In recent years, Taiwan Railways Administration (TRA) has introduced Taroko and Puyuma Express trains for the Eastern Route in order to improve the service level of trains. During peak hours, booking seats on the Eastern Route is very difficult. However, the simplified ticket type policy, which keeps trains with different service levels have the same ticket prices, makes the problem worse. An
appropriate pricing scheme based on passengers’ willingness to pay (WTP) for the attributes of the services would improve the efficiency of resource allocation and relieve the booking problem. Nevertheless, there are limited studies on this issue. This study designs questionnaire by stated preference method and estimates multinomial and latent class logit models to analyze passengers'
preferences and willingness to pay for service attributes of trains. Furthermore, we also estimate the passengers’ willingness to pay for different trains (Puyuma, Taroko and Tze-Chiang Express) by the Contingent valuation (CV) method. Estimates of both models show that passengers prefer refundable and exchangeable tickets, and passengers on long-distance trips, comparing to those
on short- or medium-distance trips, care more about travel time. Two classes, identified by passenger’s income, age and trip type, are suggested for the latent class logit model. Passengers with lower income, age between 18 to 24 and over 65, non-business trip are more sensitive to fare-related attributes, while passengers with higher income, age between 25 to 64, business trip consider
more about time-related attribute. The standard deviation of WTP estimated by the tri-bounded dichotomous choice model is smaller than that of the double bounded model. This indicates the improvement of method efficiency. The differences of passenger’s WTP among three express train types obtained by logit models are larger than those of CV models. The results of this study would
could be applied to design TRA’s express train services, such as combinations of different travel times and ticket refund mechanisms with different ticket prices.for revenue management.