| Blockbuster Inc. is the leader in rental DVD and game media services. Blockbuster supports consumer demand through a highly specialized supply-chain network. This network consists of 5,600 retail stores that are supported by a single retail store Distribution Center (DC), and an online rental operation supported by 38 regional DCs. The DVD and game products that Blockbuster supplies to consumers have a high stock-keeping unit (SKU) base, and are time-sensitive with a short life cycle. This research addresses problems that Blockbuster experiences in product demand forecasting, product allocation, and production scheduling. First, we present a sales forecast model for DVD and game products at Blockbuster. This model assumes that there are three sales components: (i) consumers who have already committed to purchasing (or renting) a product (e.g., based on promotion of, or exposure to the product prior to its launch), (ii) consumers who are potential buyers of the product, and (iii) consumers who are enticed to purchase a product due to an influencing effect from "closely-tied" (as in a social group) previous potential buyers (in the case of movie rentals and all retail products) or re-rents (in the case of game rentals). Further, we explicitly formulate into our model dynamic interactions between these sales components, both within and across sales periods. This important feature is motivated by realism, and significantly contributes to the accuracy of our model. Second, we present models developed for the online distribution supply chain that (i) forecast the demand of new DVD products for customer rentals, (ii) determine the initial order quantity to be purchased, and (iii) accurately position the DVD products among the 38 regional DCs in order to prevent costly misallocation. In our work, we consider movie-specific characteristics and consumer behavior. Lastly, we describe the creation and implementation of a Mixed Integer Programming model for the scheduling of short-range order processing operations at Blockbuster. This application, in daily use since January 2007, has resulted in annual cost savings of over... |
