

Successfully Utilising Machine Learning and AI at a Transmission Operator
Amjad Karim reviews some of the important lessons we learned whilst implementing the KAI platform at the UK’s National Grid.
We’re building an AI-powered operating system for visual asset management.
Asset managers spend millions collecting visual condition data, use it poorly and then make bad asset management decisions. Our technology helps customers unlock value, reduce costs and lower friction by applying AI-powered decision-making to visual data – leading to better asset management decisions.
We survey assets using the most appropriate method; Helicopter, UAV or BVLoS.
Experts assisted by AI undertakes fittings assessment, defect detection and steelwork corrosion analysis.
Risk models predicts the future state of steelwork and the risk of fitting failure using FMEA and environmental data.
Generate investments plans by combining current and future state with financial and operational contraints
Using LiDAR and high-res imagery, our Deepsteel app identifies the location and extent of corrosion on a pylon. Results are fast, consistent, and repeatable.
Asset Managers gather images. We use AI to generate a high-resolution asset inventory from the latest images.
We use AI to label components and identify defects and assess state. This increases the accuracy, minimises the risk of defects being detected too late which leads to better asset management decisions.
Enable asset managers to easily organise, share and access visual survey data for thousands of assets and locations. We use AI to evaluate the quality of incoming footage to ensure it meets the standards required for assessment.
Our KAI platform uses deep learning to extract objects of interest and data engineering to ensure this visual data is easily available to anyone who needs it.
Bespoke edge detection algorithms are used to identify regions of the image likely to contain steel and thereby separate background from tower.
Eliminated 25% of towers from the manual review process by marking them as clean.
We have developed a vehicle mounted solution for collecting high resolution imagery of roadside vegetation. AI is then used for detecting the presence of invasive plant species or Ash trees.
This provides a rapid, high-quality vegetation survey methodology, resulting in cost and time savings for our customers.
Knowing the types of habitat and plants alongside rail tracks is crucial for Network Rail.
We have deployed camera systems to the front of trains for collecting images. The KAI platform analyses data gathered and presents an overview to NR staff.
Digitising networks requires an inventory of what’s installed and where. This could be done manually by reviewing historic images and video. With thousands of structures and millions of images this is an exhausting manual process but an ideal opportunity for digitalisation and a practical use case for AI.
We are working with customers to complete detailed asset inventories using millions of images.
The KAI platform extracts key components from footage presenting them for review by a condition assessment engineer. Our models aim to extract over 50% of defects automatically.
Using KAI reduces assessment time by 66% and minimises unforeseen asset failure.
Amjad Karim reviews some of the important lessons we learned whilst implementing the KAI platform at the UK’s National Grid.
18k Videos over 10 Years weren’t Assigned to an Asset Our customer has been collecting
Britain’s 20,000-mile rail network requires reliable monitoring of lineside vegetation. Together with the UK Centre
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.