Abstract: Partial or total horizontal evacuation of populations in urban areas is an important protection measure against a natural or technological risk. However, casualties during massive displacement in a context of stress and in a potentially degraded environment may be high due to non-compliance with instructions, accidents, traffic jams, incivilities, lack of preparation of civil security or increased exposure to hazards. Working in evacuation plans is therefore fundamental in avoiding casualties caused by improvisation and in promoting self-evacuation whenever possible. Since it is impossible to re-create the conditions of a crisis on the ground to assess such evacuation plans, there is a need for realistic models in order to evaluate them using simulations. In this paper, we present the ESCAPE software framework that helps in the development of such plans and testing them. In particular, ESCAPE, which uses the GAMA open-source platform as a core component, provides an agent-based simulation tool that supports simulation of the evacuation of a city's population at fine temporal and Geographical scales. The framework was developed such that it works for a wide range of scenarios, both in terms of hazards, geographical configurations, individual behaviors and crisis management. In order to show its adaptability, two applications are presented, one concerning the evacuation of the city of Rouen (France) in the context of a technological hazard and the other pertaining to the evacuation of the district of Hanoi (Vietnam) in the event of floods.

Abstract: The number of smartphone applications to alert and inform the population in a risk situation in France is too large
and these solutions are still unknow by the population. This study proposes an evaluation protocol based on various
indicators, which take into account the capacity of the applications to send a targeted alert, their attractiveness, the
ability of individuals to emit information and number of hazards considered. The results obtained on 50
applications deployed in France show that very few of them meet the objectives of the alert, in the sense defined
by civil security, because of a single-risk approach, a unique sense of communication, and the low acceptance of
these solutions by citizens.

Abstract: This paper discusses the development of the CEM-DIT (Communication in Emergency Management through Data
Integration and Trust) system, which allows decision makers in crises to send out automated data requests to multiple
heterogeneous and potentially unknown sources and interactively determine how reliable, relevant and trustworthy
the responses are. We describe the underlying technology, which is based partially on data integration and matching,
and partly on utilisation of provenance data. We describe our cooperation with the Urban Observatory (UO), which
allows us to develop the system in collaboration with developers of the kind of crisis-relevant data which the system
is designed for. The system is currently in development, and we describe which parts are fully implemented and
which are currently being developed.

Abstract: Crisis Management (CM) refers to the ability to deal with crisis tasks in different phases and iterations. People working in a crisis are generally under stress to make the right decision at the right time. They have to process large amounts of data and to assimilate the received information in an intuitive and visual way. Visual Analytics (VA) is potentially useful to analyze and understand the huge amount of data in several areas including in a crisis. We designed a survey protocol to understand which themes influence visualizations to support CM. In previous work, we carried out systematic mapping studies, analysis of official documents, ethnographic studies, questionnaires during the large events held in Brazil in recent years. In this work, we interviewed eight CM specialists. We analyzed this data qualitatively with the coding technique. Then we evaluated the coding results with the focus group technique. With the results, we identified the relationships between the visual needs and other main themes of influence for CM. This thematic synthesis enabled us to build a draft model based on VA.
We hope that, after future cycles of validations and improvements, the agencies that manage crises might use this model as a reference in their activities of knowledge production and decision-making.

Abstract: This paper reports on the added value of combining different types of sensor data and geographic information for fire incident management. A survey was launched within the Belgian fire community to explore the need of added value and the use of new types of sensor data during a fire incident. This evaluation revealed that people are visually-oriented and that video footages and images are of great value to gain insights in a particular problem. However, due to the limited available time (i.e., fast decisions need to be taken) and the large amount of cameras it is not feasible to analyze all video footages sequentially. To solve this problem we propose a video summarization mechanism and a video highlight selection tool based on the automatic generated image and video tags.

Abstract: With the growth of social media and crowdsourcing in disaster response, further research is needed on the motivations
and contributions of digital volunteers. This study applies a user-centered design approach to understanding how we
might make better tools to support digital volunteers. This user-centered design approach involves stated preference
elicitation methods through an online survey to understand what digital volunteers want in such tools. Through
choice-based conjoint analysis, we contribute to mixed-methods research to gain additional insight into motivations
and user preferences for a set of design features that might be incorporated into an online tool specifically for digital
volunteers. Initial results show preferences for measures of success that were not monetary, which aligned with
directly stated motivations for volunteering. Our findings corroborate with previous research in that feedback to
volunteers is very important, as well as being able to measure the impact of their work.

