Rail user project aims to provide ‘concierge’ service
A new research project funded by the Department for Transport could make train journeys less stressful and more comfortable for passengers.
The one-year project, Onward Journey Planning Assistant, began in November 2017, supported by funding of around £230,000 from the Department for Transport through Innovate UK. It is being led at Surrey by Dr Sotiris Moschoyiannis of the Department of Computer Science, in collaboration with GTR (Govia Thameslink Railway) and SMEs CommuterHive, ManagePlaces and SnapOut.
The main aim of the research is to support passengers’ end-to-end journeys by recommending a range of options that will help them make informed decisions in order to reach their destinations as easily as possible. This might include taking an alternative rail or bus route, or sharing a taxi with other passengers travelling on the same train. The system will tailor the options to different types of passengers – for example people whose primary drivers are cost and time, or people who are more motivated by comfort and avoiding overcrowding.
The project will draw on the Department’s expertise in three areas – data clustering (to segment the passenger groups), machine learning (to create options for the onward journey), and decision control (to influence customer choices and service provision) – to create prototype software.
Dr Moschoyiannis explained: “Individual rail passengers have different needs: some want to arrive exactly on time, some want a stress-free journey, and some are more cost-conscious than others. Our research will bring this data together with real-time status of onward modes of transport and station facilities to offer a proactive concierge-like service.”
When developed, the app is likely to be offered by rail companies, enabling passengers’ needs to be tracked from the first time they show an interest in a particular journey, while in the future it could be integrated into mobile devices within train seats.
Explore our courses in Computer Science.