AI and machine learning (ML) in fault delineation, interpretation, and modeling is accelerating prospect identification and reservoir development. Our customer, an IOC in South East Asia, saved 80% of the time it previously took to complete seismic interpretation.
The seismic interpretation tools and workflows embedded with AI and machine learning from Schlumberger are using machine learning for fault identification to significantly improve the domain-science driven workflows. These solutions delivered substantial value to the interpretation loop, rapidly accelerating seismic interpretation time.
Extending the application to a user-driven ML model, with user-defined fault labels, yielded a significant increase in the accuracy and repeatability of the fault prediction. A subset of faults characterizing less than 4% of the entire seismic volume was validated by the seismic interpreter and used to infer the remaining un-interpreted 3D-cube in a few minutes, yielding significant performance improvements.
To understand how much you can rely on computers to do geophysical interpretation, we brought together a panel of geoscientists to discuss the use of AI and machine learning in seismic fault interpretation and fault extraction, during a Living Digital podcast
