A total of 1430 lactation records of 434 daughters sired by 78 bulls from 1999-2014 collected from 3 Military Dairy Farms of Southern Command were analysed to compare various sire evaluation methods for Calving Interval (CI), Lactation Length (LL), Standard Lactation Period Yield (SLPY) traits. Least Squares method (LS), Best Linear Unbiased Prediction (BLUP) method and Average Information Restricted Maximum Likelihood (AIREML) method (both univariate and multivariate model) was used for Sire evaluation. Spearman’s rank correlations were estimated. Coefficients of determination and coefficients of variation were used to measure the accuracy and stability of models. The least squares means for CI, LL and SLPY were 443.81±20.41 Days, 329.91±11.56and 3136.38±224.11 kg respectively.Most efficient model for CI was univariate BLUP as evident from error variance 2417.10 Day2.The most efficient model for LL was multivariate BLUP of CI-LL with error variance 1001.43Day2. The most efficient model for SLPY was univariate AIREML with error variance 118173 Kg2.