Applied Computing Develops Foundation AI Model Tailored for Oil and Gas Plants
London-based startup Applied Computing is deploying its "Orbital" foundation model to help energy operators analyze fragmented sensor data, claiming it can turn weeks of investigation into seconds.
Oil and gas facilities often contain thousands of sensors measuring variables such as temperature, pressure, velocity, and viscosity, yet operators struggle to utilize this information. According to TechCrunch, facilities typically make operating decisions using less than 8% of available data because they cannot quickly synthesize sensor readings, engineering documentation, and physics-based constraints. London-based Applied Computing is attempting to solve this with "Orbital," a foundation AI model. Unlike standard large language models that predict text, Orbital integrates time series data, physics-based modeling, and language processing to predict the state of a facility. It allows technicians to run simulations on how specific changes might impact the broader facility.
The core value proposition of the system is speed. Applied Computing claims that Orbital can identify anomalies, diagnose their root causes, and model the impact of potential fixes within minutes. CEO Callum Adamson states that this capability can compress investigations that previously required days or weeks into mere seconds, assisting operators in maintaining output and reducing energy consumption. By analyzing sensor readings while respecting the laws of physics and chemistry, the system offers a way to navigate the complexity of industrial environments where asset reliability and safety are paramount.
The startup has moved quickly since its 2023 founding. According to TechCrunch, Applied Computing has reached double-digit millions in annual recurring revenue in under 18 months. While the company has not disclosed its total customer count, it reports that Orbital is currently in use at several large, publicly listed upstream oil and gas, downstream refining, and petrochemical companies. The firm is also expanding its physical footprint, recently opening an office in Houston to support its existing North American client base, complementing its headquarters in London and an operational hub in Bengaluru.
Applied Computing is integrating its technology into existing industrial ecosystems through strategic partnerships. Engineering firm KBR has incorporated Orbital into its INSITE 3.0 digital platform for energy projects, specifically utilizing the product for ammonia production. Additionally, the startup is working with Indian consulting firm Wipro. CEO Callum Adamson indicated that the company is currently collaborating with a major U.S. upstream operator. These partnerships serve as a conduit for both customer introductions and access to operational data, which the company says is necessary to refine its models.
The company enters a crowded market populated by established industrial software providers. Competitors such as AspenTech offer AI-powered modeling for upstream and refining operations, while AVEVA provides physics-based process simulation and "what-if" modeling. Other firms like Cognite and Seeq focus on the data layer and workflow design. However, Adamson argues that Applied Computing’s competitive advantage lies in its ability to attract top-tier AI researchers, rather than proprietary access to industrial data. He contends that the challenge is an AI problem, not an energy or data problem, positioning the startup to compete against legacy industrial software vendors.
Sources
- 01 Applied Computing wants to give oil and gas operators an AI model for the entire plant — TechCrunch — Startups