SAFENET: Seismic Risk Assessment for Transportation Networks

SAFENET — Seismic risk Assessment For transportation NETworks is an FCT research project that uses machine learning and big data to assess how earthquakes affect transportation networks and the buildings around them. It is led by the Faculty of Engineering of the University of Porto (FEUP), with Fundação Ensino e Cultura Fernando Pessoa (UFP) and Instituto Superior Técnico as partners.

Why it matters

Transportation networks are fundamental to a country’s economy, and they play a pivotal role after a destructive earthquake — during the emergency and throughout recovery and reconstruction. Understanding disaster risk and building resilient infrastructure is a priority of the Sendai Framework for Disaster Risk Reduction and of the UN Sustainable Development Goals.

Network risk assessment is harder than assessing single structures: it has to account for dependencies across the whole system, for how population and traffic change during the day and the year, and for hazards beyond ground shaking, such as liquefaction and landslides.

What the project does

SAFENET is building an open-source, dynamic platform that brings these pieces together:

• Infrastructure and building stock — characterised with artificial intelligence and open data from OpenStreetMap;
• Population dynamics — modelled with big data from mobile phone towers, as a proxy for population density at different times;
• Risk and traffic — built on openly available seismic risk and traffic simulation tools, to avoid duplicated effort.

The platform is tested on two case studies: the district of Lisbon, at the urban scale, and the south of Portugal, at the regional scale. It is co-designed with national stakeholders, with international support from the Humanitarian OpenStreetMap Team and the SimCenter artificial intelligence group (Berkeley and Stanford).

At UFP, the team works on computer vision for infrastructure: classifying bridges and viaducts from images, so that they can be included in the risk model.

Team at UFP

• Prof. Feliz Gouveia
• Prof. Rui Silva Moreira
• Tomás Oliveira, MSc student
• Beatriz Neves, MSc student (past member)

Outputs

• Tomás Lopes Oliveira, Rui Silva Moreira, Feliz Gouveia, Catarina Costa, Vítor Silva. SAFENET: Image-Based Infrastructure Classification of Bridges and Viaducts for Seismic Risk Assessment. IEEE SmartNets 2026, Rome, Italy, 7–9 July 2026. doi.org/10.1109/SmartNets69662.2026.11604925
• Code: github.com/TomasLopes27/SAFENET.

Funding

Project 2022.07628.PTDC, funded by national funds through FCT — Fundação para a Ciência e a Tecnologia, I.P. It runs from March 2023 to November 2026. DOI: 10.54499/2022.07628.PTDC.