Marine Bacillus circulans used in this study produced 2.58 g L-1 of biosurfactant when grown in unoptimized modified marine medium (MMM). The biosurfactant production was enhanced by optimizing the critical components of the MMM using statistical modeling based response surface methodology (RSM). From the results of single-factor-at-a-time optimization experiments; glucose, urea, SrCl2 and MgSO4 were identified as critical components that affected biosurfactant production significantly. In RSM optimization, experiments were done based on the 24 full factorial central composite design to generate responses, which were used to fit the quadratic polynomial model. Statistical analyses of the data of model fitting were done by using Design Expert version 7.0. The optimum values obtained for the test components for the maximum predicted biosurfactant concentration of 3.04 g L-1 were found to be (in g L-1): glucose = 35.5, urea = 1.23, SrCl2 = 0.11 and MgSO4=0.53. Validation experiments using the optimum values of the critical media components showed that the biosurfactant yield was enhanced to 3.01 g L-1, which was in close agreement with the model prediction. Thus, the biosurfactant production was increased by 17% in the optimized MMM, when compared to that in unoptimized medium.