Supply Chain Planning Puts Garmin on Course to Improved Forecasting.
by Tony Kontzer, February 2011
For a company like Garmin, a manufacturer of global positioning system (GPS) devices, charting an accurate course isn’t just good sense—it’s at the core of the company’s business. But when it came to forecasting demand across the company’s supply chain, Garmin employees used to rely on technology that’s a far cry from the advanced navigation and communications devices the company manufactures for automotive, aviation, marine, and fitness use.
The company had financial and distribution operations located around the globe, and each had separate and independent data sets and no built-in integration between the systems housing that information. “It was a manual process to integrate data between the systems,” says Brian Pokorny, vice president of operations at Garmin. “The lack of good querying tools and a centralized database made analyzing the data and getting it to a state that was meaningful very time consuming. We couldn’t react quickly enough to make timely business decisions.”
But as Garmin executives prepared for further global expansion of the company, IT leadership made serious efforts to upgrade its manufacturing resource planning (MRP) systems to support planning and forecasting capabilities that desktop analytics could not provide. And Oracle software became the cornerstone of the system that would help Garmin planners navigate the complexities of the global marketplace.
Surveying the Territory
Until 2002 Garmin managers relied on a patchwork of mainframe-based MRP systems to get a view of supply and demand. But those systems—then located at manufacturing and distribution facilities in Taiwan and the U.K.; the Cayman Islands financial headquarters; and the company’s Olathe, Kansas, operating headquarters—weren’t integrated. So companywide forecasting was only possible by exporting the data into Microsoft Excel spreadsheets that served as a sort of makeshift unified database. Updated once a month and lacking adequate analytical capabilities, these documents required extra effort from staff to get any insight about inventory and shipping decisions—such as how much product was needed and where. This lag between the reporting process and marketplace realities made it impossible for Garmin managers to effectively exchange information with supply chain partners.
Simply put, it was not a tenable situation for a fast-growing company that was headed for an inevitable expansion into new global markets.
The consequences of lacking supply chain visibility were numerous. Accurate demand snapshots were a challenge to obtain, making it difficult to source the right amounts of materials and produce the right amount of product to meet demand. This hampered management’s ability to get the right quantity of devices to the global markets where they were needed. The lack of integration between the various MRP systems—which are typically designed to help manage all aspects of a manufacturing operation—also resulted in slow manufacturing schedule adjustments to changes in market forecasts, which in turn resulted in excess inventory being manufactured.
“The first thing we wanted to do was try to get rid of that spreadsheet as the database for forecasting,” says Pokorny. “It wasn’t fast enough to make changes and adjustments or to see real-time inventory positions versus demand. We were planning the business in monthly forecast cycles, with midcycle changes managed via e-mail.”
With the obvious need for a comprehensive, integrated solution, Garmin leaders deployed the Oracle Advanced Supply Chain Planning and Oracle Demand Planning modules of Oracle E-Business Suite in 2003. Almost immediately, the cumbersome Excel approach was phased out, with data from the various MRP systems being uploaded into the Oracle E-Business Suite applications. That meant forecasting data was now being stored in one database, with the forecast loaded into an Oracle Advanced Supply Chain Planning material plan to help managers better coordinate the output of manufacturing facilities.
But the lack of integration between Garmin’s multiple MRP systems continued to prevent that improved visibility from spanning the company’s global operations. The following year, Garmin completed the integration of disparate MRP systems, opening the door to truly global forecasting and demand planning. With Oracle Demand Planning being fed comprehensive data from Garmin’s manufacturing facilities via the newly integrated enterprise resource planning (ERP) environment, the company was able to start generating weekly forecasting cycles—even tweaking the system daily to reflect more-pressing market changes.
“We absolutely recognized the value of integrated information,” says Ed Link, Garmin’s vice president of IT.
The solution churned out quality demand forecasts for several years, which allowed Garmin to deliver—and build upon—a new responsiveness to supply chain planning. Improvements in data accuracy and integrity gave managers previously unattainable levels of forecasting precision, simultaneously reducing inventory and stock shortages. Managers could now pull inventory forecasting data from distributors around the world, working to anticipate what those distributors would buy and using that insight to inform sourcing and manufacturing operations.
A New Position
But the real test of Garmin’s new supply chain efficiency came in 2007, when Garmin experienced a period of what Pokorny calls “hypergrowth” due to the rapidly expanding consumer market for automotive GPS devices. To fully capitalize on this new market opportunity, Garmin acquired distributors in nine European countries. Those distributors had previously exclusively supplied GPS instrumentation for maritime applications—a distribution channel that Garmin management felt was necessary to own to effectively expand into the consumer market. These business developments dramatically increased the number of product SKUs that Garmin’s system had to manage, adding further complexity to an environment that was already being stretched to the limit.
In 2008 Garmin’s leadership selected Demantra, a demand management, sales, and operations planning technology that Oracle acquired in 2006, to further augment forecasting capabilities. Indeed, the Demantra solution delivered the forecasting, scalability, and stability features to contend with Garmin’s rapidly expanding global footprint. For the first time, Garmin’s demand planning staff—which relies on IT systems to make decisions about allocating product to achieve a healthy balance between supply and demand—had a holistic view of the end-to-end supply chain. Staff was now able to match forecasts with customer orders, making sure inventory was delivered in a timely fashion to meet that demand.
