Click for next page ( 10


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 9
The Changing Face of U.S. Manufacturing JOSEPH F. SHEA How can education contribute to the revitalization of American manufacturing industry? This issue is central to the competitive position of the United States in the world economy, and to the direction in which U.S. society will evolve in the decades ahead. This paper does not dwell on how U.S. industries have become noncompetitive. Rather, it attempts to indicate what can be done, indeed, what is already being done, in many factories. There is growing evidence that much im- provement is possible in the short term, and that American factories of the future can be competitive in most basic industries if national technological and management resources are harnessed. Over the last five years, the National Academy of Engineering and the National Research Council have addressed ways to improve the competitive position of U.S. manufacturing industries. The Research Council established the Manufacturing Studies Board in 1980, and the Academy devoted its eighteenth annual meeting in November 1982 to U.S. Leadership in Manufacturing. The keynote speaker at that meeting, Professor James Brian Quinn of Dartmouth College, docu- mented the declining competitive posture of U.S. industries in the world market and made a strong case that, as a nation, the United States cannot afford to let itself become a service economy with production limited only to high-technology products.) He ended by voicing a guardedly optimistic view of the future. Joseph F. Shea is senior vice-president of engineering, Raytheon Company, Lexington, Massachusetts. 9

OCR for page 9
10 SlIEA In broad terms, the solution lies in taking the following steps: Enhance the prestige of manufacturing as a profession and as an intellectual challenge. Involve, once again, the top management of our corporations in the process of production and quality. Break down the artificial barriers which exist in most companies between design and production. Treat the manufacturing process as a system, not as a collection of discrete, loosely coupled functions. Increase the commitment of our engineering schools to manufac- turing technology. Increase the interaction between industry and universities in manufacturing education and research. Provide economic incentives from federal, state, and local gov ernments. Share information on what can be and is being accomplished. The details of implementation will vary by industry, but most of the above steps will be prerequisites for anv significant improvements ~ ~ , ~ ~ . ~ ~ , . . . ~ . . Before elaborating on these points, it is useful to consider two examples which illustrate both the nature of the problem and the path to a solution. In the first example, a defense electronics contractor improved yield from about 15 percent to over 75 percent through a complex printed circuit line, and found that the labor required for the same operation could be reduced by almost 50 percent. The stimulus for improvement came from visits to Japanese companies producing similar products, where equivalent yields were well over 90 percent, with no apparent difference in technology or tooling. Japanese management would not accept the amount of rework which had become the norm in the United States, and their workers responded by controlling in-process defects. When American management realized that they could do what others had done, the gains were dramatic. In a second example, a major U.S. electronics company, which found itself not cost-competitive, cut the product cost of a line of displays by a factor of 2, increased inventory turns from about 5 to 50 (and expects to reach 80), and plans to use present floor space to produce 5 times the originally planned volume. The company had found that the Japanese produced an equivalent product with less than half the support labor, required fewer kinds of parts because of effective standardization, and based design of a production line on a close working relationship between design and manufacturing engi

OCR for page 9
THE CHANGING FACE OF U.S. MANUFACTURING 11 neers. By emulating the Japanese, the American company was com- petitive in less than three years. These are not isolated instances. Examples abound in a broad range of American industries, including automotive, appliance, hand tool, and electronic companies. U.S. industry became noncompetitive be- cause designs were not readily manufacturable and because quality standards that were much too low were tolerated in factories. WHAT IS POSSIBLE In many areas of engineering, one can evaluate how close a design- for example, for a combustion cycle, an amplifier, or a structure- comes to a theoretical limit. There is no such theoretical basis, however, for producibility of a design and achievable quality levels. Companies tend to set standards based on past performance of similar products and whatever they know about domestic competition. From that point of view, cost reductions or quality improvements of a few percent can seem like major accomplishments. But now there is hard empirical evidence in many sectors that much more is possible. New standards have been demonstrated, and one must note the magnitude of the improvements being discussed: factors of 2 or more in cost and factors of 10 to perhaps 100 in reject rate, which has a direct bearing on quality of the delivered product. Much of industry has grown sluggish with past success. Achieving anew the manufacturing excellence for which America has long been known will be difficult because many managers do not start from fresh ground. They must first rid themselves of outdated assumptions, practices, and prejudices. There is evidence that the work force will respond to new management leadership, such as the success achieved in color television manufacture when Sanyo management took over the old Warwick plant with many of the same employees and U.S. middle management. Improving the factories of today is but one more step in the continuing industrial revolution. The first phase, from the 1780s to the 1840s, was based on the application of steam power. The second phase, between 1860 and 1910, was based on new forms of power from oil and electricity. The third phase, beginning in the 1950s, was assumed to be based on nuclear energy; however, for a complex series of reasons this has not happened. Rather, this phase is based on the application of electronic systems~omputers and automation to widening areas of data handling, automation, and control. Manufacturing is a process which transforms information into a