Abstract: During a natural disaster or emergency event, individual social media posts or hot spots may not necessarily correlate
to the most devastated areas. To better understand the correlation between social media and physical damage, we
compare Tweets, data about the physical environment, and socio-economic factors with insurance claim information
(as a proxy for physical damage) from 2017 Hurricane Irma in the state of Florida. We use machine learning
to identify relevant Tweets, sensitivity analyses to identify socio-economic factors, and statistical regression to
determine the predictive capability of insurance claims as a proxy for damage. We find that Tweets alone result in a
poorly fitted regression model of insurance claims, but the inclusion of physical features (e.g., power outages, wind
level) and socio-economic factors (e.g., population density, education, Internet access) improves the model?s fit.
Such models contribute to the knowledge base that may allow social media to predict damage in real-time.

Abstract: Mass Casualty Incidents (MCIs) may occur with no notice and require a rapid response to manage the casualties and arrange their transportation to hospitals. MCIs may result in different numbers of casualties and fatalities. Further, response time can play a crucial role in reducing fatalities and protecting lives. This paper reports on a preliminary optimisation-based approach, termed MCIER, which has been developed to co-ordinate the response of ambulances to multiple MCIs. In this approach, a realistic representation of the road network is modelled for the geographical area of interest. Also, a Neighbourhood Search Algorithm (NSA) has been developed in order to find the optimum solution to the problem under consideration. A hypothetical case study of a MCI in Newcastle-upon-Tyne has been considered to investigate the effect on response time of the time of day, and day of week, on which the incident occurs.

Abstract: Several research studies have shown the importance of social media data for humanitarian aid. Among others,
the issue of missing and lost people during disasters and emergencies is crucial for disaster managers. This work
analyzes Twitter data from a recent wildfire event to determine its usefulness for the mitigation of the missing and
found people issue. Data analysis performed using various filtering techniques, and trend analysis revealed that
Twitter contains important information potentially useful for emergency managers and volunteers to tackle this
issue. Many tweets were found containing full names, partial names, location information, and other vital clues
which could be useful for finding missing people.

Abstract: Crowd simulation is suitable to evaluate evacuation strategies but its validity strongly depends on the quality of input
data. The acquisition of adequate input data is particularly challenging when simulating the evacuation of public
buildings such as universities. As they are publicly accessible, the exact number of persons on site is unknown.
Yet, to investigate specific emergency situations by means of simulation, e.g. amok or fire, information is required
about distribution and amount of people within the building at a specific point of time. Due to data privacy, public
buildings do not implement access control. However, data artifacts are available in various information systems,
e.g., wifi data, room administration. Our hypothesis is, that the acquisition and fusion of such data artifacts is
sufficient to enable data-based ad-hoc simulation of evacuation scenarios as decision support for the operations
management. To this end, we introduce a procedure for the situation-dependent collection fusion of simulation
input data. Furthermore, a case study is provided to demonstrate the feasibility of the approach.

Abstract: Social media enables fast information exchange and status reporting during crises. Filtering is usually required to
identify the small fraction of social media stream data related to events. Since deep learning has recently shown to
be a reliable approach for filtering and analyzing Twitter messages, a Convolutional Neural Network is examined for
filtering crisis-related tweets in this work. The goal is to understand how to obtain accurate and robust filtering
models and how model accuracies tend to behave in case of new events. In contrast to other works, the application
to real data streams is also investigated. Motivated by the observation that machine learning model accuracies
highly depend on the used data, a new comprehensive and balanced compilation of existing data sets is proposed.
Experimental results with this data set provide valuable insights. Preliminary results from filtering a data stream
recorded during hurricane Florence in September 2018 confirm our results.

Abstract: We argue that current technology for crisis training does not explicitly cater well enough for managing training
objectives and skill building metrics throughout the lifespan of training. We suggest how successful crisis training
may be enabled by interoperating next-generation exercise management tools with mixed-reality simulations. We
propose an architecture consisting of (1) a front-end in which training objectives, essential skills, corresponding
events and metrics can be declared, (2) a back-end consisting of simulations that implement the events and metrics
and (3) a middleware which transfers information between the front-end and back-end to enable semi-automatic
composition of the simulations and performance analysis. The purpose of this architecture is to facilitate learning
through the principles of deliberate practice. We indicate where emerging technologies are necessary to achieve this.