Multimodel agent-based simulation environment for mass-gatherings and pedestrian dynamics

TitleMultimodel agent-based simulation environment for mass-gatherings and pedestrian dynamics
Publication TypeJournal Article
Year of Publication2018
AuthorsKarbovsk, ii, V, Voloshin, D, Karsakov, A, Bezgodov, A, Gershenson, C
JournalFuture Generation Computer Systems
Date PublishedFebruary

Abstract The increasing interest in complex phenomena, especially in crowd and pedestrian dynamics, has conditioned the demand not only for more sophisticated autonomous models but also for mechanisms that would bring these models together. This paper presents a multimodel agent-based simulation technique based on the incorporation of multiple modules. Two key principles are presented to guide this integration: a common abstract space where entities of different models interact, and commonly controlled agents–-abstract actors operating in the common space, which can be handled by different agent-based models. In order to test the proposed methodology, we run a set of simulations of cinema building evacuation using the general-purpose {PULSE} simulation environment. In this paper we utilize crowd pressure as a metric to estimate the capacity of different emergent conditions to traumatically affect pedestrians in the crowd. The proposed approach is evaluated through a series of experiments simulating the emergency evacuation from a cinema building to the city streets, where building and street levels are reproduced in heterogeneous models. This approach paves the way for modeling realistic city-wide evacuations.