The digitalisation, automation and robotisation of road inspection and maintenance technologies make it possible to collect bigger volumes of data and additional types of information about road infrastructure. Methodologies and tools to support road asset management decision-making are needed to exploit this new information, progressing towards predictive maintenance and improving different aspects of road asset management. This study presents a Digital Twin-based Decision Support Tool to assist road operators in road inspection, maintenance and upgrade. The goal of the paper is twofold. First, the architecture of the Digital Twin-based Decision Support Tool is presented, describing the main components and functionalities. The system is based on a Digital Twin (DT) that mirrors real road assets to integrate different sources of data and support the processing of low-level data into high-level information. The decision support tool (DST) is able to analyse the collected information and compute the road pavement condition to derive optimal intervention plans, addressing road section conditions, human and technical resources and other external constraints. Second, the application of the proposed architecture to road pavement maintenance is described, considering the Italian highway A24 and its connections with Rome´s ring road, managed by Strada dei Parchi SpA. Road pavement data, such as the International Roughness Index (IRI) and the Sideway Force Coefficient (SFC), are integrated into the DT to be analysed through Artificial Intelligence-clustering techniques to perform the sectioning and clustering of road sections according to their status and quality index. The paper shows the benefits derived from the integration of DT technologies with DSTs for improving processes of road maintenance.