PSI - Issue 84

Martina Caruso et al. / Procedia Structural Integrity 84 (2026) 143–150

144

1. Introduction PAPERS (Piattaforma Analisi Pericolosità E Rischio Sismico) is a one-year project funded under the cascade call of the High-Performance Computing, Big Data and Quantum Computing Research Centre, which is one of the five strategic national centres established under Italy’s National Recovery and Resilience Plan (PNRR). PAPERS has been developed within the Spoke 5 thematic research node dedicated to Environment and Natural Disasters, with a focus on advancing digital tools and methods for environmental monitoring, climate modelling and disaster risk reduction. The primary objective of the project has been to develop an open-source interactive map platform for the visualisation, exploration, and analysis of earthquake scenarios at local and national levels in Italy. Developed by the Global Earthquake Model (GEM) Foundation, it leverages the OpenQuake (OQ) Engine (Pagani et al., 2014; Silva et al., 2014; Rao et al., 2025) to support both pre-computed and user-defined earthquake scenario calculations. The platform integrates pre-loaded earthquake ruptures and ground motion models, building and infrastructure exposure models, and fragility and consequence functions. High-resolution building-by-building exposure models have also been developed for selected areas, demonstrating the value of integrating detailed local datasets for urban scale scenario assessment. A catalogue of earthquake scenarios, including historical and hypothetical events, has been assembled, with a subset pre-computed to estimate damaged buildings, economic losses, and potential fatalities at multiple administrative scales. The platform accommodates a broad user base, supporting both basic and advanced workflows. Basic users can visualise and inspect input models and pre-calculated scenarios, whereas more advanced users can extend the analysis by including novel input models and running new earthquake scenarios. Compatibility with external tools has also been incorporated to support model generation and post-processing workflows. This paper presents the methodological framework, datasets and models developed throughout the project, while illustrating the structure and functionalities of the developed platform. It first introduces the interactive map platform, outlining its implementation and use cases. It then describes the development of national and urban exposure models, the earthquake catalogue, the ground motion and site modelling approaches. 2. Interactive Map Platform The PAPERS platform leverages several open-source components along with custom code tailored to the specific use cases and workflows of the project, with the aim of guaranteeing interoperability, scalability, and user accessibility. Scenario analyses at multiple geographic scales are supported by the OQ Scenario Hazard and Damage Calculators. 2.1. Platform implementation A high-level overview of the software architecture adopted for the platform is presented in Fig. 1. On the left-hand side of the diagram, we show two different types of clients. Human users interact with the platform via a web browser over a secure HTTPS connection for an interactive user interface. Additionally, custom client applications can connect to the platform using a REST API. This is intended for expert users who wish to develop new code or adapt existing applications to interact with the platform, enabling, for example, third-party simulation software to upload input data. The central column of the diagram in Fig. 1, features the NGINX web server which functions as a reverse-proxy gateway to the rest of the platform. NGINX handles HTTPS encryption and provides a convenient point of configuration for security and network aspects. Custom Django applications, in the centre of the diagram, implement the business logic required to integrate the platform’s components into a cohesive system. These applications manage maps, layers, projects, users, calculations and support the creation of new maps and layers based on calculations results. Lastly, the OQ Engine is used to carry out earthquake scenario analyses. On the right-hand side, the Docker system manages a number of containers, each delivering a service required by the custom code. Using Docker “compose” feature, an internal virtual network of collaborating services is created and scaled as needed (e.g., increasing QGIS Server instances to support more concurrent clients). PostgreSQL, extended with PostGIS, provides a relational database for the Django applications, enabling advanced geospatial queries. QGIS is used to build projects, maps and layers from geospatial data, including outputs of OQ calculations. Lastly, the MapProxy service works in conjunction with QGIS and Redis (an in-memory cache service) that generates PNG image tiles that are provided to interactive users to visualise maps via a web browser.