Agent-based Models in Supporting Pedestrian Transportation Planning and Design

Abstract

Agent-based models offer a new approach to understanding human-urban interactions in transportation systems, allowing individual entities within a system to be characterized with cognitive and behavioral properties. This paper discussed the role of agent-based representations of pedestrian transportation systems, detailing the underlying assumptions and techniques behind different types of pedestrian models and illustrating the differences between aggregate and individual agent representations. It then turns attention to the case study and the development of a cognitive pedestrian model as a way to illustrate the spectrum of potential spatial behaviors that are enabled by material changes to the transportation network. The paper concludes with a discussion and specific frameworks for employing agent-based models to support transportation planning decisions.