Pollution and global warming are current issues of concern, and their mitigation is a challenging problem. Trans-portation and the livestock industry are large generators of Greenhouse Gases (GHG), and therefore environmental criteria must be considered in the decision-making. This paper, part of an ongoing research on the design of milk and dairy supply chains with environmental considerations, proposes a mixed integer programming model for the milk collection problem with heterogeneous fleet with CO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> minimization criterion. Validation of the model was made using the Gurobi solver over 15 randomly generated instances, optimally solved, and compared to a distance minimization model. The results show the prevalent use of the largest and most polluting vehicles, in particular, to cover routes with a greater number of farms far from the factory. Also, results suggest that the distance of routes can increase by up to 70% if emissions are optimized, and emissions can increase by up to 20% if distances are minimized. Finally, future research work is proposed.