Within the management of railway infrastructures, rail defect monitoring is one of the most significant task since it affects maintenance interventions, on one hand, and safety of train operation, on the other. In recent years, several high performance measuring devices, providing a huge amount of data that have to be carefully analyzed, have been developed and employed to accomplish this aim. There is therefore the need to develop efficient and reliable procedures able to reduce them into efficient synthetic descriptors and to detect significant variations. In this paper a preliminary analysis following a large experimental measure-ment campaign carried out on the Circumvesuviana railway network operating in the Naples metropolitan ar-ea is reported. Track cant, level, gauge and twist data have been collected with a 25 cm sampling interval within a two year long period by using the Plasser device (POS/TG System), directly mounted on an conven-tional vehicle. Preliminary analysis show that by combining track alignment data with operational infor-mation, a more reliable prediction of the evolution of defects can be derived

No abstract