With assignment methods are generally indicated several statistical approaches that use genetic information to identify or verify membership of individuals or groups of individuals. In the present research, the stepwise discriminant analysis (SDA), the canonical discriminant analysis (CDA) and the discriminant analysis (DA) were used to develop an assignment method based on genotypes obtained from the Illumina's OvineSNP50 BeadChip. The three statistical techniques were used to select the most informative markers for discriminating 21 different sheep populations (20 Italian and 1 Slovenian) and their 5 geographic areas of origin. The procedure selected two reduced pools of markers, 110 and 108, that were able to exactly discriminate the 5 geographic areas and the 21 sheep populations, respectively. Moreover, all animals were correctly allocated in the validation procedure. The obtained results indicated that the combined use of the SDA, CDA, and CD techniques represents an efficient strategy for selecting a reduced pool of highly discriminant markers. These SNPs could be further used to develop a low cost customized essay to trace the breed or the geographic area of animals or derived foodstuffs.