As data continues to be amassed via ongoing operations and analyses, discovery of new backlogs of data and acquisition of new data, the need for an orchestrated and automated process transforming and synchronizing these datasets across applications and repositories into OSDU becomes significantly important. This work presents an evergreen data solution developed to address these challenges by ensuring that both newly generated (and/or discovered) and historical data are effectively curated, integrated, and utilized to support business objectives. At the core is a flexible, event-driven Consumer that connects and listens to changes across diverse data sources, platforms, and storage locations available within the operating environment. Detected changes initiate end-to-end orchestration of automated file classification and modular data transformation, contextualization, and ingestion into OSDU. The transformation modules perform an automated ETL process of pre-mapped files, application datasets, and structured databases into domain data models, ensuring full lineage, governance, and traceability maintained. The initial modules focus on interpretation data sourced from Petrel—such as horizons, faults, and wellbore markers—where changes occur continuously within the application by end-users and must be captured, technically assured, and integrated with other data in the platform. The architecture is designed for continuous extension, enabling the integration of new sources, transformation modules, and functional components over time, so that newly acquired and discovered datasets remain live, reliable, and readily accessible across OSDU-native applications and workflows. The first deployment demonstrates how the approach can move rapidly from concept to operational use, while also guiding its ongoing evolution.

As industries scale AI and data-driven operations, the real challenge is no longer data integration, but the ability to share meaning consistently across systems and over time. Yet most so-called “ontologies” remain limited to taxonomies, schemas, or loosely structured graphs, failing to capture and govern meaning, and ultimately reproducing silos. This talk argues that true interoperability is not about connecting systems, but about aligning meaning under explicit and controlled conditions. Without formal semantics, interoperability degrades into mappings and transformations that are fragile, non-scalable, and semantically ambiguous.

Building on industrial deployments in complex engineering and operations contexts, we present a formal ontology stack combining Top-Level Ontologies (e.g., BFO: ISO 21838-2, 2021), industrial core ontologies (IOF), and domain business objects aligned with standards such as CFIHOS. The key differentiator lies in axiomatized semantics grounded in firstorder logic, enabling machine-verifiable constraints, consistency checking, and cross-domain reasoning.

However, technology alone is insufficient. Interoperability requires governance of meaning: formal rules to define, validate, version, and reuse semantic models; business objects treated as controlled semantic assets; and standards integrated at the semantic level: not just referenced. In a world where AI models are increasingly commoditized, competitive advantage shifts to the ability to formalize and govern knowledge. Organizations that control their semantic layer establish a shared human–machine language, enabling reusable knowledge, scalable interoperability, and explainable AI. This approach transforms knowledge into a persistent, governed, and platform-independent asset, ensuring long-term consistency, reducing technical debt, and enabling open, standard-based interoperability.

Artificial lift systems generate large volumes of operational data across ESP, gas lift, rod pump, and other production technologies, yet many optimization workflows remain reactive and dependent on static models, historical analysis, and manual intervention. As a result, performance issues are often identified only after production losses occur.

This session explores how physics-informed artificial intelligence can transform artificial lift optimization by combining real-time operational data with engineering-based physical models. Unlike purely statistical approaches, physics-informed AI incorporates pressure-flow relationships, pump performance behavior, and reservoir interactions to create dynamic digital representations of production systems that better reflect real-world operating conditions.

Greg Fallon will discuss how these hybrid AI models enable earlier detection of equipment degradation, improved forecasting of system performance, and optimization of operating parameters across interconnected wells and production networks. The presentation will also examine the data and information management challenges associated with deploying AI at scale, including data integration, interoperability, model governance, and operational adoption.

Through practical examples, attendees will learn how AI-enabled optimization can support more proactive decisionmaking, extend equipment run life, reduce unplanned downtime, and improve production efficiency. The session will provide insights into how production data can be transformed into predictive, decision-ready intelligence that supports more resilient and data-driven operations across the energy value chain.

Energy companies hold decades of legacy subsurface data – acquired at significant cost but often trapped in inconsistent formats, disconnected from modern databases, or stored on aging physical media. With rising energy demand and increasing acquisition costs, recovering value from this existing data is a strategic priority. This presentation shares practical experience from two complementary projects tackling different facets of the legacy data challenge with a major operator’s data quality programme.

The first project addressed the identification problem: thousands of legacy seismic survey records with inconsistent naming conventions, missing identifiers, and format variations could not be reliably matched to the master database. Manual matching yielded accuracy in the single-digit percentages. An iterative approach was adopted – combining fuzzy string matching algorithms with systematic ID normalization rules. This pragmatic approach delivered a step change from single digits to near-complete match accuracy – without even requiring generative AI.

The second project tackled the verification problem: hundreds of physical data tapes required full reconciliation against a multi-million-row digital archive prior to safe decommissioning. The tooling evolved from naïve linear scanning to hash-indexed lookups, establishing efficient large-scale verification across hundreds of tapes and ensuring no data was lost before physical media disposal.

Together the projects demonstrate a full legacy data lifecycle: first establishing what you have, then confirming it is all accounted for – enabling confident migration to modern platforms and responsible decommissioning of physical media. Key takeaways include the power of well-engineered deterministic methods before reaching for AI, the importance of domain expert validation alongside automated pipelines, and the transferability of these approaches to similar legacy data challenges across the energy sector.

