I.M.P.A.T.T.O. – Spatiotemporal GeoBigData and Predictive Models for Urban Planning andpolicies support in the Municipality of Udine

Abstract. As urban centre face growing challenges from climate change, air pollution, and complex mobility patterns, city governments increasingly seek tools to plan more sustainably and act more responsively. Despite advances in urban sensing and data availability,most mid-sized cities struggle to integrate diverse data sources into a single platform for anticipatory decision-making. Moreover,traditional smart city approaches often lack citizen participation and educational value. I.M.P.A.T.T.O. addresses this gap by designing and deploying an end-to-end urban intelligence system in Udine, a mid-sized Italian city known for its proactive environmental and digital policies. The project integrates mobility, environmental and presences data, low-cost environmental sensors developed by students. A web-based dashboard will be developed with predictive models that allow scenario-based planning and early-warning capabilities. Unlike many top-down initiatives, the system is co-designed by researchers, technicians, students, and municipal officers, allowing for long-term ownership and institutional integration. In this paper, we describe the architecture, methodology, and operational model of I.M.P.A.T.T.O. We outline the key technologies used, discuss the participatory sensor deployment and educational modules, and present results from predictive modelling and dashboard integration. The paper concludes by discussing the alignment with international SDG frameworks and the potential for replication in other cities.