thumbnail group

Connect With Us:

Manufacturing Engineering Media eNewsletters

ME Channels / Lean
Share this

Supply Chains: Tightening the Links


Your lean journey must include suppliers

Richard J. Schonberger
Schonberger & Associates Inc.
Seattle, WA

The hard part of lean management resides in no-man's land, the space between the supplier plant and the downstream final user. It is the last, worst haven for inventory and attendant stretched-out lead times and wastes. And, as has been said many times, no one is responsible for it. It's a hot potato, each entity wanting it off the books.

That last (the books) is a good part of the problem. Most companies are managed by "the numbers" — accounting data, that is. As always happens, numbers management inspires numbers gamesmanship, which rises to high intensity at the end of each week, month, quarter, and year. The game has two goals: 

  • Shoving not-yet-wanted—or not-wanted-at-all—finished goods onto customers.
  • Getting purchased materials off your balance sheet and onto that of suppliers. The common practice of automakers to require suppliers to hold five—or often 10—days of finished parts is a case in point.


In all regions except Japan, inventory-turnover scores have been getting worse year by year. Japan's overall score remains the lowest by far among the nine regions.

Playing these games is a sinkhole in the path of supply-chain management (SCM). It helps explain why industry is underperforming in its SCM endeavors. Hard research data show that, for most companies, finished goods have grown greatly as a percentage of total inventories, and purchased materials have grown as well. That lack of success in getting lean in the pipelines is despite the wide availability of potent mapping, tracking, and IT tools, and consultants skilled in their use.

Filling the sinkhole requires special measures. For one thing, manufacturers would do well to take some lessons from retailing, generator of innovations in continuous replenishment. Within manufacturing, there is a lot to learn from electronics, whose SCM accomplishments stem in large part from having massively restructured itself. In another re-engineering, the formerly chopped-up logistics sector has fused itself into different kinds of all-in-one providers. As for numbers management, a simple aid presents itself: a unique metric developed years ago at IBM. I'll attend to that metric; then discuss what retailing, electronics, and logistics are doing that cinches up their supply and customer chains. A final topic is how to lay the groundwork for successful SCM via internal, then external collaboration. But first, I want to review evidence that industry is, indeed, stumbling in supply-chain and overall lean management.

I've been studying inventory-turnover trends for years. Steady reduction of inventory is the most objective way to see "lean" achievements, in plants and in supply chains. With all data from audited financial statements, our database numbers more than 1200 companies in 32 countries. To rule out short-term flurries of success in clamping down on inventories and related wastes, we've tracked at least 15 years of inventory turnovers for each company. Until the late 1990s, the multiyear trends looked fairly good in most parts of the world. In the past four or five years it's the opposite: most companies have lost ground. The accompanying table summarizes our findings, with the companies grouped by global regions. The arrows show that, in all regions except Japan, inventory-turnover scores have been getting worse year by year, with more companies bulking up than reducing inventories. Japan's rising turnover scores are recent, after worsening though most of Japan's 1990s "lost decade" of economic malaise. And Japan's overall score remains the lowest by far among the nine regions.

Even Toyota, for most of us revered as a paragon of lean, has faltered seriously. Its inventory turnover (annual cost of goods sold divided by value of inventory) has been halved from 22.9 in the early 1990s to 11.1 in 2005.

Is work-in-process the culprit? Clearly not. Manufacturers have been aggressive in converting from batchand-queue to lean production. Meanwhile, pipelines are bulging. Why?

Perhaps it's risk protection: companies adding protective raw and finished inventory buffers in case of political strife, insurgencies, border delays, and natural disasters. Perhaps it's also globalization and outsourcing, which extend supply lines. Yet a few companies, Wal-Mart being one of the best examples, keep reducing their inventories year by year, while aggressively expanding globally. And the massively outsourced electronics industry beats most other sectors in inventory reduction. More on these matters later; first let's look at IBM's astute performance metric.

In 1985, Raleigh, NC was the site of one of IBM's first lean (then called just-in-time) production transformations. The far-flung batch processes for one major product line were reconstituted as a tightly linked assembly cell; JIT deliveries of purchased materials arrived through an adjacent dedicated dock. The application was advanced for its time, and even for today. A unique feature, conceived by the task force in charge, was how they would assess the cell's relationship with its JIT suppliers.