The legislation that regulates recording and use of bathymetric data or information about the seabed has been updated and is not well known among some stakeholders. This presentation is primarily aimed at those who map bathymetric data or collect any type of information about the sea bed. The presentation will provide an overall introduction to the new regulation and what it entails, what type of mapping activity that requires an application or report, how the actors should relate to the legal framework and how to apply, how to get access to Kartverkets batymetric data, as well as a demo of our new digital application service. We belive this presentation will be useful for ensuring compliance with the regulation, while also contributing to safe and efficient development within the industry.

We introduce a federated European innovation ecosystem for subsurface data and intelligence, enabled through use case orchestration as a coordination mechanism, to address fragmentation in AI‑enabled subsurface innovation without centralisation or loss of national or organisational autonomy. Rather than creating new institutions or shared operational platforms, the proposed ecosystem establishes a joint innovation capacity that enables organisations to collaboratively develop reusable capabilities while preserving national mandates, infrastructures, and data sovereignty. The novelty of the approach lies in combining a federated ecosystem model with use case orchestration as the primary means of alignment. Concrete subsurface use cases, rather than projects or programmes, form the unit of coordination across organisations. This enables early identification of shared needs, data and technical dependencies, and reuse opportunities, while allowing each organisation to deploy and operate solutions independently within its own context.

Within the ecosystem, participants jointly develop shared components, standards, and AI‑enabled workflows that strengthen individual organisational capabilities while improving interoperability and coherence at European scale. Practical applications demonstrate that use‑case‑driven orchestration significantly reduces duplicated research and innovation effort, enables effective co‑funding, and accelerates the delivery of high‑impact subsurface intelligence capabilities.

By providing an operationally feasible alternative to both isolated national initiatives and centralised European solutions, the proposed ecosystem offers a scalable and governance‑compatible pathway for coordinated subsurface data and intelligence innovation across Europe.

The oil and gas industry continues to digitise subsurface data, yet a significant portion of high-value information enriched by subject matter experts remains trapped in unstructured formats such as scanned reports, well documents, and legacy PDFs. While the OSDU® Data Platform standardises structured datasets, unlocking value from unstructured content requires both semantic enrichment and rigorous, enterprise-grade security.

This presentation introduces a scalable, entitlement-first Retrieval Augmented Generation (RAG) architecture developed in collaboration with a major operator, transforming unstructured OSDU-referenced content into actionable, contextaware intelligence. The approach combines document reconstruction logic and header-aware chunking with a hybrid retrieval strategy, integrating vector search and keyword-based methods to improve retrieval quality. Evaluation on a curated set of subsurface-specific questions is used to guide the tuning of key system parameters and quantify performance improvements.

A key architectural principle is the enforcement of access control and watermarking through mapping user identity to OSDU Access Control Lists (ACLs), ensuring that only authorised data is retrieved and that both outputs and source documents remain traceable. The solution is designed for seamless integration into existing subsurface workflows, allowing users to access OSDU-sourced insights directly within their daily tools without introducing additional friction, thereby reducing the time spent on searching and validating data.

Beyond retrieval, the architecture establishes a foundation for future agent-based workflows, where AI systems can reason over both structured and unstructured data while maintaining full compliance with enterprise security requirements. The presentation highlights key lessons learned, challenges encountered, and the practical considerations required to scale the solution in an enterprise setting.

The Energy Industry is entering a transformational era in how data is delivered, governed, and leveraged to make business decisions. At the center of this transformation is OSDU, the industry’s open, vendor-agnostic trusted data ecosystem designed to preserve and propagate the context of trust and uncertainty across applications, workflows, technologies, as well as across domains, scientific and technical disciplines, and organizational structures. OSDU demonstrates that standardization is not a constraint on innovation, but rather an enabler of trust, interoperability, scalability, and accelerated learning. As organizations increasingly adopt Artificial Intelligence, Machine Learning, and advanced analytics in their decision-making processes, the need for trusted data with embedded quality, usability, lineage, and purpose has become essential - because “garbage in, garbage out” still holds true, even in AI￾driven environments, particularly when data quality is poor, unknown or when input lacks sufficient context. In modern decision ecosystems, false confidence derived from poor-quality or context-deficient data is significantly more dangerous than acknowledged uncertainty – uncertainty that is understood, quantified, managed, and explicitly considered in decision-making processes.

Our technical presentation advances the concept that trust must extend beyond data accuracy to include contextual dimensions of quality and utility, enabling both human intelligence and artificial intelligence systems to learn, reason, and make decisions with measurable confidence — where those decisions are traceable and auditable, allowing continuous learning through analysis and reflection on past data, processes, and outcomes. We will showcase the foundation for modern decision ecosystems where standards improve decision quality, reduce uncertainty, enable interoperable data within a shared trust context, and accelerate enterprise learning from business decisions at scale.

In geotechnical engineering, the challenge is often not a lack of data, but the ability to find, access, and integrate it across workflows. This is particularly evident in offshore projects, where ground models depend on combining seismic data (2D and 3D) with large volumes of in-situ geotechnical measurements and laboratory results.

Field Manager is a cloud-based platform that supports the full lifecycle of geotechnical survey data, from collection and quality assurance to visualization. In this talk, we present recently developed data structures and workflows for linking geotechnical data (0D/1D) with spatial references to seismic data (2D and 3D), enabling these datasets to be explored and interpreted in context. The focus is on the design and implementation of the underlying data model, and on how to balance flexibility and structure when building systems for complex, multi-source data.

