Statistical Analysis of the Variability Experienced in DLA administrative and Production Lead Time

摘要

This study examined the variability experienced in the administrative and production lead times of consumable items managed by the Defense Logistics Agency (DLA). Assets were categorized based on an item manager's determination that an asset was a problem part, and whether the asset was purchased for stock replenishment or direct vendor delivery. The methodology was a two-sample t-test of aggregated data to determine if significant differences existed between the mean deviations from the system lead times of various combinations of categories. The study produced results indicating that a significant difference did not exist between the mean deviations of problem parts and non-problem parts, but substantial variability did exist for all categories of consumable items. The degree of variability was such that the author suspected this as a possible cause for individual assets entering problem part status. The study also confirmed that while the DLA methodology used for forecasting lead times of stock replenishment purchases was accurate over aggregated data, the variability present is, to some extent, contributing to excess inventory levels.