Lean Thinking

by Daniel T Jones

  The fact that Wiremold freed up approximately 50 percent of the space in all of its operations (excepting the central warehouse, which was totally eliminated) greatly helped the acquisition campaign. While Art Byrne’s philosophy is to retain and upgrade existing management, several of the companies purchased were available because the family management could no longer run them successfully and wanted out. This provided consolidation opportunities.

  For example, two of the firms purchased were consolidated into Wire-mold’s Brooks Electronics operation in Philadelphia. Prior to the acquisitions, the three companies had operated independently, utilizing 114,000 square feet of space. Now, the combined operation has increased its total sales significantly, yet is located in Brooks’s original 42,000 square feet of space. Inventory has been reduced by 67 percent, the number of employees needed to run the combined operation has been reduced by 30 percent, and the surplus buildings have been sold.

  In effect, Art Byrne and Wiremold are running a lean vacuum cleaner across the world of batch-and-queue thinking in the wire management industry. Each time Wiremold’s vacuum sucks up a batch-and-queue producer it spits out enough cash to buy the next batch-and-queue producer! Because of Wiremold’s need to grow to utilize freed-up resources, this process can and must be repeated indefinitely. (As we will show in Chapter 11 , the first firm to adopt lean thinking in any industry can and must perform this same feat.)

  The third and final element of the Wiremold growth strategy is the rapid introduction of new products, utilizing the new product development system with its dedicated teams and Quality Function Deployment methods described earlier. For example, the new product line described in Chapter 1 has increased sales by 140 percent, both by creating a new niche in the market and by stealing sales from competitors unprepared to match Wire-mold’s pace of product introduction.

  All three strategies are critically dependent on the lean techniques introduced in production, order-taking, and product development. Indeed, the rapid introduction of these techniques is Wiremold’s fundamental strategy. Art Byrne remembers that in previous jobs he had often wanted to go faster with application of these techniques but his senior bosses were usually more interested in massive long-range “strategic” planning efforts which they believed should take precedence. “To my way of thinking, this is exactly backwards. Introducing lean techniques in every business activity should be the core of any company’s strategy. These provide both the opportunity and the resources to generate and sustain profitable growth. Profitable growth is what the strategic planners of the world are always seeking, but find hard to achieve because their company’s operations can’t deliver on their strategies.”

  The Box Score After Five Years

  As we will see in Chapter 11 , three years is about the minimum time required to put the rudiments of a lean system fully in place and two more years may be required to teach enough employees to see so that the system becomes self-sustaining. Wiremold’s performance over the five-year period from the end of 1990 to the end of 1995 is therefore a good test of the potential of lean thinking. The results are quite striking.

  To begin with product development, time-to-market has been consistently reduced 75 percent, from twenty-four to thirty months down to six to nine months. Sixteen to eighteen new products are being introduced each year (compared with two to three in the period through 1991), but the engineering/design headcount has stayed the same.

  Several new computer-aided design technologies might be assigned some of the credit for these gains, except that these techniques were adopted in 1990–91, before the time-to-market and productivity gains. As we have emphasized throughout this volume, advances in hard technologies can be useful and in many cases are very important, but they are unlikely to yield more than a fraction of their potential unless they are incorporated in an organization which can make full use of them. By placing product designs in single-piece flow with a dedicated, multiskilled, co-located team and no interruptions, Wiremold has eliminated backflows and rework in the development process while also reducing manufacturing costs and dramatically spurring sales with products which accurately address customer needs.

  The rethinking of order-taking, scheduling, and shipping has produced the same results. The old batch system, which needed more than a week to receive, process, and ship a typical order, now needs less than a day. Past-due orders are now less than one tenth of their 1991 level and continue to fall as Wiremold refines its pull system through all six product teams. Order entry errors have been practically eliminated, and misrouted or unanswered queries in the much smaller Customer Service Department have fallen from 10 percent to less than 1 percent .

  In physical production, the results are exactly as we would expect. The amount of plant space needed to produce a given volume of product has been cut by 50 percent and productivity has been increasing at a rate of 20 percent per year. The time for raw materials and components to travel from the receiving dock to the shipping dock in Wiremold’s plants shrank from four to six weeks to one to two days. Inventory turns have increased from 3.4 in 1990 to 15.0 in 1995.

  To make this possible, Wiremold has continued to reduce setup times on all of its machines and to convert all production activities for its product families to single-piece flow. For example, punch presses with large progressive dies that used to take two to three hours to change are now changed in one to five minutes; rolling mills which took eight to sixteen hours to change over in 1991 can now be changed in seven to thirty-five minutes; plastic injection molding machines that took two to four hours to change over in 1991 can now be done manually by one Wiremold employee in two to four minutes. As a result, machines that previously shifted from one product to the next two to four times per week, now change products twenty to thirty times a day.