What they devised was a performance measure they called joint inventory—simply the inventory produced for IBM by each supplier, plus that already received at IBM. This metric would help dissuade managers from the nefarious practice of pushing inventory back on suppliers in order to look good. Joint inventory is a simple, effective way to deal with common supply-chain misdeeds.

That was 20 years ago. Most of the methodologies for getting lean in the supply chains are of recent origin. In large part, the methods have been honed in retailing—practices that are equally effective in manufacturing, but not quickly adopted.

In retailing, Wal-Mart is the world champion of lean. Its inventory turnover, 4.1 in 1990, had risen to 7.6 in 2005, an improvement rate of 3.1% per year over that 15-year time span. That betterment took place while Wal-Mart was branching out into 15 countries, greatly extending its supply pipelines. Wal-Mart's SCM practices have served as a model for other major retailers. A few years ago, Nordstrom saw its sales, profits, and stock price slipping. The company threw out its mess of an inventory management system, and installed a lean one more like Wal-Mart's. Now, Nordstrom is again the top-performing upscale department-store chain.

As a retailer, Wal-Mart has no work-in-process or finished-goods inventories. It is all purchased materials. Behind Wal-Mart's sharp trimming of those stocks is a continuing series of innovations that, for other retailers, become the next thing to learn.

A movement called quick response (QR) got its start in 1986 when textile manufacturer Milliken brought together key department stores and apparel houses to agree to share demand, production, and shipping data. While a manufacturer got QR going, retailer Wal-Mart soon took it over, placing new requirements on its suppliers, such as vendor-managed inventory. Under VMI, suppliers were obliged to manage Wal-Mart's inventory replenishment via electronic access to the retailer's point-of-sale data.

Next, Wal-Mart elected to phase out of dealing with middlemen, requiring that producers ship direct to Wal-Mart distribution centers or to stores. Instead of large-batch shipping to fill downstream warehouses, the new mode became continuous replenishment: shipments synchronized to daily barcode-scanned retail sales. For manufacturers, the system calls for extending the synchronization to their production schedules and purchasing. Upping the ante, Wal-Mart is the current world leader in upgrading to RFID, opening up more ways of tightening supply and synchronizing to real demand.

For its own distribution centers, Wal-Mart pioneered cross-docking, in which a truckload of an incoming item goes not into storage, but directly into multiple trucks in small lots for immediate transfer to retail stores. Wal-Mart reportedly delivers 85% of its merchandise using crossdocking. Topping all this off, more than 2000 of Wal-Mart's suppliers have seen fit to locate empowered staff to the mega-retailer's Bentonville, AR, home base, to collaborate on both strategic and tactical issues. The ring of 20 office parks housing these suppliers has acquired the nickname "Vendorville."

This kind of close-quarters, company-to-company collaboration has a precedent: Toyota from the late 1950s to the early 1970s, the period in which the majority of practices called the Toyota Production System were perfected. Toyota and most of its suppliers were tightly clustered in metropolitan Nagoya, Japan—nicknamed Toyota City. Toyota, though, has long since lost the comfort of having suppliers close underfoot. Its formula for rapid growth has entailed opening plants in countries around the world; its suppliers scatter similarly. A breakdown of inventory data from Toyota's annual reports show that its worsening inventory performance over the last 15 years has occurred mostly in the growth of pipeline inventories.

While the automotive industry in general, and Toyota in particular, is the origin of lean manufacturing, it's not the place to look for best practices in lean supply chains. What is—besides retailing?

  • Electronics, excelling in SCM by allowing ownership of production to migrate.
  • Logistics, bent on integration.

The re-engineering of electronics began in the mid-1980s with small moves toward modular suppliers. Companies such as Hewlett-Packard and CalComp were inviting their sheetmetal fabricators to load wiring and other components into metal frames and skins. Before long, electronics OEMs were requiring other kinds of component makers to do the same thing. This greatly reduces the OEMs' burden of too many hundreds of loose parts from too many suppliers. The production cycle shrinks, while providing value-added strengthening of the supply base.

No one could have foreseen that these beginnings of modular supply would grow into structural transformation of the electronics industry. In the 1980s, electronics OEMs did their own PCB assembly. By 2000 hardly any did PCB assembly; they sold it off to what became electronic manufacturing services (EMS). From virtually nothing, EMS has grown to look as if it might, before many years, rival the size of all the OEMs put together. Such giants of EMS as Solectron and Flextronics have moved up the food chain into production, and even design, of whole PCs for Hewlett-Packard and NEC. Injection molders such as Nypro are assembling phones for Nokia and Verizon, and single-use cameras for Kodak.

