TLDR: Saudi Aramco, in collaboration with Japan’s Yokogawa Electric Corporation, has successfully deployed multiple autonomous control AI agents at its Fadhili Gas Plant. This marks a significant advancement in industrial artificial intelligence, with the AI agents optimizing operations in the plant’s gas treatment unit, leading to notable reductions in energy and chemical consumption.
In a groundbreaking development poised to redefine the oil and gas sector, Saudi Aramco has partnered with Japan’s Yokogawa Electric Corporation to implement advanced autonomous control AI agents at its Fadhili Gas Plant. This deployment, announced in late October 2025, is being hailed by industry experts as a historic milestone in the application of industrial artificial intelligence. The AI agents are now actively optimizing operations within the plant’s critical gas treatment unit, demonstrating a bold leap towards fully autonomous industrial control.
The core of this innovation lies in Yokogawa’s proprietary Factorial Kernel Dynamic Policy Programming (FKDPP) reinforcement learning algorithm. This sophisticated multi-agent architecture enables the AI to autonomously manage and enhance complex control tasks, specifically targeting the acid gas removal (AGR) process at the Fadhili plant. Unlike traditional model-based controls, FKDPP learns optimal policies through trial and error, making it highly adaptable to the dynamic and often unpredictable conditions inherent in gas processing.
The rollout of this cutting-edge solution was executed in three progressive phases. To ensure the highest levels of safety and reliability, Yokogawa meticulously developed a simulation system to train and validate the AI agents before their integration into the plant’s existing CENTUM VP production control system, which incorporates established safety mechanisms.
Initial results from the ongoing evaluation phase at the Fadhili Gas Plant are highly promising. The plant has recorded significant operational efficiencies, including a 10% to 15% reduction in amine chemical and steam consumption, and approximately a 5% decrease in electricity usage. Furthermore, the deployment has led to improved process stability and a marked reduction in the need for manual operator intervention, even when faced with fluctuating environmental conditions.
This initiative extends beyond mere automation; it represents intelligent, real-time decision-making capabilities that adapt without constant human oversight. Khalid Y. Al Qahtani, Senior Vice President of Aramco Engineering Services, emphasized the company’s strategic vision, stating, ‘Aramco has embarked on an ambitious plan to unlock value by deploying a wide range of industrial AI applications across our operations.’ This move aligns with Aramco’s sustainability goals, contributing to Saudi Vision 2030’s emphasis on green initiatives through optimized chemical and energy usage. Economically, it positions Aramco to maintain a competitive edge amidst volatile oil prices.
Also Read:
- AI’s Transformative Impact on Industrial Automation: ISA Urges Strategic Implementation Amidst Vast Opportunities
- AI Drives Unprecedented Productivity and Operational Excellence in Convenience Retail
Yokogawa Electric Corporation defines autonomous control AI as a system capable of independently determining optimal control methods, demonstrating resilience, and adapting to unfamiliar situations. The FKDPP algorithm, co-developed with the Nara Institute of Science and Technology (NAIST), received Japan’s prestigious 2023 Prime Minister’s Award for Industrial Technology Innovation, underscoring its groundbreaking nature.
