Aïdin Sumic, Emna Amdouni, Thierry Vidal, & Hedi Karray. (2022). Towards Flexibility Sharing in Multi-agent Dynamic Planning: The Case of the Health Crisis. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 274–284). Tarbes, France.
Abstract: Planning problems in a crisis context are a highly uncertain environment where health facilities must cooperate in providing health services to their patients. We focus on the health crisis in France due to the COVID19 pandemic. In fact, the lack of appropriate scheduling tools, resources, and communication leads hospitals to be submerged by infected patients and forced to transfer them to other hospitals. In this work we aim to provide a global solution to such planning problems to improve the current French health system. We introduce a cooperative approach called OPPIC (Operational Planning Platform for Inter-healthcare Coordination). OPPIC is based on a decentralized system, where health facilities plan is dynamic, flexible, robust to uncertainty, and respond to goals and optimization criteria. This paper proposed a first planning model to OPPIC and provided a first way of negotiation between health facilities based on their plan’s local and global flexibility.
|
Antonio De Nicola, Maria Luisa Villani, Francesco Costantino, Andrea Falegnami, & Riccardo Patriarca. (2021). Knowledge Fusion for Distributed Situational Awareness driven by the WAx Conceptual Framework. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 79–85). Blacksburg, VA (USA): Virginia Tech.
Abstract: Large crisis scenarios involve several actors, acting at the blunt-end of the process, such as rescue team directors, and at the sharp-end, such as firefighters. All of them have different perspectives on the crisis situation, which could be either coherent, alternative or complementary. This heterogeneity of perceptions hinders situational awareness, which is defined as the achievement of an overall picture on the above-mentioned crisis situation. We define knowledge fusion as the process of integrating multiple knowledge entities to produce actionable knowledge, which is consistent, accurate, and useful for the purpose of the analysis. Hence, we present a conceptual modelling approach to gather and integrate knowledge related to large crisis scenarios from locally-distributed sources that can make it actionable. The approach builds on the WAx framework for cyber-socio-technical systems and aims at classifying and coping with the different knowledge entities generated by the involved operators. The conceptual outcomes of the approach are then discussed in terms of open research challenges for knowledge fusion in crisis scenarios.
|
Audun Stolpe, & Jo Hannay. (2021). On the Adaptive Delegation and Sequencing of Actions. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 28–39). Blacksburg, VA (USA): Virginia Tech.
Abstract: Information systems support to crisis response and management relies crucially on presenting actionable information in a manner that supports cognitive processes, and does not overwhelm them. We outline how AI Planning can be used viably to support the \emph{delegation and sequencing} of tasks. The idea is to use standard operating procedures as initial specifications of plans in terms of actors, actions and delegation rules. When expressed in the AI planning language \textit{Answer set Programming} (ASP), machine reasoning can be used in a \textit{pre-incident review} to display relevant delegation and sequencing inherent in a plan. % together with measures of goal achievement. The purpose of this is to uncover weaknesses in the initial plan and to optimize sequencing and delegation to increase the likelihood of achieving goals. Further, adaptive planning can be supported in \textit{during-incident reviews} by updating the current status, upon which ASP will then compute new alternatives. % and corresponding goal achievement measures. At this point, initial goals may no longer be viable and the explicit suggestion of prior sub-optimal goals now worth pursuing can be a game-changer under stress. The conceptual basis we lay out in terms of delegation and sequencing can be readily extended with further planning factors, such as resource requirements, role transfer and goal achievement.
|
Cendrella Chahine, Thierry Vidal, Mohamad El Falou, & François Pérès. (2022). Multi-Agent Dynamic Planning Architectures for Crisis Rescue Plans. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 243–255). Tarbes, France.
Abstract: We are interested in rescue management in crises such as in terrorist attacks. Today, there are emergency plans that take into account all the stakeholders involved in a crisis depending on the event type, magnitude and place. Unfortunately, they do not anticipate the evolution of the crisis situation such as traffic and hospital overcrowding. In addition, decisions are taken after the information has been passed from the operational level to higher levels. This work focuses on the operational level of the emergency plan. What will happen if the actors at this level, can make certain decisions without escalating the information to higher levels? To answer this question, a multi-agent dynamic planning approach is proposed and it will be tested in two different architectures in order to see how much autonomy can be given to an agent and how they coordinate to save the victims.
|
Christian Iasio, Ingrid Canovas, Elie Chevillot-Miot, & Tendry Randramialala. (2022). A New Approach to Structured Processing of Feedback for Discovering and Investigating Interconnections, Cascading Events and Disaster Chains. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 285–298). Tarbes, France.