  By aggressively implementing single-piece flow, operations requiring five to eight operators in 1991 are conducted with one to three employees today. By utilizing single-piece flow, JIT, and Total Productive Maintenance in the largest and most complicated assembly operations, productivity has been increased by 160 percent over three years. Equally important, single-piece flow has been instrumental in reducing defects by 42 percent in 1993, another 48 percent in 1994, and another 43 percent in 1995, almost at Wiremold’s target rate of 50 percent per year indefinitely. At the same time, standard work, takt time, and visual control have been slashing accidents and injuries, which are down by more than half compared with 1991.

  Putting the improvements in product development, order-taking, and physical production together, we find that sales per employee more than doubled, from $90,000 in 1990 to $190,000 in 1995. However, this and the figures just cited are all relative to Wiremold’s previous performance. The indicators which truly count in the marketplace are sales, profits, and market share. Happily, between 1990 and 1995, Wiremold’s sales in its core wire management businesses—owned before the lean vacuum cleaner was turned on—more than doubled in an otherwise stagnant electrical equipment market and profits of the whole firm—including the new businesses—increased by a factor of six. What’s more, the growth rate, including acquisitions of related businesses, is picking up, in line with Wiremold’s strategy of doubling its sales every three to five years for the foreseeable future.

  All of these indicators are summarized in Figure 7.2 , a “box score” for Wiremold under lean management .

  T ABLE 7.2: W IREMOLD U NDER L EAN M ANAGEMENT

  But What About Firms with More Severe Problems?

  The Wiremold story is extraordinary. The firm has been transformed in a remarkably short time and now gives every prospect of growing rapidly into an industrial giant. What’s more, we could repeat this story in dozens of medium-sized firms we have discovered across the United States during research for this book.

  Wiremold was a greater challenge than Lantech, given the age and narrow skills of its workforce, the stagnation of its core market, and the entrenched us-versus-them mentality of the old ma
nagement and union, but is it still a fair test of lean thinking? After all, Wiremold has only fourteen hundred employees, operates primarily in two neighboring countries (the United States and Canada), and has relatively simple product and process technologies. What about the aging industrial giants who present the most visible managerial challenges? What about the publicly traded, mass-production firm with tens of thousands of employees, global operations, complex technologies housed in deep technical functions, and a complex network of component systems suppliers? Can lean techniques produce the same results, and in the same time frame? We turn now to Pratt & Whitney, which is truly the acid test of lean thinking.

  CHAPTER 8

  The Acid Test

  On June 1, 1991, Mark Coran drove across town, from the Hartford, Connecticut, headquarters of the United Technologies Corporation, to the East Hartford headquarters of Pratt & Whitney, UTC’s largest subsidiary and the world’s largest builder of aircraft engines. UTC chairman Bob Daniell had just given him a new assignment—one for which his background as UTC’s corporate controller and star cost-cutter seemed ideal preparation.

  The problem at Pratt appeared to be structural and substantial but not desperate. As the world’s largest builder of military jet engines 1 (accounting for a third of its total business in the 1980s), Pratt was faced with the end of the cold war, a reality to be confirmed shortly with the collapse of the Russian countercoup in August 1991. It suddenly seemed likely that much of the military engine business was gone for good.

  In the short term, the loss of military business was offset by an extraordinary boom in orders for commercial engines. As the world’s market-share leader 2 in commercial aircraft engines, Pratt had ridden the wave and racked up a record operating profit of $1.01 billion in 1990 on a record $7 billion of military and commercial sales. However, anyone familiar with the roller-coaster demand cycle in the commercial engine business knew that sales at this level couldn’t be sustained for long, and in fact, orders for spare parts had already started to fall. Therefore, Mark Coran’s job, as the new executive vice president for operations at Pratt, was to prepare the manufacturing operations in a massive company with 51,000 employees for a perhaps 10 percent permanent reduction in the size of the business, and to do this before the commercial-order boom collapsed.

  As it turned out, Mark Coran had no time to work with. June 1991 would prove to be the peak month of production volume in the history of Pratt & Whitney, with “shop hours” of work—the conventional Pratt measure of production activity—running at an annual rate of 11 million. Soon, commercial jet aircraft orders, which had reached a record high of 1,662 in 1989, started to drop steeply as the world recession set in, falling to a low of only 364 in 1993.

  Much worse for Pratt’s finances, the airlines were dipping into their inventories of jet engine parts to repair their fleets, rather than ordering new parts from Pratt. Orders for Pratt spares were sliding rapidly by the fall of 1991, and by 1992 were running at only 63 percent of the 1989 peak. This was a crushing blow because spare parts account for the great majority of the profits of every aircraft engine company, due to the industry practice of selling new engines at substantial discounts in order to capture market share and create a large user base for their highly profitable, captive spare-parts businesses.

  To make matters worse, Pratt and its two global rivals—General Electric in the United States and Rolls-Royce in the United Kingdom—were locked into spending large sums right away—$3 billion in total among the three firms—on development of the next generation of jet engines. These are the 84,000 to 100,000 pounds of thrust “monster motors” needed for the Boeing 777 and possibly for the proposed 600-seat Airbus A3XX. (The first of these, the Pratt PW4084, entered airline service on the Boeing 777 in June 1995.)

