15:00 - 15:20  Access and preservation of metadata with automatic archiving from SharePoint

- Hans Petter Gathe, Digital Strategirådgiver, iO Data AS

Automatic archiving from SharePoint to limit data volume, and also improved control, less complexity, scalability, interaction between all types of metadata, and achieve a much lower carbon footprint. A presentation with reference to a customer case.

15:30 - 15:50  Accelerating OSDU Implementation: Enhancing AI Performance Through Efficient Data Ingestion and Accessibility

- Ekaterina Sørensen, Chief Growth Officer, Kadme AS

Implementing the OSDU platform is a crucial step towards effectively managing your subsurface, drilling, and wells data. LUMIN can assist in populating the OSDU platform with clean data, ensuring your organization is well-prepared to maximize the potential of your AI tools.

16:00 - 16:20  Accelerating Source to Target Attribute Mapping Using AI and Machine Learning

- Matthew Holsgrove, Energy Consulting Partner, Wipro

Source to target data attribute mapping in support of data migration can be time-consuming. To accelerate this process Wipro created a demonstrable workflow solution using AI and Machine Learning to train a model to generate initial source to target attribute mappings for well master and wellbore data with OSDU schemas as the target.

16:30 - 16:50  Streamlining Petrel data management - A ConocoPhillips' 10 years+ success story

- Adam Watt, Senior Data & Digitalization Consultant, Cegal

Exploring ConocoPhillips' achievements through automated data management. ConocoPhillips has set a benchmark in the industry for leveraging automation to enhance data integrity and workflow efficiency.  Learn how ConocoPhillips has optimized workflows and improved data accuracy over 10 years as the inaugural user of Blueback Project Tracker.

17:00 - 1720  Machine Learning-Driven Well Log Imputation for Enhanced Reservoir Characterization in Oil & Gas Exploration

- Chafaa Badis, Data Science Advisor, Halliburton - Landmark

In oil and gas exploration, well logs are crucial for understanding subsurface geology and evaluating hydrocarbon reserves, but incomplete data poses challenges. This project uses advanced machine learning to predict missing log sections within wells and log data absent in some wells, utilizing a dataset from 107 wells. The methodology involves a Multivariate Imputer for initial data imputation and ensemble machine learning techniques to ensure accurate and reliable predictions of reservoir properties.