The determination of optimal feed rate profiles for fed-batch bioreactors with more than one feed rates is a numerically difficult problem involving multiple singular control variables. A solution strategy based on genetic algorithm approach for the determination of optimal substrate feeding policies for fed-batch bioreactors with multi-control variables is proposed. The multiplier updating method is introduced in the proposed method to handle inequality constraints on state variables. The efficiency of the algorithm is demonstrated for two case studies on fed-batch bioreactors with two control variables taken from literature. The control policies obtained retain the characteristics of the profiles generated through rigorous application of control theory. (C) 2004 Elsevier Ltd. All rights reserved.