A mixed-integer linear programming model for automatic routing decisions in oil production optimization

Production optimization of gas-lifted oil wells with routing and pressure constraints is a complex problem, which has attracted the interest of engineers and scientists alike. To this end, a mixed-integer nonlinear model is developed to automatically decide upon flowing well production to a single or multiple manifolds (in the paper referred to as automatic routing). To bypass the complexity of the problem due to its nonlinear and discrete characteristics, a mixed-integer linear reformulation is proposed using piecewise-linear approximations. The resulting formulation splits flows in the pipelines based on the pressure differences which is also observed in commercial multiphase flow simulators. Finally, a computational analysis shows that significant gains can be obtained in oil production by considering automatic routing in the optimization model.