Was the Dutch Railways? New Off-peak Infrastructure Maintenance Experiment a Success?

Was the Dutch Railways? New Off-peak Infrastructure Maintenance Experiment a Success?


R Jansen, Goudappel Coffeng, NL



What do train passengers do when confronted with a fifty percent reduction of train services one day a week during off-peak hours? A new off peak rail infrastructure maintenance experiment on three railway tracks in the Netherlands enables a study of behavioural change resulting from major shifts in regular train services once a week. In contrast to much research on behavioural change in transportplanning, which is based on stated preferences of travellers, this study focusses on real behavioural change revealed by train passengers travelling on railway tracks under investigation. A three wave panel survey is applied to assess and evaluate the behavioural impacts of a transportation policy change involving new legal and stricter security requirements with regard to visual rail inspections and small repairs by maintenance personnel.

In the Netherlands the Dutch organisation Prorail is responsible for the quality of the rail infrastructure. The railways have to be constantly maintained. Part of the maintenance consist of weekly visual inspections and small repairs of all railway tracks in the Netherlands. Until now these maintenance activities do not interfere with regular train services. However, because of stricter security requirements this will change. Only a single track instead of two tracks will be available for train services during off peak inspection hours. It will reduce regular train services by an approximate fifty percent: e.g. instead of two train services an hour only one train service is performed. How are train passengers going to react to these changes and how will it influence their travel behaviour?

The Dutch Railways is a commercial enterprise. It is granted a concession to operate most of the railways in the Netherlands. It is also responsible for this evaluation study. A range of behavioural reactions will be investigated. We hypothesise that if there is a reduction of train services by fifty percent, approximately half of the train passenger will be confronted with adjustments of regular timetables. The questions to be answered are how these passengers distribute over groups of frequent and non-frequent train users and which groups change their time of departure, change routes or their means of transport. Change of departure time and route will presumably not result in a decrease of trips and transport revenues, but change of transport means will. The study assesses the consequences of these changes in terms of number of trips and passengers and transport revenues. It will also assess whether observed changes actually result from the maintenance experiment or are just part of regular dynamics of train passengers. Crucial for instance is whether train passengers are aware of the maintenance experiment.

In addition the Dutch Railways is interested in attitudinal change of train passengers regarding general service levels (i.e. imago) but also regarding maintenance and its consequences. Furthermore, train passengers are asked to give their opinions about the quality of communication of the Dutch Railways such as announcements of timetable shifts and travel information at stations and in trains. The relationships between changes in opinions and changes in travel behaviour will be studied. Finally, a statistical model of the expected loss of transport revenues caused by the new maintenance scheme will be constructed and estimated. The model will be used to forecast losses in case of a national introduction of the new maintenance scheme.

By means of a three-wave panel survey the effects of the maintenance experiment are studied. In the first wave, before the start of the experiment, 1.639 train passengers were interviewed on three railway tracks during off-peak hours and on days on which the new maintenance scheme would be introduced. In the second wave, after the start of the maintenance experiment, 759 of these respondents were interviewed again by means of a telephone questionnaire. In the third wave the respondents of the first and second wave will be interviewed again in case they can be reached by telephone. In the third wave the panel will be refreshed with an additional random sample of train passengers that travel on the railway tracks that are under investigation. The research data will be supplemented with data on travelers such as counts per train and ticket sales.

We expect the data collection period to end in February 2005. After electronic entry and cleaning of the data, it will be extensively analyzed. The results of the analyses will be presented in the ETC paper.


Association for European Transport