Article ID: | iaor20114023 |
Volume: | 8 |
Issue: | 4 |
Start Page Number: | 403 |
End Page Number: | 424 |
Publication Date: | Apr 2011 |
Journal: | International Journal of Logistics Systems and Management |
Authors: | Rossetti Manuel D, Achlerkar Ashish V |
Keywords: | statistics: empirical, datamining, management |
This paper evaluates methodologies for the grouping of items and the setting of inventory policies in a large‐scale multi‐item inventory system. Conventional inventory segmentation techniques such as ABC analysis are often limited to using demand and cost when segmenting the inventory into groups for easier management. Two segmentation methodologies, (Multi‐Item Group Policies (MIGP) and Grouped Multi‐Item Individual Policies (GMIIP), that use statistical clustering were developed and compared to ABC analysis. An evaluation of these techniques via a set of experiments was performed. The analysis indicates that these techniques can improve inventory management for large‐scale systems when compared to ABC analysis.