The company’s model combines time-series modeling, physics-based modeling and a language model. Co-founder and CEO Callum Adamson has argued that the technical challenge is not simply gathering data, but connecting different kinds of operational information fast enough to support decisions inside complex plants.
“It's getting those three data sources to talk to each other in real time. That's the real key,” Adamson said.
Applied Computing says Orbital is already being used by large publicly listed upstream and downstream operators. The company also says it moved from stealth to double-digit millions in annual recurring revenue in under 18 months.
The startup’s pitch is built around speed and context. Orbital is meant to detect anomalies, identify possible causes and test whether a proposed response could create new issues elsewhere in a facility. Applied Computing says that kind of investigation can be compressed from days or weeks into minutes.
That matters in industries where unplanned outages can be extremely expensive and operational errors can carry serious consequences. A refinery or petrochemical plant may track temperature, pressure, velocity, viscosity and other readings across thousands of sensors, but Adamson says operators make decisions using less than 8% of the data they collect.
KBR’s role in the round gives Applied Computing more than capital. KBR has integrated Orbital into its INSITE 3.0 platform and is using it for ammonia production, while also providing operational data, domain knowledge and a channel into customers it already serves. Applied Computing also names Wipro as a partner.
The funding will support international expansion, research and engineering hiring and deeper customer deployments. Applied Computing opened a Houston office this week, while keeping its headquarters in London and an operational hub in Bengaluru. The company is also planning expansion in the Middle East.
The opportunity is large, but the bar is high. Industrial AI is not judged like a chatbot. A poor recommendation in a plant can affect equipment, output, safety and environmental risk. That makes transparency, auditability and human oversight central to whether systems like Orbital can be trusted beyond advisory use.
Applied Computing is also entering a market with established industrial software companies. AspenTech, AVEVA, Cognite and Seeq already sell simulation, modeling, data and analytics tools into many of the same industrial environments. Those companies have existing contracts, integrations and long relationships with operators that typically buy around reliability rather than novelty.
Adamson argues the company’s edge is not just access to data, but the AI talent required to build models for this kind of industrial environment.
“It's an AI problem. It's not a data problem, and it's not an energy problem,” Adamson said.
The central question for Applied Computing is whether Orbital can prove that a model built for whole-plant understanding can become a meaningful layer in industrial operations. If the company can show that its system improves decision-making without weakening safety controls, its Series A could mark a step toward AI moving from digital productivity tools into the physical infrastructure that runs heavy industry.
This analysis is based on reporting from Times of AI.
Image courtesy of Schmooly / Applied Computing.
This article was generated with AI assistance and reviewed for accuracy and quality.
