Mesoscopic modeling framework for public transport traffic

Abstract

We present a general framework for incorporating the passenger-centric viewpoint in a mesoscopic traffic flow model. We modify the Open Traffic Models (OTM) platform to take into account stops along links. In contrast to agent-based models that focus on interactions and mechanisms at a small scale, our mesoscopic queuing model framework scales well to urban-scale applications. Starting with a scenario that demonstrates the Braess paradox for a 100% car system, we show how prioritized public transportation on the link that causes the Braess paradox leads to a satisfactory solution.