This paper, with reference to the activity of the National Agency for the Safety of Rails and Road and Highway Infrastructures (ANSFISA), describes the application of digital methods and tools (BIM modeling; collaboration platforms, Machine Learning / Artificial Intelligence, Digital Twin, etc.) for the management of existing road infrastructure networks, on the basis of the principles contained in the current Italian Guidelines for risk classification and management, safety assessment and monitoring of existing bridges.

The safety of infrastructures passes through the digitization of processes and tools aimed primarily at the knowledge of the assets, the evaluation of the risks (structural-foundational, seismic, hydrogeological) and the conscious optimal decisions. The optimal decision making have to be based on structured, quantitative and reliable information.

The Agency's objective is to improve the digital information management of road infrastructure assets, in order to enhance the information, making it more accessible and with interoperable languages and consequently also provide a monitoring platform, as well as structuring the bases for subsequent modeling also, possibly, to support the training activities of inspectors using both information models and VR simulations in order to create an educational environment in which to train inspectors.

The first step to be taken concerns the identification and insertion of data, by the managers, in the IT Archive of Public Works (AINOP).

In this work, the Guidelines of the existing bridges have been imagined in an overall framework of digital information management of infrastructures, where BIM, understood as "multi-BIM" according to a multi-level approach, is one of the main tools, with a view to conscious risk mitigation.