Evaluating Forecast Performance in an Inventory Control System Author(s): Everette S. Gardner, Jr.
Source: Management Science, Vol. 36, No. 4 (Apr., 1990), pp. 490-499 Published by: INFORMS
Stable URL: http://www.jstor.org/stable/2632012 .
Accessed: 16/02/2013 09:31
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact email@example.com.
INFORMS is collaborating with JSTOR to digitize, preserve and extend access to Management Science.
This content downloaded on Sat, 16 Feb 2013 09:31:55 AM
All use subject to JSTOR Terms and Conditions
Vol. 36, No. 4, April 1990
Printed in U.S.A.
EVALUATING FORECAST PERFORMANCE IN AN
INVENTORY CONTROL SYSTEM*
EVERETTE S. GARDNER, JR.
College of Business Administration, University of Houston, Houston, Texas 77204-6282 This paper analyzes the impact of forecasting on inventory decisions in a large physical distribution system. Alternativeforecastingmodels are evaluated by developing tradeoffcurves between inventory investment and customer service. The results demonstrate that the choice of forecasting model is an important factor in determining the amount of investment needed to support any target level of customer service.
SERIES, APPLICATIONS; INVENTORY/PRODUCTIONPARAMETRIC ANALYSIS; SIMULATION-APPLICATIONS; MILITARY-LOGISTICS)
Forecasting is a prerequisite to inventory decisions in practice. Unfortunately most research in inventories ignores forecasting altogether and simply assumes that the distribution of demand and all its parameters are known. Only a few studies are available on the interactions between forecasting and inventory decisions. Lee and Adam (1986) show that the size of forecast errors influences the choice of lot-sizing rule in material requirementsplanning systems for manufacturinginventories. In distributioninventories, Croston (1972), Brown (1982), Watson (1987), and Eppen and Martin (1988) show that forecast errors can seriously distort projections of customer service. From the research to date, it is not clear how managers should evaluate alternative forecasting models in the inventory context. This paper is a study of the impact of forecasting on inventory control in a large physical distribution system. We show that alternative forecasting models define unique tradeoff curves between aggregate inventory investment and customer service. The differences between the tradeoff curves are significant. Careful selection of the forecasting model for an inventory system can increase the customer service provided by a fixed investment. Another possibility is to reduce investment while maintaining previous levels of customer service. The plan of this paper is as follows. ?2 introduces the concept of tradeoff curves for inventory analysis. This is followed in ?3 by a summary of the forecasting and inventory decision rules used in the physical distribution system. ?4 analyzes the characteristicsof the time series of inventory demands in order to identify alternative forecasting models. These models are reviewed in ?5, while ?6 develops the research design for testing the models. The savings available to management from improved forecasting are discussed in ?7. Finally, conclusions are offered in ?8. The results of this study are presently under implementation and should be useful in other inventory systems. The results also point to further research opportunities in forecasting for operational decisions. 2. Tradeoff Curves in...
Please join StudyMode to read the full document