Analyzing long-term travel behaviour: A comparison of smart card data and graphical usage patterns

Abstract This study attempted to validate the trend of long-term usage of public transport based on the information of the smart card holder and “long-term graphical usage patterns” that are obtained via a questionnaire survey. The study was conducted in Shizuoka prefecture, Japan, where the smart card “LuLuCa” was introduced in 2006. In this paper, the monthly usage (revealed usage) of smart card data was traced back from October 2011 to February 2017; on the other side, the usage survey was distributed in early 2017 to the targeting users (monitors) to obtain their “stated usage” over time. We found that the, from the smart card data, observed long-term usage dynamics are fairly in line with their chosen pattern, though biased to more recent observations. Further, regression analysis of individual patterns suggests the trend of “actual usage” can be explained with the chosen pattern (stated usage). This suggests that obtaining pattern information can be simple way for analysts to classify users and potentially predict their future demand.