In this work, the use of artificial neural networks (ANNs) as an alternative tool for modelling and predicting the optimum conversion of the unsaturated fatty acid to epoxide in comparison with the response surface methodology (RSM) was developed. In the present investigation, waste soybean cooking oil (WCO) as biolubricant basestock was prepared via structural modification of unsaturated fatty acids (in situ epoxidation). Optimization of the effect of process parameters on maximum oxirane oxygen content (OOC) was studied using RSM. Interaction among the process parameters, such as C=C bonds to H2O2 molar ratio, catalyst loading, and reaction time was examined by ANOVA. The main focus of this study was to establish optimum OOC conditions using sulphuric acid (H2SO4) as a homogeneous acid catalyst. Optimum OOC of epoxidized waste soybean cooking oil (EWCO) was found to be 4.69 mass% under the experimental conditions of 60 degrees C temperature, 6h reaction time, 1.5g of catalyst loading, and 1:2 molar ratio of C=C bonds to H2O2. The resultant epoxide product was confirmed with the help of Fourier transform infrared spectroscopy (FTIR) (at 844.82cm(-1)) and nuclear magnetic resonance spectroscopy (NMR) (at 2.8 to 3.1ppm) analysis. Significant physicochemical properties of the prepared lubricant basestock were evaluated at optimum conditions using standard methods. Further, ANN modelling and genetic algorithm (GA) optimization were carried out by using an identical dataset. The results of the study revealed that the chemically modified WCO derivatives also can act as a potential biolubricant basestock.