"It's not like the automotive industry has all this inventory lying around," says Karen Peterson, vice president and research director for Gartner Inc. (Stamford, CT). In fact, adds Paul Hebeler, automotive industry director for Oracle Corp., from his office in Troy, Michigan, "The industry has inventory as tight as it can get without sacrificing cost and service." The "sweet spot" nowadays in automotive, he says, seems to be in chasing premium freight cost reductions. But there is room for improvement in automotive's management of inventory. First, automotive should do unto logistics management as it's done to production management. Second, automotive should "infuse the supply chain with more information—replace inventory with information," says Peterson. Third, automotive should do a better job collaborating with its supply chain partners. Last, automotive should implement new software.
Implement "lean" appropriately
Inventory results from at least one of two conditions, says Kevin Prouty, research director at AMR Research (Boston, MA). "Ignorance equals inventory. Or variability equals inventory." (Even the leanest of automotive manufacturing operations, namely Toyota, keeps inventory, he adds.)
In the first condition, if you don't know what the guy upstream or downstream is going to do, you keep inventory just in case. (Optimizing asset utilization is another reason, but this article will skip over that.) In the second condition, you can have all the inventory in the world, yet you still might not have enough for optimized manufacturing because of the variability driven by demand spikes (namely customer orders), by the manufacturing processes themselves (such as the effect of automotive options on assembly schedules), by logistical upsets (think September 11th), or a combination thereof.
When companies try to circumvent Prouty's Little Law of The Conservation of Inventory, somebody gets caught holding the, uh, inventory. "If you push inventory away from one partner operation, it'll move out in either direction—either to finished inventory or down to the suppliers." That's fine in the short term for reducing localized costs; namely, for the partner pushing inventory somewhere else. But in terms of lifecycle costs, the carrying costs add up and too many instabilities creep into the supply chain.
The best bet is to "lean" out business operations. North American automotive companies have done this in manufacturing, says Prouty, but they haven't leaned their supply chain processes. "And I don't mean ‘lean' to the point where you don't have inventory; I mean lean in the way you operate the company." For instance, instead of fighting change and trying to create a steady-state supply chain, come up with ways to better respond to change faster and more efficiently—even if that means building up inventory.
Implementing supply chain collaboration, says Jim Kowalski, group vice president of automotive for Manugistics Group, Inc. (Rockville, MD), is a "great opportunity for taking some of the empty costs out of the supply chain." By the way, Kowalski defines "collaboration" as "real, two-way communication, where there are discussions about what should be done." For instance, Kowalski suggests joint capacity planning, where supply chain partners share in calculating capacities across multiple suppliers and tiers. Helping such collaboration is, for example, the Supplier Network Collaboration that Manugistics helped implement at DaimlerChrysler. Rather than transmit a series of ever-truncated electronic data interchange (EDI) commands from one tier to another over what used to take 14 days, DaimlerChrysler broadcasts demand and production information down through all the tiers of its supply chain simultaneously using email. All partners in the supply chain get notification within 24 hours—and can reply in kind, and in time to offset any anomalies coming down the supply chain.
Oracle offers similar capabilities. Suppliers can surf over to the OEM's Web portal to see data at the granular level, including customer demand, production data (line, date, and time), and billing. "There's no software to load," explains Frank Prestipino, vice president of SCM and worldwide marketing for Oracle Corp. (Redwood Shores, CA).
What are these companies doing? Answers Oracle's Hebeler, "Upgrading their technology to be more collaborative."
Better decision support
"Pure visibility by itself is real-time access to bad data," points out Prouty. Advanced planning and scheduling (APS) tools can filter those data to improve inventory visibility. In the past, optimization and constraint-based planning, says Hebeler, focused on machine capacity and production bottlenecks. Now these same tools are being applied to inbound logistics.
Deere & Co. (Moline, IL), for example, recently implemented a transportation command-and-control center using Manugistics' supply chain applications. This center helps Deere optimize its logistics scheduling and maximize transportation loads based on real-world data, such as truck trailer size, transportation rates, and day of the week. As a result, Deere has reduced logistics costs approximately 10%.
By including event management and workflow, users can set up thresholds and other parameters for a wide variety of variables associated with forecasts, customer demand, production, material movement, and logistics management. These tools watch the incoming data and then send alerts (email and even voice mail) when appropriate. For instance, if you're a carrier, the software can tell you when new loads are tendered. If you're a supplier, the software can alert you to major changes in the forecast so you can adjust your production plan accordingly.
