3D INDOOR-PEDESTRIAN INTERACTION IN EMERGENCIES: A REVIEW OF ACTUAL EVACUATIONS AND SIMULATION MODELS
- School of Built Environment, UNSW Sydney, Red Centre Building, Kensington NSW 2052, Sydney, Australia
Keywords: Evacuation movement, Evacuation behaviour, 3D space, Assessment, Evacuation models
Abstract. Pedestrian motions and behaviours in evacuations are compactly interrelated indoor environments. According to how pedestrians interact with indoor environments in three-dimensional (3D) space, 3D indoor-pedestrian interaction is defined as five sorts of specific pedestrian motions, i.e., stepping over, crawling, bent-over walking, jumping over and climbing over. However, the occurrence and prevalence of the interaction in actual evacuations have not been identified, and the degree of effectiveness and practicality of the evacuation simulation models regarding the interaction is yet to be evaluated. Understanding the interaction and simulating it realistically bring apparent benefits to support informed policymaking, building safety optimization and emergency relief efforts. Therefore, this article aims to review the 3D indoor-pedestrian interaction in actual evacuations, assess the capabilities and performance of the evacuation simulation models for the interaction, and identify their limitations and challenges for future work. We first demonstrate that the fundamental relationships between the decision-making process of a pedestrian, object distance and object size/shape in 3D space are unclear yet, and the scarcity of experimental investigation for local routes choices and each motion is vast. Subsequently, the comparisons between Social Force models, Cellular Automata models and Agent-based models suggest that these microscopic models can yield the 3D indoor-pedestrian interaction, but the height dimension of 3D space is not explicitly considered during the evacuation modelling. This article is expected to advance the understanding of pedestrian evacuations in 3D space, ultimately improving pedestrian safety in indoor environments.