OCR for page 9
12 SHEA product. The information includes design data, quantities required, and delivery dates. The transformation involves developing tools and processes, obtaining material, processing material, assembly, testing, and delivery. The factory of the future will be an integrated system with a common engineering and manufacturing data base. Data proc- essing will be used extensively to receive design information without having to reconfigure for manufacturing, estimate and order material, control inventory, program machines, monitor yields, and program test equipment. Automation will be extensive, encompassing material handling, numerically controlled machines, and closed-loop process control. Robots will function as welders, painters, assemblers, and inspectors. New materials with advanced properties will displace conventional products and processes. For example, the silicon revolution in digital electronics is known to all. Monolithic gallium arsenide microwave circuits will have an equally dramatic effect in radio frequency devices over the next decade. Composite materials, including carbon fibers imbedded in resin, will change structural designs. One general aviation manufacturer has already wound a complete fuselage from carbon fiber tape in less than a day and a half. Although the details will vary by industry, the factory of the future will challenge our long-held belief that high-volume runs of identical products are required to achieve low cost. It is conceivable that early in the next century computer-controlled flexible manufacturing systems will produce virtually all of the material goods required by society, except those with high artistic content. The companies that master this transition will gain nearly unassailable positions in the world market through their ability to produce quality products tailored to special customer requirements on a very short lead time. As the examples cited above indicate, however, a major portion of the gains to be achieved can be realized today, not in the twenty-first century, with existing technology. One approach, well established in Japanese firms and successfully employed by several American companies to improve quality and productivity while reduc- ing lot sizes, is the "just-in-time" production concept. This concept is based on the notion of producing only in response to customer demand and on short lead time. Design and operation of a manufacturing plant capable of efficiently producing any and all of its products on demand and with short lead times while conforming to quality standards require: Plants with well-defined product lines;

OCR for page 9
THE CHANGING FACE OF U.S. MANUFACTURING 13 Tight pull scheduling-that is, production responsive to customer demand; Efficient, flexible layouts and balanced process capabilities; Well-developed processes operating under statistical control; Small lot sizes; High employee involvement; and Continuing training and investment in employees throughout their careers. Finely focused factories were found in America in the nineteenth century. Today, they imply standardization of elements within a limited product family, close integration of product and tooling design, and discipline in design evolution to maximize the use of proven tooling and production processes. They will force a restructuring of the relationship between a manufacturer and the supporting vendors. Hewlett-Packard, among many others, is particularly well known for its work in this area. Flexible layouts combine group technology that is, part families funneled through a complete machining center- with production lines that enable manning in response to production demand, rapid com- munication among operators, and efficient material movement. Black and Decker has successfully responded to offshore competition by pioneering these concepts. In recent years, it has been rediscovered that the defect level must be reduced to as near zero as possible for critical functional tolerances. Even acceptable quality levels of 99 percent or so will not produce cost-competitive products. The percentage of defects can and must be driven down toward the parts per million range. This requires processes capable of statistical control, with operator responsibility for self- inspection and authority to shut down the machine whenever there is evidence that it is out of control. This key to Japanese quality is being adopted in the United States, and the General Electric dishwasher plant in Louisville, Kentucky, is a good example. The Ford Motor Company has published an excellent booklet on the subject.2 Efficient processing of small lot sizes requires minimal set-up times. A prime example is the Toyota hood and fender plant where a line consisting of a 500-ton toggle press and three 300-ton single action presses can be set up in less than 10 minutes. Many American companies are finding that set-up times can be reduced by 90 percent or more. Four Deere & Company plants, including a foundry, and plants manufacturing diesel engines, garden tractors, and heavy farm equip