Plug & Abandonment (P&A) represents a major technical, operational, and regulatory challenge for mature oil and gas provinces such as the Norwegian Continental Shelf, where large inventories of aging wells must be safely decommissioned under stringent environmental and regulatory requirements. A key obstacle to efficient P&A execution is the limited accessibility and quality of historical well information, which predominantly exists in unstructured legacy formats, including scanned reports and handwritten notes stored in heterogeneous file types (PDF, Word, PowerPoint). These limitations extend planning timelines, introduce uncertainty, and increase the risk of costly reabandonment. This paper presents a Generative AI driven solution that transforms legacy well documentation into a structured, engineering grade digital data foundation built for P&A workflows. The approach enables automated extraction and standardization of P&A relevant data and supports generation of consistent digital well schematics to inform risk based P&A design and regulatory compliance.

Extracted data is curated, quality controlled, standardized, and ingested into subsurface and engineering data platforms, ensuring traceability to source documents and enabling automated generation of P&A ready digital well schematics. The methodology was validated through a P&A case study involving more than 12,000 pages of multilingual legacy documentation across 9 wells from the Netherlands and the Gulf of America. Tasks that previously required 3 to 4 weeks of manual effort per well were completed in less than 2 days while maintaining high extraction accuracy. For operators and regulators, this GenAI enabled approach supports faster historical well assessments, earlier identification of integrity risks, standardized P&A decision making, and increased confidence in regulatory compliance, while converting legacy well data into a reliable, engineering grade foundation for risk based P&A execution.

Upstream oil and gas built its digital landscape application first, and every discipline ended up with its own tools and its own legitimate definitions of the same real-world objects — the same wellbore carries different names, identifiers, and boundaries depending on which system you ask. OSDU inverts that pattern by sharing the data instead of integrating the applications. Every application now has to answer the same question: how should I engage with the platform?

When a petrophysicist makes an interpretation from a log, what belongs in OSDU and what stays in the application? What should be written back, with what context, and linked to its sources how? And who is responsible for access, legal context, lineage, and quality — the application that creates the data, the one that uses it later, or a separate governance role?

OSDU provides rich machinery and guidance but stops short of enforcement: required content is minimal, several mechanisms can express the same idea, and responsibility between platform, operator, and application is not clearly drawn. The forum is openly candid that integrity depends on disciplined use. Without a shared protocol, the inversion OSDU was built to deliver is undone one application at a time.

This presentation argues that the industry needs to settle on shared rules of engagement, and opens the conversation — with concrete positions where the answer seems clear, and guidelines where it does not. The aim is to make explicit the discipline the platform assumes but does not impose, so it can be discussed, agreed, and adopted across organizations.

Presented jointly by Equinor and Tieto Tech Consulting, drawing on experience from building SMDA as a subsurface master data system and the current work of extending that effort onto OSDU.

Offshore exploration, site characterization, and inspection surveys are increasingly executed through remote operating models. The goal is to enable distributed teams to make timely, informed decisions from a shared operational picture.

In practice, this remains challenging. Operational data and acquired geodata are often fragmented across systems, updated asynchronously, and interpreted differently by offshore crews, onshore teams, and clients. The result is decision latency, lower confidence in remote risk assessment, and greater exposure to reruns, incomplete acquisition, and nonproductive vessel time.

Fugro Norway has addressed this in live offshore campaigns through an integrated operational and data management approach. Using ESRI-based dashboards, teams combine vessel position and track history, survey line status, weather forecasts, live video feeds, and near-real-time seabed visualizations in one operational view. This allows deviations, data gaps, and quality issues to be identified while acquisition is underway.

Building on this foundation, VirGeo® extends these capabilities through a connected data management environment acting as a single source of truth. In addition, VirGeo® provides connected data exchange, near real-time monitoring, operational data analytics, and geodata insights and analytics. By aligning operational context with incoming geodata in a decision-ready form, validation and reconciliation move closer to acquisition and reduce late manual verification and fragmented handovers.

This shortens decision cycles, reduces rerun exposure and non-productive vessel time, and speeds first usable data review. As offshore operations become more remote and make greater use of USVs, this shift from data availability to data alignment strengthens situational awareness and reduces operational uncertainty.

Operational reporting is some of the highest-volume data a drilling operation produces, and some of the least governed. Daily activity remarks are entered by hand under time pressure, in free text, with missing dates, inconsistent depths, merged operations, and activity codes that are frequently wrong or missing. The result is a record that is hard to standardize, hard to compare across wells, and hard to load into a corporate data platform.

We built a set of grounded GenAI agents that turn this raw operational narrative into standardized, OSDU-compliant well operations records, piloted with Equinor FLX across four North Sea wells (a representative well file holds around 175 operational rows spanning roughly a month). The agents reconstruct a coherent chronological daily report from scattered remarks; ground each activity to the correct code within a three-level controlled vocabulary of roughly 3,700 codes (99 main activities, 590 sub-groups) using retrieval and a confidence score; and derive risk-aware performance targets from the statistical distribution of historical execution data instead of subjective benchmarks. The output maps directly to OSDU WellOperationsReport 2.1.0, verified against OSDU reference-data manifests.

