Two sequential statistical experimental designs were used to screen and investigate the dependence of the amount of biodegradation of Procion Red MX-8B (PR-MX8B) on the fermentation variables. Fourteen factors were screened using the Plackett-Burman design. Among these factors, the most significant variables which included yeast extract, corn steep solids and starch influencing PR-MX8B decolourisation were statistically elucidated for optimization. The optimum concentrations of 5.00 g/l yeast extract, 2.99 g/l starch and 1.89 g/l corn steep solids were predicted by applying the Box-Behnken design to the second order polynomial model fitted to the results obtained. The best predicted optimal conditions verified experimentally yielded 72.11% while the predicted value from the polynomial model was 79.17%. The experimental values were in good agreement with the predicted values with a 90.81% degree of accuracy.