Master of Science in Chemical Engineering (MSChE)
This research is focused on the modeling and optimization of the crude oil scheduling problem in order to generate the most appropriate schedule for the unloading, charging, blending, and movement of crude oil in a refinery, which means obtaining the schedule that generates the lowest costs. Uncertainty, which is often present in these types of optimization problems, is also analyzed and taken into account for the resolution of crude oil scheduling problem. A comprehensive novel model is proposed to describe the upper level crude oil scheduling problem, generate an optimal solution for the mentioned problem, and allow integration with the lower level production optimization problem of a refinery. This integration is possible due to the consideration of total flows of the different types of crude oil instead of flows of a particular key component in the crude oil to linearize the upper level problem and generate a less complex model. The proposed approach incorporates all the logistical costs including the sea waiting, unloading and inventory costs together with the costs associated with the transfer of crude oil from one to another entity. Moreover, this model also offers the possibility of considering multiple tank types including storage and blending tanks throughout the supply chain and the incorporation of the capability of storing more than one crude oil type in the storage tanks during the schedule horizon. A comparative analysis is performed against other models proposed and preliminary results of integration with a lower operational level are provided. In order to take into account the possibility of uncertainty or fuzziness in the scheduling problem, for the first time an approach is proposed to face the resolution of this problem in order to obtain a more realistic scheduling of the allocations of crude oil. Fuzzy linear programming theory is used here to represent this uncertainty in order to find an optimal solution that takes into account the lack of precise information on the part of the decision maker without losing the linearity of the original system. Uncertainty in the minimum demand to be satisfied in the distillation unit according to the necessities of the market and the lack of precise information about certain costs involved in the operations throughout the supply chain are separately considered. Among the different approaches utilized in fuzzy linear programming, the flexible programming or Zimmermann method and its extension to fuzziness in objective functions are implemented. A comparison between the two cases studied and the crisp model is performed with the aim of determining the effect of these uncertainties in the schedule of the crude oils movements between the different entities in the supply chain and the total cost generated.
Document Availability at the Time of Submission
Student has submitted appropriate documentation to restrict access to LSU for 365 days after which the document will be released for worldwide access.
Chebeir, Jorge Asis Charbel, "Modeling the Crude Oil Scheduling Problem with Integration with Lower Level Production Optimization and Uncertainty" (2015). LSU Master's Theses. 31.