Integrating QI Automation and ML Workflows for Accelerating Subsurface Rock-Properties Prediction

Over the last years, the geophysical applications have been transitioning from human-intensive traditional workflows to Automated and Machine Learning (ML) based approaches supporting the geoscientists in the evaluation of the subsurface challenges. The industry goal is to accelerate the exploration, primarily near existing infrastructure, also known as infrastructure-led exploration (ILX), as well as the field development planning and decisions. Hereafter, we discuss the considerations emerged from a collaboration project between OMV Norge and SLB with the objective of establishing maturity and value potential of Automated and ML algorithms for exploration workflows to enhance geological understanding on reservoir targets in mature acreage of the North Sea.

The implemented approach appears to have a promising potential to deliver robust rock property prediction compressing the execution timeline and exploring multiple scenarios to capture the subsurface uncertainties.