Last Updated on November 19, 2022 by maximosecrets

On 28^th October 2015 at the time of the announcement that IBM was to acquire The Weather Company I was an IBMer. It certainly seemed a strange acquisition, but knowing that asset deterioration and performance can be effected by weather, that the efficiency of field working and worker safety are similarly affected my reaction was a positive one, but the question had remained a mystery until I recently did a bit of research.

• What did IBM acquire?
• Some interesting statistics about The Weather Company
• Is the weather forecast provided by The Weather Company accurate?
• What is Weather Underground?
• What industries does The Weather Company typically operate in?
• What B2B Weather Data Packages are available?
• How can Weather Alerts help businesses?
• What is IBM Deep Thunder all about?
• How has Maximo been effected by the acquisition of The Weather Company?
• Why is weather so important to asset health?
• Has The Weather Company data been used elsewhere within IBM Watson IoT?
• Some YouTube clips
• Final thoughts

What did IBM acquire?

IBM acquired The Weather Company’s B2B, mobile and cloud-based web properties, including WSI, weather.com, Weather Underground and The Weather Company brand. The TV segment – The Weather Channel – was not acquired but now licenses weather forecast data and analytics from IBM under a long-term contract. The acquisition closed on 29^th January 2016.

https://www-03.ibm.com/press/us/en/pressrelease/48884.wss

Some interesting statistics about The Weather Company

The Weather Company is the world’s most widely recognised weather media company across many platforms with weather data being used by 750+ broadcasters globally. But what else did I learn:

• It handles 26 billion inquiries on its cloud based system daily, this is several times more than the world’s leading search engine.
• It powers the 4^th most-used mobile application in USA and is the native weather application on Apple, Google and Yahoo.
• It has sophisticated weather modelling that analyses data from 3 billion weather forecast reference points at 500m grid resolutions across the globe.
• It collects barometric readings from millions of smart phones worldwide, it continuously collects weather data from 50,000 airplane flights daily and from over 200,000 personal weather stations.
• It uses 100 terabytes of third-party data daily and is generating 4GB of new data every second.
• It updates weather conditions every minute and weather forecasts every 15 minutes.

In USA 23% of car accidents and 70% of flight delays are weather related. Half a trillion dollars of economic impact annually is due to weather, impacting every sector of the economy, but most companies do not have a weather strategy.

Is the weather forecast provided by The Weather Company accurate?

According to ForecastWatch.Com The Weather Company was rated as #1 for forecast accuracy in three global regions and on three levels of forecasting.

Some of this accuracy is down to the number of weather stations.

What is Weather Underground?

In the period 2010-2015, severe weather events caused more than $100 billion in damage in the U.S. alone and the U.S. has modern warning systems. Across the world, the World Bank reports that natural disasters over the last 30 years have cost an estimated 2.5 million lives and more than US$4 trillion. There are 70 developing countries with no robust early warning system.

In 2001 Weather Underground was formed to tackle this problem, it is a subsidiary of The Weather Company. They developed the Personal Weather Station Network which now has over 200,000 stations in 195 countries providing The Weather Company with hyper-local forecasts to millions around the world with unparalleled accuracy. Each station is equipped with multiple sensors for detecting barometric pressure, humidity, temperature, wind speed and direction, and other factors. Greater personal weather station data leads to more precise and enhanced forecasts, which can help governments and communities better anticipate and act on severe weather conditions.

In the above graphic the blue dots represent personal weather stations and the red dots the public weather stations. The number of personal weather stations is increasing all the time providing The Weather Company local data to feed into its models. The picture from France is around Paris, there must be a lot of weather enthusiasts there.

This a good example of crowdsourcing – obtaining information from a large number of people via the internet. In the center graph, since November 2014 there has been a sudden rise in personal weather stations with over 100,000 added to the network in a year.

The UK Met Office has between 200-300 weather stations servicing the whole of the UK. The weather station at Heathrow will sometimes be several degrees different from the temperature in central London. Hyper-local weather accuracy is the key to making meaningful business decisions based on the weather.

What industries does The Weather Company typically operate in?

Weather effects many industries, perhaps most industries, but the ones where The Weather Company has developed specific products include:

• Media
• Aviation and Airports
• Energy including energy trading
• Insurance
• Government agencies
• Retail

The Weather Company started work with television broadcasters more than 35 years ago and there are now over 750 broadcasters using weather data globally. Weather presenters are provided with the tools to create graphics based on weather, graphics they can interact with including augmented or virtual reality, for example weather mixed with visual traffic feeds. Media companies use the weather and traffic data to provide mobile applications or to publish to social channels.

