A wholesale level consumable item inventory model for non-stationary demand patterns

Abstract

The U.S. military presently manages about 88 billion dollars in spare and repair parts, consumables, and other support items. Department of Defense (DOD) inventory models which help wholesale item managers make inventory decisions concerning these items are based on the assumption that mean demand remains constant over time. In DOD this assumption is rarely met. During periods of declining demand, such as that associated with force reduction or equipment retirement, the inventory models usually keep stock levels too high, generating excess material. Recently, the amount of excess in DOD was estimated to be as high as 40 billion dollars. On the other extreme, during periods of increasing demand, the models generally provide too little stock, resulting in poor weapons system support. The purpose of this research was to develop an inventory model which does not rely on the assumption that mean demand is stationary. Use of the model would be appropriate when a known or predictable increase or decrease in mean demand is forecasted. Through simulation the model's performance was evaluated and compared with that of the Navy's Uniform Inventory Control Program (UICP) model. The results indicate that the proposed model significantly outperforms the existing model when mean demand is non- stationary. Additionally, the results indicate that the proposed model's performance is equal to or better than the existing Navy model under many stationary mean demand scenarios.