Abstract :
This paper evaluates alternative forecasting methods based on operational performance in a global specialty chemicals supply chain. The forecasting methods evaluated in this study are simple exponential smoothing, exponential smoothing with additive trend, and exponential smoothing with damped additive trend. These methods are evaluated at various target customer service levels based on the operational performance of the supply chain. The operational performance measures used in this study are total supply chain costs, broken down into inventory, production and shipment costs, and also four different measures of customer service level. The company’s supply chain operations are simulated and the results are analyzed using the Analysis of Variance (ANOVA) procedure followed by the Tukey’s means test. The effects of using the competing forecasting methods on each of the performance measures are evaluated and discussed individually. Our results underscore the importance of incorporating all relevant costs and customer service measures into the forecasting method selection process, which is only possible with a thorough understanding of the supply chain dynamics. In the environment we modeled, exponential smoothing with damped additive trend was by far the best in terms of both resulting supply chain costs and customer service levels, followed by simple exponential smoothing and exponential smoothing with additive trend.
NaturalLanguageKeyword :
supply chain , demand forecasting , simulation , exponential smoothing
