Pre-Event-Reference-Free TLS-Based Geometric Condition Assessment of Post-Fire RC Flexural Members

Post-fire assessment of reinforced-concrete (RC) members is commonly based on concrete compressive strength measurements, a local property that may not fully represent the residual behaviour of flexure-governed members. This study addresses this limitation through a non-destructive methodology based on TLS, aimed at estimating the apparent relative loss of flexural stiffness from observed differences in residual geometry. The point clouds were filtered, rasterized, and subsequently normalized using regression planes fitted to each member. Orthogonal deviations from these planes were statistically analysed to define a geometric condition indicator (DI) based on the difference in point concentration between homologous members within a central deviation band. By comparing structurally equivalent members, increases in beam deflection and camber losses in prestressed joists were calculated and used to derive an apparent deformation-based stiffness-change index (KΔ). The methodology was applied to a one-way ribbed slab affected by a real fire. DI showed strong monotonic association with KΔ, with Spearman coefficients of −0.82 for the beams and 0.92 for the prestressed joists. Beams exhibiting increased deflection and joists exhibiting reduced camber relative to their homologous counterparts showed mean apparent relative stiffness losses of 32.7% and 36.7%, respectively. The results show that the statistical distribution of ge-ometric deviations provides information on the global response of the member that is not captured by the local compressive strength of the concrete. The proposed indica-tors also enable the identification of geometric anomalies consistent with post-fire structural deformation, providing a non-destructive screening and prioritization tool that complements inspection, material testing, and structural analysis.