Pico-Scale Science for Pedestrian-Scale Solutions (PSS4PSS): A Computational Toolbox Leveraging Molecular Simulation for Pedestrian Dynamics

Abstract

Efficient and accurate simulations of pedestrian dynamics are critical for the smart cities of the future. In this work, we present a computational toolbox that accelerates such simulations relative to a popularly used pedestrian simulation tool by leveraging computational frameworks initially developed for molecular simulation. We make the argument that the field of pedestrian dynamics could benefit to a significant extent from a serendipitous interdisciplinary synergy with the molecular-simulation community. We provide arguments and representative examples in support of this premise, demonstrating that molecular simulation tools can be repurposed to solve precisely the same governing equations as traditional pedestrian-dynamics simulation tools, yielding the same results in significantly reduced computational time. We also describe a computational tool that we have developed that streamlines the conversion of indoor maps into boundary conditions for pedestrian simulations.