Honeywell brings Forge Plant Cognition to Dangote's refinery, targeting $60M a year in AI gains

• Who: Honeywell Technologies (NASDAQ: HON), launching with Dangote Petroleum Refinery, using NVIDIA AI and Dell Technologies infrastructure.
• What: Forge Plant Cognition, an industrial AI platform for refining, petrochemical and liquefied natural gas (LNG) operators.
• When: launched October 26, 2026, first at Dangote's refinery in Lekki, Nigeria, one of the world's largest single-train refineries.
• Economics: a preliminary estimated operational benefit of more than 29 cents a barrel, or more than $60 million a year in gross benefit at full-capacity throughput — gross of implementation costs and not guaranteed.
• Scale: Honeywell is helping Dangote expand the refinery's capacity from 650,000 to 1.4 million barrels per day within three years.
• Workforce: the platform supports Dangote's goal of building a workforce that is 70% Nigerian, reducing reliance on external expertise.

Honeywell Technologies on Monday launched Forge Plant Cognition, an industrial AI platform built with NVIDIA AI and Dell Technologies infrastructure, at Dangote’s petroleum refinery in Lekki, Nigeria — one of the world’s largest single-train refineries. The system is designed to help refining, petrochemical and liquefied natural gas (LNG) operators squeeze more value from plants they already own, according to Honeywell’s announcement.

The launch is the first deployment of the platform, and it comes with a headline number Honeywell is careful to qualify. Based on preliminary data, the company estimates AI-enabled optimization with Forge Plant Cognition could deliver an operational benefit of more than 29 cents a barrel at Lekki. At full-capacity throughput, that would mean more than $60 million in gross benefit per year for this one facility — gross of implementation costs, specific to this refinery, and not guaranteed.

What the platform actually does

Forge Plant Cognition is built on Honeywell’s recently launched Forge Cognition platform, and it is pitched as a step past dashboards that only watch and warn. It combines real-time operating data, engineering knowledge and Honeywell UOP’s process expertise with trusted calculations and physics-based models, aiming to turn insight into actions operators can take. Honeywell frames the value in three areas: capturing hidden margin by finding bottlenecks and optimizing throughput, yield and product value; protecting production by detecting emerging equipment and process problems and recommending corrective action to cut unplanned downtime; and scaling industrial expertise by embedding specialist process knowledge into AI-powered workflows so engineers make faster, more consistent decisions.

The technology stack leans heavily on its two partners. The platform is powered by Dell AI Factory with NVIDIA and supported by Dell Services. NVIDIA Nemotron open models power agents that understand and interpret operational data, Omniverse libraries help build a digital twin of the plant for simulation-based training and planning, Cosmos open models with the Metropolis VSS Blueprint enable vision AI agents for real-time alerts and anomaly detection, and the PhysicsNeMo framework builds the physics AI models behind the digital-twin optimization workflows.

Why Dangote goes first

Dangote’s Lekki refinery will integrate its operating data with Honeywell’s process expertise and optimization tools to find constraints, speed up troubleshooting and inform decisions. The choice of first site is no accident: the two companies have worked together for nearly a decade, and Honeywell is already helping the refinery raise capacity from 650,000 to 1.4 million barrels per day within three years by optimizing existing assets. “We chose Honeywell Technologies’ Forge Plant Cognition because the platform can deliver unrivaled industrial expertise combined with differentiated AI capabilities,” said David Bird, CEO and managing director of Dangote Petroleum Refinery and Petrochemicals FZE.

Beyond throughput, Honeywell positions the platform as a workforce play. By embedding specialist process knowledge into AI tools, it aims to help local engineers and operators access expert guidance, solve problems faster and build technical capability — supporting Dangote’s goal of developing a workforce that is 70% Nigerian and reducing reliance on outside expertise over time.

What to watch

The $60 million figure is an estimate, not a result — and Honeywell says so itself. Actual benefit will depend on refinery configuration, operating conditions, crude slate, utilization, baseline performance, data quality, product margins and how well the customer implements the system. The more durable test is whether a first deployment at one of the world’s biggest refineries becomes a repeatable product for the refining, petrochemical and LNG operators Honeywell is targeting. Reuters also reported the launch on Monday, underlining that industrial AI’s next proving ground is not the chatbot — it is the plant floor.