A modified version of the existing Cellular Automata (CA) model is proposed to simulate an evacuation experiment conducted in a classroom with and without obstacles. This work present the use of CA with neural network decision-making capabilities to simulate an exit-selection phenomenon in the experiment, and an intelligent exit-selection behavior was observed in our model. The experimental and simulation results are reasonable, while our simulation results agree with the experimental results quite closely. From the simulation results it is observed that occupants tend to select the exit closest to them when the density there is low, but if the density is high, they will go to an alternative exit so as to avoid a long wait. This reflects the fact that occupants may not fully utilize multiple exits during evacuation. The improvement of our proposed model is valuable for further study and for upgrading the safety aspects of building design.
Key words: Cellular automata, probabilistic neural network, floor-field model, output flux, intelligent agent.
Copyright © 2023 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0