Inventory: Getting more from less with inventory optimization
Early adopters of inventory optimization management systems are improving service levels while reducing the amount of inventory in the supply chain.
By Bob Trebilcock, Executive Editor -- Modern Materials Handling, 8/1/2009
Outside the four walls of the distribution center, inventory is the hot potato of the supply chain. No one wants to hold it for longer than they have to, and no one wants to own more of it than is absolutely necessary.
At the same time, many of today's manufacturing and distribution strategies mean that someone somewhere is going to end up holding that potato whether they like or not. Take lean manufacturing. Sure, the manufacturer pursuing a just-in-time strategy has turned all the storage space in the plant into more productive work cells. But that same manufacturer still counts on its suppliers to hold and deliver the raw materials, parts and components needed to run the lines. Big box retailers that have eliminated back rooms still expect suppliers to make smaller and more frequent deliveries to keep the shelves stocked. The inventory hasn't gone away; it is just being held by someone else.
The challenge for the supplier left tossing the hot potato from hand to hand is: Figure out just how much stock to keep and where to warehouse it to minimize a company's investment in inventory, and still meet the service levels demanded by customers. No one benefits from stock outs, manufacturing lines that grind to a halt, or an airplane sitting on the ground because a critical part isn't on the shelf.
The old fashioned way to avoid those outcomes was to load up on safety stock at each storage location in the network. The smarter way is to use a set of software tools known as multi-echelon inventory optimization to set and monitor inventory levels.
These applications take a holistic approach to inventory management. Instead of setting inventory levels one location, or echelon, at a time, a multi-echelon optimization solution uses sophisticated algorithms that can assess demand and inventory positions across a network, and then set inventory levels at multiple locations, or echelons, simultaneously. Better visibility allows a company to improve service levels while reducing the total amount of inventory in the supply chain. While this is still an emerging supply chain software area, these applications are delivering measurable value to early adopters, including paybacks of less than a year in some instances.
"Any company that's going to invest in technology in 2009 wants that investment to be cash flow positive in 2009," says Simon Ellis, practice director for supply chain strategies at Manufacturing Insights (508-988-7900, www.manufacturing-insights.com) and the co-author of "The Modern Supply Chain: Inventory Optimization Competitive Assessment," adding that inventory optimization fits that bill. "These tools are driving some pretty significant savings, and in the current environment, more companies should be taking a closer look," Ellis says.
New approach
The catalyst for the adoption of these new systems is simple: With traditional inventory planning systems, users often ended up with too much inventory at one location, and too little at another, says Ellis. A company with multiple warehouses, for instance, might place a replenishment order when it runs low on a specific item at one facility, even though another facility might have more than it needs of that same item. "When you factor in the little bit of excess inventory that ends up on the shelf because of a reorder or as safety stock, it really adds up," says Fred Lizza, CEO of Optiant (781-238-8855, www.optiant.com). "That's the inventory we want to eliminate."
Multi-echelon inventory optimization, on the other hand, looks at the supply chain in three dimensions—it's a little like playing three-dimensional chess compared to the traditional board game. Instead of optimizing each location separately, these applications can see the whole supply chain at once. "Instead of looking at your inventory in a silo, a multi-echelon system can look at all of the inventory drivers for the network: from the forecast, the amount of safety stock you might need to meet a customer's service level, your inventory positions, and everything you have on hand," says Rod Daugherty, senior director of product strategy for Manhattan Associates (770-955-7070, www.manh.com). "It sees those up and down the network."
What's more, like a supply chain network design tool, an optimization tool allows a user to perform what-if scenarios that incorporate variability and uncertainty into the planning process to come up with an optimal plan. "The system allows me to answer questions like: What's the impact of holding inventory in a semi-finished state versus a completely finished state?" says Lizza. "I can model the impact of pre-building some of inventory I might need to meet seasonal demand."