Complementary to these changes, logistics subsectors have merged to form third-party logistics (3PL) companies. Overnight freight handlers such as UPS, FedEx, and DHL have transformed mail-forwarding centers into distribution centers for filling orders that arrive electronically from IBM, Motorola, or Hewlett-Packard. Former transport companies such as J.B. Hunt and Yellow Roadway have become multimodal, and have added distribution centers where they perform light assembly and packaging. Warehousing and distribution companies have expanded the other way, into transport.

The re-engineering of electronics has unleashed freedom for component parts to move to the most capable entity, enabling each company to focus on what it does well. The parallel merging of disparate logistics providers extends the beneficial thread to the remainder of order-fulfillment. These restructurings remove many sources of missteps and delays throughout the process stages, including removal of fat in the pipelines.

We've briefly considered what I think are among the most profound developments in the young discipline called supply-chain management. Actually, one of them, the joint-inventory metric, is potentially profound. An IBM unit came up with it, but there is no evidence that the idea has taken root. It should. Now I'd like to discuss lessons for manufacturing from retail and the electronics industry.

Wal-Mart's contributions to SCM are best summed up in terms of collaboration—which includes, but only secondarily, collaboration IT systems. As has been widely reported, Wal-Mart, along with such manufacturers as Dell and IBM, employs leading-edge information technology in their supply chains. One Wal-Mart IT system, according to Business Week, allows 2000 vendors to immediately see how their merchandise is selling. But any company can install software. The effectiveness of information technology at Wal-Mart and its successful imitators follows from extensive formation of collaborative agreements.

High-performing collaboration develops in two stages: the first is internal, requiring breaking down the company's departmental silos. The most successful Wal-Mart suppliers—Proctor & Gamble and Johnson & Johnson, for example—have managers in Vendorville from a mix of functions. That mix would at least include marketing, operations, and finance. Together they can, with counterparts at Wal-Mart, cover the essential bases: joint impacts on sales and market share, effects on operational and supply-chain capacities, and projected changes in cost and cash flows. This is quite unlike the usual approach in which, separately at each company, sales booking orders are thrown over the wall, clashing with operational realities; and, later, finance sees if they made money or not. Or, alternatively, to keep capacity busy, production makes product in large lots and sends it into storage. Sales offers price and other inducements to move the excesses; and, again, finance sees if money was made.

Creation of Vendorville (in effect, "Collaborationville") was an unnatural act: far-flung suppliers with plants around the world had to move staff, often city people, to small-town Bentonville. The geographical cluster that has been called Toyota City, on the other hand, evolved more naturally over time, a fairly typical industrial cluster.

However collaboration has developed, it cannot resolve many supply-chain problems without astute uses of information technology. Again, Wal-Mart is masterful. Though the company spends less on IT per sales dollar than major competitors, it leads them in results. Part of the reason may be that Wal-Mart develops most of its IT in-house, rather than using off-the-shelf systems. Moreover, Wal-Mart's IT personnel commonly have prior experience in merchandising and, later, are frequently promoted out of IT and into an IT-using function. This connectivity helps ensure that it is not IT for technology's sake, but for furthering the company's multiple ambitions.

The lesson here for manufacturing is not that they must develop their own software; that may not be feasible, except in small measure. Bringing functional expertise into IT, however, makes good sense for manufacturers. (The idea has often been addressed in the systems development community. Usually, though, it has involved much talk, but little change.) Bringing outside experience into IT is one more step in breaching internal silos, thus paving the way for effective external collaboration.

The electronics industry offers quite a different message. By example, it says we are in a new, hyper-competitive era with an open-system model for survival. In electronics, that means outsourcing, and offshoring, to modular suppliers. In metalworking and other industries that produce goods too expensive for air freight, the modular supplier is more likely to be within reasonably short shipping distances.