Abstract: Post-disaster information processing is relevant for the continuous improvement of operations and the reductionof risks. The current methodologies for post-disaster review suffer from several limitations, which reduce their use as a way of translating narrative in data for qualitative and quantitative analysis. Learning or effective knowledge sharing need a common formalism and method. Ontologies are the reference tool for structuring information in a “coded” data structure. Using the investigation of disaster management during the 2017 hurricane season in the French West Indies within the scope of the ANR “APRIL” project, this contribution introduces a methodology and a tool for providing a graphical representation of experiences for post-disaster review and lessons learning, based on a novel approach to case-based ontology development.
|
Dashley Rouwendal van Schijndel, Audun Stolpe, & Jo Erskine Hannay. (2021). Toward an AI-based external scenario event controller for crisis response simulations. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 106–117). Blacksburg, VA (USA): Virginia Tech.
Abstract: There is a need for tool support for structured planning, execution and analysis of simulation-based training for crisisresponse and management. As a central component of an architecture for such tool support, we outline the design ofan AI-based scenario event controller. The event controller is a component that uses machine reasoning to computethe next state in a scenario, given the actions performed in the corresponding simulation (execution of the scenario).Scenarios are specified in Answer Set Programming, which is a logic programming language we use for automatedplanning of training scenarios. A plan encoding in ASP adds expressivity in scenario specification and enablesmachine reasoning. For exercise managers this gives AI-based tool support for before-action and during-actionreviews to optimize learning. In line with Modelling and Simulation as as Service, our approach externalizes eventcontrol from any particular simulation platform. The scenario, and its unfolding in terms of events, is externalizedas a service. This increases interoperability and enables scenarios to be designed and modified readily and rapidlyto adapt to new training requirements.
|
Kenneth Johnson, Javier Cámara, Roopak Sinha, Samaneh Madanian, & Dave Parry. (2021). Towards Self-Adaptive Disaster Management Systems. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 49–61). Blacksburg, VA (USA): Virginia Tech.
Abstract: Disasters often occur without warning and despite extensive preparation, disaster managers must take action to respond to changes critical resource allocations to support existing health-care facilities and emergency triages. A key challenge is to devise sound and verifiable resourcing plans within an evolving disaster scenario. Our main contribution is the development of a conceptual self-adaptive system featuring a monitor-analyse-plan-execute (MAPE) feedback loop to continually adapt resourcing within the disaster-affected region in response to changing usage and requirements. We illustrate the system's use on a case study based on Auckland city (New Zealand). Uncertainty arising from partial knowledge of infrastructure conditions and outcomes of human participant's actions are modelled and automatically analysed using formal verification techniques. The analysis inform plans for routing resources to where they are needed in the region. Our approach is shown to readily support multiple model and verification techniques applicable to a range of disaster scenarios.
|
Koki Asami, Shono Fujita, Kei Hiroi, & Michinori Hatayama. (2022). Data Augmentation with Synthesized Damaged Roof Images Generated by GAN. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 256–265). Tarbes, France.
Abstract: The lack of availability of large and diverse labeled datasets is one of the most critical issues in the use of machine learning in disaster prevention. Natural disasters are rare occurrences, which makes it difficult to collect sufficient disaster data for training machine learning models. The imbalance between disaster and non-disaster data affects the performance of machine learning algorithms. This study proposes a generative adversarial network (GAN)- based data augmentation, which generates realistic synthesized disaster data to expand the disaster dataset. The effect of the proposed augmentation was validated in the roof damage rate classification task, which improved the recall score by 11.4% on average for classes with small raw data and a high ratio of conventional augmentations such as rotation of image, and the overall recall score improved by 3.9%.
|
Nada Matta, Paul Henri Richard, Alain Hugerot, & Theo Lebert. (2022). Experience Feedback Capitalization of Covid-19 Management in Troyes city. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 311–319). Tarbes, France.
Abstract: All countries have to face the COVID’19 pandemic and its heavy consequences. This sanitary crisis differs from all others in terms of the quick spread of contaminations, the high number of deaths (more than 5,5 Million globally and 123,893 in France) and the accrued number of patients hospitalized and induced in intensive care units. All sanitary procedures have proven to be inadequate. Several actors at different levels, whether international, European, national and local, as well as at the level of public and private organizations have been involved in the management of this type of crisis. These actors deal with different aspects of it, i.e., health, people protection, and economic and social situations. Existing procedures revealed a big lack in the relationships between different local and departmental actors. We did a number of interviews with strategic actors addressing the COVID’19 crisis in the City of Troyes. The objective of these interviews is to identify lessons learned from their experience feedback about relational problems and modifications needed. We present in this paper the first results of this study.
|
Nada Matta, Thomas Godard, Guillaume Delatour, Ludovic Blay, Franck Pouzet, & Audrey Senator. (2021). Analyzing Social Media in Crisis Management Using Expertise Feedback Modelling. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 17–27). Blacksburg, VA (USA): Virginia Tech.
