(190d) A Study for Integration of Procurement Planning and Short-Term Scheduling in Petroleum Refineries

Mathematical programming has been implemented into decision making since 1950s for crude planning to optimize both production and crude purchasing plan. Long-term planning and short-term scheduling constitute two important decision elements for petroleum refineries. Planning is responsible for creating profitable plans exploiting the constantly changing market opportunities, while scheduling is required to execute feasibly and optimally as much as possible according to obtained plans. It has been studied and practiced for a long time, especially in the last two decades driven by increasingly intensive global competitions, more volatile feedstock and product markets, as well as stricter environmental regulations. Especially, the availability of new crudes will lead to the evaluation of refinery assets to leverage refinery procurements in order to maximize refinery profits.

Unfortunately, due to the differences between planning and scheduling, such as different time scopes and process constraints, discrepancies always exist, which initiate inherent gap between generated plans and schedules. Therefore, the communications between planning and scheduling need to be enhanced in order to validate the generated plans and make any necessary adjustments for the plans to be feasible under real operation settings. Existing publications are either neglecting the ever-changing crude supplies or product demands, variable shipment sizes and demurrage costs [1] or focused on an aggregated planning problem without considering realistic operating practices and connection decisions between units [2]. And the aforementioned works disregard the continuous operation of crude distillation units. The reason for these sacrifices lies in the extreme difficulties when solving a full space problem with integrated formulation of planning and scheduling.

In this paper, an iterative framework integrating refineries crude procurement planning and detailed scheduling is proposed. The general objective of the integrated model is to maximize crude refining profits while satisfy all crude transfer and operation constrains. The long-term planning is decomposed into sub-periods representing different scheduling horizons. Hierarchically, the planning problem serves as a master sub-problem while the detailed scheduling is a salve sub-problem. The crude purchasing and refining plans from master problem are provided as inputs to the slave problem. Under the circumstances that any planned operations are impossible with detailed processing constraints, a feedback loop from the slave problem to the master problem will be added in the form of extra constraints. The goal of such feedback is to find the true optimal high-level decisions. In addition, the master problem can provide an increasing lower bound while the sub-problem can provide an upper bound. Thus, iterative methods can lead to optimal solutions if solved until the gap is closed [3]. Also a rolling forward time horizon framework will be created in such a way that once a portion of the time horizon has been scheduled, no additional modifications are allowed for the given period of time horizon. Thus, the number of iterations required to schedule the entire time horizon will be minimized. The finalized refining profits will be compromised due to consistent refinement from detailed scheduling problem. A detailed case study will be conducted to validate the efficacy of the proposed integration methodology framework.

References

1. Oddsdottir, T.A., M. Grunow, and R. Akkerman, Procurement planning in oil refining industries considering blending operations. Computers & Chemical Engineering, 2013. 58: p. 1-13.

2. Zhang, J., Y. Wen, and Q. Xu, Simultaneous Optimization of Crude Oil Blending and Purchase Planning with Delivery Uncertainty Consideration. Industrial & Engineering Chemistry Research, 2012. 51(25): p. 8453-8464.

3. Maravelias, C.T. and C. Sung, Integration of production planning and scheduling: Overview, challenges and opportunities. Computers & Chemical Engineering, 2009. 33(12): p. 1919-1930.