Modulated spatiotemporal clustering of smart card users

Smart card data offers an in-depth understanding of the travel behavior of public transport users. An efficient way to analyze public transport users is to group them into different clusters with similar behaviors. However, this clustering process should take into account space and time because both of these dimensions characterize daily trips. Depending on the outcome, we might wish to give more importance to space or to time, or we might wish to balance the two. In this study, we present a spatiotemporal clustering tool that permits modulation regarding the importance of space versus time. We then test this tool with different values for the space-time balance parameter to evaluate the influence of this parameter on the results. The method has been applied to 769,614 smart card transactions of the Reseau de transport de la Capitale (Quebec City, Canada). Results show that the influence of space and time can indeed be controlled, and that the types of clusters obtained vary whether one or both of the dimensions are considered.

Automated summary

This study presents a spatiotemporal clustering tool that allows for adjusting the importance of space and time. By testing different parameter values, it is found that the influence of space and time can be controlled and the obtained clusters vary depending on whether one or both dimensions are considered.