Vladislav Karbovskii, Daniil Voloshin, Andrey Karsakov, Alexey Bezgodov, Carlos Gershenson, "Multimodel agent-based simulation environment for mass-gatherings and pedestrian dynamics", In Future Generation Computer Systems, 2016, ISSN 0167-739X, https://doi.org/10.1016/j.future.2016.10.002.
In their article, Karbovskii, Voloshin, Karsakov, Bezgodov, and Gershenson use multimodel agent-based simulation to predict how the evacuation of a public space might affect those within. The agents move in systems that represent real-world areas, allowing the authors to explore how different structures and city layouts facilitate or hinder mass evacuations. The different factors in motion and surrounding feed into a social-force model that simulates crowd pressure, where high crowd pressure corresponds to a higher chance of harm to the agents. They check their work for plausibility by comparing it to cases from emergency crowd management planning procedures. This model may be used as a basis to design buildings or cities that facilitate safer evactuations. Extensions might include a response team of agents that act counter to the majority of agents or make the agents more intelligent in decisions while evacuating.