Conference:
31st IEEE International Symposium on Computer Communications (ISCC 2026)
Authors:
Gelenbe E, Ma Y.
Abstract:
Efficient and reliable evacuation management during emergency situations becomes very challenging, particularly in spatially constrained environments such as passenger ships. Recent advances in the Internet of Things (IoT) provide new opportunities to improve evacuation safety and efficiency through real-time sensing, communication, and decision-making. In this paper, an IoT-assisted evacuation framework is studied using an analytical queueing network model. A Pull-based evacuation policy is proposed, in which evacuees are instructed to proceed only after preceding evacuees have exited the evacuation system, thereby mitigating congestion propagation along the downstream corridors, stairways, and passageways. The proposed policy relies on active IoT devices to deliver real-time visual and auditory guidance, including waiting, movement, and routing instructions. The effectiveness of the IoT-based Pull Policy is evaluated through a realistic case study of evacuation on a commercial cruise ship. The queueing network formulation enables computationally efficient performance analysis and systematic comparison of multiple routing strategies under heterogeneous evacuee walking speeds, and shows that the proposed approach significantly reduces congestion and enhances overall evacuation efficiency.
