Researchers confirm overcrowding alerts can help maintain social distancing on UK public transport

A service that predicts passenger intent to travel can be used across the UK rail network to reduce overcrowding and help people maintain social distancing.

Researchers from the University of Birmingham’s Birmingham Centre for Railway Research and Education (BCRRE) worked in collaboration with British technology start-up Zipabout to validate the data powering the service, which was designed by Zipabout. Their results prove to industry it is ready to be used by all public transport operators.

The personalised information service, powered by patent pending technology which can identify how busy the transport network will be, will inform all UK rail passengers how disruptions and crowding may affect their journey. It will be particularly valuable in providing information to passengers through the covid-19 pandemic and increasing confidence in rail travel.

The UK Government announced in May 2020 the service would be made available to all UK rail passengers and have since expanded it to include all users of National Rail Enquiries via the ‘Alert Me by Messenger’ service.

The rail industry had asked for cast-iron guarantees that Zipabout’s crowding data is truly accurate before deploying the service at scale. Now the data has been validated, the Rail Delivery Group is preparing for a wider roll out of ‘Alert Me’ to include all train operators’ own digital platforms.

Validation requires the ability to correlate crowding predictions against how many people actually used the service. Working with researchers from BCRRE, Zipabout were able to correlate their predictions for SouthEastern services over a 30-day period with the actual average loading data.

The researchers found a significant correlation between the two data sources, concluding that Zipabout’s data is statistically valid. There only needs to be 15 passenger interactions with a train service for the multi-source data platform to make a statistically valid prediction of how busy a service will be in future.

Dr John Easton at the Birmingham Centre for Railway Research and Education at the University of Birmingham, said: “Working with Zipabout, our researchers have performed a detailed statistical analysis which indicates that their unique crowding predictions, based on passenger intent, provide a good representation of the historic loading data being provided by the rail industry. This confirms that the demand data being made available to the industry by Zipabout does in fact represent a good proxy measure for actual passenger loading.’

Daniel Chick, Technical Director at Zipabout, said: “We need to be providing passengers with the information they need to travel off-peak and make more space. We demonstrated how this can be done earlier this year and have now proven what we knew to be true all along: that Zipabout’s crowding data is truly accurate and can be used to support the UK’s entire transport network throughout this global pandemic and beyond. We look forward to working with government and industry to lay the foundations for the recovery from the coronavirus crisis which we all know transport will be a key part of.’

Zipabout’s multi-source data platform was collecting over 20 million interactions every month before the coronavirus crisis. This means the statistical model can be applied to more or less every train service in the country and be confident that it is valid.