“It was obvious Demantra was the future,” says Pokorny. “So we jumped onto the roadmap to be on the leading edge of the technology and continue to mature our demand forecasting and supply/demand balancing processes.”
Garmin partnered with Oracle advanced planning and scheduling project management experts, working on a customized deployment of Demantra applications that went live in 2008, complete with tight integration with Garmin’s advanced supply chain planning solution.
Pokorny says applying the near-real-time capabilities of the Demantra applications to the company’s new European distributors allowed managers in the region to forecast what they anticipated selling—rather than simply viewing their inventory position. That was no easy feat given the complexity of Garmin’s distribution network, in which demand projections that start at local distribution centers trickle back through the supply chain—first to regional distribution centers serving several countries and finally to one of the company’s central distribution centers.
Improved Information Exchange
This shift to a demand-side focus also gave Garmin supply chain managers the ability to exchange information more effectively with supply chain partners—speeding response to partners’ updated forecasts and providing a huge advantage in predicting demand. Troy Johnson, Garmin’s manager of customer product forecasting, says the improved ability to incorporate forecasts, sales plans, and inventory levels from its customers is enabling Garmin to perform a higher degree of analysis in developing its internal forecasts. “It allows us to collaborate with our customers to be more in tune with their demands,” he says.
Jon Chorley, vice president of supply chain and sustainability product strategy at Oracle, is well aware of the positive impact the shift to a demand-centric strategy can have on a customer’s bottom line.
“The number-one issue most customers face is the need to get a handle on demand volatility. If your demand changes dramatically, you could either be caught short in terms of being able to supply the product, or you can have an oversupply and therefore have to discount or be stuck with excess inventory,” says Chorley. “Moving into a demand-driven approach to managing your business is really the best practice to drive costs out of your supply chain and to be as responsive to the market as possible.”
Case in point: since deploying the new Demantra applications, Garmin has reduced inventory by 76 percent at the European distribution sites it acquired, a staggering figure that reflects the company’s increasingly real-time forecasting capabilities and translates to lower inventory costs. Even more remarkable is the 50 percent reduction in stock-outs—which occur when inventory of a product is completely depleted—resulting in increased sales and growing customer satisfaction. Taken together, the two numbers provide compelling evidence of the improved accuracy and, more importantly, response time of Garmin’s demand forecasting efforts.
If that isn’t enough, Demantra has also been quickly—and enthusiastically—adopted by Garmin’s demand planning staff. Earlier versions of Oracle Demand Planning required staff to toggle between separate screens to access data from different systems.
With Demantra applications, employees can perform that task much more quickly because the system’s interface presents in a single view more of the data elements needed to review forecasts and make better decisions. Johnson says he isn’t certain if this is due to Demantra having superior abilities to pull in more types of data, or if it’s because of Garmin’s refinement of the business processes related to the system. Regardless, Demantra allows users to pull more custom data streams into the system than before, giving demand analysts more data to work with.
Vikash Goyal, director of strategy for value chain planning products at Oracle, says other Oracle customers can benefit from some key lessons of Garmin’s supply chain transformation experience.
In addition to improved forecasting capabilities and effective exchange of information with supply chain partners, Goyal says Garmin offers an attractive blueprint for IT departments looking to tackle complex supply chain planning and demand management upgrades with minimal outside help. That was the result of a specific strategic decision by Garmin to de-emphasize the role of third-party consultants (see “ Getting the Most from Your Consultants”).
“It’s important to invest in the right people so that long term we have the expertise in-house to continue to expand our footprint and evolve the business processes,” says Brandy Vandiver, director of ERP applications at Garmin.
What’s more, Oracle’s Goyal says, other customers would do well to mimic Garmin’s phased, crawl-walk-run approach, which increased the odds of a successful implementation by allowing IT and business users to cut their teeth on each new tool before being asked to learn something else.
“Easing into the change and seeing what works and what doesn’t helps with the adoptability and acceptance of new tools and processes,” Vandiver acknowledges.
A Roadmap for the Future
Today, Garmin leaders continue to work on refining the company’s entire supply chain. Over the next year, they plan to go live with the Demantra Real-Time Sales and Operations Planning module, as well as Oracle’s Agile Product Governance and Compliance solution (to go along with its 2009 deployment of Agile Product Collaboration). Additionally, as Garmin’s management looks to extend the success they’ve had with their Oracle systems support of demand planning, they are examining new ways to augment the Demantra solution to add new analytical capabilities.
While Garmin was able to dispense with its previous Excel-as-database approach, the company still pulls data from its distribution centers and puts it into either a Microsoft Access database or Excel spreadsheet and uses those tools to create and analyze data tables. Pokorny admits that it’s a closed process that lacks the real-time capabilities Garmin covets.
Addressing that shortcoming is high on Garmin’s to-do list, and the company’s management team is leaning toward either deploying Oracle Business Intelligence Enterprise Edition or constantly dumping the data in question into a Microsoft SQL Server database and letting users rely on the various tools in Microsoft Office 2010 to slice and dice the data as they see fit, says Vandiver.
If it sounds as if Oracle has a leg up when it comes to filling out Garmin’s supply chain needs, that’s because it does. And so long as Oracle keeps adding best-of-breed capabilities to its stack, whether through its own product development or via acquisition, Garmin’s not likely to change that approach.
“The reason we look to Oracle is we like to have the systems all working together and talking together and exchanging information so the information is powerful,” says Pokorny. “The hope is that we get better information flow as well as faster, better response time in the business.”