Role-based urban post-disaster mobility model for search and rescue operations

Abstract

Technology can highly improve search and rescue operations especially happening after a large scale disaster. However, before novel systems can be deployed in a large scale, the concepts need to be tested and validated through appropriate modeling and simulation tools. To this purpose, in this paper we present a mobility model addressing the vital initial stages right after the occurrence of an urban disaster event. We implemented the model into a simulator in order to study and refine the model itself, and used it to test applications targeting urban disaster scenarios. In order to show the capabilities of the model and of the simulator, we propose an application where victims are passively involved in the search and rescue operations through their mobile devices. Results show that victims' involvement in exchanging information has a great benefit for search and rescue operations.

Survey on Simulation for Mobile Ad-Hoc Communication for Disaster Scenarios

“…When a disaster occurs in an urban zone, the map and the geography of the scenario influence the movements of rescue workers and victims [49,52] . The movements of people and vehicles can be described over a map that contains road segments onto a 2-D plane [49] .…”

“…The movements of people and vehicles can be described over a map that contains road segments onto a 2-D plane [49] . More elaborated maps can be found in OpenStreetMap, which is a crowdsource application that provides the layered representation of real locations including buildings, streets, water sources, potential obstacles, altitude, and ground conditions [52] . This data can be converted into a format read by the simulator ONE [54] to enable the simulation of opportunistic networks.…”

“…Event-driven mobility models for disaster scenarios [47][48]52] aim at reflecting this type of behavior with role-based approaches, which are oriented to representing a different response of an entity to different events. An entity may be a civilian, a victim, a police, a firefighter, or an ambulance, among others; entities may react differently to different types of events.…”

“…[46] X X X Role/events, Nelson et al [47] X X X CORPS, Huang et al [48] X X X X PMD, Uddin et al [49] X X Composite, Pomportes et al [50] X X Cluster, Saha et al [51] X X GoTo, Costantini et al [52] X X X HBDA, Conceição and Curado [53] X X Note: X represents that the feature is explicitly supported by the model. The attributes to model the event are: its central location with sensing range, the attending zone where first responders can move, the event time and lifetime, the attending role list, and the resolution effort needed to resolve the attention of the event.…”

“…For example, a civilian can become a first responder when an event immobilizes a victim. Costantini et al [52] proposed that when an entity works on an event, its effect decreases over time.…”