Search

Do more with less

Cloud-accelerated workflows to automate interpretation of subsurface data, reduce time-to-decision making, and increase understanding of uncertainty and risk.

Rapid data interpretation

The scalable compute capacity of cloud technology helps reduce time to decision and supports:

  • An open infrastructure for the integration, development or automation of data-driven interpretation workflows.
  • Incorporation of very large datasets for a higher resolution understanding of the subsurface and potential risks.
  • Accelerated multi-scenario testing through use of different ML models to help define uncertainty.
Digital model of assisted lithology interpretation
Rapid data interpretation

Spotlight

Optimizing seismic fault interpretation with AI and cloud technologies

The interpretation of faults in 3D seismic data is a critical component of hydrocarbon exploration and development workflows. Faults frequently control factors such as reservoir compartmentalization and fluid migration and may create drilling hazards.

Read spotlight

Expert-informed ML models

Domain knowledge is preserved within trained models​ to support consistent interpretation with reduced human bias across datasets. Users can:

  • Access and utilize pre-trained models
  • Train their own models with their own data.
  • Access insights on model performance and training data for greater confidence in model applicability.

Accelerated and simplified workflows

Workflows can be optimized through the provision of:

  • Easy data access/sharing via cloud-based data sources​.
  • Connectivity to multi-application workflows.
  • Standardised set of ML tools making traditional workflows more accessible to non-specialists.
  • Assisted ML model selection.
  • Workflow automation for repetitive tasks.

Assisted fault interpretation 3D/2D visualization
Accelerated and simplified workflows

Quantifiable confidence limits

Numerical measures of confidence can be assigned to interpretations allowing areas of greater uncertainty to be easily identified. This can be used to help:

  • Determine suitable ML models by measuring the similarity between training and test data.
  • Set uncertainty cut-offs to show only the most confident predictions.
  • Understand alternative possibilities by considering the likelihood of various outputs.

Related products & services

Assisted Fault Interpretation

Assisted Fault Interpretation

Shorten the interpretation lifecycle on seismic volumes of any size.

Explore
Assisted lithology interpretation

Assisted lithology interpretation

Uses machine learning technology to predict lithology in minutes.

Explore
icon

READY TO TAKE THE NEXT STEP?

Talk with a Halliburton Landmark expert

Get in touch