Sharpen your subsurface understanding

The volume of legacy data within the oil and gas industry is vast, with new data generated daily. This data needs to be integrated and interpreted consistently to improve subsurface understanding and reduce uncertainty.

A critical challenge for contemporary explorers and producers is not just to find the data, but find tools and techniques to analyze, interpret, and understand the subsurface. In addition, manual petrophysical interpretation of lithology from wireline data is a time-consuming process. Given the volume of data and limited in-house resources, traditional approaches simply cannot incorporate all the available data. How can you close the gaps in your understanding and reduce subsurface uncertainty?

A dynamic understanding of subsurface stratigraphy

You can address these challenges with DecisionSpace^® 365 Assisted Lithology Interpretation (Figure 1), that enables you to interpret and extract value from your new and legacy well data, and create consistent, high-quality subsurface models on the cloud. Assisted Lithology Interpretation delivers contextual and integrated interpretation using a supervised machine learning (ML) technique that predicts lithology from wireline log responses according to trained models to support rapid and consistent lithology interpretation, and faster decision-making cycles.

Figure 1: Well data can be seamlessly accessed and rapidly interpreted using supervised ML models in DecisionSpace^®365 Assisted Lithology Interpretation.

Assisted Lithology Interpretation offers well-to-well lithology correlation that can aid in better understanding of surface geology and reservoir characterization.

Trained lithological models make the most of your well data

Deliver rapid and consistent lithology interpretations using trained lithological models provided by default within Assisted Lithology Interpretation. These models have been built through close-knit collaboration between data scientists, geoscientists, and petrophysicists. The models have been trained using supervised ML techniques (Figure 2), that incorporate wireline and logging-while-drilling (LWD) data. Algorithms are encoded with intelligence to recognize combined features in well log curves and quantitatively assess the likelihood that these represent a particular lithology, based on previous examples seen by the system. The F1-score, a statistical measurement of model performance, was used, alongside other metrics, to assess the accuracy of the predictive models generated.

Figure 2: Model Training and Prediction Pipelines used in DecisionSpace^® 365 Assisted Lithology Interpretation. A library of pre-trained models exists within the application, ready to be used for interpreting datasets.

Custom models trained using proprietary data and interpretations can also be used within Assisted Lithology Interpretation. Custom models can be trained for you, or an integrated model training workflow (Figure 3) can be used by your geoscientists, petrophysicists, or data scientists within DS365.ai, a cloud-native, open architecture platform with MLOps capabilities, to easily train models.

By applying the same model across your dataset, this level of consistency helps to better constrain reservoir and seal lithofacies, and overburden, and helps you to identify potential bypassed pay. All of this can be achieved in just minutes—across hundreds, if not thousands, of wells – and can enable you to extract more value from your data in a fraction of the time and at a fraction of the cost.

Figure 3: Model training pipeline in DS365.ai, that can be used to train custom ML models with proprietary data and interpretations. These models can then be used to make predictions in DecisionSpace^®365 Assisted Lithology Interpretation.

Consistent, connected, quality data interpretation

Trained algorithms enable standardized processes to be implemented and help reduce interpreter bias, thus delivering consistent data interpretation across your enterprise. The system seamlessly connects with and saves lithological data as computed lithology against the well. This allows workflows to continue without the need to import or export. It also allows you to track and access previous runs, which you can use to monitor historical work and retrieve stored output.

You can instantly connect with and collaborate with your peers, wherever they are, with immediate shared access to consistent lithology characteristics for thousands of wells and the most current subsurface understanding.

Seamlessly integrate interpretation data in just minutes

Open architecture enables you to seamlessly connect and integrate with Data Foundation, the DecisionSpace^® 365 data platform, plus OpenWorks^® data management software and other industry-standard databases, as well as your own data lakes, giving you choice and flexibility.

The use of algorithm interpretation also means you can conduct multi-scenario analysis on the same datasets to aid lithology prediction and reduce subsurface uncertainty. The ultimate outcome is an accelerated decision-making cycle.

Understand uncertainty quantitatively

Built-in confidence measures apply quantifiable confidence limits to your data. The uncertainty of lithology predictions is captured and tracked throughout the interpretation workflow on a well-by-well basis by calculating a percentage of confidence at any point.

These calculations include Prior and Posterior Probability measures, which allow you to analyze the confidence outputs and make more informed recommendations. That makes for smart investments.

Make the most of your assets

Unlike conventional approaches, using a supervised ML pipeline to predict lithology from wireline or LWD data based on trained lithological models enables consistent interpretation from surface to total depth (TD). You can gain a high-resolution understanding of reservoir and seal lithofacies, overburden characterizations, and potential bypassed pay in just minutes—across hundreds, if not thousands, of wells. 

Assisted Lithology Interpretation’s consistent, rapid, and high-fidelity lithological predictions, plus quantitative uncertainty, means your organization’s human expertise can be applied where it is most critical: making decisions.

Figure 4: Locations of some training wells used within ML models provided by default in Assisted Lithology Interpretation.