Forecasting and Inventory Management of Short Life-Cycle Products

Abstract

In this paper, we provide an integrated framework for forecasting and inventory management of short life-cycle products. The literature on forecasting and inventory management does not adequately address issues relating to short life-cycle products. We first propose a growth model that can be used to obtain accurate monthly forecasts for the entire life cycle of the product. The model avoids limiting data requirements of traditional methods. Instead, it extracts relevant information from past product histories and utilizes the information on total life-cycle sales and the peak sales timing. Using disguised demand data from a personal computer (PC) manufacturer, we validate the model. Next, we model the inventory management problem for the short life-cycle environment. The uncertainty in demand is modeled through the uncertainty in the realized values of the parameters of the forecasting model. The high cost of terminal inventory, shortages, and rapidly changing procurement costs are all included in the model. Extensions to the basic model are also developed. Using optimal control theory, we derive a solution that provides valuable information on procurement cutoff time and terminal service levels. A detailed example explains the characteristics of the policy and its relevance in decision making. Many of the issues covered in the models were brought to our attention while implementing a forecasting model at a PC manufacturer. The benchmark monthly forecasts and the associated inventory levels provide information that can be very helpful in planning and controlling marketing, sales, and production.