The aims of this study were to evaluate the energy consumption and its modeling in industrial milk production units using data envelopment analysis (DEA) approach and sensitivity analysis. Data were collected from 44 industrial milk production units in Guilan province of Iran with face to face questionnaire method during 2012-2013. Inputs included animal feed, fossil fuels, electricity, machinery and human labor and output was milk yield. The results highlighted that the highest share of energy consumption belonged to the animal feed (82%) followed by fossil fuels (13%). The results of DEA application revealed that the technical efficiency (TE) was 44.6%, pure technical efficiency (PTE) was 74.48% and scale efficiency (SE) was 0.53 for the investigated units. The results of variable returns to scale (VRS) illustrated that out of 44 producers considered for the analysis, 15 units were purely technically efficient in energy consumption from the different sources. The greatest potential inputs for saving energy were related to animal feed and diesel fuel, respectively. About 3852.39 MJ (47.74% of total energy input) could be saved without reducing the milk yield through following recommendations resulted from this study. Moreover, sensitivity analysis results showed that with one MJ extra use of human labor, fossil fuels and animal feed energy, it would lead to an additional increase in yield by 3.14, 1.36 and 0.95 kg, respectively. Also, one MJ extra energy use for either electricity or machinery result in a decrease in milk yield by 1.03 and 7.16 kg, respectively. پرونده مقاله