Valuation of a transfer in a multimodal public transport trip : a stated preference research into the experienced disutility of a transfer between bus/tram/metro and train within the Netherlands

A web-based stated preference (SP) experiment is carried out to determine the transfer disutility of a transfer between bus/tram/metro (BTM) and train. The choice situations were described by the attributes BTM trip time, transfer time, in-train time, headway of the connecting mode, costs and station facilities. The experiment included choice situations between two BTM-train alternatives and choice situations between a BTM-train alternative and a train-train alternative. Based on the comparison of a BTM-train transfer with a train-train transfer, the BTM-train transfer disutility is estimated. Respondents are recruited from the NS panel (N = 1064). To increase the realism of the experiment, the values for the BTM trip time, in-train time and costs were adaptive to the real values of a recent BTM-train trip as reported by the respondent. A general mixed logit error component model is estimated. Furthermore, sub models based on trip motive, travel frequency and trip stage are estimated. The estimation results offer information on the trade-off between the different attributes of a multimodal trip. The estimated parameters are within a reasonable range, compared with findings from literature. A transfer between BTM and train with a transfer time of 8 minutes and a headway of the connecting mode of 15 minutes results in a transfer disutility of 29 minutes generalized travel time (GTT). The transfer disutility is highly dependent on transfer time and the headway of the connecting mode. Changing the transfer time from 8 minutes to 15 minutes increases the total transfer disutility to 39 – 51 minutes GTT. An increase of the headway of the connecting mode from 15 minutes to 30 minutes increases the GTT with 7 to 13 minutes. Only for recreational travelers an effect is found on station facilities. These travelers perceive a very large station positive with a value of 4 minutes GTT, compared to a medium or large station. The egress time by bus is valued with a factor 1.4 compared to a minute in-train time. The access time by bus and the access/egress time by tram/metro are not found to be significantly different from in-train time. Values of time are estimated for different trip motives and incomes. The obtained values are in accordance with other value of time studies. In general, the most optimal transfer time is found to be 8 minutes, but differences are found between respondents and stations. High-frequent travelers prefer a transfer time of 6 minutes, while low-frequent travelers prefer a transfer time of 9 minutes. The resulting values from this research can be used by NS to extend their route assignment model (for train trips) to BTM trips as well. Furthermore, insights into the preferences of different groups of travelers are provided. If the majority of travelers on a certain transfer have the same trip motive or travel frequency (for example work/business travelers during the morning peak hours), the transfer times can be adjusted to these types of travelers.