This paper develops an artificial neural network (ANN) model to forecast the
optimum level of raw materials inventory as a function of product demand,
manufacturing lead-time, supplier reliability, material holding cost, and
material cost. The model selects a feed-forward back-propagation ANN with
twelve hidden neurons as the optimum network. We test the model with
pharmaceutical company data. The results show that the model can be useful to
forecast raw material inventory level in response to different parameters. We
also compare the model with fuzzy inference system (FIS) and simple economic
order quantity (EOQ). It can be seen that ANN model outperforms others.
Overall, the model can be applied for forecasting of raw materials inventory for
any manufacturing enterprise in a competitive business environment.