The information-management lesson is that text generation is the easy part.The hard parts are binding model output to a controlled vocabulary, quality-assuring records that an AI wrote, and proving they conform to the schema and reference data the business depends on. We share what worked, where output had to be checked and corrected (including a cautionary case of fabricated reference values), and a repeatable path from freeform human reports to governed, queryable records. In the pilot, once the inputs are clean, report generation runs in one to two minutes, against the hours a night supervisor would otherwise spend reconstructing the day by hand.

As energy companies modernise their data estates, one challenge remains: unlocking value from legacy Petrel project data. While the OSDU™ Data Platform provides a cloud-native, vendor-agnostic foundation for data access and reuse, Petrel data is often locked inside proprietary project structures that are difficult to search, govern, and integrate beyond the desktop.

This presentation shows how Petrel projects can be transformed into OSDU-ready data through automated extraction, enrichment, and metadata normalisation. Using domain-aware tools such as Blueback Project Tracker and purpose-built extraction pipelines, large volumes of Petrel projects can be scanned, assessed, and prepared for ingestion. The session highlights common challenges, including version inconsistencies, incomplete legacy archives, embedded unstructured content, and missing spatial metadata, and explains how validation, enrichment, and reference-data mapping improve quality and alignment with OSDU domain models.

A central theme is the shift from treating Petrel projects as static containers to viewing them as structured sources of reusable domain data. By breaking projects into independently managed data objects, organisations can enable lineage, governance, cross-project comparison, and more collaborative cloud workflows.

The talk shares practical lessons from real-world deployments, including archive scanning, project prioritisation, ingestion integration, and quality assurance feedback loops. Attendees will leave with a clearer view of how to approach legacy-to-cloud transformation at scale and turn trapped project data into governed, discoverable, cloud-ready assets.

Managing OSDU/ADME at scale requires visibility across governance, security, and data flows. Existing tools target developers. We built a custom UI layer for non-technical stakeholders: data stewards, platform operators, decisionmakers who need situational awareness without CLI access or code. This enables faster onboarding, stronger compliance visibility, fewer operational gaps. We'll share the design patterns that work for technical platforms and show how making information accessible shifts adoption and governance outcomes.

To ensure alignment with Aker BP strategy to be the leading exploration and production company of the future and given the size and complexity of the Yggdrasil project, it quickly became clear that traditional approaches wouldn’t be sufficient. To address these challenges, we’ve adopted new, more digital ways of working, which have helped us streamline processes and improve overall efficiency.

A key aspect of our new approach involves making extended use of data. By leveraging dashboards, we can access, control, and quality check information in real-time, ensuring that our data remains both accurate and actionable. Additionally, we’ve incorporated AI tools to further enhance our capabilities, enabling us to handle large volumes of information and make smarter, faster decisions.

These digital advancements have positively impacted our workflow, allowing us to keep pace with the demands of the project and maintain high standards of information management.

Using Google Cloud's Gemini models with the Open Subsurface Data Universe (OSDU) platform.

* Multi-Agent Offset Well Analysis: This AI architecture is used to predict hydrocarbon production and run optimization scenarios for future field development based on multi-disciplinary datasets (geoscience, well design, operations). It uses a multi-agent architecture to orchestrate sub-agents: an Analytics Agent for data analysis and summarization, a Plotting Agent for generating Python data visualizations, an ML Agent for making predictions and evaluating ML models, and an Optimizer Agent for optimizing against constraints.

* Strata Scanner: An AI-driven multimodal intelligence solution designed to digitize dormant, unstructured historical data—such as well log images in .TIFF format—into modern OSDU formats in minutes.

* Automatic Schema Transformation: An interactive, conversational AI tool that helps users pipeline semi-structured database data into OSDU formats.

* Agents for Search & Automation: The integration deploys agents grounded in OSDU enterprise data to unlock conversational data access, allowing users to efficiently summarize complex technical documents, analyze large datasets, and automate workflows across the OSDU ecosystem.

The Geological Survey of the Netherlands (TNO-GSN), acting as the statutory delegated data custodian, plays a central role in management of data that must be provided to the state under the Dutch Mining Act. In this context well statuses like ‘drilling’, ‘closed-in’ and ‘plugged and abandoned’ are reported to TNO-GSN. Triggered by the European methane regulation and the need for consistent aftercare of wells, a comprehensive project focused on restructuring the definition and management of well statuses was undertaken. Years of evolving regulations, data standards and management perspectives had resulted in unclear terminology. This lead to inconsistent application of well statuses among various parties which in turn hindered proper data management and usability. To rectify this situation, seven well statuses were defined through systematic information analysis of international standards of the PPDM, requirements of the Dutch Mining Act and internal workflows. These seven well statuses now have clear definitions, assignment and data management criteria, and hierarchy rules. The legacy statuses were mapped to this new set and updated accordingly, removing redundant or ambiguous entries. Definitions are now publicly accessible via the TNO-GSN data portal NLOG, supporting transparency and interoperability between the various organisations active in the Dutch subsurface and TNO-GSN. Additionally, a semi-automated ETL process was implemented to support a consistent work process. The new well status definition framework streamlines data management, minimises miscommunication and facilitates effective knowledge sharing. This work can serve as best practice for organisations across the energy sector facing similar challenges related to metadata management or setting definitions and standards within existing data sets.

Data with geospatial context is the essential connective tissue linking subsurface and energy workflows, enabling efficient decision-making, operational efficiency, and supporting sustainability objectives. Despite its importance, handling geospatial data remains still one of the most complex elements within modern data ecosystems. When interoperability breaks down, the consequences may be rather silent but far-reaching, affecting data quality, workflows, and data platforms - including OSDU - and ultimately compromising the effectiveness of analytics and AI outcomes.

