A smart spontaneous crowd evacuation system for large multi-exit exhibition centers based on IoT

Abstract

Plans to mass evacuate visitors in an exhibition center in the case of emergency situations are critical for public safety and disaster management. Efficient crowd evacuation during mass gatherings has been an active research area during the past years. In this paper, we consider the challenging problem of finding in near real-time the most efficient and safest evacuation pathways in a multi-exit exhibition center while the fire hazard spreads. We first propose a system composed of sensor nodes to collect pertinent safety data associated with the changing environmental conditions. We then present a spontaneous dynamic evacuation system that considers the changing conditions in the risks associated with each hallway segment in terms of walking distance, heat, two major asphyxiant fire gases and crowd congestion. Our IoT-based system activates smart panels placed at major junctions of the hallways to visually guide evacuees towards the safest escape direction under the existing circumstances. The proposed algorithms aim to minimize the total evacuation time of all evacuees, while circumventing congested and perilous aisles, balancing traffic loads, and guaranteeing high scalability and reasonable computational efficiency. This work can pave the way towards the development of nextgeneration smart exhibition centers, where crowd safety is among the top priorities