Abstract Businesses in energy face persistent pressure to maximize return on assets while reducing project delivery cost and schedule. This paper examines how artificial intelligence (AI) with emphasis on large language models and pattern-recognition workflows—can optimize conventional engineering design processes and accelerate engineering execution across the project lifecycle. Building on qualitative interviews, case studies with consultants/contractors, and quantitative assessment of pilot deliverables, we describe an AI Engineering Design approach that: (i) extracts governed knowledge from an engineering standards library; (ii) uses pre-trained models for content and pattern recognition within typical deliverables; and (iii) operationalizes design logic through structured input forms to ensure conformance to ADNOC Group Engineering Standards. Proof-of-concept (PoC) activities indicate potential benefits on cost and schedule, with ∼10% optimization in engineering deliverable development and 2–4 months reduction in schedule for selected scopes, subject to data readiness and integration with existing CAD/CAE/simulation tools. We discuss key enablers—data standardization, interfaces to incumbent toolchains, and model-training resources—and provide a maturity roadmap, risk mitigations, and governance controls. We conclude that AI in engineering design is viable and timely for energy operators when deployed at procedural level on well-bounded tasks, supported by robust data stewardship and change management.
Satyaprasad Jadhav (Mon,) studied this question.