The most important objective of any production unit is to co-ordinate and utilize the various resources of production in such a manner that they yield the highest net returns. As farmers are shifting from low input-low output to high input-high output and traditional and subsistence to commercial dairy farming, this study assesses the impact of resources being used optimally through herd size categories by the dairy farmers for buffaloes and crossbred cows in the Middle Gujarat region. The resource use efficiency analysis helps the dairy farmers in taking appropriate decisions regarding resource allocation without using additional resources for enhancing their income. The findings show that on an average, the costs of human labour, green fodder and concentrates had significant influence on milk yield for buffalo milk production while costs of human labour, green fodder, concentrates and veterinary and medical charges had significant influence on milk yield for crossbred cow milk production implying that one per cent increase in the use of these inputs can led to increase in the gross returns from milk. Milk production for buffaloes was in decreasing returns to scale while on an overall basis, milk production for crossbred cows was increasing returns to scale. Further, the difference between MVP and its acquisition unit price for human labour in small category, green fodder in overall category and dry fodder in marginal category in buffalo milk production and price for human labour in medium and overall category and green fodder in large category in crossbred cow milk production were found positive and significant indicating that the dairy farmers have an opportunity to increase their profit by using more of these inputs on their farms. Thus, the study concluded that only in crossbred cows, milk production was size neutral.

Correlation and regression analysis were carried out with Standard week (meteorological standard week) wise data of milk yield and meteorological data (minimum temperature (MIN T), maximum temperature (MAX T), morning relative humidity(RH1), afternoon relative humidity(RH2), morning vapor pressure (VP1), afternoon relative humidity (VP2), wind speed (WS),bright sun shine hour(BSS), rain fall(RF) and evaporation rate (EP)) of 12 years by SAS (Statistical Analysis Software) programme (version 9.3). The overall average DMY of cross bred cows during 12 years period (2003-2014) was 8.73±0.26 kg. Maximum DMY was recorded during March (10.15±0.52 kg) and minimum was recorded during September (7.72±0.34 kg). Average DMY ranged from 4.39±0.37 to 13.83±0.55 kg from 12 year 2003 to 2014. Month of year had significant (P<0.05) effect on daily milk production. Correlation between Standard week wise milk yield data (of all 12 years) and MAX T was positive (0.1893) and significant, Whereas, RH1, RH2 and VP2 had negative and highly significant. RF was also found to be negative but significantly correlated. EP and BSS had positive and highly significant correlation with weekly average MY. There was no significant correlation among weekly average MY and MIN T and VP1 but WS was significantly correlated with average weekly MY. All the partial regression coefficients for independent variables were found significant except RH1, RF and VP2 but coefficient of determination (R 2 ) was found very low (0.1652). Stepwise regression analysis showed that the partial regression coefficient of MIN T and BSS were found negative and highly significant whereas, partial regression coefficient of RH1, WS and EP were found positive and highly significant. The R 2 of final model was 0.66 this indicated that the weather parameters (RH1, WS and MIN T) explained about 66% variations in average standard week wise milk production (averaged out of all 12 years data). The partial regression coefficient (R 2 ) of MIN T and RH1 were found negative and significant indicating negative impact on average standard week wise milk yield whereas that of WS had positive impact on average standard week wise milk yield.