Estimating the residence zone of frequent public transport users to make travel pattern and time use analysis

Public transport systems with electronic fare collection devices continuously store data related to trips taken by users, which contain valuable information for planning and policy analysis. However, if the card is not personalized, there is no socioeconomic information available, which imposes a limitation on the types of analysis that can be performed. This work presents a simple method to estimate the residence zone of card users, which will allow socioeconomic variables to be estimated, thereby enriching the analytical possibilities. The method, which is based on the observation of morning transactions of frequent users, is applied to a sample of over 2 million cards. The method is evaluated using a sample from the Santiago ODS where users declared their card id and also declared their home address. A sample of 888,970 cards that are observed at least three days in a week and show spatial regularity for the morning transaction is used for zone of residence estimation and analysis of travel patterns and time use. The results show that users who live in the city center or in the wealthier East zone experience lower travel time, spend more time at home and less time at work.