Wholesale Warehouse Inventory Control with Statistical Demand Information.
Abstract
Inventory managers often encounter erratic demand histories which are difficult to model. For example, periods of no demand are frequently observed, and when demand is positive, it tends to be quite large. Furthermore, periods of high demand are often followed by several periods of no demand. One possible explanation for this sporadic and correlated demand behavior is that demand originates from separate facilities which employ (s,S) replenishment policies. Each period's demand, therefore, is the sum of the replenishment order quantities received from inventory control facilities. The management at a multi-item inventory system at the warehouse level, in which demand is comprised at the aggregated replenishment orders from lower-echelon inventory control facilities, was investigated by Schultz 1979. Under the assumption that the warehouse observes only the aggregated replenishment orders, Schultz adapts an approximately optimal (s,S) policy rule (The Power Approximation of Ehrhardt 1976, originally designed for independent and identically distributed demands, for use in a warehouse demand environment. The demand requirements of this new policy rule, referred to as the Correlation-Adjusted Power Approximation, are the mean, variance, and variance over one lead time of demand. For the situation in which these demand parameters are known exactly, Schultz empirically demonstrates that the operating characteristics of the policy rule are close to the operating characteristics of simulation-derived estimates of optimal (s,S) policies for a wide range of parameter settings.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1980
- Accession Number
- ADA095304
Entities
People
- Carl R. Schultz
Organizations
- University of North Carolina at Chapel Hill