Presentation Profile
The Integration of Artificial Intelligence in Modern Petroleum Engineering Laboratories
Currently Scheduled: 10/14/2026 - 1:00 PM - 2:00 PM
Room: Exhibit Hall Entrance
Main Author
Raj Shah - Koehler Instrument Company, Inc.
- Jai Khanna - Worcester Polytechnic Institute
Abstract:
Advancements in data technologies utilizing Machine Learning and Artificial Intelligence based algorithms in the past decade have accelerated development of new computational tools for petroleum engineering laboratories. These tools serve to simplify data analytics and coincide the industrial pursuit of sustainability and cost-effectiveness as well as reducing the time needed for decision making. Previously, laboratories were reliant solely on extensive physical experimentation for crucial data such as reservoir characterization and fluid analysis. However, the historically standard methods such as physical core sampling and complex pressure-volume temperature (PVT) modeling are inefficient in both economical and temporal perspectives. This poster is a synthesis of recent literature (2022-2026) to exemplify the integration of machine learning and artificial neural networks in petroleum laboratory workflows for the purpose of alleviating these inefficiencies. One area with significant transformation is petrophysics, where multi layer perceptron networks and regression models are now predicting rock porosity, permeability, and Poisson’s ratio directly from well logs, augmenting tedious physical core processing. Then in fluid analysis, chain-based machine learning algorithms have proven capable of accurately predicting full PVT compositional properties without exhaustive laboratory testing. Finally, hybrid frameworks that synergize physical material balance models with AI are demonstrating increased accuracy in long-term gas production forecasting and enhanced oil recovery optimization. These developments demonstrate how integration of AI/ML tools in laboratory settings serve to enhance the efficiency, accuracy, and sustainability of petroleum engineering and the impact that these advancing technologies have had and will have in the oil and gas sector.
