Sixth EAGE Digitalization Conference & Exhibition

The proposed workflow for seismic data migration from on-prem data management system (DMS) to OSDU is a well-structured and innovative approach to address the challenges faced by E&P companies during their digital transformation journey. Below are a concise analysis and summary of key aspects of the workflow:

During the transition from corporate data management systems to OSDU, the challenges posed by manual, error-prone, and resource-intensive seismic data migration workflows make automation a necessity.

The methodology introduces a systematic, automated approach to seismic data migration, encompassing Metadata Migration, Seismic Data Migration, and Data Validation.

Enhanced Visualization: Metadata ingested into OSDU is visualized via OSDU native applications.

Incremental Migration: Supports on-demand, incremental migration of seismic data, offering flexibility to clients based on their business priorities.

Automation-Driven Reliability: The automated workflow reduces human intervention, minimizes errors, and ensures a scalable, efficient migration process.

Conclusion: This automated solution provides an effective framework for migrating Petabytes of seismic data from DMS to OSDU, addressing industry pain points such as data integrity, scalability, and compatibility. By leveraging open standards, integration with existing interpretation platforms, and real-time monitoring (BI dashboards), the workflow empowers E&P companies to unlock full potential of their seismic data in a cloud-native ecosystem.

This study addresses the critical industry challenge of leveraging valuable but fragmented legacy sidewall core (SWC) data. With over 100,000 pages of disparate reports, this geological resource has been historically underutilized. We present a systematic digital transformation workflow utilizing Optical Character Recognition (OCR) and Natural Language Processing (NLP) to automatically extract, standardize, and validate data from multi-format documents, including handwritten records.

The results demonstrate a high-accuracy (98.4%) and efficient process, achieving a 60% reduction in data retrieval time and a 40% reduction in processing time. This successfully converts legacy data into a structured, unified database. The key innovation is making this historical information instantly accessible and reliable, enabling its integration with other subsurface datasets and advanced applications like machine learning. This work unlocks the full potential of legacy SWC data, transforming it into a vital asset for enhanced reservoir characterization, informed operational decision-making, and future exploration activities.

Water wells are the primary source of water used in various petroleum engineering operations and hence, play an important role in the oil and gas industry. In the spirit of digital transformation, GWater, an automated platform, developed in-house, fulfilled the need for a streamlined workflow that improves efficiency, reduces heavy human involvement, and removes the subjectivity in the traditional approach to the water well drilling requirements estimation process. The platform removes the human subjectivity in the estimation of geological, hydrological, geochemical, and completion parameters of water wells ahead of drilling. The product guides the drilling crew on the possible locations for coring, casing, and target aquifers. The huge time saving could be channeled to more intensive tasks associated with providing more support to the field and laboratory personnel. The novelty in the methodology employed in the development of this product has generated a couple of intellectual properties and has huge potential for commercialization.

Fossils are crucial in geology, providing evidence for stratigraphic correlation and paleoenvironmental reconstruction. However, traditional fossil records often suffer from disconnection between specimen details and collection points, limiting their scientific utility. Large amounts of valuable fossil data remain locked in legacy reports, handwritten notes, and scanned charts. To address this, a novel AI-driven framework integrates Large Vision Models (LVMs), Large Language Models (LLMs), and GIS to extract and process fossil data from over 7,000 legacy documents. The methodology employs YOLO-based object detection to identify key terms, such as formation names and faunal data, followed by Handwritten Text Recognition (HTR) and LLM-based refinement. This approach achieved significant results, resolving 2,410 fossil localities across Saudi Arabia. Quality control measures, including OCR error correction and validation against benchmarks, ensured accuracy, with a Word Error Rate of just 5%. Further, LVMs like Qwen2-VL and TrOCR enabled extraction from both handwritten and printed records, while duplication analysis reduced redundancy. A major breakthrough was achieved through AI-assisted geographic reasoning, which integrated textual descriptions with quadrangle maps, expanding locality data by 1,474 points. Ultimately, 98.5% of localities were geographically verified, demonstrating the framework’s potential to transform inaccessible fossil archives into structured, standardized datasets for future scientific research.

In the digital transformation of Exploration and Production (E&P) companies, data serves as a key enabler of operational efficiency, informed decision-making, and innovation. As E&P companies continue to drill more wells, the volume and complexity of data have grown exponentially. This has introduced several challenges related to data management, accessibility, and usability.

