BY BRUCE GOLDMAN
Don't you just love it when the customer in front of you spends 10 minutes fishing for exact change while the folks on either side of you go whizzing past their respective checkers?
Anyone who has mastered the art of straddling two supermarket lines with one grocery cart has grasped, at a deep intuitive level, the value of keeping your options open. Companies supplying market-driven business clients are learning to keep their options open, too, by putting off the differentiation of the products they make until as late in the assembly process as possible. That's because in a fast-evolving high-tech world, businesses frequently respond to shifting demands from their customers (who may be other companies or end-users) by calling their suppliers and modifying their own orders late in the game.
The power of procrastination (in business parlance, "mass customization through postponement") was extolled by Carlos Nieva, Spain-based manufacturing operations director of Lucent Technologies, at an all-day workshop held at Stanford on April 13. The workshop, co-sponsored by the Alliance for Innovative Manufacturing at Stanford (AIMS) and the Stanford Global Supply Chain Management Forum (SGSCMF), featured representatives from several multinational corporations who provided case histories and business tips on the day's theme: "Managing Global Enterprises and Global Supply Chains." The workshop was orchestrated and hosted by Laura Kopczak, consulting associate professor of management science and engineering and director of research for SGSCMF.
AIMS and SGSCMF are campus-based joint ventures initiated by Stanford's Graduate School of Business and School of Engineering and corporate partners to promote the exchange of technical ideas and techniques between academia and industry. AIMS works with a broad array of large manufacturing companies addressing numerous concerns, while SGSCMF partners with more than 40 affiliates on advanced theory and practice specifically for the field of supply-chain management.
Lucent, once known as Bell Labs, manufactures high-tech telecommunications systems. It's not uncommon, Lucent's Nieva told the group of nearly 80 participants, for customers in this field to want to tinker with order specifications or quantities late in the game. One hypothetical way to prevent this propensity from sticking you with a load of unsalable inventory, he said, is the "take it or leave it" approach, a technique that might might have worked back in the old days when Lucent was part of the huge monopoly called Ma Bell, but in an age of fierce competition among numerous suppliers bearing no corporate ties to the order-placing entity just won't cut it.
Another approach might be to improve forecast accuracy. A company with a perfect crystal ball would always manage to have the right parts and product inventory available at just the right time to satisfy total demand, said Nieva. Unfortunately, in a dog-eat-dog, leapfrogging-technology, morphing-customer world, that's not how it works. "Low forecast accuracy is and will remain a fact of life."
The best alternative, Nieva said, is to delay the differentiation of your products (whether a serial port should be on the left or the right side of an electronic appliance, for example) for as long as possible, so you can roll with the punches when customers change their minds. He suggested several routes to achieve this delay, including simplifying the product, standardizing the components, and eliminating options that customers don't care about in the first place.
A key global market exemplifying the ascendancy of flexibility over forecasting is the semiconductor industry, noted John Hoffman, vice president and co-general manager of the etch-products business group at Applied Materials of Santa Clara, Calif. Applied Materials is a leading manufacturer of semiconductor fabrication equipment, which must meet ultra-precise specifications. "Our customers' expectations regarding the time from order to delivery are around 16 to 18 weeks," Hoffman said. "About eight weeks into that cycle is when they send their contractors and plumbers in and they change their specs. So, we want to start our customization after the eighth week."
Toward that end, Hoffman said, Applied Materials has been employing mass customization by standardizing parts and modularizing components. This is difficult, he said, due to such factors as differing safety regulations in, say, Japan, Taiwan and Germany.
Adding to the difficulty are the exquisite technical sensitivities inherent in the semiconductor fabrication business. Competitive pressures constantly force efforts to lower costs, Hoffman told the audience. "Once, we came up with a new screw and saved some money. Then one day, a customer calls us and says, 'Hey, all of a sudden, our microprocessors are running a little bit faster what changed?' It turned out our new screws were sputtering silver into their semiconducting materials and messing up their microprocessors."
Forecasting the market for semiconductors that Applied Materials' equipment is designed to produce is no easy feat. One thing is clear, Hoffman said: "Whether you're forecasting low growth or high growth, it's big growth. Demand has exploded. We'll double our output from the first to the third quarters of this year." But that growth is saltatory, doubling every year for three years during the "boom" part of the cycle, then taking a 50 percent dive during the next year when the "bust" phase hits. "We have to exercise great flexibility in the treatment of contracting our own suppliers," some of whom could be put out of business if Applied Materials were to cut its order rate too callously, he said.
Don McKay, director of logistics development at American President Lines (APL), a global shipping company, related a case history exemplifying the necessity of thinking locally.
A few years ago, a multinational automotive giant hired APL to manage the flow of vehicles from plants to dealerships in an eastern China province. But when APL went to select a local logistics partner, "some very capable local logistics providers declined to participate in the bidding process. We didn't know why, so we talked to them, and we learned that they didn't like our selection process they thought there was too much paperwork and they were uncomfortable with making presentations. They found the entire process demeaning, kind of like begging."
So APL changed its ways. "We met with each supplier individually. Then we did the research and paperwork. We made decisions, talked to the small number of finalists, described the project in more detail to them and went on site visits." The result, McKay said, was that "virtually all qualified suppliers did bid. And we made sure those that didn't win understood why, so if they were stronger in another province, they'd want to try for the job there."
Two executives from accounting and financial-consulting giant Ernst & Young, Rick Cooper and Jim Goodbout, warned participants always to consider the tax implications before making big international supply-chain decisions.
Globalization and e-commerce are changing the playing field, opined Cooper. Manufacturing and distribution centers don't stay in one place so long anymore. But once you start moving things around, there are big tax implications many tens or even hundreds of millions of dollars' worth to every move you make, or don't make.
Only about a quarter of the taxes businesses pay are on income; the rest goes for such things as value-added tax, customs tariffs, property taxes and employment taxes. Thus, it's necessary to consider which countries are going to be, for tax purposes, the "locations" handling inherently delocalized transactions or owning inventory at various phases of assembly.
But tax planning is more than just moving your operation to Singapore or Mexico, he said. "You don't always choose the place with the lowest tax rate," he said. "You have to consider where your people are now and whether they're movable, as well as what kind of telecommunications backbone a potential host country has. If you make the right decisions, you can improve your cash flow by 20 to 80 percent." SR