Honeywell Technologies launched Forge Plant Cognition, an industrial artificial intelligence (AI) platform developed using NVIDIA AI and accelerated computing capabilities, alongside Dell Technologies’ AI infrastructure. The platform aims to help refinery, petrochemical and liquefied natural gas (LNG) operators improve asset performance, increase productivity and unlock greater operational value.
The platform will be deployed first at Dangote Petroleum Refinery in Lekki, Nigeria, one of the world’s largest single-train refineries. The initiative aims to increase throughput while strengthening the capabilities of local engineering and operations teams.
Honeywell estimates that AI-enabled optimisation through Forge Plant Cognition could deliver operational benefits exceeding $0.29 per barrel at the Lekki refinery. At full-capacity throughput, this could translate into more than $60 million in gross annual benefits for the facility.
Industrial AI to Improve Productivity and Profitability
Rajesh Gattupalli, President of Honeywell Technologies UOP, said industrial facilities offer significant opportunities to improve performance and generate value through digital technologies and connected solutions. He added that the collaboration with Dangote demonstrates how industrial AI can enhance productivity, reliability and profitability at scale.
Built on Honeywell Technologies’ Forge Cognition platform, Forge Plant Cognition combines process engineering expertise, industrial automation and AI capabilities. It helps operators identify performance bottlenecks, anticipate potential problems and translate operational insights into practical actions.
Unlike conventional AI solutions that primarily analyse data and provide recommendations, the platform integrates real-time operating data, engineering knowledge, trusted calculations and physics-based models. It also draws on Honeywell Technologies UOP’s process expertise to support more informed operational decisions.
Three Key Areas of Industrial Optimisation
Forge Plant Cognition is designed to deliver operational benefits across three critical areas:
Maximising operational value: The platform identifies bottlenecks and helps optimise throughput, product yield and value across interconnected processes.
Improving plant reliability: It detects emerging equipment and process issues, recommends corrective actions and helps minimise unplanned downtime.
Scaling industrial expertise: By embedding specialised process knowledge into AI-powered workflows, the platform helps engineers and operators make faster, more consistent decisions.
With more than a century of experience in refinery design and operation, Honeywell Technologies UOP brings extensive process engineering knowledge to the platform. This expertise helps AI systems identify how changes in operating conditions affect efficiency, throughput and overall plant performance.
NVIDIA AI and Dell Technologies Power the Platform
The platform runs on Dell AI Factory with NVIDIA and receives support from Dell Services. It combines several NVIDIA technologies to analyse plant data, simulate operations and improve decision-making.
Key technologies include:
NVIDIA Nemotron: Open models power AI agents that interpret and understand operational data.
NVIDIA Omniverse: Its libraries support the creation of digital twins of plants and processes for simulation-based training and planning.
NVIDIA Cosmos and Metropolis VSS Blueprint: These technologies enable vision AI agents to detect anomalies and generate real-time alerts.
NVIDIA PhysicsNeMo: This framework supports the development, training and fine-tuning of physics-based AI models at scale.
Together, these capabilities support digital-twin-integrated optimisation workflows, helping operators evaluate potential changes and improve plant performance.
Dangote Refinery to Integrate AI-Driven Optimisation
At Dangote Petroleum Refinery in Lekki, Nigeria, the platform will combine operational data with Honeywell Technologies’ process engineering expertise and AI-powered optimisation tools. This integration aims to identify operating constraints, accelerate troubleshooting and enable better-informed decisions.
David Bird, CEO and Managing Director of Dangote Petroleum Refinery and Petrochemicals FZE, said the company selected Forge Plant Cognition for its combination of industrial expertise and advanced AI capabilities. He highlighted its potential to optimise performance, unlock operational value and support the refinery’s commitment to technological advancement and operational excellence.
Developing Local Engineering and Operations Talent
Beyond improving operational performance, Forge Plant Cognition will support Dangote’s efforts to develop a skilled local workforce. By embedding specialised process knowledge into AI-powered tools, the platform will give engineers and operators access to expert guidance, accelerate problem-solving and strengthen technical capabilities. Furthermore, as Dangote expands its workforce, these tools will support its goal of developing a workforce comprising 70% Nigerian employees, gradually reducing dependence on external expertise.
Honeywell and Dangote Expand Their Long-Term Partnership
The launch builds on nearly a decade of collaboration between Honeywell Technologies and Dangote. In November 2025, the companies announced plans to increase the refinery’s capacity from 650,000 barrels per day to 1.4 million barrels per day within three years.
The expansion initiative focuses on optimising existing assets and accelerating the delivery of refined products to the market. With the introduction of Forge Plant Cognition, the partnership is extending its focus to AI-driven optimisation, operational efficiency and workforce development. As per the press release, by combining industrial expertise, digital twins and advanced AI technologies, Honeywell Technologies and Dangote aim to improve refinery performance while establishing a more intelligent, efficient and technology-driven model for industrial operations.
