct and measurable way to congestion at airports. With accurate insight into local weather, airlines can better predict congestion to make more precise decisions about exactly how much fuel to put on any given plane.
Another case in point is precision agriculture. Pesticides and fertilizers are sensitive to environmental factors like rain — some need periods of clear weather while others require rain immediately after application.
Energy, particularly renewables, are another area where precision weather forecasts can translate directly to the bottom line.
Of course, the new Deep Thunder algorithm's overall success or failure has yet to be determined, given it's only just been launched. Nonetheless, it speaks to an important need, says Constellation Research VP and principal analyst Doug Henschen.
"The theme of precision data and precision insights is going to be a big trend and challenge in the years ahead," he says. "Of course everybody would like greater precision, and this example of fine-grained weather prediction is certainly good news. Same goes for fine-grained traffic prediction, fine-grained crime prediction, fine-grained financial market prediction and so on. All of the above are empowered by more voluminous and detailed data."
"Another enabler is more powerful combinations of algorithms," he adds. "In this example, machine learning was added, but did not replace, the techniques previously used. This is ensemble analytics, the approach of applying multiple analytical techniques rather than searching for a single, silver-bullet technique."
The full CIO piece, which is well worth reading, also goes into how IBM plans to use Deep Thunder's forecasting capabilities for still other industries, such as retail, insurance and utilities.
An algorithm alone, no matter how powerful, doesn't provide the entire answer, Henschen notes.
"Delivery of fine-grained insights and predictions is just the beginning," he says. "The next, and perhaps harder, challenge is finding ways to put the information and predictions to use. Are airline operations teams, airport management systems, farmers, retailers, insurers and utility companies equipped to make use of granular insights and predictions? Are humans equipped to absorb and discern this fine-grained information or would find-grained decision-support have to be built into their applications?"
"We’re facing much the same challenge as industries and organizations rush to make use of low-latency and real-time data," he adds. "Of course the idea of using real-time data and insights sounds appealing, but if crucial systems and other critical data sources can’t keep up, the value is wasted.
"There’s no sense offering a real-time product recommendation, for example, if your insight into customer transactions is subject to overnight batch processing," Henschen says. "It’s a real fail when you send a discount offer for a product that a customer purchased at full price 12 hours ago. It’s fine to dream big and strive for fine-grained or real-time insights, but think through what it will take to put those insights to work."
Reprints
Reprints can be purchased through Constellation Research, Inc. To request official reprints in PDF format, please contact Sales.
24/7 Access to Constellation Insights
If you’d like unrestricted access to Constellation Insights, consider joining the Constellation Executive Network for analyst advice and analyses that you can use.
