Modelling of five stations asynchronous production lines having single source of imbalance using artificial neural network

In this paper the performance of a five station asynchronous production line having a single source of imbalance i.e., the mean operation time of the different workstations (MT), has been modelled using artificial neural network (ANN). The production line has been analyzed to find out the effect of number of neurons in the hidden layer (n) over the training. The three performance parameters i.e., average error, root mean square error, and maximum absolute error; have been used. The key finding is that beyond a certain number of neurons in the hidden layer, the training is not satisfactory may be because of over fitting.