The auto industry's tentative moves in that direction began about 15 years ago, largely in Brazil. A few Western automakers opened reduced-sized final assembly plants there designed to receive, from nearby suppliers, automotive modules such as a preassembled power train or front-seat cockpit. Most of the automotive industry, however, clings to the old model: outsized assembly plants receiving thousands of loose parts from many suppliers. For their part, first-tier producers are eager to become modular suppliers themselves, but not to the point of extending the idea back to their second tiers. Cases in point:

  • So far Toyota has shown no interest in modularity, and is building ever larger assembly plants as it continues with rapid global expansion.
  • Sliding sales of the Big-Two US automakers are resulting in bankruptcies for Tier-1 suppliers, even for Dana, among the leanest and best-managed of them.

Analysts, according to the Wall Street Journal, suggest that Dana's cost problems relate to its long-standing vertical-integration practices, and reluctance to outsource and manage its supply chains. The industry plays follow-the-leader, and the leaders (e.g. Toyota and Dana), to their detriment and that of the whole industry, have stuck with old models.

Not so in aerospace/defense. Aerospace/defense was in denial about just-in-time (now called lean) in the 1980s and into the early 1990s. Now the industry is in quick catch-up mode. The big four—Boeing, Lockheed-Martin, General Dynamics, and Raytheon—have sharply improved their inventory turnovers in the last six to 13 years. In supply-chain management, Boeing is aggressive in selling off component-parts plants to first-tier suppliers. Labor unions do not like it, but are not in a strong position to interfere. So Boeing, and the other majors, seem bent on leading industrial restructuring of aerospace/defense similar to what occurred in electronics in the 1990s.

This discussion yields a raw listing of practices for dealing with chronic delays and wastes in the supply chain. These are not particular to any one industry or type of company, but should apply widely—to companies in basic metals, plastics, glass, medical devices, foods, chemicals, paper, furniture, appliances, automotive, metalworking of all kinds, and so on.

  • Internal and external collaboration—internal multifunctional teams face to face with counterpart supplier and customer teams.
  • Joint inventory—as a performance metric aimed especially at managers in purchasing (raw materials) and marketing (finished goods).
  • Vendor-managed inventory—with your inventory records open to suppliers.
  • Continuous replenishment—fed by your daily usage data.
  • Synchronization via advanced scanning and tracking systems.
  • Squeezing out the middleman—unless the middleman is a 3PL.
  • Direct shipment—bypassing others' or your distribution centers.
  • Cross-docking at distribution centers.
  • Breaking free of ingrained vertical integration to source noncore items from lower-cost suppliers more expert than you.
  • For any product line involving large numbers of component parts, establishing modular suppliers to assume subassembly responsibilities for logical groups of parts.

Companies will be familiar with most of these points. Few, though, have laid the groundwork—the within-company and company-to-company collaboration necessary for effective implementation. These items are by no means comprehensive. Rather they draw from companies (e.g., IBM years ago and Wal-Mart today) and industries (electronics and logistics) that have been consistently at the leading edge of resolving supply-chain management issues.

Lean management has enjoyed good success within factories, but little in the pipelines. Companies that pay heed to these supply-chain management issues and solutions may be able to gain significant advantages over competitors that are standing still, doing little, or just following entrenched patterns in their own industries.


Richard Schonberger

Richard Schonberger, president of Schonberger & Associates Inc. (Seattle, WA), has written and lectured on lean manufacturing since the early 1980s. His 1982 book, Japanese Manufacturing Techniques: Nine Hidden Lessons in Simplicity (Free Press/Simon & Schuster, in nine languages) was the first lean-manufacturing book by a non-Japanese author. Schonberger's 1986 book World Class Manufacturing (Free Press, available in eight languages) introduced the term world-class manufacturing to industry. Four more books on related topics followed, the most recent being, Let's Fix It! Overcoming the Crisis in Manufacturing: How the World's Leading Manufacturers Were Seduced by Prosperity and Lost Their Way (Free Press, 2001). Schonberger left the University of Nebraska in 1985, where he was the George Cook Professor of Management, teaching production management and information systems. He can be reached by e-mail at


This article was first published in the September 2006 edition of Manufacturing Engineering magazine. 

Published Date : 9/1/2006

Manufacturing Engineering Media - SME
U.S. Office  |  One SME Drive, Dearborn, MI 48128  |  Customer Care: 800.733.4763  |  313.425.3000
Canadian Office  |  7100 Woodbine Avenue, Suite 312, Markham, ON, L3R 5J2  888.322.7333
Tooling U  |   3615 Superior Avenue East, Building 44, 6th Floor, Cleveland, OH 44114  |  866.706.8665