Rotational molding is a method for manufacturing hollow plastic parts. In the work reported here, adaptive fuzzy logic techniques have been used to relate the machine oven temperature to other manipulated parameters of the process. The objective is to design a reliable control system for the rotational molding process. An adaptive fuzzy network was developed to correlate changes in oven temperature to changes in the opening of the control valve on the fuel system. The network parameters were optimized using real-valued genetic algorithms. This network gave good results when its performance was compared with experimental data from a commercial rotational molding machine. The network was successfully utilized to design a control system, which works well in regard to set point tracking and load rejection. (C) 2005 Wiley Periodicals, Inc.