OCR for page 9
14 SHEA ment have already made major gains, as has Speed Queen, one of Raytheon's subsidiaries. Employee involvement requires developing a team authorized to control quality at the source, trained in many different operations, able to move from operation to operation as demand dictates, and able to handle all routine set-ups and maintenance as a matter of course. Such teams are the natural precursor to the technician teams required to run the factories of the future. One example of the team approach is TRW's wire and cable plant in Lawrence, Kansas, which is operated by a semiautonomous team. Team members are encouraged to become qualified to operate every piece of equipment in the plant, for which they must pass both written and hands-on operating tests. They are then paid for the highest qualification achieved, regardless of the job duties being performed at the moment. The team follows the flow of work through the plant, operates different machines as required, and even makes decisions on manning and operation times to meet schedule requirements. The just-in-time concept has resulted from a reexamination of the manufacturing process as a system. The gains include inventory reduction, regained floor space in the plant, shorter schedules, lower costs, and higher quality. The results achieved by a growing number of companies demonstrate what can happen by creating an intellectual climate that challenges entrenched assumptions about how manufac- turing plants should be structured. A HISTORIC VIEWPOINT The aspects of manufacturing just discussed-flexibility, design standardization, tooling, tightly controlled tolerances, product evolu- tion, supplier base, and quality are not new. Ironically, they contrib- uted to the growth of American manufacturing from the early 1800s to the present. When Eli Whitney took a government contract in 1798 to deliver 10,000 muskets two years later, colonial manufacturing was a collection of artisans in cottage industries. Finished products varied widely in quality, and gross imperfections were common. Whitney spent a year building the tools, jigs, and other production fixtures necessary for an integrated flow of work through his factory. At each station, he located tools, machines, parts, and skilled workmen to keep the flow of muskets steady. By organizing to accommodate a regular process of manufacture, and by building machinery capable of working within

OCR for page 9
THE CHANGING FACE OF U.S. MANUFACTURING 15 fine tolerances, Whitney redefined the nature of the production task. He pioneered the first of six stages of American manufacturing: factories well suited to the sequential production of simple, imitative, not very capital-intensive products, assembled from machine-made, inter- changeable parts. The second phase of American manufacturing began when demand for volume production of consumer goods, such as sewing machines, required that products be broken down into clusters of technologically specialized components. The latter were then assigned to different factory work units which fed them as needed into the overall process flow. Isaac Singer devoted much time and energy to product design, developed standardized components, and organized his production system as a vertically integrated whole. The 32-acre plant Singer built in 1873 had a rail-supplied foundry, forging shop, milling department, and multiple facilities for inspection and testing of both components and final products. He found, by experience, that it paid to put just as good parts into the cheapest machine as into the highest priced pearled and ornamented cabinet machine. Across the product line only the decor changed; all the working parts were the same. Highly specialized, vertically integrated factories tended to resist model change, however. Many companies which emulated Singer fell into the trap of manufacturing a product with increasing efficiency until it became obsolete, but Samuel Colt, the legendary arms maker, confronted the issue directly. He took American manufacturing into its third stage by institutionalizing constant improvement in process and product technology as a path to achieving competitive advantage. The central reality of the fourth stage was the new-found importance of suppliers. The end of the nineteenth century saw a rapid proliferation of machine shops, die makers, and technology base suppliers an infrastructure which helped prepare the ground for the first generation of automobile manufacturing. The existence of this supplier base lent support to managers who were personally experienced in process technology and understood sources of components. Allan Nevins writes of Henry Leland, who supplied engines to Ford as well as to the Olds Motor Works before forming the Cadillac Motor Car Company in 19023: To work to a 1/10,000 of an inch was not exceptional in that factory, and Henry M. Leland could supervise production requiring 1/1,000,000 of an inch. The firm had devised or improved some of the machine tools, and had worked out the revolutionary methods which produced the gears for the Columbia bicycle and other metal products combining great delicacy, strength, and precision.

