1887

Abstract

Summary

An Al-powered geosteering sandbox simulator builds upon a new ensemble-based workflow, featuring a multi-step Al-powered modeling sequence that combines generative geomodeling with machine-learning-based simulation of modern ultra-deep electromagnetic logs.

These models bring high realism and geological complexity while running efficiently on a single computational node. In the open-source web prototype, users can “drill” through generative geomodel ensembles and observe how the Al-driven data-assimilation workflow updates uncertain predictions hundreds of meters ahead of the measurement location. They can also compare their steering intuition to trajectories optimized by a dynamic programming algorithm informed by the uncertain predictions.

The early-access reveal of the fully featured interactive simulator invites early feedback from geoscience practitioners, which can guide the development of the next generation of AI-assisted drilling workflows.

Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202639094
2026-03-09
2026-02-06
Loading full text...

Full text loading...

References

  1. Abdellatif, A., Elsheikh, A.H., Busby, D. and Berthet, P. [2025] Generation of non-stationary stochastic fields using generative adversarial networks. Frontiers in Earth Science, 13.
    [Google Scholar]
  2. Alyaev, S. and Fossum, K. [2025] Distinguish Workflow Demo. GitHub repository: https://github.com/geosteering-no/DISTINGUISH-WF.
    [Google Scholar]
  3. Alyaev, S., Fossum, K., Djecta, H.E., Tveranger, J. and Elsheikh, A. [2024] DISTINGUISH Workflow: a New Paradigm of Dynamic Well Placement Using Generative Machine Learning. In: ECMOR 2024, 2024. European Association of Geoscientists & Engineers.
    [Google Scholar]
  4. Alyaev, S., Ivanova, S., Holsaeter, A., Bratvold, R.B. and Bendiksen, M. [2021a] An interactive sequential-decision benchmark from geosteering. Applied Computing and Geosciences, 12.
    [Google Scholar]
  5. Alyaev, S., Tveranger, J., Fossum, K. and Elsheikh, A.H. [2021b] Probabilistic forecasting for geosteering in fluvial successions using a generative adversarial network. First Break, 39(7).
    [Google Scholar]
  6. Canchumuni, S.W., Emerick, A.A. and Pacheco, M.A.C. [2019] Towards a robust parameterization for conditioning facies models using deep variational autoencoders and ensemble smoother. Computers & Geosciences, 128.
    [Google Scholar]
  7. Chan, S. and Elsheikh, A.H. [2019] Parametric generation of conditional geological realizations using generative neural networks. Computational Geosciences, 23(5).
    [Google Scholar]
  8. Expert PanelAlyaev, S., Antonsen, F., Holbrough, D., Kuvaev, I. and Rabinovich, M. [2026] Where Geosteering is Heading by 2035: Insights from an International Expert Panel and Participant Survey at a Norwegian Workshop. In: Accepted to the SPWLA 2026 UDAR Topical Conference. SPWLA.
    [Google Scholar]
  9. Fossum, K., Alyaev, S., Tveranger, J. and Elsheikh, A.H. [2022] Verification of a real-time ensemble-based method for updating earth model based on GAN. Journal of Computational Science, 65.
    [Google Scholar]
  10. Lee, D., Ovanger, O., Eidsvik, J., Aune, E., Skauvold, J. and Hauge, R. [2025] Latent diffusion model for conditional reservoir facies generation. Computers & Geosciences, 194.
    [Google Scholar]
  11. Tadjer, A., Alyaev, S., Miner, D., Kuvaev, I. and Bratvold, R.B. [2021] Unlocking the human factor: Geosteering decision making as a component of drilling operational efficacy. In: Unconventional Resources Technology Conference, 26–28 July 2021. Unconventional Resources Technology Conference (URTeC).
    [Google Scholar]
/content/papers/10.3997/2214-4609.202639094
Loading
/content/papers/10.3997/2214-4609.202639094
Loading

Data & Media loading...

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error