(72c) An Integrated MINLP Model of Simultaneous Crude Procurement Planning and Movement Scheduling for Refineries

The petroleum refining process is very complicated, since huge volumes of crudes should be continuously handled through a highly integrated network with extremely intensive units. As one of the key factors that could significantly affect the refining process, the procurement of feedstock, including detailed types, volume and timing, is still vulnerable to restrict operational requirements of refinery and various uncertainties. Different crudes may possess distinct properties, thus require additional operations or even different processing procedures. On the other hand, the crude procurement cost is comparable to the operating cost in quantity, taking up a significant proportion of the overall cost [1]. And the crude market has become highly volatile since the last two decades [2]. Improper crude procurement plan will inevitably cause serious operating problems and result in economic loss. Thus, both process feasibility and refinery profit margin will be affected. Note that after the crude procurement plan has been made, crude movement schedules, covering from vessels, berths, storage tanks, pipelines, charging tanks to CDU have to be well scheduled to accomplish the smooth and stable crude transfer. In previous studies, these issues were addressed in a sequential manner, generating respective crude purchase plans and crude movement schedules [3,4]. Some works studied a partial combination of crude purchasing and blending operations [5,6]. However, the two crucial elements have not been simultaneously considered in a full scope, which could result in the gap between current solutions and real optimality in terms of both operation feasibility and process economy.

In this study, a new methodology has been proposed to tackle this tough challenge. Both crude procurement and oil movement will be simultaneously considered in order to eliminate the optimality gap that could obstruct the processing efficiency and profitability of refinery. Accordingly, a novel MINLP model has been developed. Through this model, the optimal crude procurement plan and oil transfer schedule can be efficiently determined towards the maximal profit margin. More importantly, the integrated method will ensure that decisions for the two sectors can harmoniously cooperate and eventually benefit the processing of refinery. In addition, realistic requirements have been taken into account in this work, including purchasing constraints, such as crude characteristics and crude delivery time and process requirements, for instance tank inventory and blending specs, so that the model applicability can be guaranteed. The developed model has been tested through various case studies to demonstrate its efficacy and performance.

References

[1] Ji, X., Huang, S., & Grossmann, I., (2015) Integrated operational and financial hedging for risk management in crude oil procurement. Industrial & Engineering Chemistry Research, 54, 9191-9201.

[2] Chen, R., Deng, T., Huang, S., & Qin, R., (2015) Optimal crude oil procurement under fluctuating price in an oil refinery. European Journal of Operational Research, 245, 438-445.

[3] Kallestrup, K., Lynge, L., Akkerman, R., & Oddsdottir, T., (2014) Decision support in hierarchical planning systems: The case of procurement planning in oil refining industries. Decision Support Systems, 68, 49-63.

[4] Zhang, S., & Xu, Q., (2015) Refinery continuous-time crude scheduling with consideration of long-distance pipeline transportation. Computer and Chemical Engineering, 75, 74-94.

[5] Oddsdottir, T., Grunow, M., & Akkerman, R., (2013) Procurement planning in oil refining industries considering blending operations. Computer and Chemical Engineering, 58, 1-13.

[6] Zhang, J., Wen, Y., & Xu, Q., (2012) Simultaneous optimization of crude oil blending and purchase planning with delivery uncertainty consideration. Industrial & Engineering Chemistry Research, 51, 8453-8464.