The aim of this research is to measure, using smart card data, how much do
public transport users change their behavior through time. To quantify the
change in behavior, we split the smart card records of a user into a set of
separated time windows. Then, we measure the variation between each pair of
time windows. Three algorithms that calculate the variation in users' mobility
are assessed. Using data from a Canadian transport system, we show that
measuring the stability of user behavior at an individual level provides new
insights for public transport operators, e.g., it can be used to measure users'
adaptability to changes in the transport system.