In aviation if an aeroplane hits turbulence not only is there a risk to passengers and crew but the aircraft may need to be taken out of service for inspections and this can create delays or cancellations. The Weather Company uses real time observations from aircraft to predict areas of flight risk so that planes can be rerouted to avoid severe flight hazards. This has reduced crew injuries by more than 50% and a similar percentage reduction in mandatory inspection checks. Applications exist for pilots and operators.

Airports use weather data to understand when there will be an increase in the number of passengers in the airport due to likely delays or cancellations due to the weather. Weather alerts are used by airports to ensure that staff are not outside during severe weather conditions. Forecasts are used to ensure there are enough staff to handle de-icing requirements or for snow clearance.

Utilities use The Weather Company data to determine demand based on forecasted weather data. Long term forecasts to help trading, short term forecasts to understand where crews or engineers should be positioned in the event of damage to critical infrastructure.

Insurance companies use accurate weather pictures to identify likely fraudulent claims from severe weather. They can also pay out quickly to claimants who are in the known envelope of the weather event. Knowing the weather at the time and location of a vehicle accident can assist with risk scoring as part of the claim investigation.

Government agencies use probabilistic forecasts to gain early warning of severe weather conditions and then to track the most likely path, expected impact and timing of these weather events.

Accurate weather forecast can be used by retail clients to predict when products will be needed in store, the first storms or snow of the season creates a lot of activity. The same weather conditions at two places may result in completely different outcomes. 85F in Minnesota and the shopping malls will be empty as everyone is heading for the lakes. 85F in Atlanta and the shopping malls will be full with people trying to cool off. Analytic insights can have local variations. Retailers can see cost savings of up to 12% by not over-manning during quiet periods and an increase of up to 7% by having adequately manned stores during busy periods. Historical weather data is used to model predictions and calculate impact. For example DIY stores can see a massive increase in footfall on public holidays but sales are weather dependent. Just comparing how a store faired on the same day the previous year is not a good way to predict what will be achieved this year.

What B2B Weather Data Packages are available?

There are many weather data packages available from The Weather Company including:

• Historical weather
• Current conditions
• 2-day hourly forecasts
• Immediate future out to 15 day forecasts
• Seasonal and sub-seasonal forecasts
• Severe weather forecast alerts
• Health and lifestyle indices, e.g. pollen, air quality, frost potential, tides, ski, golf, etc
• Alerts leveraging weather, traffic, road and location data
• Solutions for Aviation, Energy, Media and Insurance
• The Weather Company data is also currently available in North America and Europe through the IBM Watson IoT platform, Bluemix

https://business.weather.com/writable/documents/Data-Packages/DataPackages_2016-2.pdf

ONGOLF is a company that has used a weather data package to enhance their golf solution by providing local weather data to each of their 100+ golf course customers allowing more efficient course maintenance and improved course health.

Weather data sets can be made available to work with analytics tools, for example Cognos or SPSS, or asset management systems like Maximo or Tririga. Weather data when used with analytics and historical data gives you insights which can be used to help make operational and strategic decisions. For example, does the weather pattern in the location of a set of assets lead to deterioration faster than similar assets at another location? What is one of the first things you do when you wake up? You look at the weather forecast and make decisions based on that information. Would forecast weather data change where you plan to do work next week or cause you to reassign work if severe weather was forecasted?

How can Weather Alerts help businesses?

In order to improve customer loyalty and also to help mitigate against adverse weather conditions companies can subscribe to weather alerts which they can then push to their customers. This combines geo-location and accuracy of weather predictions to only inform those customers that would be affected. The accuracy of the weather prediction is a key point as false-positive alerts can negatively impact the customer experience.

Insurance companies use these alerts. The Weather Company’s data modelling can predict weather to within 30 minutes which is invaluable in lightning or hurricane prone areas as sometimes minutes can count. Depending on geography text alerts can be provided for Hail, Lightning, Wind, Rain or Snow using predefined templates. Insurance companies have seen a 5% increase in customer loyalty by providing weather alerts.

Pantene has created a frizz index based on weather conditions allowing them to more accurately target when and where to advertise a particular product. Weather-related geo-located advertising created a 24% increase on sales YoY.

