Smart card systems have been spreading globally not only for purchasing general goods and services, but also for fare payment of public transport. When an individual traveller is holding a public transport smart card, the precise times of his/her passing through ticket gates are recorded in the smart card system. The longitudinal smart card data of a large number of passengers have been stored by a public transport company. The aims of this paper are to analyze the behaviour of railway passengers using the smart card data, and to evaluate the effects of train operation policies. The analysis is focused on the before and after comparison of travel choice behaviour of passengers when the railway company changed the train timetable. The individual passengers passing times at the entrance and exit ticket gates of origin and destination stations were aggregated for a small discrete time interval in a day. Through analyzing the departure and arrival time distribution and the travel time distribution, it was found that the passengers smoothly adjusted their travel behaviour for the improved train timetable. When the passing times of ticket gates of both origin and destination stations are combined with the train timetable, it becomes possible to identify the train that each traveller was likely on board. Through the before and after comparison, it is shown how the railway passengers have changed their train choosing behaviour for the improved time schedules.