Information inaccuracy in inventory systems

Abstract

It is critically important for inventory-carrying facilities to provide high availability of products at the minimal operating cost. To achieve this objective, many companies have automated their inventory operations and rely on the information system in critical decision makings. However, if the information is inaccurate, it may lead to high out-of-stocks and/or excess inventory. This thesis examines what the primary causes of the inaccuracy are, how and to what extent they degrade the inventory system performance, and what can be done to compensate for the inaccuracy. Analytical and simulation modelling demonstrate that the inventory system performance is highly sensitive to the inaccuracy caused by stock loss, which is the disappearance of items (such as due to theft) not detected by the information system. That is, even a small level of stock loss accumulated over time can lead to inventory inaccuracy that disrupts the replenishment process and creates severe out-of-stocks. In fact, revenue losses due to out-of-stocks can far outweigh the property losses due to the disappearing items.

(cont.) One way to deal with the inaccuracy problem is the use of RFID-based automatic product identification technology under development at the Auto-ID Center, which can provide the real-time and accurate information regarding the location and quantity of objects in supply chain. It is found that even when this technology provides imperfect measurement of the stock quantity, dramatic performance improvement can be achieved using an inventory control scheme based on dynamic programming. Various other methods of compensating for the inventory inaccuracy are presented and evaluated. Analysis of each method reveals that the inventory inaccuracy problem can be effectively treated even without automatic identification technology in some situations. However, each method has weaknesses.