PSI - Issue 78

Mariano Di Domenico et al. / Procedia Structural Integrity 78 (2026) 1237–1244

1241

4. Results Fig. 3 shows the normalized debris volume and ejection distance for all ground motion records in the cases of 3P70 and 6P70 buildings. These two were selected as representative examples; results for the 1950s and 1990s buildings are not shown for brevity, as no additional trends were observed. On the graphs, debris volume is normalized by the total volume of infill walls on a single façade, while ejection distance is normalized by the total building height (9.5 m or 18.65 m). Spectral acceleration Sa(T = 1 s) is reported along the Y-axis.

3P70 6P70

Fig. 3. Normalized ejected volume and ejection distance for all records.

As expected, at low seismic intensities, debris volume and ejection distance are both zero. These quantities increase with seismic intensity but tend to stabilize at higher levels. This "saturation" effect is visible as a nearly flat vertical line in the plots and is attributed to the onset of nonlinear behaviour in the structural system (Di Domenico et al., 2021), which limits further increases in floor accelerations and, in turn, in debris velocity and volume. There appears to be no clear difference between infills with and without openings, likely due to normalization. Fig. 4 presents summarized results by stripe: average values and 16th/84th percentiles of normalized debris volume and ejection distance, assuming a lognormal distribution of values across records. Key observations include: 1. Expected debris volume does not exceed 30% of the total for older buildings and 20% for newer ones. 2. Expected ejection distance reaches up to 10% of the height for older buildings and up to 25% for newer ones. 3. A saturation effect is observed for both volume and distance at high seismic intensities, as nonlinear response flattens acceleration demands. 4. Solid infills (without openings, blue curves) tend to be ejected further than infills with openings (red curves), due to their greater strength and later collapse with higher ejection velocities. However, this result may also depend on the dynamic response of the structure and is not always consistent. 5. There is no clear hierarchy between 3-storey and 6-storey buildings due to normalization: height and infill volume both scale with the number of storeys. 6. Construction age shows clearer trends: newer (1990s) buildings have lower normalized debris volumes but higher normalized distances. This is likely due to their stronger frames and infills, which collapse less frequently but with higher velocity once failure occurs.