H2020 beAWARE Project: Improving decision-making support and management services in extreme weather events
- Type Project
- Status Filled
- Execution 2017 -2019
- Assigned Budget 6.725.209,00 €
- Scope Europeo
- Autonomous community Comunitat Valenciana
- Main source of financing Horizon 2020
- Project website https://beaware-project.eu/
Progress beyond the state of the art can be summarized as follows:
- Obj1: A set of approximately 100 user requirements was formulated, corresponding to 3 relevant use cases.
- Obj2: a) An adaptive speech recognition component implemented with the ability to transcribe calls from existing call center solutions b) A broad coverage multilingual text analysis component to detect from text a wide range of incidents and impacted objects, locations and states associated with incidents
- Obj 3: beAWARE can collect, harmonize and store information from the following sources: meteorological, hydrological, risk maps, multimedia, tweets.
- Obj4: Visual analytics components extract high-level information by extracting low-level features and translate them into high-level concepts based on DL techniques. The extracted concepts lead to a real-time, comprehensive understanding of the situation from visual information (affected people, traffic conditions, etc.). In addition, new methods for combining multimodal information in early and late fusion schemes along with dimensionality reduction algorithms were proposed.
- Obj5: The backbone of beAWARE's reasoning mechanisms consists of the use of an ontology to map input, estimate the severity level of an incident, group relevant incidents, link heterogeneous information, and reveal hidden relationships. In addition, a validation mechanism was included to control system traffic and minimize the likelihood of malicious data impeding the system's effectiveness.
- Obj6: A multilingual report generator for generating multilingual reports derived from beAWARE's ontological representations. The system can issue two types of multilingual reports: situational and summary reports that cover incident occurrences in chronological order. Entities and concepts beyond those modeled by the ontology can be verbalized, thus facilitating the portability of the developed components to new emergency scenarios.
- Obj7: A PSAP that involves visualization and interaction techniques to enhance situational awareness, including situation assessment, GIS analysis, and mapping.
- Obj8: Three large-scale pilots successfully carried out, involving a total of more than 600 participants.
In terms of social engagement, in addition to the participatory stages of design, testing, and evaluation, which resulted in the participation of more than 600 people (FR and general public), the project reaches various communities through participation in conferences, workshops, exhibitions, and seminars. A Stakeholder Network was established to support direct liaison with relevant stakeholders, reaching 149 members. A detailed list of dissemination and social engagement activities can be found in D8.3(M36) available online at: https://beaware-project.eu/results/public-deliverables (Opens in a new window). Despite the project's end in December 2019, the consortium is moving in several directions with the aim of promoting the activities carried out during the project, further exploiting the project results, and maximizing future impact.
During the course of the project, we developed the beAWARE Platform. The platform is a combination of different components that offers a variety of tools and functionalities, some of which represent promising innovations for current SoA technologies.
The exploitable parts of the Platform are the following: -Frost Server: It is a standards-based sensor data store, used in beAWARE to store and retrieve heterogeneous time series data and sensor metadata in a uniform manner.
- Crisis Classification API: Identify and classify crisis events with early warning and real-time risk assessment capabilities - Analytics Module: Semantic data collection and integration.
- The module is based on an ontological framework and consists of three analysis components:
- Multimedia Analysis subcomponent to analyze visual and audio content and automatically understand and detect crisis events in it.
- Text analysis subcomponent to analyze and extract information from the text such as the type of crisis, the objectives involved, the geolocation of an event, etc.
- Social media analytics subcomponent to continuously collect and analyze tweets from Twitter and classify them as relevant or not.
- Drone Platform and Dashboard: Service for connecting drone service providers with customers to configure, launch, autopilot, and monitor drone missions.
- Crisis Management Knowledge Base: A comprehensive system for accessing, aggregating, and visualizing different sources of information to improve situational awareness.
- beAWARE PSAP Map and Dashboard: Involves visualization and interaction techniques to enhance situational awareness.
All developed components were created using a participatory design approach, in a spiral SCRUM development/evaluation process, from the first prototype to the final system. The prototypes were demonstrated in three pilot projects involving a total of more than 600 people. Ultimately, the platform fulfilled its original objective and met user expectations. Furthermore, all project results were communicated and disseminated according to the original plan, developed in the first phase of the project, and updated regularly.
beAWARE (Enhancing Decision Support and Management Services in Extreme Weather Climate Events), funded by the European Union's H2020 research and innovation programme (GA-700475), explored the combination of innovative technologies to create a holistic approach to implementing a crisis management framework that supports all phases of an emergency call sequence. The project's main objective was to build on existing platforms, theories, and methodologies for disaster management and add the necessary elements to make them work efficiently under a single objective.
The key deliverable is an integrated beAWARE platform encompassing high-end technologies and machine learning capabilities to extract insights from text messages, social media, and voice calls; a classification mechanism to process weather and other multimodal data and generate early warnings and alerts in real time; deep learning techniques to detect crisis events in visual content; drone routing and automated piloting to receive valuable information from aerial imagery; case-based reasoning and decision support algorithms for crisis management; and automatic multilingual report generation to transform all of the above into linguistic information for authorities, creating a usable system for action in real-life case scenarios. Its usability and interoperability have been stress-tested throughout the project's lifetime through three major field trials and will undergo a permanent stress-testing environment (demonstration) after the project's official completion.
