Based on four levels of full factorial design, a statistical design of experiments is used to investigate two cases of dibenzothiophene biodegradation in batch processes using Bacillus sphaericus HN1, involving as factors, yeast extract and dimethylsulfoxide or magnesium sulfate for first and second cases, respectively. Predictive models have been correlated finding out how significant the effects of these variables (factors) and their interactions are in practice. Also, response surface methodology has been applied to visualize the effect of the studied factors and LINGO software was used to find out the optimum values of the variables for enhancing the process.