A few key challenges are:

1. Vast data volumes: Over the years, E&P companies have amassed enormous datasets, often reaching petabyte scales. Seismic data alone accounts for 85–90% of this data, underscoring its critical importance in exploration workflows.
2. Siloed data storage: Data is often stored in disparate systems such as shared drives, SharePoint, OneDrive, and other storage solutions. These systems are typically not integrated, leading to data silos that hinder collaboration and accessibility.
3. Unindexed and Unstructured data: A significant portion of the data remains unindexed and unstructured, making it difficult to search, retrieve, and utilize efficiently.

To address these challenges, the paper focuses on indexing, metadata extraction, and seamless integration with advanced data discovery and analysis tools.

The workflow supports automated ingestion of indexed data into the OSDU platform. This integration ensures interoperability and facilitates data sharing across the organization.

Seismic data processing is challenged by limited domain expertise and manual, labor-intensive workflows, which restrict scalability and consistency. This work introduces an agentic artificial intelligence framework that automates seismic noise attenuation outcomes by embedding expert judgment within three autonomous agents: one for diagnostics, one for remediation, and one for quality control. Unlike previous approaches that focus on scripting workflows or enhancing tool usage, this solution operationalizes domain expertise, enabling agents to collaborate and deliver noise-attenuated seismic data. The system employs image similarity and feature embeddings to diagnose, remediate, and validate seismic data in a self-regulating quality loop. This proof-of-concept demonstrates practical decision-making and establishes the feasibility of fully agentic seismic processing pipelines. By automating outcomes rather than tasks, this approach addresses critical bottlenecks and provides a foundation for future development of transformative subsurface workflows.

This abstract discusses the implementation and evaluation of Retrieval-Augmented Generation (RAG) systems in the subsurface domain, where professionals face challenges with large volumes of unstructured data and the need for effective knowledge transfer. The RAG tool, developed by a cross-functional team, leverages text-based experiences from subsurface professionals, enriched with metadata, to enable efficient information retrieval and support faster decision-making.

Three key lessons emerged from the evaluation phase:

1. Metrics Alone Are Insufficient: Quantitative metrics like retrieval recall and faithfulness provide valuable diagnostics but do not fully capture system performance. Qualitative user feedback is essential to interpret results and guide improvements.
2. User Feedback Is Critical: The system incorporates a feedback loop where users evaluate queries, retrieved documents, and generated answers. This feedback refines both the content and structure of responses, ensuring the tool meets real-world needs.
3. Curated Evaluation Datasets Are Crucial: Domain-specific datasets, informed by user interactions, are necessary for robust evaluation. Building such datasets is challenging but vital for meaningful assessment and continuous improvement.

The abstract concludes that RAG systems, when combined with expert feedback and tailored evaluation data, can enhance knowledge retrieval and decision-making in safety-critical subsurface applications.

We present the Scampi system, a digital solution designed to automate and accelerate species identification in palynology. Traditional biostratigraphy is hindered by slow, manual fossil identification and a shrinking pool of specialists. Scampi addresses these challenges by leveraging content-based image retrieval (CBIR) using vision transformers, enabling efficient and accurate search of microfossil images.

The software allows experts to scour large image databases, refine results through active learning, and visualize depth distributions of identified specimins. A case study on 23 slide scans from a North Sea well showed that Scampi gave a 96% match to traditional methods, delivering interpretations ten times faster and halving the overall workflow time. The approach is expected to scale well for larger studies.

Scampi’s integration of advanced machine learning and user-friendly interfaces promises to transform biostratigraphy, making analyses faster and more cost-effective.

Artificial intelligence has rapidly entered geoscience interpretation, but most commercial implementations remain automation-first—closed, opaque, and detached from human reasoning. These systems accelerate processing but erode scientific accountability by producing untraceable results that interpreters must validate without understanding.

This presentation argues for a paradigm shift toward interactive deep learning (IDL), in which interpreters remain within the learning loop through continuous, real-time feedback. Such systems promote transparency, reproducibility, and data-centric iteration, allowing experts to guide model behavior instead of merely assessing its output. Quantitative comparisons demonstrate that interactive workflows achieve faster network convergence, drastically lower discard rates, and significantly reduce post-processing workload relative to automated “black-box” methods.

Theoretical foundations draw from human–computer symbiosis, emphasizing that cognition and decision-making should remain distributed between human and machine. Responsible adoption of interactive deep learning in seismic interpretation rests on three principles—transparency, accountability, and synchronous guidance – which together reestablish the interpreter as a scientific partner rather than a passive consumer of algorithmic output. Re-centering AI on interactivity transforms it from a tool of automation into an engine of collaboration that enhances, rather than replaces, human expertise in subsurface interpretation.