The aim of this study was to investigate optimum conditions for biological removal of flue gas pretreatment wastewater and achieve maximum elemental sulfur yield. A three-factor, three-level Box–Behnken design was used to derive a second-order polynomial equation and construct contour plots to predict responses. The independent variables selected were hydraulic retention time (X1), inlet sulfate concentration (X2), and air flow (X3). Fifteen batches were done in a biological united system and evaluated for elemental sulfur yield (Y1). The transformed values of the independent variables and Y1 were subjected to a full-model second-order polynomial equation. The equation was modified based on Fisher's F- and probability P-values. The computer optimization process and contour plots predicted the values of independent variables X1, X2 and X3 (16 h, 1,348 mg L−1 and 165 L h−1 respectively), for maximized response of Y1. The experimental results at predicted conditions demonstrate that the modified model equation has good applicability to the practical system.