IADC/SPE 2026 Day one: Drilling technology continues a relentless advance
(WO) - Across several panels on day one of IADC/SPE International Drilling Conference and Exhibition 2026, hosted this year in Galveston, Texas, automation powered by machine learning was a recurring theme. As operators across the drilling value chain seek greater efficiency, A.I. and its many applications in automation offer a path to a new era of energy efficiency.
Lee Womble, the 2026 IADC/SPE Chairperson, and Jason Gahr, ExxonMobil, opened the first general keynote with a focus on the conference’s tagline “intelligent energy.” Gahr noted that aside from development, one of the major challenges facing A.I. as an industry tool is wider adoption.
Technical panel highlights. Even so, later technical panels demonstrated just some of the many use cases for A.I., or more specifically, purpose-built machine learning systems. These systems have the capacity to predict drilling equipment wear, give advance warning for drilling events and enhance downhole data logging. One such system, presented by Lucas Katzmann, Baker Hughes, offers the capability to identify stuck pipe events up to hours in advance, using a combination of data-driven machine learning (ML) models and more traditional physics-based models. The method offers real-time analysis and prediction while avoiding the heavy and time-consuming computational costs of a purely ML-based evaluation.
Additional ML examples. Other notable technical panels included Michael Zhang, Patterson-UTI, who presented a ML-based means of managing generators for drilling operations. This ML-based management system, trialed on nine rigs in the field, offered operators the ability to save five to eight gal of diesel fuel per hour, per rig—elimination of which would save an estimated $100,000 in fuel costs. Bruno Reinoso, Nabors Industries, also demonstrated the power that ML models bring to maintenance with mud pump health monitoring. With his proposed model, an ML system trained on various pump efficiency data can allow operators to detect power-end and fluid-end failures up to hours before the actual event, enabling operators to take decisive action to mitigate potential non-productive time (NPT).
In both panels, minimizing unplanned NPT was the main target of these ML systems. For Reinoso, in reference to Nabors’ operations, mud pumps account for 15% of all failures within the company, and of those, 80% of the issues are triggered by fluid issues—the very focus of his early warning system. Similarly, Zhang pointed out the needle that needed to be threaded in using ML systems to pursue efficiency: to an unsupervised A.I. agent, the most fuel-efficient generator is one that never turns on. Guardrails were a cornerstone of the training process to ensure that fuel waste from permanently online generators was eliminated without causing unplanned NPT through a lack of power during demand surges.
Predictive equipment wear was also a focus within technical panels. Shilin Chen, Halliburton, presented a novel algorithm aimed at real-time prediction of PDC cutter and bit wear. Many of the challenges in such wear prediction boil down to the sheer variety of cutter bit shapes and rock types and shapes, which make predictive analysis computationally heavy and therefore unfeasible in real time. With Chen’s proposed algorithm, an offline model is used to build 3D mesh bit-rock interactions, using an algorithm that is fully independent of rock type and strength and near-agnostic in regard to cutter shape, thus eliminating much of the computational demands for real-time information for operators.
PDC bit cutters. Bonar Noviasta, SLB, similarly brought PDC bit cutters to the forefront in a paper on novel multi-ridge cutter geometry. Tested in both an unconventional onshore U.S. formation and an onshore Middle East formation, the ridged cutters showed impressive gains in ROP for each at 20% and 10%, respectively. With additional gains in the way of reduced orientation sensitivity and improved wear resistance, the cutters represented another milestone in the relentless pursuit of greater drilling efficiency.