OCR for page 9
16 SHEA During times of technological ferment like that characterizing the first few decades of the automobile industry and challenging us now, management skills and technical understanding on the order of Leland's are invaluable to competitive success. The fifth stage resulted in the strength of our industrial capacity during and after World War II. A flow system, producing technologi- cally complex products at high volume, was mastered. Henry Ford organized operations at Highland Park strictly in terms of the necessary flow of work by using separate production lines for each component to reduce process bottlenecks, by applying conveyors and other techniques of line-flow management, and by driving inventories down to acceptable levels. One unsolved problem was the integration of the work force into the production process not as a faceless mechanism, but as a reservoir of competitively valuable human strengths. U.S. industry is late coming to a sixth phase of American manufac- turing, perhaps because its very success has led it to believe that it is as good as it could be. For several decades, in all too many industries, management effort has been directed away from production and toward marketing and finance. It is time to redress that neglect and reap the benefits of creative integration of a skilled labor force, data processing, and advanced technology into the production process.4 Plato wrote in The Republic: "The direction in which education starts a man will determine his future life." Accordingly, in 1984 the Manufacturing Studies Board of the National Research Council com- missioned a study of industry-academia cooperation in manufacturing, recognizing that creation of an intellectual climate to carry out the changes discussed here requires that industry and universities focus together on manufacturing technology. This is easier said than done in the academic world, because many problems in manufacturing are applied research at best and may not rank high on the tenure criteria. Until manufacturing curricula are developed by universities and become an attractive option for the better students, the issues of competence of manufacturing personnel and their ability to adapt to technological opportunity will continue. Schooling is necessary but not sufficient. Industry must change the employment practices for manufacturing professionals, and provide both financial incentives and intellectual challenges so that better candidates will opt for careers in manufacturing. In the short run, the obvious route is for industry to encourage changes in university curricula and to supplement those changes with applied research support related to the specifics of individual industries. The issues become how to convince management that such investment

OCR for page 9
THE CHANGING FACE OF U.S. MANUFACTURING 7 1 is prudent, and how to bring engineering faculty up-to-speed fast enough so that they are indeed useful in either training or consulting. The Research Council study on university-industry cooperation in manufacturing chaired by this author has not yet finalized its recom- mendations, but its initial conclusions are summarized here. The study has concluded that three segments of society must work together to reinvigorate American manufacturing: industry, universities, and gov- ernment. Actions appropriate to each are suggested below. WHAT CAN INDUSTRY DO? Management must be convinced that significant changes are possible. In the short term, quality and productivity can be improved by focusing on details within the manufacturing process. In the long term, invest- ments in technology, both process and system, and in the people who operate that technology can result in factories of the future which retain or regain a competitive position in world markets. Achieving these ends will require increased technical strength in manufacturing organizations. Recruiting for manufacturing will have to be put on an equal basis with engineering; manufacturing salaries will have to compete with engineering salaries; and continuing edu- cation programs must be developed for manufacturing personnel. Organizational reforms must force a closer relationship between en- gineering and manufacturing to develop producible designs and the restructuring of factories to reflect the systems nature of manufacturing operations. More important, manufacturers must be convinced that universities can contribute and must be willing to explore modes of cooperation. Obviously, the conviction will vary from industry to industry, with major differences from company to company within a given sector. WHAT CAN UNIVERSITIES DO? Manufacturing curricula must receive peer and administrative ac- ceptance, requiring a strong champion within the institution. Univer- sities that choose to strengthen or initiate manufacturing-related pro- grams must define the criteria by which those efforts will be judged against more traditional research activities. Manufacturing systems engineering curricula are being developed. There appears to be no general agreement on what the course content should be, or how it can be applied to a given industry. Examples stressing manufacturing applications should be introduced into the

