Identifying Main Drivers on Inventory using Regression Analysis

Konferenzbeitrag

Inventory management is essential for satisfying customer demands and reducing logistics costs. Extensive material portfolios, balancing inventory costs versus customer service levels as well as fluctuating demand are factors that influence inventory levels. Inventory management is often considered as inventory reduction; quick reduction activities are carried out without knowing the exact root causes of nontarget inventory levels. A sustainable and comprehensive approach could be, to consider those factors which have a strong impact on inventory and to use this information for further inventory optimization activities. This paper therefore gives an answer to the question, how to systematically identify main drivers on inventory, using multiple linear regression analysis and how to quantify their impact considering company-specific data and structures. The described approach is applied in a case study at a company in the commercial vehicle industry. Data sets from different locations are analyzed and compared. It will be shown in a methodical way that few factors have a strong linear influence on inventory level and differ depending on the characteristics of the respective location. Companies can thereby analyze main drivers on inventories e.g. per location, region, sales channel or company-wide, depending on the chosen data set. The results can be used to identify root-causes for nontarget inventory levels and form the basis for company-specific inventory optimization activities.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: 10419/209191

Classification and Topics

• Classification: Management
• Subject: Inventory Management ; Regression Analysis ; Main Drivers ; Root Cause Analysis

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Hoppenheit, Steffi ; Günthner, Willibald A.
• Event: Veröffentlichung
• (who): epubli GmbH
• (where): Berlin
• (when): 2015

Further information

• Handle: http://hdl.handle.net/10419/209284
• Last update: 10.03.2025, 11:44 AM CET