Enterprise resource planning (ERP), says Prestipino of Oracle, "was supposed to deliver this wonderful thing called ‘ATP'—available-to-promise. Up to now, that was impossible to do. At best, ERP could inspect your warehouse. If the inventory requirement wasn't in your warehouse, then ERP would backorder the requirement. Full stop." Now enterprise systems comprising ERP, APS, and warehouse and logistics management can "peer" into your warehouses and your suppliers' warehouses around the country, even the world. Moreover, they can analyze manufacturing schedules and actually look inside the supply chain. From this, they can respond with the particular day inventory will be available—even if the physical inventory is nowhere to be seen.
The lesson here? Evaluate—and implement—the integrated enterprise applications coming onto the market today. Or, at the very least, implement individual software applications that follow those standards that make seamless software integration a reality.
It's a fact of life, explains Karen Peterson of Gartner: Projected demand in the automotive industry can change 400% up or down—daily. That has to be smoothed out—from the very beginning. One way is to limit the proliferation of options. Asks Peterson, "I mean, who's going to want a pink car with yellow seats?"
Mitsubishi America, for example, rationalized the number of options it was providing. Doing that improved the OEM's ability to forecast, as well as get more accurate forecasts. Likewise, continues Peterson, better market analysis is needed to gain a better idea of what customers actually want. "If the dealer has only yellow pickups and that's what people buy, that doesn't necessarily mean that's what people want."
Another approach to inventory management is to apply service-parts technology to the inbound supply side, namely, "vendor managed inventory" (VMI). For proof of concept, look at Dell Computer. Dell has virtually no inventory other than the finished product shipped to customers. All the inbound supply-side inventory is maintained by Dell's vendors. Granted, this approach makes Dell's no-inventory claim somewhat disingenuous—the suppliers are carrying the inventory in plants and warehouses camped around Dell's assembly plant in Austin. However, somewhere a happy medium exists between Dell's VMI and Japanese-style Keiretsus for North American automotive industry to further investigate.
Tighten partner relationships
"I'd love to say it's all technology and technology is going to win over everything else, but it's not," says Peterson. This is not necessarily an inventory or technology issue, she continues, but OEMs and suppliers must create win-win, or partnership, relationships in sharing information. Doing that requires another "implementation": Trust.
"A vast difference exists between collaboration and dictation," explains Kowalski of Manugistics. Most automotive manufacturers, particularly the OEMs, tend to view their relationship with their supply chain as adversarial, command and control: "Here's our forecast, here's our production schedule, now meet it." Too often, meeting those demands is done at great expense, such as through premium freight.
How can automotive companies establish trust? "Not easily," says Peterson. One way is through better contracts that are less abrasive, better at quantifying the value for suppliers to implement changes, and more apt to actually share the benefits.
Kowalski has these suggestions. Initiate programs that embrace suppliers and wherein you work cooperatively with your suppliers. "The last time I checked, that business model is much more popular than the Lopez model," adds Kowalski. Next, let people act on the information they're collecting and supposedly collaborating on. Such "empowerment" goes beyond typical procurement versus supplier relationships. Last, realize that trust is built upon sharing information. In the past, suppliers tended not to say too much because they were afraid that what they said would be held against them "in the court of purchasing," says Kowalski. Conversely, OEMs didn't say too much because they wanted something to hold over their suppliers.
In the final analysis, three things become obvious. First, sharing data is a competitive advantage. Second, actions speak louder than words. Third, OEMs tend to drive the supply chain, so the onus is on them to initiate, facilitate, and then stand by this collaboration with their supply chain partners.
Get back to basics
North American companies too often try shortcuts. "They'll take a technology, say it looks like a ‘best practice process,' put it in place, and make their people work around it with little regard to the legacy and human resource issue around it," says AMR's Prouty. "Usually some piece progresses the company, but the entire implementation never lives up to the expectation."
So, start small, have a vision, get your own house in order. Look internally. Determine what's the variability that's driving you to maintain inventory. Is it because of internal business processes or external? Too many companies collaborate with their supply chains before figuring out how to collaborate internally. Last, don't spend a ton of money on a product development system. Instead, focus on getting the tools to make you faster and better.