This work analyzes the daily behaviour of users by using their mobile data. In this paper, we define the routine to be the pattern of users appearing at certainplaces at certain times of day. The aim of this article is to extract routine patterns by analyzingthe location sequences obtained from users' mobile data. Since the users' location sequence issparse and we want to obtain the users' activity pattern during a single day, we analyze theusers' location sequence for each day of the week. First, the mobile data is pre-processed toobtain the average one-day location sequence of each user. Then, a model is built to obtainthe dissimilarity between the one-day location sequences of users. Finally, all one-day locationsequences are clustered based on the obtained dissimilarity.