AI/machine learning
The gap between machine learning research and effective deployment in the oil and gas industry is an alignment challenge between research questions and real decisions, between model design and operational constraints, and between innovation and the people expected to use it.
Technology and partnerships remain important, while phased approaches may supplant lengthy appraisal programs, experts said during CERAWeek.
CERAWeek panelists see AI as a way to leverage data and people in interpreting data for exploration, but a cultural shift at companies may still be needed.
-
This paper investigates the use of machine-learning techniques to forecast drilling-fluid gel strength. -
Machine learning and a decade of gas composition records helped the operator identify wells that were most likely to produce paraffins.
-
The companies plan to develop new artificial-intelligence-powered processes and workflows to optimize oil and gas production.
-
Routine status reporting often presents a challenge because of its intimidating and time-consuming nature for both employees and supervisors. With large language models, a system was developed to generate coherent artificial-intelligence-driven reports. The goal is to enhance the understanding of overall insights and reduce the time required for individual report read…
-
This paper aims to emphasize the importance of decision-making based on quantitative monitoring outputs, from both a business perspective and an ecosystem-service perspective, in future offshore projects.
-
Experts at SPE’s Annual Technical Conference and Exhibition say that despite AI’s great potential, it’s important to be realistic about AI’s capabilities and to remember that successful projects solve specific business problems.
-
This paper presents the processes of identifying production enhancement opportunities, as well as the methodology used to identify underperforming candidates and analyze well-integrity issues, in a brownfield offshore Malaysia.
-
This paper delves into the evolving landscape of drilling automation, emphasizing the imperative for these systems to go beyond novelty and deliver quantifiable financial value.
-
This paper presents a workflow that combines probabilistic modeling and deep-learning models trained on an ensemble of physics models to improve scalability and reliability for shale and tight-reservoir forecasting.
-
This paper describes a new application that leverages advanced machine-learning techniques in conjunction with metocean forecasts to predict vessel motions and thruster loads.
