University researchers develop algorithms to reduce train delays

Researchers at the University of Portsmouth have developed a new smart tool capable of automatically detecting lane delays.

Developed in conjunction with First MTR South Western Railway through a two-year Knowledge Transfer Partnership (KTP), the new algorithm will help minimize disruption to rail travel on the network by detecting problems faster.

The South Western Railway (SWR) franchise network is the UK’s largest and busiest, with more than 1,700 trains in daily operation and serving around 235 million passenger trips per year in the south of England .

When delays occur in current systems, it can be difficult for controllers to detect them early, resulting in further delays in the selection of contingency plans.

Although many rail operators have sought to overcome some of these challenges, much of the technology and systems used have remained the same for decades, despite a dramatic increase in the number of passengers, trains and crews on the train. network.

KTP Academic Supervisor Dr Edward Smart said: “As a commuter myself, I am delighted to be able to contribute to this project which will enhance the customer experience. It highlights the impact that machine learning algorithms can have for real world applications. “

University researchers automatically analyzed the data to determine the point of delay, identify which trains would be affected, and select the appropriate contingency plans to get services back on track. The smart tool is designed with machine learning techniques to dramatically reduce the time for data analysis and processing.

KTP Academic Manager Professor Chris Simms added: “Automatic delay detection is the future of the rail industry. This project took an important first step in realizing the potential represented by machine learning to mitigate rail delays. “

The University of Portsmouth and the SWR project were funded by Innovate UK.

Currently, the new algorithmic tool is being tested within the SWR Control Center, which is responsible for controlling the movement of trains on the network.

Chris Prior, Control Projects Manager at SWR, said: “Working with the University of Portsmouth has been a great experience for SWR and has transferred the understanding across the company on systems development and AI.

“Together, we have developed a system that improves response speed to recover operational delays, learn and continually improve the SWR customer experience. “

RTM365 is hosting a virtual event on improving the passenger experience on August 18, 2021. Register here.

Previous Global Proteomics Market - Analysis by Component
Next CHEYENNE MOUNTAIN GUIDE: Pikes Peak Library District's Stroll-a-Story Program Provides Family Fun | Special sections