PSI - Issue 84

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

148

sectorial employees as a proxy for economic activity. Building height and construction periods are mapped using Global Human Settlement (GHS) datasets, and building typologies are assigned according to height, epoch, and use category. Seismic code levels are derived from construction period, and replacement costs are assigned per building category, ranging between 1,200-1,800 €/m² of built-up area. At the national scale, the commercial exposure represents a total built-up area of 258 million m² and 480 thousand buildings, with a total replacement cost of 625 billion EUR, of which 152 billion EUR are structural, 200 billion EUR non-structural, and 273 billion EUR contents. The industrial building exposure model is based on the combination of ISTAT-defined industrial and energy production areas (including quarries and mines) with building-specific data from the GHS-OBAT database. All non residential structures located within those industrial areas are classified as industrial, whose footprint geometry, height, and construction epoch are retrieved. Building typologies are assigned based on ESRM20 assumptions, with approximate shares of 70% precast concrete frames, 20% steel frames, 5% cast-in-place concrete infilled frames, and 5% unreinforced masonry, further refined spatially using building height and construction epoch. Seismic code levels are mapped according to construction epoch. Replacement costs are assigned per building type, with an average unit cost of approximately 840 €/m² of built-up area. At country level, the industrial exposure accounts for a total built-up area of 901 million m² and 297 thousand buildings, with a total replacement cost of 1.8 trillion EUR, of which 280 billion EUR are structural, 477 billion EUR non-structural, and 1.1 trillion EUR contents. As a proof of concept for integrating urban exposure data into the platform, building-by-building exposure models have been developed for Villa D’Agri (Potenza, Basilicata) and Rieti (Lazio). The raw datasets, provided by Spoke 5 colleagues, include building footprints and attributes such as geographic coordinates, construction year, height, number of floors, footprint area, occupancy type, and soil type. Villa D’Agri is shown as an example in Fig. 5. In parallel, the infrastructure exposure model for the country is based on OpenStreetMap (OSM) data (OpenStreetMap contributors, 2025) via Geofabrik (2018). Roads are categorised using standardised OSM tags that indicate function and relative importance, with additional information identifying bridges. The dataset includes highways, motorways, and primary to tertiary roads, processed in GIS format and converted into OQ format as a graph of nodes (vertices) connected by edges (links). Each edge is weighted by road length, and the network is modelled as undirected for the simplified impact assessment applied here. Fig. 6 illustrates the total road length per province, while road density and the number of bridges are also available at the provincial level.

Fig. 5. Footprints of the investigated buildings in Villa D’Agri (Potenza, Basilicata), shaded according to their year of construction.