Modelagem e avaliação da integração temporal em uma rede de transporte de ônibus usando grafos temporais

Abstract

Urban mobility in large cities has faced growing challenges with the introduction of new private transport options and progressive decrease in the number of users of public transport. In this context, public transport is updated for including new information and communication technologies to increase the attractiveness of users. However, handling the large volume of data provided by the operation, particularly from location sensors installed in vehicles, is one of the greatest challenges in the area of transport management, planning and operation. This work aims to develop a time-varying graph model for the operation of bus transport. A graph database is generated from open data of daily operation of the Curitiba public transportation. By using this database, complex network metrics are then applied to evaluate the transport network topology, highlighting the importance of different bus stops. A methodology was proposed to identify possible areas for temporal integration, i.e., for creating integrated “virtual terminals” in order to allow passengers to change buses without paying a new fare. Based on this methodology, the temporal integration of bus stops in four areas was proposed. The impact of the temporal integration on metrics of the static network was presented. Finally, the quality of temporal integration was evaluated through the correlation of stopping times of buses at the same area. It allows to identify the times at which the connection of passengers can be made in a reduced time. Although the results were carried out for the transport system of Curitiba, the model can be adapted for bus transport of other cities.