Weather alerts could be used by electric utilities wishing to alert their customers to potential disturbances due to high winds. It could be used by utility companies generally or other companies with field service workers to inform their employees of adverse weather conditions depending on their current locality or expected route.

What is IBMs Deep Thunder all about?

On 15^th June 2016 IBM announced plans to advance the precision and accuracy of weather forecasting by combining hyper-local, short-term custom forecasts developed by IBM Research with The Weather Company’s global forecast model. The powerful combination of the two models will be called Deep Thunder. Historical weather data will be used to train machine learning models that will help businesses predict the actual impact of weather. The combined models are expected to provide accurate weather predictions around the world to within a 0.2 to 1.2 mile resolution and also take into account other relevant environmental data such as vegetation and soil conditions to better understand the impact of a weather system. The Weather Company will integrate this capability and broaden access to it on a global scale. This will allow companies to understand the impacts of weather and identify the actions to take.

http://www.theweathercompany.com/DeepThunder

How has Maximo been effected by the acquisition of The Weather Company?

In the first year since the acquisition we now see in Maximo:

• As part of Maximo Scheduler Plus there is a Weather Configuration Manager application to enable the configuration of weather data on the Graphical Scheduling, Graphical Assignment and new Graphical Appointment Book applications. The Current, Hourly and Daily forecasts can be configured with up to 54 weather related attributes.
• The new Work Center applications include weather data.
• Maximo Asset Health Insights uses current and historical weather data to understand the conditions in which the asset is and has been operating and uses this as a driver in assessing the health of assets.

Why is weather so important to asset health?

This is best illustrated with an example. In electric utilities ambient temperature can be an important factor in estimating transformer life particularly when coupled with the load through the transformer. A study of historical records across 200 locationsA physical place where assets exist and where work can be performed. More in mainland China concluded that transformer expected asset life varied from 10.6 years to 149.3 years.

http://www.hindawi.com/journals/tswj/2013/125896/

Weather data can provide value by:

• Reducing the frequency of asset failures. The cost through operational disruption of unplanned asset failures is far more expensive than performing just in time maintenance.
• Safely increasing the periodicity between inspections and preventive maintenance. If there are less inspections and maintenance then there is more time for other asset management activities that can also prolong asset life, for example predictive maintenance.

Has The Weather Company data been used elsewhere within IBM Watson IoT?

Looks as if IBM has been busy. This is probably not a complete list but I have found references to weather being integrated with:

• IBM Insights Foundation for Energy
• IBM PMQ (Predictive Maintenance and Quality)
• IBM Insights for Weather for Bluemix (the IoT platform)
• IBM IoT Real-Time Insights

And no doubt weather data is being integrated in many other places as well, not just that related to IBM Watson IoT. For example, extending weather alerting to all populations by using mesh networks, peer to peer smartphone connections.

http://www.twice.com/thewire/ibm-and-weather-company-unveil-world-s-first-mobile-weather-alerting-platform-underserved-populations-emerging-markets/64350

Some YouTube clips

62 miles separate us from space. What if you could map the atmosphere?

https://www.youtube.com/watch?v=gsdR_w2ITOg

The Weather Company, an IBM business

https://www.youtube.com/watch?v=gPOKHT3zWLs

Harnessing the power of Weather: IBM and The Weather Company

https://www.youtube.com/watch?v=2hWQtey8D5w

Combining weather insights with cognitive IoT

https://www.youtube.com/watch?v=mu9JQMmgO8c

Weather Company Data for IBM Bluemix: Deep Dive

https://www.youtube.com/watch?v=pE0fnnW7wt0

The Weather Company Alerts for Worker Safety

https://www.youtube.com/watch?v=8vrc8ERuc7o

The Weather Company Alerts for Engagement (Customer Loyalty)

https://www.youtube.com/watch?v=Jlk3ILCf0ow

The Weather Company You Tube channel

https://www.youtube.com/channel/UCEsWthnLbQ3O9xovUmU2_qQ

Final thoughts

Prior to the acquisition by IBM The Weather Company was already a global platform providing local weather with sophisticated models and analysis being created by 400 meteorologists and data scientists. It had big ambitions to map the weather to 62 miles of atmosphere and to provide accurate weather forecasting and weather alerts to all populations.

Combining with IBM will extend the use of weather data to provide economic value to more businesses in a greater number of industrial sectors than The Weather Company could have done alone. The combination of weather and IBM Watson will ultimately lead to cognitive decision making based on precise weather data. This is cloud computing, crowd sourcing, big data and cognitive all working together.