- 11 Jan 2007
- Working Paper
A Perceptions Framework for Categorizing Inventory Policies in Single-stage Inventory Systems
Executive Summary — In research surrounding inventory policies, there is a prevailing assumption of completely rational agents. In practice, however, deviations from the optimal policy abound, and analytical models to understand the effects of inventory dynamics on practice may require ways to model these deviations. Modeling deviations from the optimal policy is also important for better understanding inventory systems and supply chains. The term "perceptions" in Watson's research is not meant in its conventional sense, as in the perceptions of individual managers, but rather forms the basis for a framework for modeling and categorizing a range of inventory policies, including optimal inventory policy. This paper, which is a technical article meant more for an academic audience, explores the usefulness of his framework for categorizing the range of inventory policies that can be employed in a single-stage supply chain. Key concepts include:
- This is a technical article intended for an academic audience.
- Current models track forecast errors and use them and past demand realizations to predict future demand. But the results may not accurately reflect the reality of the demand process.
- This proposed framework categorizes a range of inventory policies in a single-stage supply chain.
- The framework will be useful for more robust examinations of supply chain management dynamics.
In this paper we propose a perceptions framework for categorizing a range of inventory policies, including optimal inventory policies, that can be employed in a single-stage supply chain. The perceptions framework is based on forecasting with Auto-regressive Integrated Moving Average (ARIMA) time series models within the context of a single stage stochastic inventory system with periodic review, constant leadtimes, infinite supply, full backlogging, linear holding and penalty costs and no ordering costs. Forecasting ARIMA time series requires tracking forecast errors (interpolations) and using these forecast errors and past demand realizations to predict future demand (extrapolating). Categorizing deviations from optimal inventory policies is possible if we allow the perception about demand implied by the interpolations or extrapolations to be different from the actual demand process. We do not use perception in its more conventional sense; we are not making any claims about the actual perception of any manager. Rather the perceptions here serve as a device for modeling and categorizing a range of inventory policies.