Abstract: Disaster managers report that several disasters would have turned out on a dramatic scale without spontaneous unaffiliated on-site volunteers (SUV). Since SUVs are usually not integrated in chains of command and behave in a certain pattern of its own, coordination of SUVs becomes a challenge for disaster management. One key to coordination is communication and adequate support by information systems. However, real disasters or field tests are usually too expensive, elaborate, and partly impossible when coordination of SUVs is to be exercised or novel tools and methods must be evaluated. Simulating the SUV's behavior by software-agents is considered a constructive solution, however, the specification of simulation settings is an open research field. Therefore, this paper aims at identifying relevant attributes affecting SUVs behavior by a state-of-the-art literature review, classifying and discussing the attributes. Our results provide a sound basis for defining SUV-agents and performing suitable simulations in the future.

Abstract: Recent disasters have revealed growing numbers of citizens who participate in responses to disasters. These so-called spontaneous unaffiliated on-site volunteers (SUVs) have become valuable resources for mitigating disaster scales. However, their self-coordination has also led to harm or putting themselves in danger. The necessity to coordinate SUVs has encouraged researchers to develop coordination approaches, yet testing, evaluating, and validating these approaches has been challenging, as doing so requires either real disasters or field tests. In practice, this is usually expensive, elaborate, and/or impossible, in part, to conduct. Simulating SUVs' behaviors using agent-based simulations seems promising to address this challenge. Therefore, this contribution presents a conceptual model that provides the basis for implementing SUV agents in simulation software to perform suitable simulations and to forecast citizens' behaviors under a given set of circumstances. To achieve adequate simulations, the conceptual model is based on the identification of 25 behavior-affecting attributes.

Abstract: The evaluation of an emergency scenario is often based on the use of simulation models. The specificity of these models involves the need for a complex evaluation of the problem domain, including the physical conditions behind the considered threat. Based on emergency occurrences data, provided by the Portuguese National Civil Protection Authority, we are currently developing a methodology for evaluating a real situation, based on past occurrences. The aim is to develop a platform that will enable the evaluation of a risk scenario based on existing civil protection data. The methodology under development should enable the evaluation of different scenarios based on the collected available data. This will be achieved thanks to the facilitated configuration of several aspects, such as the geographical region and relevant properties of the considered threat. In this paper, we describe the methodology development process and the current state of the platform for risk evaluation.

Abstract: In this article, we present a novel approach, which allows not only to identify potential cascading effects within a network of interrelated critical infrastructures but also supports the assessment of these cascading effects. Based on percolation theory and Markov chains, our method models the interdependencies among various infrastructures and evaluates the possible consequences if an infrastructure has to reduce its capacity or is failing completely, by simulating the effects over time. Additionally, our approach is designed to take the intrinsic uncertainty into account, which resides in the description of potential consequences a failing critical infrastructure might cause, by using probabilistic state transitions. In this way, not only the critical infrastructure's risk and security managers are able to evaluate the consequences of an incident anywhere in the network but also the emergency services can use this information to improve their operation in case of a crisis and anticipate potential trouble spots.

Abstract: Situational awareness (SA), the ability to perceive, comprehend and predict situation around you and it is a key in attending any incident as critical foundation for successful decision-making. Because incidents are solitary events, development and assessment of SA presents a significant challenge. In this article we analyze the authoring process of twenty-two scenarios implemented in the XVR on-scene virtual simulation software used to assess rescue incident commanders' (ICs) SA. To allow the scenarios to be used by different assessors, the Collaborative Authoring Process Model for Virtual Simulation Scenarios (CAPM) was developed. In Estonia, 473 assessments were recorded in Effective Command database and analysed by all three levels of SA as recommended by Endsley (2000). Introduction of CAPM resulted in scenarios being re-used by different assessors for authentic SA measuring. In the last sections of this article, we introduce our suggestions to improve virtual scenario design and SA research.

Abstract: A primary component of disaster response is training. These educational exercises provide responders with the knowledge and skills needed to be prepared when disasters happen. However, traditional training methods, such as high-fidelity simulations (e.g., real-life drills) and classroom courses, may fall short of providing effective and cost-efficient training that is needed for today's challenges. Advances in technology open a wide range of opportunities for training using computer-mediated simulations and exercises. These exercises include the use of mixed reality games and wearable computers. Existing studies report on the usefulness of these technologies for training purposes. This review paper synthesizes prior research and development of disaster response simulations and identifies challenges, opportunities, and lessons learned. Through this review, we provide researchers and designers with an overview of current practices in designing training simulations and contribute practical insights into the design of future disaster response training.

Abstract: In aiming to decrease the number of casualties and people with difficulty in wide-area evacuations due to a large earthquake, it is highly important to visualize and quantify the potential danger in residential areas. In this paper, we construct a multi-agent simulation model, which describes property damage (such as building-collapse, the spread of fire and blocking of streets) and people?s evacuation behavior after an earthquake occurring. Using this simulation model, we quantify the wide-area evacuation difficulty in densely-built wooden residential areas, and evaluate the past project to improve buildings and streets based on this indicator. Furthermore, we demonstrate the effects of adding new evacuation routes between two intersections of streets with narrow width and long distance. Through these case studies, the effectiveness of our simulation model on urban disaster mitigation planning is shown.

