Urban travel dynamics usually demonstrate a mix of sustainable and changing behavioral patterns that reflect local specifics, such as habits of residents, public holidays, traffic density, seasonality, and weather. This paper presents an original approach for modeling the passenger flow to and from subway stops in suburban areas of cities based on an anonymized and statistically aggregated telecom operator dataset. The primary data source analyzed in this study was obtained through a collaboration agreement with one of the three main mobile network operators in Bulgaria and contains information about trips from the “Manastirski Livadi” residential complex, located in the suburbs of the city of Sofia, to several different destinations in the capital city. The data are grouped into three independent clusters of travelers: (1) within the “Manastirski Livadi” residential complex, (2) to destinations that are accessible by subway or tram, and (3) to destinations to which there is no convenient “green” public transport. For each of these three groups, we developed Markov regime-switching models suitable for analyzing passenger flow behavior. We tested the model for different clusters of travelers, specifically in the city of Sofia. The results obtained can be used to create optimal routes and timetables for public transport in suburban areas and to analyze and improve suburban route networks in other regions.