화학공학소재연구정보센터
Industrial & Engineering Chemistry Research, Vol.47, No.21, 8282-8285, 2008
Optimal Solution-Range Analysis in Production Planning: Refinery Feedstock Selection
This paper presents a new way to expand and analyze the optimal solution of a process planning model. Feedstock selection in a refinery is a complex process, accomplished in many steps that typically involve numerous optimization executions that aim at providing a "value" for each potential crude oil feedstock to the refinery for a given operational period. Traditionally, the optimal crude oil slate as determined by the planning system is a point solution that does not provide detailed information regarding potential (finite) changes in the optimal crude oil selection without degradation of the overall plant economics., In this work we define the optimal selection range for each feedstock as the result of a Pareto-type analysis including all feedstocks. Two flexibility indices are defined for each crude oil feed providing a quantitative flexibility metric in the context of optimal economics. The proposed approach provides a useful and practical way to quantitatively evaluate each potential crude feed. This is expressed in terms of an optimal range as defined by the ability of a feedstock to replace and be replaced in the optimal Solution without significant economic change. In addition, this analysis provides a map of the optimal surface in terms of near-optimal points around the base solution.