The archive, called WAI (Weather for AI), is intended to enable weather-sensitive industries and businesses to easily access massive amounts of unique hyper-local historical weather data from every point on earth, and generate tailored AI-driven insights. Now, says the company, decision makers will be able to predict the specific impact of selected weather parameters on their operations, as well as gain a better understanding of the connection between weather and business performance.
“WAI’s hyper-accuracy and unique historical reanalysis of gridded data will catalyze real time, data driven action,” says Shimon Elkabetz, CEO and Co-Founder of ClimaCell. “But beyond these game changing benefits users will be amazed by the seamless, intuitive way the service runs. Not only will using ClimaCell’s historical datasets create tailored insights to predict business outcomes and drive better informed decision making, all this will be done in a fraction of the time that people today spend trying in vain to plan around the weather.”
Until now, says the company, industries and businesses most vulnerable to weather have encountered two significant obstacles when trying to train their AI models based on historical weather data: First, they’ve had to rely exclusively on traditional tools of the weather trade such as satellites, radar, and weather stations – standard data sources that produce vague results, and often times the data that’s used isn’t accurately calibrated to a desired location.
Second, the data that’s generated is frequently too massive and complex to be of any practical value. As a result, users can’t customize this information to their specific business needs. These two major issues, says the company, have prevented businesses around the world from generating the actionable insights needed to stay one step ahead of Mother Nature, train their AI models, and better predict their future business outcomes.
ClimaCell’s approach applies AI and machine learning technology to its unique historical gridded weather data reanalysis to produce what it claims is unparalleled accuracy. WAI’s ultra-high resolution datasets date back years, and can be quickly and fully customized to train AI models by desired location, coverage, resolution, as well as specific weather and air quality parameters. Customers can then use the company’s MicroWeather API to use real-time and forecast data that matches the resolution of WAI’s archive.
The company’s hyper-local historical data is derived from a global network of Weather of Things (WoT) virtual sensors – wireless signals, connected cars, airplanes, street cameras, drones, and other Internet of Things (IoT) devices. By fusing this data into cutting-edge AI-driven modeling techniques, says the company, WAI uniquely uses a hyper-local global grid of less than 500 meters to create the world’s first data archive to provide ultra-accurate historical results for any location on earth.
ClimaCell, Google collaborate on free access microweather forecasting
IBM AI weather tool enables predictive business forecasting
‘Weather of Things’ app offers consumers a street-by-street forecast
Supercomputer-based weather forecasting system rolls out worldwide