A manufacturer can simplify the forecasting process by pooling demand forecasts for a product group and then disaggregating demand to the item level based on historical proportions. This approach leverages past data to predict future demand distribution, which is typically more reliable and objective than relying on marketing input, as historical data reflects actual observed demand patterns.