A sales forecast model for short-life-cycle products: New releases at Blockbuster

摘要

Sales forecast model for short-life-cycle products is challenging because 1. They have highly compressed short active life, especially for entertainment media product, 2. No reliable benchmarks can be used to predict sales of specific products, 3. Under-forecasting or over forecasting could impact the profitability of the product or in poor product utilization and unnecessary stocking and holding costs. The research work has tried to develop a sales forecast model for movie and game products at Blockbuster, a rentable-DVD and game-media firm. The model formulated divides sales components into three categories and allows dynamic interactions between the sales components within each sales period as well as period-by-period updates of model parameters. The sales model is tested against extensive sales data and was validated using publicly available environmental data. In continuation, additions have been incorporated to adjust and improve the initial forecasts based on actual first-period sales, thereby allowing the firm to respond to any discrepancies in the initial order. (36 refs.)