StormGeo is the global provider of advanced analytics and meteorological services delivering decision support for weather sensitive operations. Artificial intelligence and deep learning, or machine learning, are an increasing part of the advanced analytics StormGeo offers to shipping, renewable energy, corporate enterprise and offshore industries.
In this article, StormGeo’s CEO, Kent Zehetner, discusses how the company utilizes advanced analytics to benefit its customers.
We have heard a lot about advanced analytics and the amazing results it can deliver, but what is it really?
Advanced analytics are sophisticated methods to discover deeper insights, make predictions, or generate recommendations by examining vast amounts of data. With advanced analytics you can uncover new opportunities in all types of data from a variety of sources. For example, using advanced analytics, StormGeo can reduce a vessel’s fuel consumption and predict a helideck’s heave or pitch and roll movements on an oil rig.
What are the benefits of using advanced analytics?
Machine learning, and in particular deep learning, provide us with actionable decision guidance so that our customers can manage risk and operations, control costs and increase revenue. Since its inception, StormGeo has analyzed petabytes of data and transformed it into actionable decision guidance.
Can everyone get hold of data and provide these advanced analytics?
Yes, in theory, but raw data in its unprocessed state does not offer much value. It is with the right analytic techniques that StormGeo offers rich insights to aid in providing true businesses value. StormGeo has access to more than 20 years of high quality historical data.
“We are constantly trying to take our business one step further, and we see a great potential for future products and services based on artificial intelligence,” said CEO Kent Zehetner.
A word that often appears when we talk about advanced analytics is “deep learning.” Can you explain what deep learning is?
When the data volume gets very complex, unorganized, and comes from many different sources it is impossible for humans to comprehend and understand relationships, while machines will be able to process and put the data into context. The technical term for using machine learning in this way is called "deep learning."
Where do people and the human factor have a role? Is StormGeo moving away from being a company whose focus is on its experienced staff towards being an automation, data-driven company?
To classify the dataset, we depend on our experts to transfer their knowledge about the datasets. It is important that our team of data scientists "train" the algorithms so we can apply that learning to the entire dataset. The computer is then able to recognize complex patterns and train the algorithms so they support the decision process – and in some cases automate it.
But even when you have the best automated solutions, the human decision maker will still want to consult with experts when challenging situations are present. That will remain the situation for considerable time. Our forecasters and route advisors are and will be highly valued by our customers.
I had the pleasure to give a lecture to future data scientists at my old institute, Information Science in Bergen. When meeting young and tech-savvy people I always have to run an unofficial poll: You have to choose between your doctor, who at 88% of the time gives a correct diagnosis, or choose an automated online doctor using artificial intelligence that is correct 91%. Whom would you pick? Close to 90% preferred their human doctor. I had expected that to be lower. It has not changed much since I was a student and read about a similar survey in 1995. The customer wants the best of both worlds, just as we want our human doctor to have the best diagnostic tools available today. A captain on a vessel or on a rig in bad weather wants a human expert to discuss the situation.
Can you tell me how StormGeo utilizes machine learning?
Our system works on comprehensive amounts of data. Our essential data is made up of several petabytes. With machine learning we are able classify and recognize new patterns in the data. To illustrate this we can use meteorological data as an example. The meteorological data consists of many layers that are put into a matrix with numbers (e.g. temperature, wind speed, precipitation). If you couple this with historical observation data, we can predict what will most likely happen next.
How does advanced analytics benefit your customers?
Our customers share the fact that their businesses are affected by weather. Through advanced analytics services, StormGeo can help customers gain new insights to make their operations more efficient and automated. If we couple our weather data with our clients data we can help them to find new solutions. The patterns and insight found can help to decrease the fuel consumption cost for a commercial ship. StormGeo’s reporting tool daily collects operational data from clients’ vessels. By analysing the ship and fleet performance, StormGeo can manage fleet efficiency and give owners and operators’ insight to determine the efficiency of their vessels compared to sister ships.
Our team of data scientists will help move our customers' decisions from historical trends and "gut feel" toward data-driven decisions.
“Through advanced analytics services, StormGeo can help customers gain new insights to make their operations more efficient and automated. If we couple our weather data with our clients' data, we can help them to find new solutions,” said Zehetner.
Can you provide examples of how advanced analytics is being applied in different industries?
For shipping, we have developed algorithms for efficient and safe weather routing of ships, which reduced overall fuel consumption by approximately 600,000 tons per year. This is approximately 1.9 million tons of CO2 per year, which is equivalent to a CO2 consumption of 430,000 cars. StormGeo routes approximately 6,000 ships per month and has on-board systems installed on more than 5500 vessels.
In the renewable energy sector, the demand for systems to analyze vast amounts of data is also great. Due to the volatility of renewable energy, particularly wind and sun, this is a far more complex market. Wind force has a major impact on the balance of the European electricity market. Weak wind, in addition to low temperatures, will increase the electricity demand. To secure the power gap, other power sources must be planned to increase production.
StormGeo has models that can predict disruptions in the electricity grid. We can predict, with great probability, when and where there will be potential power outages or grid disturbances so the grid operators’ proactively. By using deep learning and artificial intelligence, we can also improve the wind park energy production predictions.
Contact us to discuss how Advanced Analytics can help your business.