Silicon carbide (SiC) was etched in a NF3/CH4 inductively coupled plasma. The etch process was modeled by using a neural network called generalized regression neural network (GRNN). For modeling, the process was characterized by a 2(4) full factorial experiment with one center point. To test model appropriateness, additional test data of 16 experiments were conducted. The GRNN prediction performance was optimized by means of a genetic algorithm (GA). Compared to a conventional GRNN model, the GA-GRNN model demonstrated a significant improvement of more than 85%. Predicted model behaviors were highly consistent with actual measurements. From the GA-optimized model, several plots were predicted to examine etch mechanisms. The model predicted that parameter effects are a complex function of plasma conditions. The etch rate was strongly correlated to the variations in the pressure-induced dc bias. This was also illustrated for the variations in the gas ratio. (C) 2004 American Vacuum Society.