Abstract: Currently social media are largely used in interactions, especially in crisis situations. We note a big volume of interactions around events. Observing these interactions give information even to alert the existence of an incident, event, or to understand the expansion of a problem. Crisis management actors observe social media to be aware about this type of information in order to consider them in their decisions. Specific organizations are founded in order to observe social media interactions and send their analysis to rescue and crisis management actors. In our work, an experience feedback of this type of organizations (VISOV, a crisis social media analysis association) is capitalized in order to emphasize from one side, main dimensions of this analysis and from another side, to simulate some aspects using TextMining that help to explore big volume of data.
|
Nilani Algiriyage, Raj Prasanna, Kristin Stock, Emma Hudson-Doyle, David Johnston, Minura Punchihewa, et al. (2021). Towards Real-time Traffic Flow Estimation using YOLO and SORT from Surveillance Video Footage. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 40–48). Blacksburg, VA (USA): Virginia Tech.
Abstract: Traffic emergencies and resulting delays cause a significant impact on the economy and society. Traffic flow estimation is one of the early steps in urban planning and managing traffic infrastructure. Traditionally, traffic flow rates were commonly measured using underground inductive loops, pneumatic road tubes, and temporary manual counts. However, these approaches can not be used in large areas due to high costs, road surface degradation and implementation difficulties. Recent advancement of computer vision techniques in combination with freely available closed-circuit television (CCTV) datasets has provided opportunities for vehicle detection and classification. This study addresses the problem of estimating traffic flow using low-quality video data from a surveillance camera. Therefore, we have trained the novel YOLOv4 algorithm for five object classes (car, truck, van, bike, and bus). Also, we introduce an algorithm to count the vehicles using the SORT tracker based on movement direction such as ``northbound'' and ``southbound'' to obtain the traffic flow rates. The experimental results, for a CCTV footage in Christchurch, New Zealand shows the effectiveness of the proposed approach. In future research, we expect to train on large and more diverse datasets that cover various weather and lighting conditions.
|
Oussema Ben Amara, Daouda Kamissoko, Frédérick Benaben, & Ygal Fijalkow. (2021). Hardware architecture for the evaluation of BCP robustness indicators through massive data collection and interpretation. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 71–78). Blacksburg, VA (USA): Virginia Tech.
Abstract: Recently, the concept of robustness measurement has become clearly important especially with the rise of risky events such as natural disasters and mortal pandemics. In this context, this paper proposes an overview of a hardware architecture for massive data collection in the aim of evaluating robustness indicators. This paper essentially addresses the theoretical and general problems that the scientific research is seeking to address in this area, offers a literature review of what already exists and, based on preliminary diagnosis of what the literature has, presents a new approach and some of the targeted findings with a focus on the leading aspects, having a primary objective of explaining the multiple aspects of this research work.
|
Rocco Sergio Palermo, & Antonio De Nicola. (2022). A Simulation Framework for Epidemic Spreading in Semantic Social Networks. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 266–273). Tarbes, France.
Abstract: Epidemic spreading simulation in social networks denotes a set of techniques that allow to assess the temporal evolution and the consequences of a pandemic. They were largely used by governments and International health organizations during the COVID-19 world crisis to decide the appropriate countermeasures to limit the diffusion of the disease. Among them, the existing simulation techniques based on a network model aimed at studying the infectious disease dynamics have a prominent role and are widely adopted. However, even if they leverage the topological structure of a social network, they disregard the intrinsic and individual features of its members. A semantic social network is defined as a structure consisting of interlinking layers, which include a social network layer, to represent people and their relationships and a concept network layer, to represent concepts, their ontological relationships and implicit similarities. Here, we propose a novel epidemic simulation framework that allows to describe a community of people as a semantic social network, to adopt the most commonly used compartmental models for describing epidemic spreading, such as Susceptible-Infected-Susceptible (SIS) or Susceptible-Infected-Removed (SIR), and to enable semantic reasoning to increase the accuracy of the simulation. Finally, we show how to use the framework to simulate the impact of a pandemic in a community where the job of each member is known in advance.
|
Rouba Iskandar, Julie Dugdale, Elise Beck, & Cécile Cornou. (2021). PEERS: An integrated agent-based framework for simulating pedestrians' earthquake evacuation. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 86–96). Blacksburg, VA (USA): Virginia Tech.
