In this study, treatment of a local South African oil refinery effluent using a coagulation flotation process is designed using response surface methodology (RSM). A Box-Behnken design (BBD) implementing the RSM is applied to evaluate the effects and interactions of three operating parameters, viz., pH, coagulant dosage and flotation time, on the treatment of mineral oil wastewater (MOW). Polyacrylamide (Zetag-FS/A50), which is a water-soluble compound, is applied to enhance the adsorption mechanism and intermolecular bridging to minimise the amount of oil droplets. In addition, due to the monomeric nature and the charge density of the Zetag-FS/A50, its efficiency was evaluated to serve as an alternative coagulant for the pretreatment of the MOW. The removal of chemical oxidation demand (COD), soap oil and grease (SOG), total suspended solids (TSS) and turbidity from the MOW were used as the response variables for the coagulation flotation process. This was done with a standard dissolved air flotation jar test. The results show that the actual COD, SOG, TSS and turbidity percentage removal at optimised conditions with a coagulant dosage of 50 mg/L were 82%, 83%,70% and 83% respectively, while the predicted response was 92%, 96%, 73% and 87% for COD, SOG, TSS and turbidity, respectively. The analysis of variance (ANOVA) showed that the proposed models are significant at a 95% confidence level. A quadratic model was generated for response variables COD and SOG, while TSS and turbidity produced a linear model. The models fitted well with the experimental data with correlation coefficients (actual R2) of 0.94 for COD, 0.91 for SOG, 0.81 for TSS, and 0.75 for turbidity.The outcome of the study shows that the RSM has merit to optimise and identify the most important factor to control and the Zetag-FS/A50 coagulant has the potential to adsorb the oil droplets in order to enhance the treatment efficiency of the process.