This presentation delves into the causes and failure points behind these geospatial interoperability challenges. Using seismic position data and metadata as an example, current data workflows - including data quality considerations - will be examined, along with their impact on diverse stakeholders, ranging from data and solution architects to end-users reliant on the data.

Attendees will gain insight into risks within geospatial workflows. Additionally, strategies to prevent the compromise of AI-driven insights, along with practical approaches to ensure data integrity, interoperability, and readiness for analytics, decision-making, and sustainable operations, will also be discussed.

This presentation showcases Equinor’s initiative to strengthen well data reporting by taking ownership of and improving Standard Operating Procedures (SOPs) with Service providers.

Through extensive insight work, including interviews with more than 50 stakeholders across disciplines and companies, we identified key pain points across the data lifecycle. These included poor communication, insufficient quality-control routines, inconsistent file structures, unclear task ownership, and excessive iteration cycles in end-of-well (EOW) deliveries and authority reporting.

The findings highlight systemic challenges throughout the process including lack of standard templates, insufficient metadata governance, fragmented workflows, and limited transparency in data sharing and version control.

To address these issues, Equinor has established a comprehensive SOP framework with clearly defined requirements, standardized data formats and workflows, improved QC processes, better role clarity, and enhanced collaboration between all stakeholders. New elements include structured data delivery overview, standardized plots and metadata, shared platforms, defined timelines, and aligned QC responsibilities (Equinor, LogQC and Service providers).

The overall objective is to enable “correct data delivered the first time,” reducing rework, improving efficiency, ensuring compliance with Equinor and authority requirements, and ultimately making well data reporting simpler, faster, and more reliable.

As organizations invest in OSDU to modernize subsurface data management, a key question remains: how well does the platform perform with real, complex data—and what does that mean for business value?

This presentation shares practical insights from testing the OSDU data platform using actual well log datasets, including high-volume array data. The work focuses on understanding how platform capabilities translate into operational efficiency, scalability, and usability in real-world scenarios.

We explore the end-to-end lifecycle—from ingestion to consumption—highlighting the impact of data complexity on throughput, data accessibility, and downstream workflows such as interpretation and analytics. The session will demonstrate where OSDU enables faster access to subsurface insights, and where challenges—such as ingesting large datasets, managing complex schemas, and retrieving array-based data—can introduce friction and cost.

By connecting technical findings to business outcomes, this talk provides actionable guidance on how to optimize data onboarding strategies, improve user experience, and reduce time-to-value when working with subsurface data in OSDU.

Critical information for the evaluation of carbon storage prospects is held in the well abandonment documentation for historic wellbores. But extracting that information is a slow, and laborious exercise. Can LLMs do it better, faster, and cheaper? The NSTA and Flare Solutions have been working together to find out. We will share the outcome of work across over 1,000 well abandonment documents to extract cement plug and casing cutting parameters critical to understanding store integrity risk and the possibility of wellbore remediation. We'll also share what we've learned along the way about using commodity LLMs to answer domain-specific questions quickly, cheaply, and reliably.

Moving from applications to a shared data platform changes access control at its core. When data lived inside an application, access to the application was effectively access to the data. On a platform like OSDU, many users need the platform but must not automatically see all data on it. Access must instead be derived from the context and obligations attached to each dataset.

That context is rarely simple. Access decisions in the energy domain are shaped by sensitivity, licence and joint venture terms, service provider agreements, national regulation, operatorship, and commercial constraints. These dimensions overlap, evolve, and do not map cleanly to static roles or groups — leaving a gap between how legal and governance stakeholders reason about access and how platforms enforce it.

Equinor and Tieto Tech Consulting are closing this gap with a rule-based approach in OSDU. Legal and commercial intent is expressed as declarative rules. For each request, the platform gathers the relevant context across related data, evaluates the applicable rules, and produces an outcome that can be explained and audited. Rules are treated as data — authored, versioned, and refined iteratively as real-world complexity emerges.

The session will demonstrate the concept through a purpose-built application based on work underway at Equinor. Attendees will leave with a clear mental model: platform access is not data access; data access must follow governance context; and rules can bridge legal intent and technical enforcement.

Presented jointly by Equinor and Tieto Tech Consulting, drawing on work currently being implemented and validated against real-world data.

Two years ago, at this conference, the argument was made that file format compliance remains indispensable even in the era of AI – that the foundation of reliable automation is a reliable benchmark. This argument holds, and we are beginning to see the tangible benefits of automation in data loading and QC.

Compliant and regularised subsurface data enables automated file identification and ingestion, systematic geodetic parameter validation, repeatable QC procedures, structured error reporting, and export to common exchange formats – all without manual intervention. The data lifecycle becomes more manageable, and each dataset is assigned an integrity indicator so that informed decisions are made. This process is established and in practical use today.

In this talk, we present practical experience of operating such workflows, examining both what is working and where the boundaries remain. Non-compliant data continues to present genuine challenges: CRSs that must be inferred when not explicitly declared, input files in formats that resist automation, poor quality scanned legacy data, and fundamental errors embedded invisibly within otherwise plausible datasets. In each of these cases, human expertise remains critical – and likely will for some time.

