PSI - Issue 84

Alessandro Scala et al. / Procedia Structural Integrity 84 (2026) 489–496

493

often represents the clearest evidence, particularly if the deformation is horizontal, as can occur when the bearing is made of neoprene (Fig. 2b). Expansion joints can also provide useful information for detecting landslide-induced stress, especially if evidence such as offsets, separations, or deformations is present. It is also crucial to recognize signs present in the terrain surrounding the bridge, as they can not only indicate the presence of slow-moving landslide phenomena nearby but also provide information about their characteristics. Ground cracks, leaning poles or trees (Fig. 2c), areas devoid of vegetation, or actual detachment niches (Fig. 2d) can help in gaining an initial understanding of the landslide’s surface extent, its direction, and possible kinematics. However, it should be noted that such signals can be difficult to detect, particularly in areas that are inaccessible or covered by vegetation.

Fig. 2. Examples of signs of slow landslide-structure interaction (a) missing and damaged masonry elements; (b) horizontal deformation of neoprene bearing; (c) tilted poles; (d) visible detachment niche. 4. Some applications The method was applied to a sample of 23 bridges, which differ both in terms of structural type and in the type of landslide phenomenon with which they interact.

a) b) Fig. 3. Graphs showing the calculated values of (a) the pre-inspection index and (b) the post-inspection index for the 23 bridges. Yellow bars represent the final value computed for the selected bridge, while the green portion represents the contribution of categories with negative weights (e.g., information related to knowledge level and mitigation measures), which reduce the final index value.

Table 1. Case studies.

Name of the viaduct

Static scheme Hyperstatic Hyperstatic

Landslide type

PI pre

PI post 0.234 0.516 0.448 0.425

Landslide CoA

A (Viaduct 21) B (Viaduct 18) C (Viaduct 13) D (Viaduct 12)

Rapid Slow Rapid Slow

0.649 0.353 0.080 0.599

Medium

Medium-High Medium-High

Isostatic Isostatic

High

Fig. 3 presents the results obtained for the entire sample. This representation allows the contribution of both positively weighted categories and negatively weighted categories to be clearly identified. At first glance, the coefficient may appear to decrease after inspection. However, this effect is due to the fact that the pre- and post-