Five different control strategies for controlling a complex nonlinear and time-varying fermentation process are compared. The main objective of the paper is to determine conditions under which neural-network-based controllers may prove superior to conventional linear and nonlinear adaptive controllers. Extensive computer simulations were performed under identical conditions using the five methods and were evaluated using the same set of criteria. Neural networks are found to be superior when adequate prior information concerning the dynamics of the process is not available and accuracy and robustness are critical issues.