Application of a Mixed Integer Nonlinear Programming model for the optimization of the production of a Brazilian refinery

The oil industry comprises processes that involve calculations for mass and energy balance,
making activities such as production planning and scheduling quite challenging. Problems
involving these activities are generally as NP-Hard due to the various constraints involved in
the process, necessitating the application of operations research techniques to generate high-quality
solutions within feasible computational time. Therefore, this work proposes a Mixed-Integer Nonlinear
Programming model to optimize the transfer of loads (mixture of crude oils) between Storage
Tanks and the Atmospheric Distillation Unit of a Brazilian refinery, aiming to maximize the processed
load while considering operational and physico-chemical constraints. The obtained results
were satisfactory, as it was possible to achieve feasible solutions in a short computational time with
higher load processing compared to the company in the case study.