Understanding the Influence of Exogenous Factors on Rail Demand in UK



Understanding the Influence of Exogenous Factors on Rail Demand in UK

Authors

Bhanu Patruni, RAND Europe, Andrew Daly, RAND Europe, Charlene Rohr, RAND Europe

Description

This paper presents analysis of National Travel Survey to understand how factors external to the rail industry influence the demand for rail travel and provide recommendations on how the current forecasting framework can incorporate these factors

Abstract

Understanding demand for rail travel in UK
Andrew Daly, Bhanu Patruni, Charlene Rohr, RAND Europe
Mark Wardman SYSTRA
William Hawkes, Department for Transport, UK
Péter Connell, Alex Coulthard, LeighFisher

The demand for rail travel in UK is forecast using models based on ticket sales data to obtain the influence of exogenous factors as laid out in the Passenger Demand Forecasting Handbook (PDFH) and the Transport Analysis Guidance (WebTAG) recommendations. However, over recent years the rail growth figures derived from the PDFH and WebTAG recommendations have not performed well in explaining the increase in rail demand. A study was commissioned by Department for Transport (DfT) to better understand how factors external to the rail industry influence the demand for rail travel and provide recommendations on how the current forecasting framework can incorporate these factors.
Although variables like income and employment are covered in the existing forecasting framework their impacts are often estimated using aggregate data. Moreover, there are other important influential variables that are not covered in PDFH. Therefore, as a part of this study, an extensive analysis of National Travel Survey (NTS) was undertaken to understand factors influencing the demand for rail travel. Discrete choice models were developed from NTS data (1995-2014) to understand and quantify how different socio-economic factors influence the rail demand.
The models are structured to understand two issues related to rail demand: who are rail users and how many trips rail users make. They can therefore quantify which of these are more important in understanding growth in rail demand. The approach used, the travel frequency model including a ‘stop-go’ sub-model (Daly and Miller, 2006), has been successful in modelling numbers of trips made in a wide range of study areas.
Models were estimated for three travel purposes: commute, business and other travel. To further understand the variation in rail trip making by geography, separate models, for each purpose, were estimated for rail trips originating or ending in London and for rail trips originating and ending elsewhere. This paper presents the results from these models, specifically discussing the impact of the following characteristics on rail trip rates:
• Age of the traveller
• Household and personal income
• Car availability - the number of household cars and company cars
• Licence holding
• Economic status of the traveller
• Occupation status of the individual
• Sector in which the individual works
• Time effects
Knowledge of the development of these exogenous factors over the period helps to explain how rail demand has changed and projections of these variables for the future will help in making better rail forecasts.

References
Daly, A. and Miller, S. (2006) Advances in modelling traffic generation, presented to European Transport Conference, Strasbourg

Publisher

Association for European Transport