Saudi Aramco deploys AI refinery system with Emerson

Saudi Aramco has deployed an artificial intelligence (AI)-driven refinery optimisation platform developed with Emerson, as the energy giant accelerates the use of industrial AI across its global refining operations.

The deployment integrates Emerson’s Aspen Hybrid Models into Aramco’s refinery planning environment, creating what the companies describe as one of the world’s largest multi-site, multi-period refinery optimisation models. The initiative forms part of Aramco’s wider digital transformation and AI strategy aimed at improving operational efficiency, increasing refinery yield volumes and reducing planning-to-execution gaps.

At the centre of the project is the use of Aspen Hybrid Models, which combine first-principles engineering models with industrial AI and operational plant data to capture complex non-linear relationships in refinery performance. According to the companies, the system has already achieved prediction accuracy levels of up to 98.5% in key refinery units.

The AI models have initially been deployed across Continuous Catalyst Regeneration (CCR) and Platformer Units, helping optimise feedstock blending, improve product quality forecasting and enhance margin prediction accuracy across Aramco’s global refining network.

The deployment highlights the growing role of AI in industrial process optimisation, particularly in the energy and petrochemical sectors, where traditional modelling approaches often struggle to adapt to changing feedstock mixes, fluctuating operating conditions and increasingly complex sustainability requirements.

“This deployment represents a significant milestone in Aramco’s AI strategy and our long-standing relationship with Emerson,” said Ahmad Alkudmani, director of the global optimiser department at Saudi Aramco. “We are committed to leveraging innovative technologies for smarter, more efficient refining optimisation. With improved model accuracy, we are enhancing planning decisions, reducing the manual adjustments required from our engineers and uncovering new value across our global assets.”

The companies said the hybrid modelling approach enables Aramco to improve refinery yield and stream quality across a wide range of feedstocks and throughput conditions, while also automating model updates and reducing the need for manual engineering intervention.

Another key focus is reducing discrepancies between refinery planning assumptions and actual plant performance, a long-standing challenge in downstream operations. By aligning planning models with operational reality, Aramco aims to achieve more reliable production forecasting and greater operational consistency across sites.

The AI deployment also supports broader refinery flexibility goals by enabling more dynamic feedstock selection and blending strategies. This is expected to improve profitability while supporting more sustainable, efficient operations.

Aramco is now expanding the use of Aspen Hybrid Models into hydrocracker units across its assets, in a move designed to validate the scalability of the AI optimisation framework across additional refinery processes.

The technology stack is being deployed through Emerson’s AspenTech Performance Engineering and Manufacturing and Supply Chain suites. The platform uses thousands of converged simulation cases built on first-principles engineering models calibrated with real plant data to create nonlinear optimisation models capable of operating at enterprise scale.

“Aramco continues to set the standard for operational excellence through digital innovation,” said Claudio Fayad, chief technology officer at Emerson’s Aspen Technology business. “This deployment of AI-driven Aspen Hybrid Models to optimise complex, multi-site, multi-period planning workflows demonstrates the tangible value of combining deep domain expertise with advanced AI.”

The project reflects a wider trend across the Middle East energy sector, where national oil companies are increasingly investing in AI, automation and advanced analytics to improve operational resilience, maximise asset performance and support long-term digital transformation programmes.

As industrial organisations seek measurable business outcomes from AI investments, refinery optimisation is emerging as one of the most commercially significant use cases for industrial AI in the region.