Abstract: In the recent months, many measures have been taken by governments to fight the COVID-19 pandemic. Due to the unknown properties of the disease and a lack of experience with handling pandemics, the effectiveness of measures taken was often hard to evaluate the effectiveness of measures, leading to inefficient measures and late execution of efficient measures. Many models have been proposed to evaluate the performance of these measures on the spreading of a pandemic, but these models are commonly vastly simplified and, thus, limited in expressiveness. To extend the expressiveness of the models, we developed a epidemic simulation inside of a flexible and scalable city simulation game to analyse the counter measures to a pandemic in this city and spot common places of infection on a microscopic level. The configurability of our developed epidemic simulation will also be useful for potential future pandemics.

Abstract: Concepts such as trust, shared understanding, cultural differences, mental workload, and organizational structure all impact upon the effectiveness of an organization (e.g., Tindale & Kameda, 2000), and even more so in the context of large scale multinational operations (e.g, Smith, Granlund, & Lindgen, 2010). In order to study these concepts we plan a multinational, distributed experiment with participants from three nations collaborating in the same virtual environment: Canadian, British, and Swedish participants will work together as part of a multinational MTS to deal with a complex task and gain control of a crisis situation. Empirical research on MTS remains limited (see, e.g., DeChurch & Marks, 2006) particularly at the multinational level where the investigation of MTS has been so far focused on case studies and exercises (e.g., Goodwin, Essens, & Smith, 2012). Therefore, there is a need to empirically study multinational MTS in order to assess the specific issues that multinational operations face, notably cultural and languages differences. The simulation environment used as experimental platform for this project is C3Fire (www.c3fire.org, Granlund & Granlund, 2011). C3Fire creates an environment whereby teams must work together to resolve a crisis in the firefighting domain, with the goal of evacuating people in critical areas, putting out the forest fire, and protecting buildings and other areas of value from the burning forest fire. This platform makes it possible to study participants' collaborative processes when dealing with a set of crisis scenarios in the context of a simulated emergency response situation. To deal efficiently with the crisis management operation, participants need to prioritize between different objectives, identify and protect critical areas, and plan and implement activities based on given resources. All these tasks are distributed between team members, compelling participants to exchange information and coordinate within and between teams to execute the task. The task is divided into three areas of responsibility as follows: 1) Information and Planning, responsible for situation assessment and providing the operating picture; 2) Operation and Logistic, responsible for intervention and resource management; and 3) Search and Rescue, responsible for research and management of civilians. C3Fire is designed to: 1) achieve an optimal compromise between internal and external validity; 2) show flexibility in scenario configuration (spectrum of units and roles – including search and rescue functions; Tremblay et al., 2010), allowing researchers to capture emergency response and crisis management and rapid response planning; 3) be highly configurable for testing many different types of teams (e.g., hierarchical vs. horizontal organizations); and 4) readily provide objective, non-intrusive metrics for assessing teamwork effectiveness (including macrocognitive functions and team processes) as well as quantitative measures of task performance (that take into account conflicting mission goals). © 2012 ISCRAM.

Abstract: This paper describes the field tests of a simulation based game aiming at raising awareness and creating a deeper understanding of the dynamics of the comprehensive approach (CA). The setting of this game is that of a failed state where an UN intervention takes place after massive conflict that requires a CA to stabilize the situation. That is, the civil and military actors need to collaborate effectively, taking into account their respective strengths, mandates and roles. Underlying the game is the Go4it CA simulation Model (GCAM2.0). GCAM2.0 was extensively field-tested in eight sessions with about 16 persons each, aiming at assessment of the perceived realism and learning effects. It was found to provide a sufficiently authentic experience to obtain awareness of the CA in novices. With regard to improving the deeper understanding of the dynamics and complexity of the CA, in a cooperation-oriented setting only deeper learning can be reached.

Abstract: Automated coordination is regarded as a novel approaches in Emergency Response Systems (ERS), and especially resource allocation has been understudied in former research. The contribution of this paper is the introduction of two variants of a novel resource allocation mechanism that provide decision support to the centralized Emergency Operations Center (EOC). Two quantitative models are computationally validated using real-time, data-driven, Monte-Carlo simulations promoting reliable propositions of distributed resource allocations and schedules. Various requirements are derived through a literature analysis. Comparative analyses attest that the Monte-Carlo approach outperforms a well-defined benchmark.

Abstract: This paper describes some of the lessons learned from the FireGrid project. It starts with a brief overview of the project. The discussion of the lessons learned that follows is intended for others attempting to develop a similar system, where sensor data is used to steer a super-real time simulation in order to generate predictions that will provide decision support for emergency responders.