Network Rail in the UK has collaborated with technology start up nPlan to deploy machine learning technology on 40 rail projects initially and all projects by mid-2021.
The technology roll out follows the successful completion of a trial with nPlan. During the trial, Network Rail tested the start up’s risk analysis and assurance solution on two of its largest rail projects, the Great Western Main Line and the Salisbury to Exeter Signalling project.
Network Rail hopes that leveraging machine learning will enable it to transform the way major rail projects are delivered throughout the UK.
The nPlan algorithm compares what was planned against what actually took place at an individual activity level.
The data will be used to produce accurate cost and time predictions to ensure efficient planning and implementation of rail schemes. It will also help to prevent work from over running.
Network Rail said that by using data from more than 100,000 programmes, it will increase prediction accuracy, reduce delays, allow for better budgeting and unlock early risk detection, resulting in better certainty in the outcome of these projects.
Network Rail’s programme director (affordability) Alastair Forbes said: “By championing innovation and using forward-thinking technologies, we can deliver efficiencies in the way we plan and carry out rail upgrade and maintenance projects. It also has the benefit of reducing the risk of project overruns, which means in turn we can improve reliability for passengers.”
Infrastructure manager Network Rail owns, operates and develops 20,000 miles of track, 30,000 bridges, tunnels and viaducts and thousands of signals, level crossings and stations.
Network Rail has recently completed a 12-month programme of electrical works, infrastructure upgrades and operational changes to increase reliability on a critical section of track between Liverpool Street station and Bethnal Green.