Modelling Long-distance Travel in the UK

Modelling Long-distance Travel in the UK


C Rohr, J Fox, A Daly, B Patruni, S Patil, F Tsang, RAND Europe, UK


This study provides evidence on the importance of destination choice modelling in long-distance travel models, by testing the relative importance of these different responses in (simultaneous) models of mode, destination and frequency choice.


Detailed study of existing data indicates that travellers? behaviour in long-distance journeys differs substantially from routine journey patterns. Not only is the set of available modes different, but we observe different trade-offs being applied, while the profile of the travellers themselves is substantially different. Additionally, long-distance travel, although typically a small share of the number of total trips, accounts for a substantial share of all passenger kilometres travelled, and their emissions. For these reasons, development of a specific model of long-distance travel is essential for appraising the impact of new transport infrastructure aimed at this market, particularly high-speed rail, and of transport policies on long-distance travel. In such models, there is no doubt that the accurate treatment of mode choice is vital; it is also clear that accessibility changes, such as low-cost airlines, have significantly increased the number of long-distance journeys being made. What is not clear, however, is whether it is reasonable and necessary to model destination choice fully or whether it is possible to represent destination-specific effects through differential generation.

The aim of this modelling study is to provide empirical evidence on the importance of different responses, specifically mode, destination and frequency responses in models of long-distance travel behaviour. Moving on substantially from a previous paper , the models of disaggregate travel choices are estimated using information on long-distance tour trips from the UK National Transport Survey data, rail-only data and newly collected household survey data. Separate models are developed for three journey purposes: (i) commuting, (ii) business travel, and (iii) visiting friends and relatives and other long-distance travel. The model estimation procedure explicitly examines the role of income on cost sensitivity and makes detailed tests of cost and time non-linearities. Different nesting structures are tested and assessed and recommendations on the importance of destination choice in multi-modal long-distance models are made.

The results presented throw light on the specific aspects of long-distance travel, model structures and specifications. They form the basis for a forecasting model that will be used for the appraisal of a wide range of transport policy aimed specifically at long-distance journeys.


Association for European Transport