Angelo Amorosi

1. Masonry buildings are particularly vulnerable to ground-induced deformations due to their low tensile strength and consequent susceptibility to cracking (Burland and Wroth 1975; Boscardin and Co...

This paper investigates the influence of tunnel-building configurations on tunnelling-induced damage on masonry buildings subjected to shallow tunnelling in soft ground through three-dimensional finite element analyses. Tunnel excavation is simulated using a step-by-step procedure to reproduce the progressive advancement of the tunnel and to appraise the associated transient soil-structure interaction effects. Four representative configurations are examined: centred and orthogonal (CO), centred and skewed by 60° (CS), eccentric and orthogonal with a 10 m offset (EO), and eccentric and skewed by 60° with a 10 m offset (ES).The constitutive model adopted for the soil is the Hardening Soil with small-strain stiffness (HSsmall), while for masonry the Enriched Jointed Masonry Model (JMM-E) is used. This model captures the anisotropic and irreversible behaviour of masonry within a multilaminate framework and perfect plasticity. The analyses focus on the initiation, development, localisation, severity, and type of damage during tunnel advancement, with emphasis on the role of tunnel eccentricity and skew angle. In particular, the use of JMM-E allows the strain-based assessment to be complemented by a mechanism-based interpretation, distinguishing between tensile opening and shear sliding on either bed or head joints. To this end, selected damage proxies are used. A combination of strain-fields and plastic points allows for a more insightful interpretation of the type of damage affecting the structure.The results show that the tunnel-building relative position significantly affects both the magnitude and the spatial distribution of damage. The centred and orthogonal configuration leads to the most severe scenario, according to the adopted damage indicators and for the sample set of parameters adopted. The study ultimately aims to highlight the importance of explicitly modelling progressive excavation and of resorting to comprehensive damage descriptions beyond conventional metrics solely based on tensile strains. The proposed approach was developed within a custom, semi-automated model generation procedure and output processing implemented in Python®, and provides a useful basis for more reliable assessment of tunnelling-induced damage on masonry buildings in urban contexts.

The increasing complexity of climate-driven multi-risk scenarios requires a paradigm shift from isolated, sector-specific hazard management to holistic, integrated frameworks. This paper presents a scalable methodology developed within the RETURN project to model transiently triggered ground instabilities (GIs), specifically landslides and soil liquefaction, through integrated, multi-stage computational workflows, termed "toolchains”. These toolchains logically sequence static susceptibility assessments, cyclic preparatory processes and rapid triggers to dynamically generate high-fidelity, single and multi-hazard impact scenarios over regional areas. The proposed framework overcomes the impracticality of deploying highly data-demanding, physically based models over broad regional areas, while preserving computational efficiency. The toolchains are applied within a Virtual Test Bed (VTB), a highly parametrised digital ecosystem of a generic high-risk territory to test the algorithmic robustness and the capacity of the tools to quantify impacts on synthetic exposure zones under severe seismic actions. Furthermore, we show the methodology integrates and combines the cascading effects of spatially and temporally independent processes triggered by the same factor, to be possibly further coupled to study multi-hazard scenarios. This adaptable framework aims at bridging the gap between advanced scientific modelling and practical decision-making, providing an operational toolset capable of supporting both long-term, non-structural mitigation strategies and rapid, near-real-time hazard assessments under ongoing transient forcings.