EAGER: GeoAlerting: An Exploratory Case Study in Socio-Technical Alerting

University Of California-Irvine, Irvine CA

Investigators

Abstract

This project focuses on exploratory research towards the next generation of adaptive socio-technical crisis alerting and warning systems. Alerting systems inform entities affected by crisis ? schools, businesses, hospitals, and the public at large ? about impending dangers, the status of infrastructures, life lines, and available help and actions designed to reduce exposure to natural and human-induced threats ? e.g. evacuate, shelter-in-place etc. The eventual goal is to develop alerting systems that deliver an accurate message to the right targets in the right format and at the right time, however this requires a synergistic exploration of the challenges at multiple levels ? physical/geographical, network and user levels. This effort is an initial exploratory effort that aims to meaningfully combine the three levels into effective alerting schemes. Intellectual Merit: Specifically, this project offers a new model of geography based alert dissemination that incorporates both the geographical aspects and societal needs. While information needs are strongly correlated to the geographical location in the context of a natural or human induced disaster, this insight has never been fully exploited in the alert dissemination process. The principal investigator (PI) conjectures that societal/user information in combination with geographical context can be used to improve the alert dissemination process. In particular, the research aims to enhance the structural properties of a network to represent both the societal and physical connectivity. This requires an understanding of which aspects of a social network are useful to map and maintain as the situation evolves. The research addresses the above interdisciplinary research issues via the following 3 steps. The first research task develops mechanisms to create and maintain a geo-aware overlay network, in order to model geographical correlations. In the second step, the geo-aware overlay network is enhanced to capture and represent societal characteristics by using a geo-social mapping process, and the outcome of this process is a geo-social overlay network. The final research task is that of effective utilization of the geo-social overlay network in the alerting process, which is adaptive to the given knowledge of potential failure, in order to support reliable dissemination at the societal scale in the presence of geo-correlated failures. The project includes a prototype implementation to test/validate the research on a campus testbed. Broader Impact: The research paves the way towards a new generation of socio-technical alerting systems that are far more effective than alerting mechanisms in use today and opens new avenues for interdisciplinary research on technologies for alerting systems. If successful the project will lead to new ways of leveraging the capabilities of existing network infrastructures and quickly repurposing them in emergency situations to provide critical situation awareness and disaster management information to diverse organizations and individuals. The PI has strong ties with entities that have a stake in the outcome of this research, including local governments, emergency management agencies and organizations at the state and federal level, and university- and agency-based researchers. A workshop with stakeholders including academic researchers from multiple disciplines, industry participants and government agencies is part of this project to discuss current processes, systems and challenges in disseminating emergency alerts in public. Outreach efforts will also raise awareness of the upcoming communication models for socio-technical alerting systems; it will help demonstrate the possibility of synergistically combining information about the disaster, infrastructure and the public to achieve the desired level of response.