At these boundaries, there are feasible approaches to reduce the triage burden when data loading. Probabilistic, fuzzy CRS identification uses spatial and contextual clues to arrive at a reasoned prediction of positioning. Pattern recognition can identify fields within non-compliant files, and format fingerprinting can inform and refine future loading decisions.

Each improvement introduced into the automation workflow reduces the load on the analyst, allowing expertise to be directed toward genuine ambiguity rather than routine compliance checks. We explore the current industry state, emerging developments, and the challenges that still require solving before full automation can be achieved.

As organizations adopt OSDU as the foundation for energy data management, establishing scalable governance and data quality processes becomes increasingly important. While OSDU provides standards for metadata, Data Quality Rules, RuleSets, and quality assessments, organizations still face challenges in operationalizing these capabilities across enterprise data platforms.

This presentation shares lessons learned from a joint Equinor–Microsoft collaboration focused on bridging governance and data quality within an OSDU system.

The initiative explored two complementary areas. The first focused on governance alignment, evaluating how OSDU governance concepts, including Data Quality Rules, RuleSets, scoring mechanisms, and validation requirements, could be represented within an enterprise governance platform. This work provided insights into the alignment between OSDU governance requirements and enterprise governance capabilities and identified key considerations for future integration.

The second focused on implementing an end-to-end data quality workflow using OSDU, Microsoft Fabric Lakehouse, and Microsoft Purview. OSDU data quality rules were translated, managed, and executed through Purview's Data Quality capabilities, while assessment results were generated and written back to OSDU. This established a closed-loop process linking governance definitions, quality execution, and quality reporting.

The session will present the architecture, implementation approach, lessons learned, and practical considerations from integrating OSDU, Fabric Lakehouse, and Purview. It will discuss how governance and data quality capabilities can work together to support trusted, reusable, and well-governed energy data, while highlighting opportunities and challenges in aligning enterprise governance platforms with OSDU standards.

tbd

Reservoir simulation is the integration point where static models, dynamic data, and subsurface uncertainty converge into decisions for field development planning. To be operational, simulation workflows must be reproducible, auditable, and connected to governed source data such as rock and fluid properties, well configurations, and constraints. This presentation summarizes the evolution of dynamic reservoir simulation support in the OSDU Data Platform through a staged, industry-collaborative program involving operators, ISVs, and system integrators, meeting twice weekly alongside multi-day working sessions.

The first MVP focused on simulation initialization. It enabled loading the core inputs required to start reservoir models, including rock and fluid models developed in parallel efforts for the Rock & Fluid Samples domain (RAFS). In practical usage, this allowed models to be initialized consistently and original fluids in place to be computed from the same governed data foundation.

The second MVP (started 2026) extended capabilities toward history matching in a deliberately limited scope, supporting single-segment wells, numerical aquifers, and historical wellbore constraints needed to reconcile model response with observed data.

MVP3 completes the field-management trajectory by enabling prediction workflows and more realistic well modelling, including multi-segment wells for horizontal and multilateral architectures and support for inflow control devices (ICDs). It also adds supporting capabilities such as hydraulic tables and extends aquifer functionality to analytical models. Across the MVPs, the intent is consistent: represent simulation inputs and outputs with clear semantics, identifiers, and provenance so updates remain traceable and interoperable across tools. Together, these increments demonstrate how OSDU can host simulation-critical data and workflows in a governed, interoperable platform while iteratively expanding realism and decision readiness

The adoption of OSDU has opened new opportunities for sharing seismic interpretation data across applications and analytics. Realizing these opportunities requires robust, vendor-independent software tools that can address the technical and architectural complexities of OSDU, its Reservoir Domain Data Management Service (RDDMS), the RESQML data exchange standard, and the Energistics Transfer Protocol (ETP). Equinor found the available open-source Python tooling for RESQML, RDDMS, and ETP to be insufficient.

This presentation shares Equinor’s experience developing rddms-io and pyetp, two open-source Python libraries designed to simplify interaction with the OSDU RDDMS using ETP and RESQML. Pyetp provides a modern, asynchronous ETP client implemented in pure Python, along with Pydantic models for all RESQML objects. Rddms-io offers higher-level abstractions for constructing RESQML-based objects from geophysical models. It leverages pyetp to manage their upload and download to and from RDDMS services. Together, these libraries aim to lower the barrier for Python-based applications and analytics to exchange interpretation data via RDDMS while remaining interoperable with established domain standards.

Furthermore, we describe the motivating use case of making interpretation data available to downstream applications and analytics through OSDU. We share practical lessons learned when modeling common geophysical objects in RESQML, where flexibility often conflicts with interoperability. Particular attention is given to the interaction between RDDMS and the OSDU Core Catalog, including trade-offs between metadata duplication, discoverability, and overall system complexity.

We also showcase results from an ongoing pilot implementation of automated delivery of interpretation data from vendor seismic interpretation software via OSDU to a third-party application used in Equinor’s well-planning workflow.

In this presentation we will introduce the new Equinor Data Catalog based upon Microsoft Purview, and show how it can help to manage our data estate, govern data assets, monitor quality & health – and make data available as data products for consumers across the company.

As organizations transition toward Data & AI–driven workflows, seamless data access across applications is becoming a strategic priority. OSDU is widely recognized as a platform capable of addressing these integration needs. However, from an operational perspective, challenges remain. Despite growing adoption of OSDU as a common data platform, true integration between subsurface applications and cloud ecosystems continues to be a barrier.

