Agent-based transport models require detailed disaggregated information on individuals, households, and daily activity patterns. However, such data are not available mainly due to cost of collecting and privacy issue. Census data are either aggregated information, or it lacks detailed residential or in general activities location. Mobility surveys are limited in size and also mostly lack activity’s location due to privacy. This thesis addresses this issue in a specific case study by developing a synthetic population for the city of Torino to support the eventual implementation of an agent-based model. To achieve this result, this study combines three data sources, namely (1) the 2011 census aggregated dataset at census tract level, (2) the 1% census individual and household microdata sample and (3) a very detailed mobility survey that was administered by Politecnico di Torino in the Turin metropolitan area in 2017.