The management of safety on railway construction sites represents a particularly complex challenge due to the presence of existing infrastructures, work vehicles, simultaneous work and operational constraints typical of interventions near railway lines. The present work proposes an integrated methodology for the preventive management of safety based on the integration of reality-based survey by UAV, Building Information Modeling (BIM) and simulation of construction site activities. The workflow begins with the acquisition of existing conditions through drone photogrammetry, the generation of a point cloud and the creation of an interoperable BIM model of the railway infrastructure. The model is subsequently integrated with parametric models of the work vehicles, their operational envelopes and with the time planning of the works, allowing 4D simulations and clash detection analysis for the preventive detection of spatial and temporal interferences. The methodology was applied to a case study relating to a real railway section, demonstrating how the integration of reality-based survey, parametric modeling and time simulation makes it possible to identify critical issues that are difficult to detect using traditional planning tools. The results highlight the potential of the proposed approach in supporting the planning of site activities, the comparison of alternative operational scenarios and preventive safety management. The main contribution of the research consists in the development of an interoperable workflow that integrates reality-based surveying, Scan-to-BIM modeling, 4D simulation and semantic representation of safety information through BIM objects, laying the foundations for future automatic rule checking procedures and for integration with Digital Twin environments.