Effective Crude Blending for Improving Naphtha Productivity Using Linear Programming Modelling

Abstract

Petroleum refining is a complex industry that involves a wide spectrum of activities starting from crude purchase, transportation to refineries, refining operations, product transportation and delivering the product to end users. Its nature encounters complex economies that are heavily linked, that is, the selection of the right crude is not linked only to transportation costs, else related to considerations of refinery configuration, capabilities and constraints in converting the crude into products, as well as the product volumes and prices fluctuation modes. Solution packages based on Linear Programming (LP) technique have evolved through years among various mathematical optimization techniques, throughout the process routing from crude evaluation, selection, scheduling and product logistics planning. The objective of this study is to develop a mathematical programming model for solving a blending problem in a major refinery in Alexandria, Egypt with the objective of maximising Naphtha productivity. Refinery planning and optimization is basically addressed through special purpose linear programming software packages that remain a black box for the users and that are very cosdy for the organizations. The model developed in this work was proved to be highly effective at the level of solving the blending problem. This research has shown that the developed linear programming model for the blending problem has yielded better overall Naphtha productivity for the case of the oil refinery studied, as compared to results obtained by the commercial software.