Regression Based Integration of Demand Forecasting and Inventory Decision

In the traditional inventory decision-making, demand forecasting and inventory control decisions are made independently, so it is difficult to ensure that the inventory cost is minimized. To make the demand forecasting and inventory control be consistent, this paper proposes a cost regression model, in which the demand forecasting is combined with the inventory decisions, aiming to minimize the inventory cost rather than the forecast error. A computational example is presented, to show the difference of the integrated decision-making model and the decentralized decision-making model.