Skip to main navigation Skip to main content
The University of Southampton
Engineering

Forecasting Passenger Numbers at New Railway Stations

Brief description

Our integrated rail passenger demand and station choice model can quickly and accurately forecast passenger numbers for proposed new railway stations at any site in Great Britain.

Fig 1

Explore this case study

The challenge

The UK Government has recently set out an ambition of ‘reversing the historic contraction of the rail network’ with an emphasis on new local connections and stations that support housing development or economic growth, or that address urban congestion. There will, therefore, be a continuing need to assess proposals for new railway stations and lines. A crucial part of this evaluation process is to obtain accurate demand forecasts, as predicted station patronage is a key driver of the benefits that will determine whether a scheme is considered viable. The models that are conventionally used to predict passenger numbers at new stations have not performed well in practice, with real-world usage often very different from the forecasts. Industry guidance suggests that bespoke models should be developed for each new station, which makes validation of models difficult and adds considerably to scheme development costs.

What we do

University staff have developed several generations of demand models to predict passenger numbers at new railway stations. Our latest models forecast both railway station choice and railway station demand. They take account of the often complex processes that determine which railway station people will choose to use, and consider the impact of a range of different factors in determining total passenger usage at a railway station. The models were calibrated based on data on actual passenger behaviour and can produce demand forecasts for potential new stations at any site in Great Britain. Model forecasts have been validated against actual passenger usage at a number of recently opened stations, demonstrating that the model is capable of producing accurate predictions in a range of different contexts.

Our impact

The latest generation of our station demand model has been used to undertake demand forecasts for several industry partners, including the Welsh Government, where it was used to help prioritise a number of potential new station schemes, and Transport Scotland. Following a range of dissemination activities to industry and the wider public, we have received enquiries about the model from Network Rail, transport consultants, local authorities and groups campaigning for the reopening of lines and stations. This has led to the model being used to generate forecasts for customers on a consultancy basis.

A tool incorporating the model is also available as an open-source project, enabling anyone to generate a forecast. This has the potential to democratise the planning process by opening up the ability to produce station demand forecasts to a wider range of stakeholders. Earlier generations of our new station demand models have been used to produce forecasts for a large number of potential stations across Great Britain, with many of these stations now open and providing a crucial transport service for local communities.

The facilities we used and partners we work with

Our rail station demand model is hosted on the Data and Analytics Facility for National Infrastructure (DAFNI), enabling multiple forecasts to be quickly produced using DAFNI’s supercomputing facilities.

D AFNI on Track with Railway Station Demand Planning | Data & Analytics Facility for National Infrastructure - DAFNI

Key Publications and media

A video walk-through of the station demand model hosted on DAFNI can be viewed here:

https://www.youtube.com/watch?v=q0CmY5lilWg

Full details of the new station demand model and how to use it can be found here:

About | Station Demand Forecasting Tool

The open-source project can be found here:

https://github.com/station-demand-forecasting-tool

Key publications are as follows:

Young MA, ‘Essential tool to aid rail campaigners and politicians to reverse Beeching’, published in Railwatch, https://www.railwatch.org.uk/backtrack.php?mag=rwm&issue=165

Young MA et al, ‘An automated online tool to forecast demand for new railway stations and analyse potential abstraction effects’, presented at 17th Annual Transport Practioners’ Meeting, https://eprints.soton.ac.uk/cgi/eprintbypureuuid?uuid=53e613c2-f320-4917-bc9f-2301910605a7

Young MA & Blainey SP, ‘Development of railway station choice models to improve the representation of station catchments in rail demand models’, published in Transportation Planning and Technology, https://eprints.soton.ac.uk/cgi/eprintbypureuuid?uuid=5ed28026-f09b-4b19-a221-874208eff894

Blainey SP & Preston JM, ‘A GIS-based appraisal framework for new local railway stations and services.’, published in Transport Policy, https://eprints.soton.ac.uk/cgi/eprintbypureuuid?uuid=30161dd0-713e-4025-932b-019d8d684fff

Blainey SP, ‘Trip end models of local rail demand in England and Wales’, published in Journal of Transport Geography, https://eprints.soton.ac.uk/cgi/eprintbypureuuid?uuid=d9d0a984-1215-4c25-868e-c99b5e333675

Related Staff Member

Related Staff Member

Related Staff Member

Privacy Settings