Speaker
Description
In emergency evacuation analysis, the geometric characterization of walking surfaces is a key factor influencing safety conditions and pedestrian behavior. While floor inclination is commonly evaluated from point cloud data, this parameter alone may not adequately reflect a surface's actual walkability, particularly in the presence of local irregularities. This work introduces a metrological approach that combines slope estimation and flatness evaluation using point clouds acquired by a Terrestrial Laser Scanning (TLS). Slope is derived from the orientation of locally estimated surface normals, whereas flatness is quantified through deviations from locally approximated planes. The analysis deepens understanding of factors that can affect measurement uncertainty, including the choice of algorithm, the setting of hyperparameters, and the study strategies for the floor surface. In particular, both quantities are evaluated at different spatial scales by modifying the size of the “neighbourhood” used in the calculations. The methodology is applied to a point cloud dataset acquired on the rooftop of an italian Basilica open to the public. Preliminary findings indicate that areas with similar inclinations may differ significantly in surface regularity, highlighting the importance of adding flatness into the analysis. The proposed work provides a framework for a more complete geometric characterization of evacuation paths.