Are we in an AI bubble? The question may seem academic to petroleum engineers who are already capitalizing on the momentum of digitalization across the industry, yet any engineer, regardless of their career stage, could be forgiven for feeling overwhelmed by the sheer scope of specialized skills now demanded in this rapidly evolving digital landscape.
Paper SPE 226792 articulates the core skills and technologies needed on the journey to becoming a digital petroleum engineer. Of course, a foundation in classical petroleum engineering skills is indispensable. What really sets this synoptic paper apart from the ever-growing din of digital noise is its approach to recapping petroleum engineering history, the current exploitation of large language models and, most importantly, looking ahead to anticipated future challenges such as solution scalability and opportunities such as quantum computing.
Not every problem demands a machine-learning or artificial-intelligence solution. Fit-for-purpose analytics solutions, which can be easily and sequentially followed on a process flowchart, remain relevant and are relatively cheap to develop and deploy. Paper SPE 226368 provides an elegant template for those seeking to understand how to correctly frame problems to be addressed through large-scale data analytics.
Well placement for optimizing reservoir performance is a tough problem. Paper SPE 223867 reports an 80% reduction in required simulations within the design-of-experiments stage. This approach delivers a notable improvement over conventional global-search methods or proxy models that apply constraint penalties after sampling, often wasting computational effort on invalid scenarios.
Summarized Papers in This January 2026 Issue
SPE 2267792 Rise of the Digital Petroleum Engineer Reshapes the Oil and Gas Industry by Babak Moradi, SPE, Lasse Hermansson, and Tor Ellingsen, THREE60 Energy, et al.
SPE 226368 Data Analytics Leveraged in Designing Flaring-Reduction Strategy by Farras Sailendra and Bugi Setiadi, SPE, BP
SPE 223867 Well-Placement Optimization Workflow Blends Gradient-Free Algorithm, Physics-Informed AI by Kheireddine Redouane, SPE, and Ashkan Jahanbani Ghahfarokhi, SPE, Norwegian University of Science and Technology
Recommended Additional Reading
SPE 228097 Building EnergyLLM: A Domain-Specific Large Language Model Trained on SPE Content by J. Eckroth, i2k Connect, et al.
SPE 227088 Unsupervised Machine-Learning Workflow for Identifying Microresistivity Borehole Image Features by G. Keretchashvili, SLB, et al.
SPE 227244 Integrating Machine-Learning Clustering Into PVT-Based Reservoir-Fluid-Characterization Workflows by S. Bestman, Saudi Aramco, et al.
Kamlesh Ramcharitar, SPE, is a data and insights analyst with the Northern Alberta Institute of Technology (NAIT). He worked at Shell Trinidad and Tobago as a technical data manager before moving to the NAIT. Ramcharitar has more than 16 years of experience in reservoir, production, and process engineering across multiple private and state-owned companies. He holds BS and MS degrees in chemical and petroleum engineering, respectively, from the University of the West Indies. Ramcharitar has published SPE papers on synthesizing sonic log data sets and applied machine learning for inferring subsurface compartmentalization. He has been awarded the Regional Young Professional Outstanding Service Award and has served as technical program chair for the SPE Energy Resources Conference and on the Trinidad and Tobago section board. Ramcharitar is currently the chairperson-elect for the SPE Edmonton Section.