Abstract: Interorganizational coordination is crucial among human services providers responsible for responding to both personal and widespread crises. Too often, however, agencies providing disaster relief, shelter, and connection to other social service systems operate in information silos. Moreover, organizations that assist the same people may be duplicating services or ineffectively providing services to those in need. In the past, there has been no easy way for human service organizations to share information about clients, resources, and services. Over the last decade, distinct initiatives have begun to standardize data collection, storage, and transmission standards within human service domains. This paper describes several human services standards currently in use or under development in the United States and discusses how each support distinct, yet related, human service information management during disasters. The paper concludes with a call for the development of an overarching human services data interoperability standard.

Abstract: In this paper we describe our work in progress on RAVEN, a framework, which makes it possible to build applications for collaborative editing of structured data on Android. RAVEN offers developers compile time tools, which use only the schema to generate all database handling components, edit and list user interfaces, as well as those needed for data synchronization, significantly reducing development effort. In addition, RAVEN also offers the ability to do the same work, entirely at runtime, using only a smartphone. With RAVEN it is possible to construct data oriented applications on phone at any time, including during a disaster. Users can share their applications simply by sharing the database and corresponding schema. Thus, RAVEN enables completely decentralized application creation, sharing, and data distribution, avoiding issues of connectivity to centralized resources. In this paper we show that with RAVEN it is possible to construct a new application at runtime and compare the results with an equivalent custom-built application. © 2012 ISCRAM.

Abstract: This paper describes an information system designed to be deployed in emergencies caused by sudden onset natural disasters. The aim is to streamline the communication flow and collaboration between media, aid workers and government agencies with the affected population, to help the latter get verified, accurate and actionable information that will enable them to make decisions and recover from the disaster. The Emergency Information Service (EIS) system also provides means for affected population and field workers to channel vital data back up into aid response. This tool is part of a free information service run by Thomson Reuters Foundation to help survivors of natural disasters. It will serve the affected populations, local media and relief responders by providing fast, practical and verified information in local languages through the best means available.

Abstract: Social media applications such as Twitter and Facebook are fast becoming a key instrument in gaining situational awareness (understanding the bigger picture of the situation) during disasters. This has provided multiple opportunities to gather relevant information in a timely manner to improve disaster response. In recent years, identifying crisis-related social media posts is analysed as an automatic task using machine learning (ML) or deep learning (DL) techniques. However, such supervised learning algorithms require labelled training data in the early hours of a crisis. Recently, multiple manually labelled disaster-related open-source twitter datasets have been released. In this work, we create a large dataset with 186,718 tweets by combining a number of such datasets and evaluate the performance of multiple ML and DL algorithms in classifying disaster-related tweets in three settings, namely ``in-disaster'', ``out-disaster'' and ``cross-disaster''. Our results show that the Bidirectional LSTM model with Word2Vec embeddings performs well for the tweet classification task in all three settings. We also make available the preprocessing steps and trained weights for future research.

Abstract: When disaster strikes, the emerging task environment requires relief agencies to transform from autonomous mono-disciplinary organizations into interdependent multidisciplinary decision-making units. Evaluation studies reveal that adaptation of information management to the changing task environment is difficult resulting in poor information quality, indicating information was incorrect, outdated or even unavailable to relief workers. In this paper, we adopt a theory-driven approach to develop a set of information management roles and dynamic capabilities for disaster management. Building on the principles of advance structuring and dynamic adjustment, we develop a set of roles and capabilities, which we illustrate and extend using two field studies in the Netherlands. By studying regional relief workers in action, we found that in tactical disaster response decisionmaking units, several information management roles are not addressed and that information managers are preoccupied with information gathering and reporting, whereas information quality assurance is not on the agenda.

Abstract: Relief agency managers show growing interest in dashboards for assessing multi-agency disaster preparedness. Yet, there is a dearth of research on the development and use of dashboards for disaster preparation. Consequently, information system architects in the disaster management domain have little guidance in developing dashboards. Here, dashboards refer to digitalized visualizations of performance indicators. In this paper, we discuss the experiences gained from an action research project on the development of dashboards for assessing disaster preparedness. The objective of this paper is to discuss experiences and tradeoffs extracted from the development of dashboards in practice. We organized a two-day gaming-simulation with relief agency managers for the evaluation of the dashboards. While the relief agency managers acknowledged the usefulness of dashboards in the disaster preparation process and expressed their intention to use these in practice, they suggested that the formulation and clustering of performance indicators requires further research.

