Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
INTEGRATED MANUFACTURING SYSTEMS: AN OVERVIEW JAMES J. SOLBERG Manufacturing stands on the threshold of a new era in which all manufacturing en- terprises must compete in a global econ- omy. A gloomy view of this new economic environment can project dismal conse- quences for the quality of life in America. A more optimistic view recognizes that de- spite challenges to the nation's economic and technological strength, this fundamen- tal change in the competitive environment presents opportunities as well as challenges. Indeed, it is possible to imagine a future of economic growth and prosperity on a global scale. One thing that can be said with certainty is that all sectors of society will undergo profound changes. Consequently, many manufacturing practices that were effective in the past will no longer be so in the fu- ture. Those companies that prosper and those that fail may well be distinguished principally by their ability to plan for change. Moreover, we can expect the eco- nomic environment to remain highly dy- namic, so changes must be confronted with the expectation that they will be continual. Fortunately, there are many options open to companies that understand the dynamics of this situation. The problem that they face is to identify, from among the rich universe of possibilities, those opportunities that rep- resent the best investment of limited re- sources. There are many conflicting voices speak- ing to manufacturers. Some emphasize au- tomation equipment; others say that the data network is key. Noting delays in pro- cess flows, some suggest that attention to material handling is essential; others argue that it is only the processing steps that add value. Some put the primary burden for reform on management, some on the work- ers, some on suppliers, and so forth. CAD, CAM, CAE, JIT, SQC, MAP, MRP the 1
2 list of abbreviations for promised solu- tions- goes on and on. How can we make sense out of all of this? Numerous conferences and groups have addressed issues related to the changing na- ture of manufacturing. The National Re- search Council, through the Manufacturing Studies Board, recently completed a broad examination of the current status of U.S. manufacturing, covering technology, em- ployment, and government policy (NRC, 1986~. Several workshops have attempted to define research needs in more specific technical areas (ASME, 1986, NRC, 1984, Thomas, 1983; Volz and Naylor, 1985~. Each of these reports provides useful and important background material. The papers contained in this volume were prepared for a conference that was con- vened by the National Academy of Engi- neering to assess the strengths and weak- nesses of current information and tools for planning and controlling integrated manu- facturing systems, and to identify critical needs and opportunities for future study. Although the scope of inquiry included all discrete product manufacturing, it was lim- ited to the technical aspects of the problem. In the conference the emphasis on tools for planning and controlling operations re- flected a concern for the adequacy of avail- able methodology. The attention given to integration indicated that existing manu- facturing technology is too segmented gen- erally. With fragmentation of the research community mirroring the situation in in- dustry, a principal objective of the confer- ence was the creation of a synthesis of needs and understanding from a wide range of individual views. The dialogue between representatives of the user community who are involved in manufacturing and mem- bers of the research community in univer- sities, industry, and other laboratories gen- erated a deeper understanding of needs and opportunities. The organization of the papers in this volume reflects the structure of the confer- JAMES J. SOLBERG once. In their paper, Erich Bloch, director of the National Science Foundation, and Kathy Prager Conrad emphasize the im- portance of gaining an understanding of current needs and future prospects in inte- grated manufacturing. The next two papers are authored by representatives from com- panies who had achieved notable success with automation. Arthur I. Roch, jr., LTV Aircraft Products Group, describes the LTV experience with flexible machining and Laurence C. Seifert, AT&T Network Sys- tems, discusses the AT&T integrated man- ufacturing system for electronics. The sys- tems aspects of integrated manufacturing systems are described by UIrich Flatau, Digital Equipment Corporation, and by John A. White, Georgia Institute of Tech- nology. The human role in these systems is discussed by William B. Rouse, Search Technology, Inc. Technical issues in prod- uct design are discussed by Herbert B. Voelcker, Cornell University, and Daniel E. Whitney and coworkers, Charles Stark Draper Laboratory. The issues that relate to processing are discussed by Vijay A. Tip- nis, Georgia Institute of Technology. Sys- tem design issues are discussed by Rajan Suri, University of Wisconsin, and by Arch W. Naylor and Richard A. Volz, University of Michigan. System operation is discussed by Floyd H. Grant, FACTROL, Inc. Not surprisingly, the papers present dif- fering points of view. Also as expected, the authors tend to propose their own perspec- tives as the correct ones from which to ad- dress all other issues. Both the speakers and the participants recognized this during the conference and openly acknowledged their biases with good-natured humor. More than once, participants likened themselves to the blind men trying to describe an ele- phant based on which part of the animal each of them explored. Although a full consensus does not emerge from this set of papers, important issues that transcend the narrow view of any individual discipline are clearly identi-
