Supply Chain Management

Tuesday, February 4, 2014

How new computing technology helps manage infrequent and highly variable demand for military inventory?

As the largest logistics combat support agency in the Department of Defense, the Defense Logistics Agency is partnering with LMI, a private and not-for profit consulting corporation, to implement a more efficient system for managing military inventory. DLA manages nine supply chains and 5 million items, including clothing, subsistence, medical and spare parts. In Fiscal Year 2012, DLA had $53 billion in sales and revenue that could rank it 53^rd on the Fortune 500 list. Many DLA items have highly variable and infrequent demand. It is not unusual to see the variance of some of DLA’s items to be 10 to 100 times of its average demand, which is considered extremely large variable compared to commercial inventory systems. It is very challenging to manage DLA’s inventory.

In response to declining defense budgets, the Defense Logistics Agency is trying to cut the costs of storing items and procurements in their inventory. Traditionally, DLA used the same demand planning software as the other commercial inventory system to determine the time and size of purchased items that has variability far different from what the software can understand. Often time that leads to excessive inventory for some items but short of the others, excessive procurement workload, and depleted working capital, says Robert Carroll, a DLA logistics Operations employee. The adaptation of new inventory management technology, called Peak Policy and Next Generation Inventory Model (known collectively as PNG), enables DLA to skip the part of forecasting demand for many of their “unforecastable” items and employ risk-based hedging strategies to set levels directly from the empirical data of a particular items. The tool allows users to see a three-way tradeoffs among inventory investment, procurement requests, and customer wait time, depending upon their organization’s specific supply chain objectives. As a result, the costs of DLA’s inventory is significantly reduced with an estimation that its inventory will be driven down by $1 billion within three years, while improving customer service and decreasing procurement workload.

PNG is a new inventory management technology that allows us to manage inventory without adding too much forecasting, as forecasting is always a kind of gambling. There are always risks and errors incurred in each estimation. PNG is only made available due to new computing technology and complex algorithm models people developed nowadays. Compare to Dabawalla case study, DLA is a good case of applying technology on their supply chain system. It poses the question of what are the circumstances and characteristics for us to determine the use of advanced technology in the supply chain management?