PSI - Issue 78

Marco Civera et al. / Procedia Structural Integrity 78 (2026) 1783–1790

1788

(a) (b) Fig. 2. Uncorrected RN efficiency for a scenario with return period =475 years and considering DS3 – (a) new simulations and (b) previous ones, retrieved from (Miano et al., 2024) . The differences are due to the use of Monte Carlo simulations; both graphs converge to the same value of =0.49 . 4.4. Damage State 4 and 5 (DS4, DS5) At this point, most of the simulations return no instances of > , as shown in Figure 3. Simply put, the seismic hazard in the selected area is not high enough to likely cause any DS4 level of structural failure. Only after circa 1500 simulations, the results begin to converge towards =0.615 , which is very close to the pre-earthquake maximum efficiency of the RN (-3.9%). Once more, it is important to stress that this does not mean that no asset was damaged; it only means that very few assets suffered extensive damage (i.e. partial collapse). Instead, if the corresponding corrective factor ( =0.008 ) is applied, one can obtain , =0.005 , following the monotonical downward trend already established so far.

Fig. 3. Uncorrected RN efficiency for a scenario with return period =475 years and considering DS4. Finally, for DS5, the uncorrected mean efficiency is stable and never deviates from the upper bound ( = , =0.64 ). Simply put, in not even one simulation, out of 2000, a single bridge of building suffered total collapse – that is due to the low seismicity of the area surrounding Turin. Of course, as mentioned earlier, this case cannot be investigated any further because the DS is deemed to have (almost) zero probability of occurring. 4.5. Results for Sc50 For completeness, the same analyses described in Sec. 4.1 – 4.4, considering Sc475, were also performed for Sc50. The results are qualitatively similar and, therefore, do not require much commentary. As before, DS5 was never