A New MILP Formulation for Scheduling of Crude Oil Operations

Scheduling of crude oil operations is one of the most critical and challenging problems in petroleum refineries. Crude oil operations have more impact on the profitability of the whole refinery as these operations are the first steps of refinery operations. Crude oil operations scheduling is a complex problem as many complicated relationships and many conditions must be considered. In a refinery, a planner tries to schedule all crude oil operations by using a wide variety of complex and very dense data sets. Daily planning is mainly based on manual scheduling by trial and error method and operational experiences. Therefore, as a planner, it is almost impossible to reach the optimal solution or even a satisfactory feasible solution by handling all the data sets in a short period. Unfortunately, there is no efficient and adaptive tool or software that can be used for different refinery scheduling cases. In this study, an efficient scheduling and allocation model for an oil refinery located in İzmir is developed for unloading crudes into multiple storage tanks from the vessels arriving at various times, blending crudes, and feeding the crude distillation units from these tanks at various rates over time. To tackle the scheduling problem effectively, a Mixed Integer Linear Program (MILP) is formulated. The model is solved for the refinery, and satisfactory solutions are gathered.