Two studies, those of Maquaire et al. (2004) and Ferlisi & Pisciotta (2007), cluster the exposed buildings to units of equal vulnerability. The first authors present a semi-automatic procedure using Hierarchical Ascending Classification to aggregate the buildings at risk based on their geometrical and contextual characteristics. Their methodology is compared with an expert’s vulnerability zonation and the two results are in accordance for the majority of the buildings. Also, Ferlisi & Pisciotta (2007) homogenized the buildings at the Lazio and Abruzzo regions in Central Italy according to their use in order to reduce the time for data collection in a working scale of 1:25,000. Each homogeneous unit consists of buildings with specific use (residential, commercial and industrial) whose distance between them does not exceed 100 m. In terms of loss index, Kappos et al. (2009) estimated the risk in urban blocks; buildings that belong to the same urban block form a cluster and the loss index of this cluster is the average of the loss indexes of each building weighted upon the built area.