OCR for page 9
18 SHEA core technical courses in the established disciplines. Faculty must be given release time for curriculum development. It is hoped that the issue of curriculum content will receive reasonable attention from this symposium, particularly as related to manufacturing as a system. This topic must not be confused with courses related to manufacturing processes which should be taught within the traditional engineering school structure. The physics, chemistry, metallurgy, instrumentation, and control courses required for, say, submicron semiconductor fabrication, fiber optic communications, composite structures, and synthesis of new chemical products, are subject matter for the electrical, aeronautical, mechanical, and chemical engineering faculties. The tougher question is how and, frankly, whether to teach manufacturing as a system. The traditional industrial engineering programs are not, in general, held in high esteem by either industrial or academic peer groups. Since a fundamental principle of management should be "You cannot manage that which you do not understand," a student must come to a manufacturing systems engineering (MSE) curriculum with a strong engineering foundation perhaps augmented by a year or two of industrial experience. The seeds of a manufacturing systems curriculum may lie in providing courses which apply the principles of data processing, information systems, data base feedback and control, employee utilization and motivation, and system engineering methodology to management of a manufacturing system. Since manufacturing must work closely with design, the principles of design for manufacturability must also be included, as well as the use of automation together with cost estimating in the design cycle. Quality must be a required subject not just the usual principles and statistical methodology, but emphasis on what quality levels can be and have been achieved. These experiences can set the standards by which students judge the future performance of their plants. This is a lot to pack into a degree program, and some of it may be better learned if it is deferred to continuing education. At the least, the MSE student should take away a vision of what factories can become, some tools with which he or she can begin to contribute, and the zeal to make the vision a reality. Universities must encourage better students to consider careers in manufacturing by raising admission standards and by stressing man- ufacturing opportunities in high school recruiting. And, perhaps, university research can develop a stronger theoretical basis for man- ufacturing. What is meant by a producible design? How can achievable

OCR for page 9
THE CHANGING FACE OF U.S. MANUFACTURING 19 quality levels be estimated? Together, industry and universities can establish research programs that address problems in manufacturing technology. Additional actions suggested for industry and universities include: Financial support by industry for manufacturing initiatives at universities including grants, equipment (and related maintenance support), and scholarships; Joint development of co-op programs and defined research pro- grams in manufacturing; Use of industry personnel as adjunct faculty; and Use of faculty as industrial consultants, and faculty sabbaticals in manufacturing assignments. WHAT CAN GOVERNMENT DO? These problems have begun to attract government attention at both the state and national levels. Several states have appropriated funds for the establishment of centers of manufacturing technology to encourage regional groups of industries and universities to focus on the generation and dissemination of knowledge in this area. The Ben Franklin Institute in Pennsylvania, the Industrial Technology Institute in Michigan, Rensselaer Polytechnic Institute's Center for Manufac- turing Productivity and Technology Transfer in New York, and programs in Ohio, Arizona, North Carolina, and elsewhere are inno- vative and promising experiments. Proof of success will be the degree to which these centers can become self-sustaining. Industry will have to provide the necessary support by recognizing the value of services received. Federal policy is still evolving. The Department of Defense, long a sponsor of manufacturing research, has increased funding in manufac- turing-related technologies, primarily related to defense needs. The National Science Foundation sponsors a program in manufacturing sciences and is in the process of creating a series of Engineering Research Centers, several of which will relate to manufacturing. The U.S. Congress is contemplating several bills, but no clear pattern has emerged. A broad cross section of industry must be motivated to improve manufacturing practices and to explore what help they can get from universities or the emerging manufacturing centers. Companies must be encouraged to find out what modern technology, applied to their particular situations, can do. Some form of tax incentive that promotes

OCR for page 9
20 SHEA cooperative programs may help align the random motions inherent in our free economy. Today, perhaps 5 percent of engineering schools stress manufactur- ing, but the problem is critical enough that probably 95 percent should be offering competent programs. It must be cautioned, however, that the assumption that universities can effectively contribute to either short- or long-term improvements in manufacturing is an intellectual act of faith. Most of the progress cited has been made in industries on the high- technology side of the national spectrum, but the actions advocated here have broader applicability. Management in many industries must be convinced that they have an alternative to low labor rate, offshore factories, or inevitable surrender to foreign competition. NOTES 1. J. B. Quinn. 1983. Overview of the current status of U.S. manufacturing. Optimizing U.S. manufacturing. U.S. Leadership in Manufacturing. A Symposium at the Eighteenth Annual Meeting, November4, 1982. Washington, D.C.: National Academy Press. 2. Available from the Statistical Methods Office, Operations Support Staff, Ford Motor Company, Booklet #80-01-251. A. Nevins. 1954. Ford: The Times, The Man, The Company. New York: Charles Scribner's Sons, p. 212. 4. This encapsulated view of American manufacturing history draws extensively on Industrial Renaissance, Producing a Competitive Future for America by W. Aber- nathy, E. Clark, and A. Kantrow of the Harvard Business School (New York: Basic Books Inc./Harper Colophon Books, 1983).