Estimation of pedestrian flows with open-source crowding data: an integrated modelin Nomentano-Tiburtina district, Rome

This research aims to investigate pedestrian crowding in the urban area of the Nomentano-Tiburtina district in Rome, through a methodology that sees an integrated approach between a configurational digital model, created with the Space Syntax tool by Qgis, and a physical model based on data collection through the mobile observer mode and therefore directly and through the use of Google Street View, in virtual terms. An innovative methodology was therefore developed to be able to obtain a physical model through virtual data collection, and this made it possible to calibrate the digital model. The calibration highlighted, with a Pearson of 0.88, the effectiveness of the methodology developed. At the same time, all the factors influencing the digital model, such as demand-driven and land-use, were also studied. From the analysis of these factors, it was found that the presence of shops on the pedestrian network plays a strategic role in the assessment of pedestrian crowding, recording a better correlation than the data measured on site. The conclusions drawn from this research, therefore, are of considerable help in urban planning, emphasizing concepts such as accessibility and pedestrian safety.