Abstract: In the preparedness activity for disasters and emergency management, serious games can help in training medical first responders by providing emergency simulations which are always available, safer and cheaper than real-world simulations. However, serious games for training emergency medical services (EMS) must take into account the presence of different actors in crisis situation like police and firefighters and the high volume of (medical as well as non-medical) expert knowledge. The aim of our approach is not only to acquire technical skills but also to develop the capability to act, to cooperate and coordinate in non-procedurally previewed emergency situations. This paper proposes both (i) a detailed prototype of a serious game's scenario that supports instructors in the training in EMS and (ii) an adaptive infrastructure A.R.G.I.L.E (Architecture for Representations, Games, Interactions, and Learning among Experts). We illustrate our ideas on an example of a complex road accident. This work is done with the collaboration of a Hospital Emergency Department implied with us in a R&D project.

Abstract: Emergency management in highly dynamic situations consists of exploring options to solve a planning problem. This task can be supported through the use of visual cues that are based on domain knowledge of the current domain. We present an approach to use visualization of critical constraints in timelines and hierarchical views as decision support in mass-casualty emergency situations.

Abstract: The design of Emergency Management Systems is an activity that requires knowledge from various related domains for providing a more complete and usable solution. In this context, design patterns including knowledge from previous experiences can be a useful source of information to support the development of this type of applications. In this paper, we introduce a catalog of design patterns for Web-based Emergency Management Systems collected from design principles, design patterns and existing implementations from involved areas, taking into account requirements particular to this domain.

Abstract: Cooperation and mutual assistance in emergency situations is one of the main objectives of the Latin-American Association of Governmental Organisms of Civil Defense and Protection. To promote such collaboration m-ARCE has been developed; an ubiquitous mobile office for disaster mitigation where users can send and receive information anywhere and anytime. When a catastrophe happens in a country, and almost all infrastructure is destroyed, mobile technology, such as mobile devices and wireless networks, offers the user resources to ask for help and to manage it. Latin-American Countries often suffer catastrophes that provoke numerous human losses and major economic and social problems. International assistance and collaboration with the affected country is necessary to help in its recovery. The Web, like Internet, offers static office services to users who can access information using an infrastructure in indoor environments. On the other hand, mobile computing and networking use the Internet, together with mobile physical devices linked to it, and software platforms built upon it, to design and coordinate systems across countries. In the ubiquitous mobile office design, we describe how services, such as chat, email and wireless communication, should be configured for emergency situation. We make use of ubiquitous hypermedia -linked nodes in ubiquitous spaces- to ensure mobility and accessibility to the mobile device interface, such as PDAs and smartphones.

Abstract: Understanding the interdependencies of critical infrastructures (power, transport, communication, etc.) is essential in emergency preparedness and response in the face of disasters. Unfortunately, many factors (e.g., unwillingness to disclose or share critical data) prohibited the complete development of such an understanding. As an alternative solution, this paper presents a conceptual model-an ontology-of disasters affecting critical infrastructures. We bring humans into the loop and distinguish between the physical and social interdependencies between infrastructures, where the social layer deals with communication and coordination among representatives (either humans or intelligent agents) from the various critical infrastructures. We validated our conceptual model with people from several different critical infrastructures responsible for disasters management. We expect that this conceptual model can later be used by them as a common language to communicate, analyze, and simulate their interdependencies without having to disclose all critical and confidential data. We also derived tools from it.

Abstract: When a disaster or emergency occurs, one of the most pressing needs is to establish a communication network for the first responders at the scene. Establishing and accessing a reliable communication infrastructure at a crisis site is crucial in order to have accurate and real-time exchange of information. Failure in the exchange of timely and crucial information or delay in allocating resources impedes early response efforts, potentially resulting in loss of life and additional economic impact. At a disaster site, the existing communication infrastructure may be damaged and therefore partially or totally unavailable; or, there may not have been previously existing infrastructure (as in the case of remote areas). A communication infrastructure within the context of emergency applications should be reliable, easily configurable, robust, interoperable in a heterogeneous environment with minimum interdependencies, and quickly deployable at low cost. A disaster scene is a chaotic environment which requires a systematic approach to abstract the system, study the flow of information and collaboration among different disciplines and jurisdictions to facilitate response and recovery efforts. We have deployed the wireless mesh infrastructure in several drills at the university campus and in the city as part of the California Institute for Telecommunications and Information Technology (Calit2) NSF-funded RESCUE project (Responding to Crises and Unexpected Events). To evaluate network performance and identify the source(s) of bottleneck, we have captured the network traffic. The lessons learned from test bed evaluations of the network based on real-world scenarios can be applied to future applications to enhance the network design and performance.