This presentation explores how SLB connects subsurface applications such as Petrel and Techlog with OSDU, enabling efficient, governed data exchange while preserving the user experience within domain tools. The session spans multiple subsurface domains. In seismic, capabilities such as streaming allow interpreters to access large datasets directly from OSDU without duplication, improving performance and collaboration. In wellbore, connectors enable bi-directional exchange of logs and interpretations, ensuring consistency with OSDU schemas. The presentation also extends to interpretation workflows, including seismic interpretation outputs and reservoir models, supporting coherent data flows across disciplines, as well as RAFS data integration to broaden access to subsurface datasets.

A key focus is the tangible value for users. By connecting applications directly to OSDU, users benefit from reduced data handling, improved quality, and faster access to trusted data—while continuing to work in familiar environments. These connectors enable hybrid workflows where applications remain the primary interface and OSDU acts as the system of record.

Through practical examples, the session will highlight how integration reduces data silos, improves collaboration, and accelerates decision-making. Ultimately, these connectors provide a foundation for scalable, data-centric subsurface workflows, delivering measurable value to interpreters, petrophysicists, and reservoir engineers.

Offset well analysis is one of the most valuable yet time-consuming workflows in drilling. Engineers often spend days or weeks selecting relevant analog wells, reviewing daily drilling reports, completion reports, end-of-well reports, and other semi-structured records, then manually translating that history into planning actions. This paper presents an AIaccelerated offset well analysis workflow that moves beyond data extraction to analysis and operational decision support.

The approach combines clustering-based well similarity methods with generative AI applied to unstructured and semistructured drilling data. Structured well attributes are first used to identify the most relevant offsets for a new plan. Large language models with OCR and document parsing then analyze drilling narratives to extract and interpret hazards, operational sequences, lessons learned, formation behavior, NPT drivers, losses, instability indicators, and mitigation actions. Rather than only producing structured fields, the workflow builds a customizable driller’s roadmap: a practical, well-specific summary of expected risks, depth-linked events, recommended mitigations, and execution guidance for the next well.

This reduces review time from weeks to days, improves consistency, and unlocks learning from large historical well populations that are too large for manual review. The paper will also present case studies showing how the workflow helps identify stuck-pipe risk, loss-zone patterns, and offset performance trends, demonstrating how generative AI can turn legacy well records into operational intelligence for faster, smarter, and safer well planning.

The study presents the development of an AI-based agent and software system designed to extract data from geothermal well reports and perform well nodal analysis. The main goal of the project was to create a tool capable of processing technical documentation (containing text, images, tables) and converting them into structured data used for optimizing the production of geothermal wells.

The system utilizes a Retrieval-Augmented Generation (RAG) approach combined with a Chroma vector database to efficiently retrieve relevant technical information. To ensure high-quality data extraction, a pre-trained YOLO model is used to filter images, focusing OCR efforts on relevant schematics while excluding unnecessary noise such as company logos. The AI agent identifies and extracts key parameters, including well trajectories and casing details, transforming them into a structured JSON format.

A dedicated nodal analysis module allows users to evaluate inflow and pressure conditions (IPR/VLP). The workflow is semi-automated, allowing engineers to review and adjust the extracted data before running calculations. This humanin-the-loop design preserves QA/QC reliability while significantly accelerating the transition from raw documentation to actionable insights.

The solution was developed during the SPE Europe Energy Geohackathon 2025, where it was awarded second place. The results demonstrate that the system effectively reduces manual workloads and supports more consistent analysis of geothermal assets, proving the practical value of combining Agentic AI tools with traditional engineering workflows.

As upstream oil and gas partnerships grow increasingly data-intensive, national oil companies face a strategic imperative to govern and leverage E&P data as a foundation for competitive partnership engagement. Kuwait Oil Company (KOC) has commissioned a digital transformation of its Data Control and Reporting (DC&R) function, establishing a five-year strategic roadmap across four capability pillars: technology, data, process, and people. The roadmap was developed through a four-phase engagement comprising discovery, assessment, analysis, and design, applying a five-level maturity model across six dimensions: organization, governance, processes, technology, data management, and reporting/analytics. The roadmap defines a structured path to full operational maturity through five progressive implementation stages, underpinned by a digital platform architecture aligned with KOC's IT strategy. KOC's DC&R transformation establishes a four-function operating model comprising Data Management, Data Governance, Data Analytics and Visualization, and Reporting and Performance Management, supported by three digital components: a governed enterprise data management system; an AI-ready analytics platform; and a Data Room environment for secure, controlled sharing of asset and partnership data. These three components are unified through the Partnership and Oversight Portal, a governed digital access layer providing IOC partners, regulators, and stakeholders with role-based access to partnership data, real-time dashboards, KPI frameworks, scenario-based insights, and Data Room content. Outcomes include compressed due diligence timelines, strengthened licensing accountability, and real-time performance oversight building IOC confidence across the partnership lifecycle. This paper presents a replicable E&P partnership data governance framework, addressing a domain underrepresented in NOC data management practice, applicable across the Middle East upstream community.

