Data-Driven Public Transport Routes and Timetables Based on Anonymized Telecom Data

Human activities, including daily urban commuting, usually demonstrate a mix of sustainable and changing behavioral patterns. The development and continuous update of optimal routes and schedules for public transport is a key element in satisfying the transport demands of dwellers. The data available demonstrates a perceived private car dominance over public transport use in Bulgaria. The limited capabilities of cities to develop and introduce optimal routes and schedules of public transport services based on citizens’ actual demands and needs is considered a major shortcoming. The aim of the article is to validate an approach for public transport optimization based on anonymous and statistically aggregated mobile phone positioning data. Our results, which focus exclusively on this novel kind of data, supplement earlier studies on the use of Seasonal ARIMA models in short-term passenger demand forecasting. This allowed us to make a statistical assessment and forecast potential future passenger flow for our chosen routes at different time ranges of the day. Data-driven public transport routes and timetables would make public transport more attractive and useful. Reducing the share of private car trips will improve the air quality and will help lower the level of noise pollution and CO2 emissions in our cities.