  Because of its four-year product development cycle for new engine designs and the eighteen-month production lead time to physically build an engine once ordered, Pratt was helpless to respond to a dramatically changed world. Capital spending on the PW4084 was locked in and many engines were already under construction for customers who suddenly no longer wanted them. What was more, the airlines were sending a clear signal that they now wanted low-cost rather than high-performance engines for the 1990s, designs that could not be ready for years.

  The first half of 1991 had continued the record profits of 1990, but the turn in the market was breathtaking and Pratt was suddenly heading for a $1.3 billion swing in its operating results within a year, culminating in a $283 million loss for 1992. As Coran remembers, “Very suddenly, just when I arrived, everything that could go wrong went wrong. Rather than a simple cost-cutting exercise to deal with a 10 percent drop in volume, I realized that we needed to rethink the whole business.”

  Fortuitously, just at the time of the crisis, several key executives in the UTC group—including Coran, George David, the president of UTC’s Commercial and Industrial Group, and Karl Krapek, the president of Carrier—had become familiar with lean principles, mainly from the accident of being located in Hartford where Art Byrne was working steadily to apply them. In addition, Coran had a major advantage. He had never had an operating job prior to arriving at Pratt and therefore had none of the biases of the traditional mass-production operating executive. He therefore resolved to implement lean thinking as the best way to save Pratt & Whitney .

  The attempt to do so represents the acid test. If Pratt can apply these principles quickly in a massive, publicly traded, high-tech organization with extraordinarily deeptechnical functions and life-or-death demands on product quality, plus all of Wiremold’s problems, then literally any American firm can.

  From the American System to Mass Production 3

  Pratt provides a wonderful example of the mass-to-lean conversion because the firm was centrally involved in creating the very mass-production system which eventually threatened its survival. What’s more, it twice went through the progression from flexible start-up to stuck-in-the-mud mass producer that we saw at Lantech.

  The original Pratt & Whitney Company was created before the American Civil War by Francis Pratt and Amos Whitney. These “Yankee mechanics” learned their trade as inside contractors at Samuel Colt’s armory, opened in Hartford, Connecticut, in 1855. They produced the individual parts needed for Colt pistols and rifles, hiring their own workforce but using Colt’s plant and tools.

  Of central importance to our story, Pratt and Whitney also built many of the four hundred machine tools and the gauges Colt needed to achieve his goal of totally mechanized gun production in which parts were interchangeable and handwork for “fitting” was eliminated. 4 This approach became known as the American System, in comparison with the European System in which parts were individually handcrafted, with each part fitted to those already in place to create a completed product.

  When Pratt and Whitney left Colt in 1860 to establish The Pratt & Whitney Company, they took with them a fundamental set of ideas about manufacturing practice which dominated the company until very recently. They believed that best practice called for the creation of special-purpose machines able to perform specific operations on specific parts, if possible at high speeds in high volumes. They further believed that machines performing similar types of tasks should be grouped together in departments and that simple logic called for setting up a machine to make a given part and then making a batch of them before setting up the machine for the next part. In other words, they built the precision machinery needed for the familiar world of batch-and-queue and, over time, organized their own factory in accord with these principles.

  Over the next sixty-five years, Pratt & Whitney grew from a small work-shop under the direct management of the two founders into a massive and highly successful organization. In its many departments focused on specific processes—casting, drilling, tapping, heat treating—Pratt produced the parts needed for lathes, grinders, millers, cutters, and borers for metal-working industries. The firm also pioneered extremely precise gauges to check the a
ccuracy of parts and sold these along with their tools. Over the years, Pratt’s machines became more complex and capable of more delicate and sophisticated tasks. In addition, advances in metallurgy made it possible to work prehardened metals so parts could be made to net shape without fear that subsequent hardening steps would interfere with interchangeability. However, the basic philosophy of production did not change.

  The Rise of the Eagle 5

  In the summer of 1924, Frederick Rentschler resigned as president of the Wright Aeronautical Corporation in New Brunswick, New Jersey, because the bankers investing in the firm would not back his idea for a radial, air-cooled engine much larger than the revolutionary Wright Whirlwind just entering production. 6 He believed this large engine would swing the military away from liquid-cooled designs and make commercial aircraft economically viable for the first time.

  With the encouragement of the U.S. Navy, Rentschler sought new financial backers and early in 1925 contacted Pratt & Whitney in Hartford, which was experiencing a slump in its business and found itself with excess plant space and tools. In addition, Rentschler noted that the Hartford area was full of Yankee mechanics skilled in operating the types of tools Pratt produced, precisely the tools needed for aircraft engine manufacture. 7

  Rentschler proposed to play a similar role at Pratt & Whitney to that Francis Pratt and Amos Whitney had played seventy years earlier at Colt’s armory. He outlined a plan to set up a company within a company, using P&W’s long-established name with its worldwide reputation for precision machinery. He proposed borrowing a million dollars from P&W’s owners (in return for giving them 50 percent of the stock in the new Pratt & Whitney Aircraft Company) 8 and using Pratt’s underutilized plant space and tools to make his new engine. An agreement along these lines was reached in July of 1925, and Rentschler was back in the aircraft engine business.