Asset Cortex – Generative AI for asset hierarchy

Asset Cortex – Generative AI for asset hierarchy

Status: Live
Reference: NIA_NGT0292
Lead Network: National Gas Transmission PLC
Start Date: 01 October 2024
End Date: 01 March 2027
Budget: Total: £ 133,260.00
Funding Mechanism: Network Innovation Allowance
Technology: Asset Management, Digital Network, Maintenance Inspections, Network Automation, Network Monitoring, Resilience
Strategy Theme: Data and digitalisation
Collaborating Networks: National Gas Transmission PLC
Project Partners: DNV are the independent expert in assurance and risk management. Driven by their purpose, to safeguard life, property and the environment, they empower their customers and their stakeholders with facts and reliable insights so that critical decisions can be made with confidence. They are also a world-leading provider of digital solutions for managing risk and improving safety and asset performance for ships, pipelines, processing plants, offshore structures, electric grids, smart cities and more. Their open industry assurance platform Veracity, cyber security and software solutions support business-critical activities across many industries, including maritime, energy and healthcare.

Abstract

The Asset Cortex project is a Generative AI initiative by National Gas Transmission (NGT) aimed at transforming its legacy 4-level asset hierarchy into a deeper, ISO 14224-compliant structure. This Proof of Concept (PoC) will explore the feasibility of using AI to infer component-level details from system-level data such as pressure and age, enabling automated hierarchy generation. The project supports RIIO-GT3 objectives, including predictive maintenance, digital twin creation, and improved asset lifecycle visibility. It will also enhance integration with systems like SAP and Copperleaf, and streamline field force operations. Key phases include requirements capture, data mapping, AI model development, benchmarking against manually collected data, and final reporting. Grasby Bottom and Hatton Multi Junction sites will serve as testbeds. The project is expected to reduce manual effort, improve scalability, and lay the foundation for broader digital transformation. It will also inform IT infrastructure needs and data governance strategies. While the current phase focuses on feasibility, successful validation could lead to full-scale deployment, supporting NGT’s strategic goals around automation, cost efficiency, and sustainability. Asset Cortex is positioned as a foundational enabler for future infrastructure planning and operational excellence across the gas network.

The Asset Cortex project will provide National Gas Transmission (NGT) with several valuable new learnings. First, it will demonstrate the feasibility of using Generative AI to infer detailed asset hierarchies from system-level data like pressure and age, offering a scalable alternative to manual data collection. Second, it will test the practical alignment of existing asset data with the ISO 14224 taxonomy, revealing gaps in data quality and structure across sites. Third, by benchmarking AI-generated hierarchies against manually collected data at Grasby Bottom, NGT will gain insights into model accuracy, confidence levels, and limitations. Fourth, the project will identify IT infrastructure and integration requirements for future scaling into systems like SAP and Copperleaf. Fifth, it will highlight governance and interoperability challenges, supporting RIIO-GT3’s goals around data consistency and the Presumed Open principle. Finally, through technical reporting and cost-benefit analysis, NGT will understand the economic value of automating asset hierarchy transformation, helping inform future investment decisions and digital strategy. These learnings will not only validate the use of AI in asset management but also lay the groundwork for broader digital transformation across the gas network.

Article metrics loading...

/content/projects/NIA_NGT0292

2024-10-01

2025-10-30