Dynamic Models Analysis for Large-scale Networks



Dynamic Models Analysis for Large-scale Networks

Authors

Martijn Breen, TSS - Transport Simulation Systems S.L., Josep Peranau, TSS - Transport Simulation Systems S.L., Jordi Casas, TSS - Transport Simulation Systems S.L.

Description

This paper presents the integration of a Dynamic Macroscopic model inside the integrated framework of Dynamic Traffic Assignment for urban network analysis within Aimsun.

Abstract

Traditionally traffic demand models require as input the impedance of a demand with respect to the network supply, for example, mode choice or departure choice considers the associated travel time to each consideration. Considering this objective, the main critique with respect to the evaluated travel times using static models is that the estimation of the travel time could diverge considerably taking into account that these models are capacity unconstrained, additionally, dynamic models, such as mesoscopic models, have an excess of detail that is sometimes unnecessary for the final requirements. The Quasi-dynamic model developed in Aimsun represents an intermediate approach between static and dynamic models which is a model with capacity constraints on a link level and that incorporates a sound node model to represent junction dynamics. This model calculates travel time by considering the free flow travel time and the average delay accumulated in vertical queues.

This paper presents the integration of a Dynamic Macroscopic model inside the integrated framework of Dynamic Traffic Assignment for urban network analysis within Aimsun. This model represents the evolution of a Quasi-dynamic model in two respects, firstly, the dynamic component that allows the evaluation of the dynamicity of the demand and supply and secondly, the queue spillback propagation.

This paper evaluates a comparison of the Dynamic Macroscopic model with respect to the Quasi-dynamic and mesoscopic model in terms of travel time estimation. The evaluation is performed using real networks validated with real data sets and finally concludes with a summary of advantages, disadvantages and strengths and weaknesses of each model.

Publisher

Association for European Transport