INTEGRATED MANUFACTURING SYSTEMS: AN OVERVIEW fled. The recurrence of these themes con- veys a sense of unity on key issues. The first of these issues is the communication gap between industry and universities. Both groups acknowledge past neglect and ex- press an eagerness to remedy the situation. Industrial and academic representatives have identified the design-to-manufactur- ing transition as a second key problem. "Collaborative manufacturing" and "si- multaneous engineering" are recently pop- ularized phrases for a very old need, which nevertheless remains as one of the major challenges. Competitive pressures have in- creased the urgency of the plea for better coordination between design and manufac- turing and for improved tools to support this coordination. To date, the research community has provided little help in this arena, leaving the practitioners to struggle as best they can with ad hoc procedures. The creation of a helpful methodology will be difficult, but at least the problem is now more universally acknowledged. Also apparent from the papers is the ex- treme complexity of the tangled web of re- lationships among the various subsystems involved in manufacturing. The differing perspectives presented in the papers leave little doubt that, although important tools and insights are included by each author, none addresses all of the important rela- tionships. One of the most interesting points of discussion in the conference was the question of how best to address this issue of complexity. Modern manufacturing sys- tems are highly complex. They share some of the complexities of biological or eco- nomic systems, such as the interactions of an enormous number of variables whose re- lationships are unknown and a mixture of time scales, including very short to very long. The physical, logical, and temporal separation of consequences of actions often confounds direct attempts to control behav- ior. Unexpected side effects may counter the very effect we desire to achieve. One approach to this problem is to acknowledge 3 that such complexity is inevitable and, therefore, to concentrate on developing far more sophisticated methods of analysis and control. An alternative approach is to con- clude that we will never learn to under- stand and manage all details of such com- plicated systems, so we must strive to simplify them. To a neutral observer, it is obvious that both approaches should be pursued as vig- orously as possible. In fact, this debate over whether it is better to master complexity or to avoid it is somewhat reminiscent of the perpetual debate over hardware and soft- ware issues. Both approaches are valid, and neither can advance for long without prog- ress from the other. Still, we can expect the debate to continue for years. The general tone of the papers contained in this volume reflects a sense of the change that is taking place. The old vocabulary of manufacturing systems seems inadequate for expressing many of the new concepts in the field. The birth of new paradigms of thought is always a bit unsettling, so it would not have been surprising if the broad range of papers included in this volume in- cluded divergent views. That this is not so indicates general agreement that serious changes in the direction of manufacturing are in order. Under these circumstances, it is impor- tant that we understand both the global issues associated with manufacturing as well as specific technical needs and opportuni- ties. Two questions might be asked. Why is manufacturing different now? Of the knowledge and methods we have acquired in the past, what should be kept and what should be abandoned? In searching for an- swers to these questions, we must not be too willing to discard what has been done in the past, for there is a danger of chasing a passing fad or misreading a cyclical change as permanent. At the same time, we cannot hope to make significant progress in the competitive race if we insist on clinging too firmly to traditional thinking.
4 Such issues as product cost and quality are, of course, timeless. The ability to pro- duce what the customer wants in a timely manner is also basic. These fundamental driving forces point to a continued need to improve the efficiency of the product reali- zation process, while ensuring that the job is done properly. But there are other, more subtle issues to consider. Proliferation of product variety, coupled with a generally faster pace of new product introduction, has led to shorter product life cycles. With the market life of many elec- tronic products such as personal computers and audio/video equipment now only about eighteen months, a new generation of a product can quickly render an earlier de- sign obsolete so that it can no longer com- pete. Mechanical products face less extreme pressures but are similarly affected. Under such conditions, it is essential to recover de- velopment costs and any product-specific capital equipment costs very quickly. An- other implication of this shorter product life cycle is that production equipment must be more flexible; one cannot assume that ma- chinery will be used to make only one prod- uct type throughout its useful life. The increased cost of money has directed new attention to inventory reduction. In- stead of relying on storage of a sufficient quantity of components and finished goods to meet all demands, manufacturers must increasingly rely on careful synchronization of supply and demand of these goods. An important consequence of this shift is a tighter coupling of the processes, which in turn implies a need for much better man- agement of a system that is becoming in- creasingly complicated. Since inventories were.originally intended to buffer, and therefore isolate, the effects of disruptions in one step of the production sequence, it is now necessary to contend with "ripple effects" that a disruption can produce throughout the manufacturing sequence. It goes without saying that the avail- JAMES J. SOLBERG ability of ever more powerful and inexpen- sive computers has permanently changed the way we do business. This trend will certainly continue. Indeed, computer tech- nology and the system options that it per- mits will soon dominate the information as- pects of manufacturing. Although most of these trends and issues are well recognized by industry today, it is important to look further ahead to changes that will alter the basic foundations of com- petitive manufacturing in the next century, and to seek answers to deeper questions. For example, how can the United States compete with countries that have abundant low-cost labor and are also