Definition and evaluation of flexible transport systems: the case of Vila do Conde, Portugal
Authors
Abstract
In the last decades, the public transport sector has been re-evaluating its role in the society. The need of social inclusion and quality of transport service, especially in a context of economic austerity, calls for innovative transportation solutions. Demand Responsive Transportation (DRT) systems can potentially address these issues with a service that only runs when and where necessary, with routes and frequencies that may vary according to the observed demand, by adapting solutions in real-time. The advantages of DRT in terms of social cohesion, mobility, traffic, environment and service quality seem fairly obvious. However, these services may lead to financial unsustainability or insufficient service quality levels, which implies a more advanced operational management with (real time) solutions involving multiple objectives.In the scope of this project, a DRT simulation software tool is used to analyse a real case study. The simulation is based in stochastic processes and enables service cost and quality evaluation with indexes as mean pickup time delay and total distance travelled. A survey was developed addressing the significant traffic generating activities, located in Vila do Conde, a small town to the north of Portugal, related with work, education, health and social action – activities that induce high levels of people flows. The resulting data was processed to obtain an estimation of the demand discretized in time, per village, in the form of a OD Matrix, which was complemented by a study of the transport needs to develop DRT design drafts for simulation.
The results highlight the importance of the simulation approach in the design of DRT services. We propose a service design for a school service for students and for a service mostly dedicated for health care centre users and for other kind of travel purposes. Both services were simulated with very distinct setups in terms of schedule, temporal distribution of requests, type of reserve and route, being each one simulated in dozens of scenarios and during hundreds of service hours.