Industrial & Engineering Chemistry Research, Vol.49, No.20, 9970-9977, 2010
Strategic Planning of Integrated Multirefinery Networks: A Robust Optimization Approach Based on the Degree of Conservatism
This paper considers the problem of strategic planning under uncertainty for optimal integration and coordination of a multirefinery network. The deterministic model proposed by Al-Qahtani and Elkamel [Comput. Chem. Eng. 2008, 32, 2189-202] was extended to account for uncertainties in raw material costs and final product prices, as well as product demand. The robust optimization methodology of Bertsimas and Sim [Op. Res. 2004, 52, 35-53] was applied, which deals with uncertainty in a tractable manner and does not add complexity to the deterministic problem. An industrial-scale study illustrated the benefits of the integrated planning and demonstrated that the modeling of uncertainty in process parameters provides a more practical perspective of the refining industry. In addition, probability bounds of constraint violation were calculated to help decision makers select appropriate parameters to control solution robustness and evaluate trade-offs between conservatism and total profit.