Case study

How a signalling company used AI to help trains run on time

Learn how a signalling company used regression to reduce train delays.

This guidance is part of a wider collection about using artificial intelligence (AI) in the public sector.

AI technique used

  • regression

Objective

A company responsible for managing railway traffic wanted to increase the proportion of trains arriving at their destination on time from 82% to 90%.

Situation

The company wanted to be able to predict when delays were likely to happen and use these insights to help controllers prepare for and minimise impact on the train network. This would allow the company to redirect traffic in a way that reduced delays.

Action

The company built 3 separate models which interacted with one another. They:

  • built a forecasting model to predict delays using historical data which included features such as train arrival times, position, lateness, and the timetable
  • built a model to learn patterns in the data to help controllers better understand how the network operates and what causes delays
  • trained a recommendation engine to suggest alternative platforms to controllers

Impact

The new system can give the signalling company warnings of delays up to an hour in advance, with an accuracy 50% higher than previously.

This has allowed the company to better redirect trains to avoid delays, reducing total lateness at all London train stations by up to 200 minutes every day.

Updates to this page

Published 10 June 2019