Extractive desulfurization (EDS) of thiophene (T), benzothiophene (BT) and dibenzothiophene (DBT) in simulated fuel using two phosphonium and two pyrrolidinium ionic liquids was investigated. A set of single-factor-at-a-time experiments was carried out to determine factors that significantly affect the EDS process. The single-factor-at-a-time experiments indicate that high sulfur removal (SR) can be achieved using long extraction time (>60 min), high temperature, low fuel-tosolvent ratio (1:4), or large number of extraction stages (N E = 5). However, the single-factor-at-a-time experiment does not take interaction between the factors into consideration and may fail in determining a suitable operating condition. Therefore, a response surface methodology (RSM) based on Box-Behnken design was employed to study and analyze the effects of time, temperature, fuel-to-solvent ratio and their interactions on the EDS process. The ionic liquid, tetrabutylphosphonium methanesulfonate [P 4444 ][MeSO 4 ] yielded the best result with %SR (DBT, BT, T) of 69%, 62%, 61% at the optimum time, temperature, fuel-to-solvent ratio and mixing rate of 15 min, 30 °C, 1:1 and 800 rpm, respectively. A similar performance was obtained during the single-factor-at-a-time experiments but at less favorable conditions indicating the superiority of the statistical approach used in this work in optimizing the liquid−liquid extraction performance.