Analysis and Improvement of Stock Inventory Management at Northrop Grumman Newport News (NGNN) – VT08-NG

Economic and Technical Feasibilities of Implementing Robotics in an Automotive Repair Environment – UA06-RRAD
June 15, 2015
Experimentation Test-Bed for Evaluation and Benchmarking of RFID Technologies – CDP-05 OSU Collaborative
June 15, 2015

Analysis and Improvement of Stock Inventory Management at Northrop Grumman Newport News (NGNN) – VT08-NG

This project analyzes a stock inventory system of a process and makes further recommendations for improvement.

Sponsor:

Northrop Grumman

Research Team:

Douglas Bish

Universities Involved:

Virginia Tech

Start Date:

01/01/07

End Date:

12/31/07

Summary:

The objective of this project is to evaluate The Company stock inventory system, along with current inventory practices and procedures, at the Company. The Company requested an assessment of their current inventory system and, if appropriate, recommendations on improvements to the system.
The purpose of this collaborative project between Northrop Grumman and Virginia Tech is to evaluate current inventory policies and metrics and to recommend changes that improve the management of stock inventory. We will examine two sets of policies, classification/allocation policies and inventory policies. Classification policies determine whether an inventory item is classified as project or stock inventory, while allocation policies govern when, and how much, stock inventory is sold to a project. Inventory policies define when, and how much, inventory is to be ordered for stock from suppliers. These inventory policies directly drive the performance measures, and determine how well Northrop Grumman balances the costs of inventory against the costs of insufficient inventory.
The inventory problems faced by the Company are quite difficult; forecasting is problematic due to high variability, there exists a fairly high probability of obsolescence for certain parts, and certain parts have long lead-times. Furthermore, each part number is managed by a controller, and thus there is a fair amount of human intervention, which implies that there are many potential strategies in use. This can make analysis and change difficult. It is recommended that only ONE ordering methodology instead of the existing three.