Algorithmic trading:
Role reversal

Knowledge is crucial to executing a good trade in any market. Statkraft uses artificial intelligence and machine learning to be able to trade quickly in the energy market.

Digitalisation changes the way we work in all areas of our working lives, and the power market is no exception.

"People and machines have simply switched roles. Up to now, people have done much of the operational work themselves; sitting in our offices, we punched in prices day after day. By retrieving market data, we can now create trading strategies driven by algorithms. Then these algorithms can tell us, for example, when we should buy power in Germany or the Nordic region, or gas in the UK," says Arne Jørund Haugland, head of the digitalisation and algorithmic trading unit in Statkraft's Markets and IT business area.

The emergence of renewable power sources such as sun and wind has made the energy market more weather-dependent and thus more unstable.

"As a result it's much more difficult to predict how much power will be produced in certain time periods. In a short-term market like this, there is a greater need to be able to make quick decisions, and machines are better suited to that than people," he says.

(Video: Oda Hveem)

Curious and creative

Haugland's digitalisation team combines the technology of algorithmic trading with artificial intelligence and machine learning to fully automate the trading process.

"It means that the machines not only tell us what to do and when, but they also make the decisions for us. Our role is thus to monitor the systems and develop them further," he says.

Algorithmic trading makes energy trading faster and more efficient, which has its advantages. Imagine an industrial company that has received a large order for aluminium, enough work to keep the plant in operation for a whole year.

"The company is dependent on purchasing the power needed to keep operations running. Previously, they called their power supplier every day and told them how much power they would need that particular day, based on planned production. Today they can use an algorithm that will not only determine how much power the plant will need from day to day, but that will also go to the market, buy the power and then report back on its results," Haugland explains.

Such a model can be applied to all players in the market, whether they sell or buy power.

"We must constantly monitor and develop the systems, so the machines are not replacing us. We just have to work in a different way," emphasises Haugland.

Illustration of different energy sourcesRenewable energy sources such as sun and wind are affected by weather conditions, making it more difficult to predict the amount of power that will be produced in certain time periods. In such an unstable and short-term energy market there is a greater need to be able to make quick decisions, and machines are better suited to that than people. (Photo: Shutterstock)

Endless possibilities

Another advantage of using algorithms is that you can easily get a bigger picture.

"What was previously often limited to a local market can be scaled up to include the entire global market. If you want to establish yourself in a new country, you no longer have to hire as many people – the algorithms can interpret the market for you. In less efficient countries, fast machines can create value that will reduce costs for both the company and the end user in the market," he says.

Digitalisation is not about drones, artificial intelligence and the Internet of Things, but about having a culture of applying new technology.

"The technology we use in algorithmic trading was originally developed for the finance industry and then transferred to the energy market simply because it happened to work there too. The fact that the same technology can be used in multiple places makes the spectrum incredibly wider and the possibilities endless. You don't need to reinvent the wheel each time; instead you can find new ways to apply it," says Haugland.

Trading room at Statkraft's headquarters in OsloDigitalisation, algorithms and artificial intelligence make energy trading faster and more efficient. The picture is from the trading room in Statkraft's headquarters at Lilleaker in Oslo. (Photo: Statkraft)

Trial and error

His team is also working on developing business examples to test how the systems can work even more efficiently. Interdisciplinary collaboration is key.

"Since the power market is regulated, we work with lawyers in the compliance department, and since we use advanced computer technology, we are dependent on close collaboration with the IT department. The analysts help us understand how the market works, and the traders provide input on how machines can perform even more tasks. Our job involves a lot of trial and error, of getting new system versions running as quickly as possible and using the feedback to develop the systems further," he says.

Applying new technology to working life and life in general is nothing new, he points out.

"This is how humanity has evolved all the time. Just think of agriculture, where farmers have gone from sowing by hand to using the horse and plough to the current situation where machines do most of the job. The combination of people and machines is not revolutionary; what is special now is that developments are moving infinitely faster than before."

Haugland believes that the technology anxiety sometimes expressed in the media is exaggerated.

"It's a natural psychological reaction to be sceptical of change, but if we are able to be open about what technology can do for us, we will see that current developments benefit everyone. I think technology helps people to truly develop themselves – using the amazing human brain for boring, repetitive tasks is a waste of time. Let the machines take care of those, so we can do things that are more fun!"

Arne Haugland av Technoport 2019Arne Jørund Haugland was among the speakers at the Technoport conference in Trondheim, Norway, on 27 March 2019. This year the conference was given the title "Nordic Deep Tech". More than 1,000 innovators meet every year in Trondheim to explore an exciting world of technological innovations. (Photo: Ole Martin Wold)

Show more

Related articles