Machine learning to improve NCS efficiency

The Norwegian continental shelf is a global leader in low CO2 emissions, and significant cost cuts have made the NCS competitive. But Norway cannot take the competitive position for granted - we must constantly improve through the introduction of new technology and new ways of working.

– The use of machine learning on the NCS is in its infancy. Applying machine learning to historical data could automate a lot of work that until now has been done by humans. What used to take two months can now perhaps be done in two days, says Chairman of the board in OG21 and Vice President Research & Technology in Equinor, Stein Olav Drange.

There is an increasing activity in the development and testing of machine learning within many subject areas.

- Machine learning is becoming part of the industry's toolbox. For some problems, machine learning will be the preferred tool, while for other problems, theoretical models and practical testing will still give us the best decision basis. The toolbox will improve if we dare to test out machine learning more, Drange continues.

In the report, OG21 provides several recommendations to industry and industry organizations on how to better succeed with machine learning. OG21 believes success is closely connected to increased collaboration - collaboration on data, on technology solutions and on knowledge and competence.

OG21's project report and the DNV GL report which it is based on, were presented at the OG21-forum, November 11, 2020.