In PETRONAS, Petroleum Arrangement Contractors are required to meet defined data management obligations as stipulated in PETRONAS data governance standards, including requirements for data submission, data quality, and data release. These obligations are critical to regulatory compliance, operational assurance, and data‑driven decision‑making. However, contractors often face challenges arising from complex submission processes and manual compliance checks. This disconnect between governance requirements, usable information, and day‑to‑day workflows increases effort and delays governance compliance execution. To address these challenges, PETRONAS implemented NDex (National Data Excellence), a unified platform designed to simplify compliance by connecting data, information, and people through embedded and streamlined processes. NDex integrates modular capabilities including StarPAC, Data Uploader, InstaPAC, and Intelligent Approval and Assurance for Release. StarPAC embeds governing standard checks directly into submission workflows to validate compliance at the point of entry. Data Uploader supports governed bulk and system‑based ingestion with metadata validation. InstaPAC strengthens data culture by promoting timely, right‑first‑time submissions and clearer ownership. Intelligent Approval and Assurance for Release applies machine learning to prioritise reviews, identify potential risks, and accelerate data release decisions. By integrating governance controls, behavioural enablement, and intelligent assurance within a single platform, NDex reduces process complexity, improves data quality, and shortens data release cycles. This presentation shares lessons from PETRONAS’ experience in simplifying compliance within contractor‑driven, regulated environments.

The Nyhamna gas plant, located in Aukra on Norway’s west coast, is a key component of the Ormen Lange development and plays a critical role in Europe’s energy supply by processing and exporting natural gas through the Langeled pipeline. Over time, the facility has evolved into a multi-field gas hub, handling large volumes of gas from multiple offshore assets.

Efficient operation and maintenance of such a complex facility rely heavily on the use of temporary equipment — non-permanent, movable assets such as electrical systems, lifting gear, scaffolding, and utility units. These assets are utilized across different departments, including electrical and mechanical disciplines, and are frequently installed, relocated, or removed depending on operational needs. In particular, electrical temporary equipment—including portable power distribution units, temporary cabling, and switchboards—directly interacts with live systems and supports critical activities such as permit-to-work and isolation processes. Mismanagement of these assets can introduce significant health, safety, and environmental risks, including potential ignition sources from batteries, diesel engines, and energized systems.

To address these challenges, a GIS-enabled inventory and tracking solution was developed using ArcGIS and Power Platform technologies. The system provides real-time spatial visibility of equipment location, status, ownership, and usage duration, enabling improved situational awareness and operational control. By integrating ArcGIS Experience Builder, Survey123, ArcGIS Online, Power Automate, and dashboards, the solution supports data capture, validation, workflow automation, and decision-making within a unified digital ecosystem. The approach enhances safety compliance, supports permit-to-work workflows, and reduces unnecessary equipment rental and manual tracking. The implementation demonstrates how geospatial technologies can enable smarter asset management & efficient operations.

A growing number of maritime industries, including offshore wind, seabed mineral exploration, fisheries, and petroleum, are increasing the demand for high-quality geospatial data.

At the sime time, data silos remain a persistent barrier in both public and private sector, limiting the efficient use of geospatial information by restricting access, creating duplication of effort and reducing the ability to integrate data across maritime domains.

This presentation explores the underlying reasons these challenges remain, with particular emphasis on the gap between technical solutions and their implementation - and demonstrates how improved data sharing can unlock significant collective value across sectors.

Drawing on experiences from public geospatial data management in Norway, and informed by the author’s role as Marine Geodata Coordinator at the Norwegian Mapping Authority, this presentation highlights how clear roles, governance structures, and a national spatial data infrastructure have played a key role in breaking down data silos and enabling cross-sector data integration.

We argue that collaborative and governance-related factors, including data ownership, incentives and trust often is more critical than the enabling technology itself.

Effective management and use of subsurface data is becoming a key enabler for improved decision-making and operational efficiency in the energy sector. This paper presents Petoro’s experience adopting SLB’s Lumi™ Data & AI platform as part of a transition to a cloud-based, OSDU-aligned data ecosystem. The implementation focuses on building a scalable and flexible foundation for subsurface data, enabling accessibility, interoperability and governance. Core capabilities for data management, seismic access and data exchange have been established alongside structured processes for data quality and lifecycle management.

By reducing manual data handling and improving data availability, the approach enables domain experts to work more efficiently and focus on high-value interpretation tasks. The platform enables integration with broader AI innovation ecosystems, facilitating experimentation and the development of new data-driven use cases in collaboration with technology partners. A key outcome is the already developed integration of AI-driven use cases leveraging Lumi’s AI Workspace and conversational data capabilities (Tela™), including well benchmarking and automated document insights, demonstrating end-to-end value from data ingestion to decision support. This case study highlights practical experience with platform adoption, demonstrating the importance of governance, standardization and incremental implementation in strengthening the integration between enterprise data platforms and AI innovation ecosystems.

Over the past two years, "Equinor’s seismic journey to OSDU® Data Platform" presentations has focused on establishing core platform capabilities, including metadata ingestion, metadata enrichment, seismic file ingestion and governance. This presentation continues that journey by exploring the use of OSDU External Data Sources (EDS) as a mechanism to extend data discovery beyond the platform boundary.

The session shares practical experiences from enabling discovery and access to third‑party seismic metadata through EDS, allowing external metadata to be exposed and searched alongside native OSDU data without full migration. EDS proved practical to adopt for early collaboration scenarios, supporting faster onboarding and improved visibility across data domains.

The experience also highlighted important considerations for production readiness, including alignment on reference data, missing metadata strategy, metadata quality, and operational governance as integrations scale. The presentation summarizes key lessons learned and demonstrates how EDS can be used as a standardized mechanism to share and discover metadata across partners, enabling efficient exchange with external data providers.