Abstract

This study analyses customer behaviour after timetable changes using panel smartcard data from Shizuoka, Japan. Whether passengers use earlier or later buses or give up using public transport is analysed, as well as how long passengers require to settle into a new pattern. It is observed that the adjustment stage is about two months, judging from aggregate analysis using a stability matrix. It is suggested that during the first two months, passengers learn to maximise their utility by trading off departure time from home and minimising the risk of being late. A bus choice model is constructed, and the factors that lead to changes in behaviour are extracted using a multinomial logit model. The findings include that those with more regular choices are also those who are more likely to stop using the bus service as a result of a timetable change. Overall, this research contributes to better understanding the demand effects of schedule changes.