aggressively developing and acquiring advanced technology? The first requirement is that we accept that changes in manufacturing technology are inevitable. Instead of resisting these changes as we have tended to do in the past, we must find ways to take advantage of them. In terms of the directions for re- search, this means that we must investigate those technologies that can operate effec- tively in a changing environment. In more human terms, it means that we must em- phasize the kind of education that prepares people for changing roles. Second, we must understand the neces- sity of relying comparatively less on experi- ence and more on sound theory. The ability to apply trial-and-error learning to tune the performance of manufacturing systems be- comes almost useless in an environment in which changes occur faster than the lessons can be learned. There is now a greater need for formal predictive methodology based on understanding of cause and effect. This methodology can be expressed in a variety of forms: equations, mathematical models, simulations, algorithms, approximations, etc. Of course, a good deal of such meth- odology already exists, but the practices of industry tend to place greater reliance on experience-based knowledge than on theory-based knowledge. This difference is
INTEGRATED MANUFACTURING SYSTEMS: AN OVERVIEW due in part to the failure of practitioners to familiarize themselves with the analytical tools that are available. In part it is due to a failure of the research community to de- velop the kinds of tools that are needed and to put them into a usable form. Although many of the computer aids as- sist individuals in doing their separate jobs, and sometimes in facilitating communica- tion among them, they generally fail to pro- vide the kind of integrated system that is needed to trace the effects of choices made in one arena of manufacturing upon another. In part it is due to a paucity of sound data on the operating parameters and perfor- mance characteristics of complete systems. In part it is because the present situation can be described as an ad hoc collection of programs that were designed separately, each tailored to specific requirements but incapable of functioning as a unit. A wide variety of software tools is needed to handle the information aspects of manufacturing so that alternative choices for actions can be assessed. These tools must extend the scope of concerns beyond traditional boun- daries. Moreover, they must be capable of functioning together as a coherent system. This cross-disciplinary integration is not likely to occur, however, without deliber- ate, focused effort. Another extremely important guiding principle for research is that we must gen- erate reusable results having broad appli- cability. The best examples of advanced manufacturing systems that have been commercially developed, including those described in this volume, are tuned to the .snec~ific set of conditions in a single Plant. 5 impact we desire on the whole of discrete manufacturing practice, we must find ge- neric solutions that can be applied in many circumstances. If future research is to develop the needed tools for design and control of integrated manufacturing, it must emphasize three critical points: (1) more direct industry in- volvement, so that the work will meet their needs and will be in a usable form; (2) a broader "systems" view, with specific at- tention to interfaces, so that pieces will not have an isolated utility; and (3) greater rigor, including better performance data, so that we can be certain of what we know and can apply the knowledge to different situations. Applied to specific cases, these principles suggest that research proposals lacking in these characteristics may be less worthy of support than those that do ad- dress them. These generalizations may appear to state no more than what was, or should have been, obvious for years. However, there is a new urgency to the need as we attempt to direct the relatively scarce resources avail- able for research in manufacturing to those key issues on which our future competitive- ness will rely. If the views expressed in this volume are a true measure, a substantial number of people are apparently ready to work together to consider collectively where the priorities lie. Each reader must contemplate the spe- cific conclusions to be drawn. The papers collected in this volume, representing the thoughts of some of the leading experts in various disciplines related to discrete man- ufacturing, can provide a stimulating be- Although these systems may provide great ~ '' benefits to the companies who own them. there are few transferable benefits to the next company wanting to do something similar or even to the same company in another plant. In effect, each new system development project starts over from the beginning. If we are to have the kind of ginning. However, there is one additional point that should be borne in mind. The message of this volume is contained in the totality of what is presented, not the sepa- rate pieces. Therefore, readers would be well advised to resist the natural tendency to consider in detail only those sections that best match their own interests. As we search
6 for the right path to a future of sustained prosperity in manufacturing, our best hope for collective wisdom lies in open-minded consideration of all of the points of view represented here. REFERENCES American Society of Mechanical Engineers (ASME). 1986. Goals and Priorities for Research in Engi- neering Design. New York: ASME. National Research Council (NRC). 1984. Computer JAMES J. SOLBERG Integration of Engineering Design and Production. Committee on the CAD/CAM Interface, Manufac- turing Studies Board. Washington, D.C.: National Academy Press. National Research Council (NRC). 1986. Toward a New Era in U.S. Manufacturing: The Need for a National Vision. Manufacturing Studies Board. Washington, D.C.: National Academy Press. Thomas, Michael E. 1983. A Workshop on Research Directions in Industrial Engineering, Georgia Insti- tute of Technology, Atlanta. Volz, Richard A., and Arch W. Naylor. 1985. Work- shop on Manufacturing Systems Integration, Uni- versity of Michigan, Ann Arbor. .
