The present work had as its main objective to identify and characterize milk production systems through methods of multivaried statistical analysis. Factorial analysis, cluster, discriminant and canonical correlation were utilized. The production gradation and defrayal were the main classification criterium among the productors. Combining the results obtained with the cluster analysis and the canonical correlation analysis, the second classification analysis aggregates the grazing lands and the care with the flock health. Three productors groups were identified, among which the first one is prominent for aggregating about 90% of the total analyzed and is composed by relatively smaller productors. The discriminant analysis and the measuring among financial indicators confirm such classification.