Selection of optimal parameters in gas‐assisted injection moulding using a neural network model and the Taguchi method

Describes how a statistical Taguchi approach and a backpropagation neural network model were devised to evaluate the effect of various parameters and identify the optimal parameter setup values in a gas‐assisted injection moulding process. In applying the Taguchi approach, an L 18 orthogonal array was employed to collect the observations, and the same collected data sets, with two additional inputs, were utilized to construct a neural network model to ascertain whether utilizing such a neural network would provide an improved generalization capability over a statistical method. The effect of the learning rate and the number of hidden nodes on the efficiency of the neural network learning algorithm was extensively studied to identify what provides the best forecasting of performance measure. In addition, to verify the generalization capability of the neural model, eight different parameter setups, which had not been included in the full factorial design, were constructed for network testing. The results revealed that the network is more efficient in identifying the real optimal parameter setup.