Not long ago, big data and weather were two concepts that you wouldn’t expect to see in the same sentence, but things are changing. In 2015, IBM snatched up The Weather Company, which owns weather.com, in a deal worth over $2 billion.
The company understood just how much businesses—and society as a whole—depend upon the weather. We may be the most advanced society in history, but our industries are still affected by wind patterns, rain, humidity, and temperature. Understanding these elements and how they affect us is a key challenge for big data scientists—and a massive opportunity.
Managing the weather’s impact on logistics organizations has a significant impact to the overall costs and operations. Weather patterns affect more than decisions about whether an aircraft can fly, or what road or path is the safest for a truck to travel due to snow, rain, or ice. The weather has a significant impact and can create a domino effect on the supply chain as well as the overall safety of employees.
One example of this is how MITRE worked with the Center for Advanced Aviation System Development (CAASD) to evaluate the safety and efficiency of U.S. national airspace systems. In doing so, it consumed vast amounts of information, including weather data.
In the trucking industry, combining big data and weather information can drive new efficiencies into the supply chain. According to the US Department of Transportation, trucking companies lose 32.6 billion vehicle hours annually due to weather-related congestion.
AccuWeather’s team of 100 meteorologists uses analytics tools to review 20 terabytes of data, producing over 100 global forecast models. The team uses these models to create thousands of local, minute-by-minute forecasts that it can then send to truckers and their dispatchers. That can help with planning routes and predicting safe speeds for specific roads at different times throughout the day.
Weather affects something that is important to both the transportation and the energy industries: the price of fuel. Natural gas prices spike during cold periods, and oil prices can suffer due to bad weather, especially when adverse conditions hit production areas. This has a knock-on effect on businesses.
Downstream energy supply companies can also use weather analytics to predict energy demand. Using data sets from the U.S. National Oceanic and Atmospheric Administration (NOAA), electricity generators save $166 million annually by producing 24-hour temperature forecasts, according to the GovLab, a nonprofit organization focused on the use of open data to solve problems for business and society.
Insurance companies stand to gain more from accurate predictions of weather risk than many other industries. Reinsurer Swiss Re predicted that natural disasters, including weather-related events, cost global insurers $95 billion in 2017.
Pacific Specialty Insurance Company is marrying big data and weather analytics tools to understand roof-level weather conditions and how they affect household damage. The concept changes how the company handles weather-related property damage claims, helping it make better decisions and reduce claims expenses by 15 percent.
Insurance companies have used weather data to deliver personalized alerts to people based on their geographic location. In one initiative, insurers told automobile owners that hail would be hitting their location, giving them the opportunity to move their vehicles and avoid damage.
Weather analytics can also create new efficiencies for the agriculture industry, another area in which large companies have been investing for long-term profit. In 2013, Monsanto acquired the Climate Corporation for its ability to digest vast amounts of weather information and use it to assess crop risk.
Similarly, Weather Analytics, a computer, data, and atmospheric science company, uses weather data to help evaluate risk for farmers by anticipating potentially damaging weather conditions and communicating significant weather events in advance. It also analyzes historical crop data in the context of historical weather data to anticipate yields and predict crop prices each year.
While we can draw easy correlations between the weather and some business outcomes, others are less obvious. Consider how some retailers use weather event information in advertising, for example. Some products sell more effectively in certain weather conditions. Companies are now mixing weather data with information about consumer behavior to dynamically change their online advertising as the local weather environment changes. More importantly, not stocking the right merchandise for specific weather conditions can lead to stock-outs or excess inventory.
The same concept works for brick-and-mortar stores. Starbucks has been using weather data to predict the proportion of hot and cold drinks that it will sell on a given day.
The Federal Reserve Bank of San Francisco studied the correlations between weather patterns and short-run employment and found that, while hot temperatures boost local job growth, heat waves drive it down. The effects can be subtle, with weather affecting some counties in different ways.
Just a few short years ago, it may not have been possible to blend big data and weather to reduce risk, grow revenues, and minimize the impact to supply chains, but thanks to increasingly accurate measurements, it’s raining data in the weather prediction business. This is creating significant benefits for companies wanting to forecast tomorrow’s fortunes—and avoid yesterday’s bad luck.
Learn more about how your organization can harness the power of advanced weather data to transform your business.