Enabling artificial intelligence in rail
Artificial Intelligence (AI) could bring huge benefits to the railway. AI-powered solutions can improve safety, cut costs, and give rail customers a better experience. But we need to ensure that we get the most from AI—and avoid the pitfalls.
In response, we initiated the ‘Enabling Artificial Intelligence in Rail’ research programme. Our aim is to give the industry a better understanding of how AI-powered solutions can add most value and be successfully adopted.
Share your experiences
Future investment needs better understanding of the applications for which AI is already being used. It also needs to know which are the most promising applications in rail. This will create synergies between similar and/or related endeavours. To build this industry-wide landscape we need information about the AI applications for rail that you are working on.
AI-related workstreams
There are four workstreams in the Enabling Artificial Intelligence in Rail programme.
Human factors principles for AI in rail
To bring benefits, AI needs to make the most of human strengths and mitigate human limitations. We have identified seven human factors principles for designing and operating rail-specific AI-powered solutions. The principles make AI more likely to succeed. They also make it more likely that rail staff will embrace these solutions as useful tools to enhance their work.
Our interactive slides introduce the seven principles, why they matter, and how they should influence the design of AI-powered applications.
We also include rail-specific case studies, showing the principles in action.
AI Change Toolkit
We are working on an AI Change Toolkit to help organisations identify and address areas of concern when introducing AI-powered solutions. The toolkit guides users through a three-step process—define, assess, and consider.
Learn about our journey so far in the report ‘Structured assessment for the safe application of AI to rail: early feasibility and proof-of-concept’.
Navigating the standards landscape
There are limited railway-specific standards on AI applications, while wider standards and guidance are rapidly increasing. This makes it hard to navigate AI standardisation and to identify relevant standards to use.
We produced a tool to navigate current and emerging standards that are relevant to AI in rail. We developed the landscape alongside the British Standards Institution. You can explore the standards landscape and there is a user guide to help you do so.
Landscape of current and emerging AI applications in rail
Are you developing or deploying AI-powered solutions within rail? We are looking into mapping the adoption of AI in rail. This potential resource would help organisations share learning, avoid duplication of effort, and prioritise investment.
