Abstract:
This thesis aims to improve the production planning of the Asbaâ Adrar refinery, a vital facility
for supplying petroleum products in southern and southwestern Algeria. Using a linear
programming model, the thesis seeks to maximize the refining margin while respecting
constraints related to material balance, quality (octane number, Reid vapor pressure), and
capacity (54,000 tons/month of crude oil). The model was implemented in MATLAB, resulting
in a 19.5% increase in monthly revenues and a 44.8% increase in net profit margin. The thesis
outlines the fundamentals of refining, presents the refinery (with a capacity of 600,000
tons/year, and units for distillation, reforming, and cracking), introduces linear programming,
and formulates the optimization model. Despite its limitations (lack of yield constraints, static
model), it offers a reusable methodology for other refineries and contributes to scientific
research by linking theory with Algeria's industrial challenges.
