Abstract

In the aim for a more sustainable transport system, governments try to stimulate multimodal trip making by facilitating smooth transfers between modes. The assessment of related multimodal policy measures requires transport models that are capable of handling the complex nature of multimodality. This complexity sets requirements for adequate modeling of multimodal travel behavior and can be categorized into three classes that are related to the range and combinatorial complexity of the available alternatives, the mathematical complexity of modeling the choice between them, and the complex effect of demand–supply interactions. Classical modeling approaches typically fail to meet these requirements and state-of-the-practice approaches only partly fulfill them. Therefore, the underlying hypothesis of this study was that the application of such models in network design implied an ill-advised decision-making process. Thus, these modeling approaches, as well as the promising state-of-the-research supernetwork approach, were conceptually compared with each other. Requirements for multimodality were constructed, and all three models were tested on the way in which these requirements can be met. The findings of this conceptual comparison were supported by realistic examples in the real-world transport network of the Amsterdam Metropolitan Area in the Netherlands. The theoretical shortcomings of the classical and state-of-the-practice approach were shown to indeed result in implausible predictions of multimodal travel behavior. The flexibility of the supernetwork approach, however, was very capable of describing the expected effect of supply changes on travel behavior in most situations. This study illustrates the urgency for applying sound multimodal modeling approaches in network design studies.