This presentation showcases a practical use case of combining Agentic AI with tNavigator for history matching. It explores how AI-driven agents can support reservoir engineers in day-to-day workflows - enhancing efficiency, accelerating analysis, and strengthening insight generation for better decision-making.

The case studied field comprises 13 stacked reservoirs over a gross thickness of approximately 600 m and is affected by a complex, non continuous and Y shaped fault system. This complexity prevented the construction of a single pillar grid suitable for dynamic simulation. An alternative modelling workflow using three separate geological models, stacked and dynamically coupled in tNavigator, enabled successful model initialization and history matching under commingled production conditions. The integrated model is now used to support robust field development planning and scenario evaluation.

The gas lift optimization design focuses on improving artificial lift performance of the well, by designing and optimizing gas lift systems to enhance well production efficiency and maximizing recovery.

To better understand the geology of the oil field, a reevaluation study was conducted using the Seismic and Geology Desginer tNavigator. The study began with the reinterpretation of well logs, faults and horizons corresponding to major geologic formations within the area. This was followed by a detailed analysis and integration of seismic attributes (RMS, Sweetness, and Spectral Decomposition) using stratigraphic slices to delineate sand presence and geometry. This analysis improved the understanding of the reservoir architecture and provided new insights into the geological evolution of the field. As a result, the study helped define and prioritise future drilling targets.

Deep geothermal projects in the Upper Rhine Graben commonly target naturally fractured reservoirs associated with major fault zones. However, fracture permeability is highly heterogeneous, and identifying optimal drilling targets remains a key challenge for successful project development. In this context, informed and data-driven targeting decisions are crucial. We present an integrated workflow to identify potential drilling targets using a composite, multi-proxy approach to reduce the existing uncertainties and support more confident development decisions. The methodology combines geomechanical, structural, and lithological indicators to derive a likelihood property for enhanced structural permeability. This composite parameter enables the identification of geothermal subsurface sweet spots, providing a foundation for defining potential well trajectories inferred from limited available subsurface data. Further well placement optimization is driven by taking dynamic geothermal processes into account while simulating the geothermal reservoir behavior. Results are essential to accurately forecast the geothermal energy production.

The Bergknapp discovery is located on the Halten Terrace in the Norwegian sea, approximately 8 km from the Maria field. The discovery well drilled in 2020 proved hydrocarbons in multiple stacked reservoirs of Early to Middle Jurassic age (Garn, Ile and Tilje formations). Drill-stem tests (DST) were performed in the Garn and Tilje formations providing valuable dynamic information. The discovery well was sidetracked, and an up-dip appraisal well was drilled in 2023, providing additional reservoir data. The well data has shown significant vertical and lateral variations in reservoir quality, scatter in reservoir pressure trends, and large spread in oil-water contacts depths.

Multi-realisation (ensemble) method was chosen for reservoir modelling and development planning to capture subsurface uncertainties, reflect observations from the data and ultimately make sound development decisions. A modular automated workflow was developed in tNavigator Model Designer (MD) comprising all steps of static model construction and dynamic model initialization. Subsurface uncertainties were modelled by varying the workflow input parameters in the Assisted History Match (AHM) module. One-at-a-time sensitivities were run to identify key uncertainties. The range of uncertainty for key metrics (STOIIP, ultimate recovery) was quantified by running a large number of model realisations. DSTs in the Tilje formation were simulated to calibrate dynamic models. Natural depletion and water injection concepts were simulated under uncertainty, providing statistical ranges of ultimate recovery and forming a basis for development decisions. The presentation will focus on the modelling workflow, results interpretation, and learnings from using tNavigator for development planning under uncertainty for a complex discovery.

Indonesia’s oil and gas industry remains vital for national energy security and economic sustainability, yet its future will depend heavily on the strength of its people, technology, and institutional collaboration. In this context, IATMI (Ikatan Ahli Teknik Perminyakan Indonesia – Indonesian Petroleum Engineering Association) serves as the nation’s leading professional organization representing oil and gas engineers and practitioners. This presentation highlights how IATMI can play a more impactful role in shaping the industry’s future by strengthening human capital, promoting innovation and knowledge-sharing, supporting constructive policy dialogue, and fostering collaboration across government, operators, academia, and service providers. By leveraging its extensive network and technical expertise, IATMI is well positioned to help ensure Indonesia’s oil and gas sector remains competitive, resilient, and aligned with the country’s long-term energy transition objectives.