In any machining process, it is important to select the appropriate machining parameters to facilitate the better Material removal rate (MRR). The generation of prediction equations are of utmost importance in the optimization of the machining parameters. In this paper, an artificial neural network (ANN) assisted improved prediction model for the MRR of Glass fiber reinforced plastic composites (GFRP) turned components is built using Multiple regression analysis (MRA). Using the process parameters namely spindle speed, feed and depth of cut, the turning of the GFRP composites was carried out on a conventional lathe using single point HSS cutting tool. By employing the Taguchi's L16 array (3 Level), the experiments were conducted and the MRA was carried out for the prediction of Material removal rate (MRR). The MRA predicted values were found to be less accurate for the test data of MRR. In order to overcome this problem, an experimentally validated radial basis ANN was used to predict the MRR values for L343 array (7 level). Further using this L343 array, MRA was again conducted to develop the expression for MRR. This expression for MRR yielded a drastically improved result. The reason for this can be attributed to the use of a higher input level Taguchi's design, which was made possible by adopting the radial basis ANN.