PSI - Issue 78

Nadia Salvatore et al. / Procedia Structural Integrity 78 (2026) 81–88

86

Fig. 4. (a) Stratigraphic log of the first 18 m under the abutment n.2; (b) plot of CSR 7.5 (cyan line) and CRR calculated from CPT (blue line) and SPT(orange points) versus depth; (c) plot of FS calculated from CPT (blue line) and SPT (orange points) versus depth; (d) plot of LPI calculated from CPT (blue line) and SPT (orange line), using the simplified methods proposed by Idriss and Boulanger (2008); (e) plot of LDI calculated from CPT versus depth. 4. Computing the GETI index The PGA and D H values were plotted in the fragility curves (Brandenberg et al., 2011) for the evaluation of damage on bridges in case of ground shaking and lateral spreading respectively, assuming that the Permanent Ground Displacement (PGD) value is conceptually the same of D H . The results are shown in Fig. 5 a and b. The obtained values of damage were put into the equations developed by Salvatore et al. (2025) to calculate the GETI, as follows: = ( � ) + � �� � + ( � ) − ( � ) � �� � − ( � ) ( � ) − � �� � ( � ) + ( � ) � �� � ( � ) (1) where ( � ) is the probability of damage from ground shaking; � �� � is the probability of damage from liquefaction and liquefaction induced lateral-spread; ( � ) is the probability of damage from landslide. GETI is expressed as the total probability of damage on a linear infrastructure, given by the sum of the probability of occurrence of damages derived from ground shaking, liquefaction, and landslide/rockfall, subtracting the product of the probability of damage derived from the simultaneous occurrence of the hazards (Salvatore et al., 2025). The input parameters for each class of damage (low, medium, high, severe) are reported in Table 1 before and after the ground improvement. Before the soil treatment the main hazard is related to lateral spreading induced by liquefaction ( � �� �=1 ), while the effect of ground shaking is almost negligible ( ( � )=0.04 ). As outlined in the Level 1 analysis, the occurrence of a landslide in the studied area is not relevant due to the distance of the infrastructure from the mountain range, so the probability of damage derived from landslide was setting to 0 ( ( � )=0 ). Table 1. Input parameters for GETI before and after ground reinforcement ( � ) � �� � before � �� � after ( � ) Low 0.04 1 0 0 Medium 0.01 1 0 0 High 0 1 0 0 Severe 0 1 0 0