Abstract: Traditional seismic risk assessment approaches focus on assessing the damages to the urban fabric and the resultant socio-economic consequences, without adequately incorporating the social component of risk. However, the human behavior is essential for anticipating the impacts of an earthquake, and should be included in quantitative risk assessment studies. This paper proposes an interdisciplinary agent-based modeling framework for simulating pedestrians' evacuation in an urban environment during and in the immediate aftermath of an earthquake. The model is applied to Beirut, Lebanon and integrates geo-spatial, socio-demographic, and quantitative behavioral data corresponding to the study area. Several scenarios are proposed to be explored using this model in order to identify the influence of relevant model parameters. These experiments could contribute to the development of improved of emergency management plans and prevention strategies.
|
Simon Mille, Gerard Casamayor, Jens Grivolla, Alexander Shvets, & Leo Wanner. (2022). Automatic Multilingual Incident Report Generation for Crisis Management. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 299–309). Tarbes, France.
Abstract: Successful and effucient crisis management depends on the availability of all accessible relevant information on the incidents during a crisis. The sources of this information are very often multiple and manifold – in particular in the case of environmental crises such as wild fires, floods, drought, etc. For the staff of the control centres it can be a challenge to follow up on all of them. In this paper, we present work in progress on an automatic multilingual incident report generator that produces summaries of all environmental incidents communicated by citizens or authorities in a given time range for a given region in terms of a text message, an audio, a video or an image and analyzed by dedicated modules into uniform knowledge representation structures.
|
Tina Mioch, Reinier Sterkenburg, Tatjana Beuker, & Mark A. Neerincx. (2021). Actionable Situation Awareness: Supporting Team Decisions in Hazardous Situations. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 62–70). Blacksburg, VA (USA): Virginia Tech.
Abstract: Situation Awareness (SA) has been recognized and studied as an important requirement for an effective task performance of first responders. The integration of increasingly advanced sensor, network and artificial intelligence technology into the work processes affects the building, maintenance and sharing of SA. Connecting SA to decision support models provides new possibilities for the development of actionable SA (aSA), entailing information that guides the momentary decision-making processes of the concerning actors. In the European ASSISTANCE project, we are developing an aSA module that displays information about gas distributions, its current and predicted future states (e.g., entailing risks of breathing-in of toxic gases), with references to effective decision-making patterns for this situation. The aSA model is continuously updated based on sensor data. This paper gives an overview of this aSA module for chemical hazard prediction and corresponding display, and presents initial team design patterns that will be integrated into this display to support its actionability.
|
Xiaojing Guo, Xinzhi Wang, Luyao Kou, & Hui Zhang. (2021). A Question Answering System Applied to Disasters. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 2–16). Blacksburg, VA (USA): Virginia Tech.
Abstract: In emergency management, identifying disaster information accurately and promptly out of numerous documents like news articles, announcements, and reports is important for decision makers to accomplish their mission efficiently. This paper studies the application of the question answering system which can automatically locate answers in the documents by natural language processing to improve the efficiency and accuracy of disaster knowledge extraction. Firstly, an improved question answering model was constructed based on the advantages of the existing neural network models. Secondly, the English question answering dataset pertinent to disasters and the Chinese question answering dataset were constructed. Finally, the improved neural network model was trained on the datasets and tested by calculating the F1 and EM scores which indicated that a higher question answering accuracy was achieved. The improved system has a deeper understanding of the semantic information and can be used to construct the disaster knowledge graph.
|
Yasas Senarath, Jennifer Chan, Hemant Purohit, & Ozlem Uzuner. (2021). Evaluating the Relevance of UMLS Knowledge Base for Public Health Informatics during Disasters. In Anouck Adrot, Rob Grace, Kathleen Moore, & Christopher W. Zobel (Eds.), ISCRAM 2021 Conference Proceedings – 18th International Conference on Information Systems for Crisis Response and Management (pp. 97–105). Blacksburg, VA (USA): Virginia Tech.
Abstract: During disasters public health organizations increasingly face challenges in acquiring and transforming real-time data into knowledge about the dynamic public health needs. Resources on the internet can provide valuable information for extracting knowledge that can help improve decisions which will ultimately result in targeted and efficient health services. Digital content such as online articles, blogs, and social media are some of such information sources that could be leveraged to improve the health care systems during disasters. To efficiently and accurately identify relevant disaster health information, extraction tools require a common vocabulary that is aligned to the health domain so that the knowledge from these unstructured digital sources can be accurately structured and organized. In this paper, we study the degree to which the Unified Medical Language System (UMLS) contains relevant disaster, public health, and medical concepts for which public health information in disaster domain could be extracted from digital sources.
|