The intercity railroad transportation in Taiwan constantly suffers from extraordinary high travel demand during long-weekend holidays. Developing a railroad train dispatching plan to fulfill such a special need efficiently and effectively has long been a major challenge for both researchers and train operators. In the real world operation, when planning railroad train dispatching we must consider three major aspects, including the preference of customers, train stopping, and train scheduling. Based on the concept of transportation demand management, the main idea of this study is to utilize the limited resources of the Taiwan Railway Administration (TRA) more efficiently and effectively by considering the preference of railroad customers, and by giving the priority to long-distance customers and high-class trains during long holidays. Following the idea, an analytical model of railroad train dispatching is proposed. The model can be used to plan a whole day train schedule including the departure time of each train and its stopping stations. A heuristic algorithm is developed to obtain the best plan of railroad train dispatching based on the objective of maximizing the combination of operation income and the saving of customer’s travelling time. An actual database provided by the TRA during a long holiday in 1997 is used to verify the model. The results indicate that the model can satisfactorily and significantly improve the train dispatching plan. It is also found that customer time-saving is the most relevant benefit provided by the model. This paper finally suggests that, during long holidays, railroad operations should be focused on serving long-distance travellers, whereas buses on short to medium-distance travellers, to achieve an optimal use of transportation resources.