The current research work has been developed within the RES-Q Living Lab located in Larissa, Greece

The paper entitled “RES-Q: Self-evolving Post-disaster Response Plans Utilizing Semantics” has been accepted to be presented at the 3rd International Conference on Natural Hazards & Infrastructure (ICONHIC 2022) held in Athens, from July 5th to July 7th, 2022. In particular, Dr. Iatrellis, Ms Bania and Prof. Samaras of University of Thessaly will introduce the RES-Q approach that proposes an information technology solution concerning the real-time recommendation and orchestration of post-disaster response plans. The RES-Q approach comprises an expert system and a workflow execution engine based on an ontological infrastructure for modeling the response actions for each type of disaster. The ontological model is designed using a multi-layer approach encapsulating the required knowledge streams and a semantic rule repository. During the execution of a post-disaster plan, the system reasons over the rules and composes the next steps of the corresponding response processes. The rule repository is able to infer new knowledge as each plan progresses, which can update the RES-Q ontology accordingly.

The contribution concerning the state-of-the-art in the disaster risk management domain can be presented in a three-fold structure:

1. Dynamic composition of post-disaster response plans: The innovation behind RES-Q approach is the fact that the predefined plan for each disaster is utilized only as a basis of the response scheme. RES-Q approach supports continuous reasoning over the semantically enhanced knowledge base and the established meta-models so as to dynamically compose each step of the response during execution. A key contribution towards the dynamic composition of the response plans is their proper computer modeling and encoding with harmonized and formalized terms across all stakeholders.

2. RES-Q ontology: The RES-Q ontology is able to nurture collaboration between the domains that are directly or indirectly linked with the disaster risk management area, bringing together stakeholders from diverse domains such as urban authorities, healthcare or seismology. To this regard, we introduce a novel ontological model for disaster risk management, which is designed based on a multi-layer approach. The RES-Q ontology has been developed as a core component of the RES-Q platform for integrating and semantically enriching disaster-related domain-specific concepts and metadata, which can be further utilized for the implementation of a repository of post-disaster response guidelines. The RES-Q ontology, however, can also be utilized as a standalone domain ontology.

3. Establishment of post-disaster meta-models: in any execution cycle, semantic meta-models are established by first executing a semantically annotated set of response guidelines for each disaster and, second, workflow execution to control the corresponding processes. RES-Q incorporates a rule repository created by utilizing Semantic Web Rule language in order to extend the expressivity of the RES-Q ontological model. The rules are used in the inference process to reveal new facts from given ones and update the ontology accordingly, something not achievable using a relational database. The workflow environment covers the complete lifecycle of response plans (both design and execution mode) by offering a toolkit incorporating the BPMN process modeling notation.

The current research work has been developed with a Living Lab approach in coherence with the INVEST4EXCELLENCE European project. Under the RES-Q pilot, the areas of the Living Lab - as micro versions of larger territories - can be seen as appropriate sites for validating and spreading new concepts, processes and technologies, serving as showcases at macro level. Thus, scalability and replicability to other regions is strongly embedded in the RES-Q project.