Raman spectroscopy as an effective tool for material identification offers benefits that are important for the process control of the fibres in terms of quality control in production, in use, but also in the field of fibre analysis for forensic identification. In particular, it is specific information on chemical composition and structure of material; rapidity of measurement, non-destructiveness, independence on chemicals, portability, possibility of automation of measurements and ability to detect structural changes. In the study, samples of natural and synthetic polymer fibres were measured and analysed. For the evaluation of large spectral data sets multivariate analysis techniques were applied to find the patterns within the data. The spectral database was created on the basis of the acquired Raman spectra of textile fibres and was successfully tested on other samples. The results show a great potential of the used method for the fibre identification.