Refineries are designed to process a specific type of crude, which has been possible with a constant supply of crude with the desired quality. However, the situation has changed over the last few decades as the sources are depleted with constant crude quality to necessitate crude blending to meet the specs. This software is a comprehensive optimizer for the crude blending recipe for batch (sequential) or inline (parallel) crude blending.

It is no secret that refineries produce 80-90% of crude oil by blending as fuel products. Fuel blending is a complex process, and any inefficacies and sub-optimal recipe leads to a loss of 25-40 M$/year, reduces the refinery’s low-margin bottom line, and needs an accurate estimation of tangible benefits. This software is a comprehensive Excel-based estimator of the fuel blending benefits using four methods, from simple to data-intensive estimation.

Fuel blending is a complex process, and refineries lose 25-40M$/year in a giveaway of octane due to inaccurate prediction of final blend quality. The Dupont coefficient is one method to estimate the interaction of blend components, but it requires a pervasive, expensive, and abrasive method. This software is a non-abrasive method to estimate offline and online non-linear regression estimation of customized Dupont coefficients using blending historical data.

Fuel blending is a complex process and requires the recipe optimization of 10-12 components using non-linear blend models and optimizers. However, some refineries still use manual blending in this age and resort to Excel to manually adjust the recipe to meet the final but not guaranteed final blend qualities. This software is an Excel-based method programmed for all non-linear blend models to give a fair estimate of the non-optimized recipe as a starting point.