The Defense Logistics Agency is using a number of initiatives to responsibly reduce inventory stored in depots across the globe, reducing the agency’s footprint and expenses while ensuring military readiness isn’t degraded.
Rightsizing inventory is one of the “Big Ideas” launched by DLA Director Navy Vice Adm. Mark Harnitchek in early 2012. Officials expect this to save significant amounts of money, specifically in storage dollars, and give the agency more flexibility in its operations. Disposing of excess stock, shrinking contingency stock, improving forecasting and planning, and reducing infrastructure that houses inventory will all contribute to overall savings.
Emptying the Attic
As anyone who watches the television show “Hoarders” knows, the first order of business in cleaning out a space is getting rid of excess material, particularly that which hasn’t been used in a long time. In DLA’s case, the excess material doesn’t equate to junk, but simply to material that isn’t needed very often or that is beyond the levels demanded by military customers. The short-term goal is to reduce DLA’s inventory by $4 billion by the end of fiscal 2014 while maintaining readiness for military customers, said Barry Christensen, chief of the DLA Logistics Operations’ Supply and Demand Planning Branch.
As DLA moves to decrease its inventory, planners in DLA Logistics Operations are looking at three categories of material: potential reutilization stock, contingency retention stock and economic retention stock, Christensen said.
Potential reutilization stock is an obvious target for disposals, Christensen said, because it is goods that the agency has already identified as excess. Potential reutilization stock is identified through monthly inventory reviews in which the agency looks at current operating requirements and retention limits for the items it manages, he said. Items identified for disposal are sent to DLA Disposition Services for reuse by military or other government customers, or, ultimately, public auction or destruction after having their military capabilities removed.
While Christensen called the disposal of potential reutilization stock a “no-brainer,” he said updating the levels of contingency retention stock DLA keeps on hand for the military services is a bit more complicated. This stock is made up of items that the military services have asked the agency to store because of their specialized nature or the strong possibility they will be needed for contingency operations. The services had some preliminary reservations about reducing these levels, but the agency saw a need to reduce the almost $2 billion worth of contingency retention stock it maintained. Christensen and his team worked closely with counterparts in the military services to reduce those levels by nearly 80 percent.
“What we’re doing is challenging the blanket philosophy that said, ‘Keep everything,’” Christensen said, noting that DLA and the services have been working on a long-term inventory improvement plan with the Office of the Secretary of Defense.
“Everybody wants sound readiness; that’s what we’re here for,” he said. “But I think we’ve made good efforts in bringing them visibility of the material that hasn’t been moving and the depth of stock that we store.”
The last category of items the agency is looking to reduce is economic retention stock, which is stock kept because of the economic viability of storing it versus the potential cost to re-purchase it in the future, Christensen said. DLA has updated its economic retention model to better reflect the costs of storage, including maintenance on storage facilities and the long-term cost of keeping infrastructure running, he said. This new model has identified higher storage costs for some items, moving them out of the economic retention category into contingency retention or potential reutilization, where they are considered for disposal.
Throughout all these efforts, DLA is working from a comprehensive inventory management plan from the Defense Department, which the military services are also following, said Joe Rodgers, an operations research analyst in the DLA Logistics Operations Supply and Demand Planning Branch.
Rodgers meets weekly with representatives from DLA’s inventory control points, DLA Distribution, DLA Disposition Services, DLA Operations Research and Resource Analysis, and DLA Information Operations to coordinate inventory reduction efforts and talk about plans and challenges.
“It’s very helpful in terms of planning and identifying any obstacles against that plan and working together as a team to try and work solutions around those obstacles,” Rodgers said. “The thing I see most is the sharing of knowledge and opportunities across the board from one supply chain to the other. If they were working in a vacuum, they wouldn’t have those opportunities.”
Having reduced its inventory by $2.2 billion since February 2012, significant progress toward its goal, DLA is a leader in the government when it comes to inventory reduction, Christensen said. Since DLA takes up about half the covered storage-space in government-run depots, the reduction of inventory should lead to a smaller footprint, as long as changes are made to ensure the space doesn’t get filled up again, he said.
Keeping the Attic Clean
Even the most diligent cleaning efforts are for naught if a room is immediately dirtied or filled, so DLA is accompanying its inventory reduction efforts with major changes in the way it plans for, develops requirements for and contracts items to better manage the overall Defense Department financial risk on items that are difficult to forecast. These efforts include using improved models that better predict what items will be demanded and for what quantities, reducing lead times by using more long-term contracts, and improving flexibility in order quantities and changes.
Just like forecasting the weather, forecasting the amount of items customers will need is not an exact science, but for years DLA has used a system that looked at order and demand history to predict future needs. The problem with that system was that it was a one-size-fits-all solution, and many items couldn’t be accurately forecasted within it, Christensen said. Items that see sporadic demand make up a significant portion of DLA’s catalogue, but because these items were statistically non-forecastable the traditional methods led to the wrong quantities being purchased and subsequently stored, increasing DLA’s storage requirements, he said.
