Determining the Costs of Delay to Different Types of Train

Named Investigators: Prof John Preston (University of Southampton); Prof Chris Nash (University of Leeds)
Researchers: Dr Graham Wall (University of Southampton); Daniel Johnson, Dr Richard Batley (University of Leeds)
Industry Collaboration and Mentors: DfT; Network Rail; ATOC; Arup

Background:
Network Rail’s train regulation objective is to achieve a “fair and reasonable balance” between the minimisation of overall delay to train movements, the minimisation of overall delay to passengers and time-sensitive goods, the maintenance of connections between passenger trains, the avoidance of undue discrimination, the protection of commercial interests and the maintenance of safety and security. However, it is also difficult to make trade-offs between delays to passengers on different types of journey (e.g. through and connecting) and time-sensitive goods without knowing the respective values (costs) of those delays. The train classification system in use by Network Rail is too limited and coarse to be of much use for detailed timetabling and train regulation.

Objectives:

To develop a better understanding of the social costs of delays to passengers and freight trains.
To establish delay cost values for different train types.

Summary of methods:
The values of time, and thus the costs of known delay, to different types of traveller (business, commuter, leisure) and different types of freight, are already known and established. There is also a body of research on the value of reliability/punctuality, which concludes that in general the value of time for an unplanned delay is greater than the value of time for known journey time. For freight services, we will use the results of recent surveys undertaken by the University of Leeds using the LASP (Leeds Adaptive Stated Preference) software tool which have sought to quantify the value of reliability. For passenger services, we are planning additional Stated Preference research to consider variations in values by passenger type and time of day/week of travel, as well as the impact of information provision on the valuation of delay. Particular attention will be paid to valuing the trade-offs in terms of passenger delays of holding trains to ensure connections.

Using these values in conjunction with data on typical or actual train loading will enable the estimation of the aggregate costs of planned and unplanned delays to different types of train. While these total train delay costs will reflect the numbers and constituencies of their users, consideration will be given to whether delays to longer-distance services may have more extensive and serious effects than those to local services, through the loss of paths at later stages in the journey. A further consideration is the social cost resulting from delays if they impact on levels of rail traffic and hence mode split. Knowledge of relevant own and cross elasticities will be reviewed and existing values of external costs for road and air transport adapted to measure these further costs of delay, which will depend heavily on not just how much traffic is lost to other modes, but also on the location and time of day of that traffic. The costs will be greatest for peak hour trains paralleling heavily congested roads.

This topic has becoming particularly relevant given debates concerning the performance targets set in the 2007 White Paper Delivering a Sustainable Railway. In its 2007 Strategic Business Plan (SBP), Network Rail suggests a reliability target of 91.6% in England and Wales by the end of Control Period 4 (2014) is more realistic than the White Paper’s 92.6%. Network Rail indicate that a 92.6% target could be achieved by expenditure of £400 million – but ‘(i)t remains unclear how highly passengers value continuous improvements in punctuality beyond around 90 per cent’ (SBP, page 12). We will investigate whether such a threshold effect exists.

Work Done:
Work at Southampton has included synthesizing relevant work from projects C2 and C3 (Preston and Wall, 2007, Wardman et al., 2007) and developing a detailed literature review with respect to costs of delays for passenger services (Wall and Preston, 2007). A similar review for freight services is being undertaken by Leeds and is nearing completion.

The proposed Stated Preference work will complement substantial work using Revealed Preference data undertaken by Leeds for the Department for Transport (Batley et al, 2007, Batley et al 2008a, b) building on theoretical developments in the marginal valuation of travel time and scheduling (Batley, 2007).

Work Remaining:
work at Southampton will focus on the design and implementation of the Stated Preference surveys.

Outputs:

Poster from RRUK Workshop 2008
Presentation from RRUK Workshop 2008
Batley, R. (2007) Marginal Valuations of Travel Time and Scheduling and the Reliability Premium. Transportation Research E, 387 – 408.
Batley, R., Nicolás Ibáñez, J., Wardman, M., Shires, J. and Whelan, G. (2007) A Discrete Choice Study to Assess the Impact of Reliability on Passenger Rail Demand. European Transport Conference, Strasbourg, France.