Lumpy demand forecasting using neural networks

Summary form only given. Forecasting is the statement about the unknown results or outputs of events that will occur at a specific time in the future. To enhance the organizational competitive advantage in a constantly fluctuating environment, an organization´s management must make the correct decision on a timely basis depending on the information available. This study presents an application of neural network methods for forecasting three different items that exhibit lumpiness in their trend pattern. Using sixty periods of simulated data for each time series, forecasts were prepared for the next thirteen periods