Cómo planificar óptimamente la cadena de suministros de una refinería de petróleo

oil refineries use linear programming applications to the difficult task of planning their supply chain. however, since those applications are of a very specific purpose, and black box type, they have an internal programming unknown to the user. so modifications to situations where, for example, there exist parameters with nonlinear functional relationships are not allowed. in this paper we propose a mathematical model to optimally elaborate the supply chain in a petroleum refinery. a simple conversion refinery scheme was selected and modeled as a flow network. the gams modeling language and it solvers cplex and snopt were used. subsequently, a case study related to certain changes in the variables and decision parameters was done, in order to evaluate the performance of the model. in every case the model was able to provide an optimal mix of the feeding flows which maximize the profit in the refinery process, and being consistent with the final product specifications.