A simple method to improve forecasting is to work more closely with customers to identify their needs and future trends. DLA has been improving customer collaboration for several years, Christensen said, but the system still needed a solution for the items that had sporadic high demand. That solution has come in the form of two new requirements models, Peak and NextGen, which were developed under DLA’s Logistics Research and Development Program. They are based on advanced algorithms and detailed modeling and aim to better set inventory levels for items that have historically been difficult to forecast.
“The items that typically cause trouble are those items that have infrequent demand or items that have frequent, but erratic demand,” said Bob Carroll, chief of the DLA Logistics Operations’ Planning Division. “So what do we do up front to make sure we don’t have to dispose of stuff? We do a better job of picking the winners and losers. We’ll still have some losers, given that we’re responsible for managing millions of items, but we do a better job of mitigating both inventory and customer support risk.”
Peak and NextGen are two very different, complex models, but their goal is the same: to set an appropriate inventory level for items that do not fit into the traditional forecasting model. Peak’s basic process is to find the peak levels of demand for each item over a 20-quarter period and then group the items into cost “buckets” based on those levels and their reorder quantities, Carroll said. The system assigns each bucket a multiplication factor of greater than or less than one and then runs through every possible combination to determine the best trade-off between inventory investment, material availability and workload for acquisition personnel, he said.
NextGen is a much more complicated model that uses a probability of how much stock of a given item will be on hand based on its order frequency and demand quantity, he said. The demand probability is then applied over the item’s lead time to determine availability, workload and investment. The NextGen process then calculates the metrics for reorders. This again allows planners to determine the best order quantity based on the balance of availability, investment and workload, Carroll said.
No forecasting model or requirements determination model will ever be perfect, Carroll said. And often with items ordered infrequently, DLA is faced with two poor choices: buying more than currently needed to get a lower per-unit cost but risking that the item might be stored for a long period of time if demand wanes, or buying a small quantity required for an immediate demand at a higher price and risking having to purchase more if demand spikes. After about six months of use, Peak and NextGen are performing as they were expected to and should start to deliver tangible benefits within the year, he said.
“Whatever we do in supply, we’re always a lead time away,” he said, referring to the time it takes to put an item under contract and then have the vendor produce and ship it. “You will start seeing benefits along the way, but we should see 70-80 percent of the benefit at 12 months and at 18 months, we should see the majority of the benefit.”
Those lead times, referred to as administrative and production lead times, represent another area in which DLA is working to ensure excess material doesn’t accumulate. DLA planners have traditionally used a forecasting technique to predict the required lead times for each item, Carroll said, but the same formula was applied to all items, leading to problems with accuracy. In the short term, the agency is refining the current forecasting model to improve lead times, but the ultimate goal is to apply the lessons learned by Peak and NextGen to lead-time forecasting, he said.
A way to combat the problems with lead times is to use more long-term contracts, avoiding the need for frequent contract workloads, Christensen said. Long-term contracts usually have a much shorter administrative lead time, he said, and vendors have a better idea of future needs, leading to better planning. DLA is also looking to improve its agility and flexibility in ordering by improving coverage durations, or the time periods covered by each order, and improving its ability to change or cancel contracts when needed, Christensen said.
Closing up Shop
Where there are fewer things to store, less storage space is needed, so as DLA rightsizes its inventory, it is also looking for ways to reduce the infrastructure used to store its stock. The push to reduce infrastructure builds on existing initiatives, like DLA Distribution’s inventory optimization efforts, to ensure the space being used fits the requirements and is reduced wherever possible.
Reducing warehouse space is a joint effort between DLA Logistics Operations, which develops the overall strategy and helps identify items for disposal; DLA Distribution, which handles the transfer of items to DLA Disposition Services; and DLA Installation Support, which works on vacating and possibly tearing down warehouses as they are emptied, said Lynne Allen, former program manager for infrastructure reduction in DLA Logistics Operations Strategic Programs and Initiatives.
“What we are looking at is working with the customer as we reduce our inventory and move it out of our storage buildings, then we are consolidating what is left to reduce our footprint. We will work with the services then to reuse or demolish the buildings so that we get infrastructure reduction for all of DoD,” Allen said.
This effort goes hand-in-hand with Strategic Network Optimization, a DLA-led program that aims to make better use of the Defense Department’s distribution network through reconfiguring the supply network, reducing inventory and eliminating excess warehouse capacity.
“What you should end up with are, for DoD, installations that are compact,” she said. “You’re not going to keep getting rid of buildings forever, but you’re going to reduce your infrastructure such that your installation becomes sized to what the true requirement is to support its tenants. Then it turns into a sustainable process that’s just reviewing your requirements and reviewing your real estate and seeing what you need.”