And, like an event management and visibility tool, an optimization application can be used to highlight exceptions to inventory levels or demand. "If everything is flowing according to the plan, and this week's safety stock target for an SKU is the same as last week's target, nothing hits the radar screen," says Lizza. "But if there is a spike in demand or something changes in production and a part is being consumed faster than was planned, the system will alert an inventory planner that the stock level needs to change."
At that point, the system can run a new plan and set new stock levels for a part.
Most importantly, because the system is tracking inventory across the supply chain, it can work with supply chain execution systems, like warehouse and transportation management systems, to transfer stock between facilities rather than order additional stock. "As you're picking in a distribution center, a warehouse management system (WMS) is keeping track of the inventory on hand," says Manhattan Associates' Daugherty. "At the end of the day, the replenishment system can compare the inventory position in a facility against the policy created by the optimization system."
When it's time to reorder, the system can look first to see if stock can be transferred between facilities. If so, an order can be sent to the WMS and transportation management system (TMS) in the facility with extra stock, just like any other order. It's all automated, so it can happen without someone intervening.
Managing spare parts
One of the first adopters of multi-echelon inventory optimization was the repair parts industry. It faces a unique set of issues: A machine or piece of equipment may have to be supported for decades, the demand for specific parts is unpredictable, and contracts may require delivery of some parts to a maintenance organization in as little as two hours. "Typically, what you find is that 80% of your parts will be used less than one time a year," says Tim Andreae, senior vice president of marketing for MCA Solutions (215-717-2180, www.mcasolutions.com). "Deciding whether you even need to stock one or two of those parts to meet a service level is an important decision—both from a service and a financial point of view."
A multi-echelon optimization tool allows maintenance organizations to balance financial considerations and service-level requirements, says Andreae, by factoring in the variability and unpredictability associated with service parts into the inventory planning equation. They do this by looking at everything from the design of the network to the parts required for scheduled maintenance programs to a multi-year history of unscheduled maintenance for every part. In addition, the system can be tied into a product lifecycle management system (PLM) to factor in any changes to the product or to be notified when a new part can be used to replace a prior part. "The system takes all of that information in to set stock levels for every part, and to determine where in the supply chain to locate those parts," says Andreae.
The payoff: By bringing both intelligence and discipline to the inventory planning process, service parts customers "have been able to reduce their inventory by anywhere from 15% to 40% while still providing the same level of service," says Andreae. What's more, by automating the process, supply chain planners are much more productive.
The next echelon
Now that some early adopters have mastered the basics of multi-echelon optimization, they are taking the systems to a new level of planning, according to Adeel Najmi, vice president of product strategy and planning for i2 Technologies (800-800-3288, www.i2.com).
Having optimized inventory levels, says Najmi, companies are now using the tools to answer questions like how to optimize the total net landed cost associated with a product. "Given the volatility in fuel costs and the globalization of the supply chain, the problems today are much more complex," he says. "The visionaries are determining whether they are better off using fast modes of transportation, like air cargo, to get frequent deliveries and carry less inventory, or using cheaper but slower transportation methods and carrying more inventory."
Increasingly, inventory optimization is working hand in hand with supply chain network design applications, and doing so on a more regular basis. "It used to be that a company would do a one-time design to optimize where the factories and DCs were going to be located and what strategies they were going to use to meet demand," Najmi says. "In the last two years, we have seen companies review their network designs and their optimization plans on a fairly regular basis."
Najmi calls this optimized performance. One company he works with has a network of small DCs close to its markets that are used to distribute promotional items. "They're changing how they utilize those facilities and what stock flows through them every week," Najmi says. "They're using these tools to get much more agile about their supply chain."
While most companies are still trying to go from a single- to multi-echelon view of their inventory, performance optimization is where the industry is heading. "Today, it's the visionaries that are going to the next level, but other companies will have to get there sooner or later in order to compete," Najmi says.
View All On-Demand Webcasts