In any disaster and crisis, incident time is the enemy, and obtaining accurate information on the scope, extent, and impact of the disaster is critical to creating and orchestrating an effective disaster response and recovery. The primary objective of beAWARE is to provide support throughout all phases of an emergency incident. More specifically, we propose an integrated solution to support forecasting, early warning, transmission and routing of emergency data, aggregated multimodal data analysis, and coordination management between first responders and authorities. Our intention is to build on platforms, theories, and methodologies already used for disaster forecasting and management and incorporate the necessary elements to make them work efficiently and toward the same goal.
The overall context of beAWARE focuses on situational awareness and command and control (C2). The first phase focuses on forecasting the extreme condition and making relevant preparations. Once a disaster occurs, an initial assessment must be conducted as soon as possible to determine the scope, geographic distribution, and scale of the incident. Situational awareness involves being able to accurately determine what has happened, what is happening now, and what... This is followed by planning and coordinating the most effective response possible with the available resources. This observation phase will give way to an orientation phase, which suggests a cognitive orientation, both individual and collective, toward the perceived and communicated data. Once data orientation (or lack thereof) occurs, a decision is made, the ultimate result of which is action. The crisis management center always strives to understand reality so that it can feel it can make the best possible decision under the circumstances.
The intensity, frequency, and economic costs of extreme weather events (Opens in new window) are increasing with global climate change. As a result, floods, droughts, heat waves, and wildfires are expected to become more common and severe in the future. The EU-funded beAWARE project (Opens in new window) is working to address this issue.
The project developed a comprehensive communications and analytics platform to support decision-makers, first responders, and citizens. It includes multilingual verbal and written communication analysis, multilingual reporting, and rich multimedia emergency communication. Smarter, faster feedback and decision-making at the local level. According to the project's deputy coordinator, Dr. Anastasios Karakostas: "In all disasters and crises, time is the enemy. Obtaining accurate information on the scope, extent, and impact of the disaster is critical to creating and orchestrating an effective disaster response and recovery effort.
The main objective of beAWARE is to provide support in all phases of an emergency incident. The beAWARE knowledge base is the semantic foundation that provides the classification scheme and inference rules, and also analyzes the information and distributes the results. Information is obtained from local citizens, first responders, social media, local weather forecasts, sensors such as static in-situ cameras to monitor water levels, and even cameras on drones. Citizens and first responders communicate with the beAWARE platform through the beAWARE mobile app. The social media monitoring module searches and validates related social media content and then analyzes it to perform spatiotemporal clustering of relevant posts.
Intelligent algorithms analyze both visual and auditory information. As Dr. Karakostas explains, "The visual analytics module can define the type of crisis, detect relevant objects (e.g., cars in a flooded area), and estimate traffic." An automated speech recognition module transcribes audio messages received through the beAWARE mobile app or phone calls in four languages (English, Greek, Italian, and Spanish). "The Public Safety Answering Point (PSAP) is beAWARE's central command and control system," notes Dr. Karakostas. "It receives information from citizens, first responders, and social media, processes it through machine reasoning engines and analytics services, and generates automatic incident reports or enriches field reports with information based on multimedia analytics, social media, and sensors." Two field pilots validated the beAWARE platform: a heatwave simulation in Greece and a flood simulation in the Eastern Alps region of Italy.
During the flood test, the mayor, as the decision-maker, was supported by various offices in the municipality and the region to staff the control room and act as first responders. Increasing awareness of the platform and its usefulness for the public good. In addition to publications and participation in numerous conferences, beAWARE also organized the first (2018) and second (2019) International Workshops on Smart Crisis Management Technologies for Climate-Related Events (ICMT) at the International Conference on Information Systems for Crisis Response and Management (ISCRAM). The widespread implementation of the beAWARE platform at the local level should empower communities to take ownership of crisis management. Community leaders, first responders, and citizens will be able to cope with increasingly severe weather events more quickly and effectively, saving lives and livelihoods.
- ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS ANAPTYXIS
- UNIVERSIDAD POMPEU FABRA
- ELLINIKI OMADA DIASOSIS SOMATEIO
- FREDERIKSBORG BRAND OG REDNING
- MOTOROLA SOLUTIONS ISRAEL LTD
- FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
- AYUNTAMIENTO DE VALENCIA
- ILMATIETEEN LAITOS
- IBM ISRAEL - SCIENCE AND TECHNOLOGY LTD
- AUTORITA' DI BACINO DISTRETTUALE DELLE ALPI ORIENTALI
- Project website (CORDIS)
- CORDIS project factsheet (pdf)
- beAWARE System
- ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS ANAPTYXIS website
- Pompeu Fabra University website
- ELLINIKI OMADA DIASOSIS SOMATEIO website
- MOTOROLA SOLUTIONS ISRAEL LTD website
- FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV website
- Valencia City Council website
- ILMATIETEEN LAITOS website
- IBM ISRAEL website - SCIENCE AND TECHNOLOGY LTD
Related projects
