A flexible framework for developing integrated models of transportation systems using an agent-based approach

Travel demand, traffic flow and land-use models are typically modeled in a decoupled way, i.e. each of the components is modeled separately assuming that parameters related to the other components are fixed. Moreover, the models are often developed by different groups for different contexts, requirements, etc. In this paper we present a prototype of a software framework which allows the user to develop an integrated simulation of a transportation system and also to link additional models to the new simulation in a standardized way. We use an agent-based approach as the basis of such a model. Integrated transportation system models allow model users to overcome the limitations of traditional aggregated, independent transportation models, particularly with respect to sensitivity to behavioral aspects of the travelers. Another requirement, which the software is to satisfy, is the interoperability of models developed in the new framework with legacy models. By interoperability we mean, that any component of the of the model can be interchanged by a legacy software and be used for the integrated simulation. This would allow disparate research groups working on modeling different aspects of a transportation model to plugnplay their models into the framework and test those as a part of an integrated model of an entire system, providing a benefit to researchers, modelers and institutional users of such models. (C) 2012 Published by Elsevier Ltd.