Applying a Bootstrap Approach for Setting Reorder Points in Military Supply Systems
This report develops and applies a nonparametric bootstrap methodology for setting inventory reorder points and a simple inequality for identifying existing reorder points that are unreasonably high. The authors demonstrate that an empirically based bootstrap method is both feasible and calculable for large inventories by applying it to the 1st Marine Expeditionary Force General Account, an inventory consisting of $20-30 million of stock for 10-20,000 different types of items. Further, the authors show that the bootstrap methodology works significantly better than the existing methodology based on mean days of supply. In fact, the authors demonstrate performance equivalent to the existing system with a reduced inventory at one-half to one-third the cost; conversely, the authors demonstrate significant improvement in fill rates and other inventory performance measures for an inventory of the same cost.