The goal of this master thesis is to understand if the use of an immersive system in the field of railway safety might improve the understanding of data by decision-makers. The driving analysis of a simulation performed for train drivers of the Schweizerische Südostbahn AG company as part of the maintenance of the Italian language skills of German-speaking train drivers was taken as a case study. The simulations required the drivers to deal with potentially dangerous situations on the Airolo-Biasca route and to react to such events by driving and communicating with the dispatcher in the best possible way. As in most rail safety cases, the simulation data recorded during the project are of different types and extracted from different sources. The audio data comes from recordings made using a digital recorder, the video data is taken from the cockpit monitor, and the numerical data derived from the train's motion recorder. As stated in the literature review, information from different types and data sources is usually represented in an overly complex manner. This complexity may be due to various design-dependent factors, such as, for example, the use of representations of data relationships that are overly complex, ambiguous, difficult to understand, or with represented designs that are understandable only by specific users. Therefore, this research uses immersive systems, Hololens1, to attempt to circumvent these problems and understand if the visualization is simpler.