Research project: An Automated Demand Forecasting Model For New Local Railway Stations
Implementing a tool for predicting passenger numbers at new railway stations.
Implementing a tool for predicting passenger numbers at new railway stations.
The considerable growth in passenger rail use in the UK in recent years has been accompanied by a large number of schemes to construct new stations and railway lines. With the number of journeys by rail expected to increase by 40% by 2040, and the UK Government’s recently stated ambition to reverse ‘the historic contraction of the rail network’ through new local connections and stations that support housing development or economic growth, or that address urban congestion, many new schemes will be proposed in the coming years. However, analysis has shown that the accuracy of the forecasting tools conventionally used to assess such schemes leaves a lot to be desired. In order to remedy this situation, recent PhD research has developed integrated railway station choice and demand models capable of producing forecasts of passenger numbers at new station locations anywhere in Great Britain, and of estimating the levels of demand abstraction at nearby existing stations. These models have already been used in practice to help the Welsh Government assess the business case for a number of proposed new railway stations across Wales. However, in their current form the models require a substantial level of input and manual processing from a highly-trained user in order to produce demand forecasts, and this limits their usefulness to industry and government stakeholders. A semi-automated modelling tool capable of being used by industry/government employees with minimal specialised training would greatly increase the positive impact of the modelling research for both transport infrastructure planners and wider society, by enabling faster and more reliable assessment of the relative merits of different new railway station schemes.
This project, funded by the university’s EPSRC Impact Acceleration Account, will develop such a modelling tool, hosted on the Data Analytics Facility for National Infrastructure (DAFNI, www.dafni.ac.uk ). It will also include work with industry partners (such as Transport Scotland) to use the tool to produce demand forecasts for proposed new railway stations or railway lines as part of the business case appraisal process. This will help ensure that investments in local rail transport provide the maximum possible level of societal benefit.
Transportation