The service life of a machined part depends greatly on the quality of its surface roughness value. Depending on the requirements of the intended service life, machined parts may be smooth or rough. Therefore, the quality of surface finish of machined parts must be closely monitored by technicians, artisans and operators in the shop floor in order to achieve optimal output from machining operations. The quality of surface finish obtained through machining operations are affected by a number of factors. However, the most important factors are the cutting speed, feed rate and depth of cut. These are the parameters that directly affect the characteristics of the surface roughness value of a machined component. In this study, an investigation was carried out on the effect of machining parameters on the surface roughness of 0.3% medium carbon steel using the Lathe. All other factors inherent in the cutting operations was held as constant whilst the cutting parameters were intermittently varied. For each combination of cutting speed, feed rate and depth of cut, the corresponding value for surface roughness was measured and recorded. The methods of analyses chosen for this study were artificial neural network, response surface methodology and the factorial design. Findings from the study show that the result from artificial neural network and the factorial design method exhibited similar patterns with the results of previous studies which recorded that feed rate was the most significant parameter affecting surface roughness. Also, findings revealed that increasing feed rate, cutting speed and depth of cut would result in a corresponding increase in the surface roughness values. Results obtained through response surface methodology generally did not agree with most of the findings of previous works in literature. Based on the predictive capabilities of the three models, statistical metrics indicated that the ANN model was found to be the best model with a coefficient of determination of 0.9979 and a mean square error of 0.003017. This was followed by the factorial design model with a coefficient of determination of 0.9298 and a mean square error of 0.0980. The model from the response surface methodology came last with a coefficient of determination of 0.9600 and a MSE of 0.1008. From the findings, it is strongly recommended that values of feed rate must be systematically selected during machining operations and the models developed could be used to predict the surface roughness of 0.3% medium carbon steel by technicians and artisans on the shop floor.