CAM: A Model - IEEE Xplore

2 downloads 0 Views 2MB Size Report
design, engineering, and drafting (grouped under the title. CAD) can be linked to various ... of the CAD/CAM technology or project management ap proaches.
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. EM-34, NO. 2, MAY 1987

101

Effective Implementation of Integrated CAD/CAM: A Model PAUL S. ADLER AND DUANE A. HELLELOID

Abstract—A model of the organizational conditions of effective implementation of integrated C A D / C A M is developed. The central proposition is that the conditions o f successful implementation cannot in general be created concurrently with the technological effort of C A D / C A M integration. The model proposes that integration effectiveness is principally a function o f the skills, procedures, strategies, and culture which preexist the integration of C A D and CAM systems. In this model, technological characteristics and project management methods play a subordinate role. I . INTRODUCTION

I

N RECENT years the use of C A D (computer aided design) and C A M (computer aided manufacturing) systems has increased substantially. C A D has helped increase the produc­ tivity of individual designers; its introduction has also served —«-

*~ :

~~~~Λ'.~η\~~

/tar!/*M

*-Λ««ν*ο.

a s ail VJppv/l lUliltjr ι υ uuptirvw uiv ^ v u i u i n u i i u n v/i uvaigu ivumo,

the total product development time has often thereby been dramatically reduced. C A M systems have helped increase productivity on the factory floor by reducing costs, improving quality, and increasing flexibility. These, along with advances in inventory and scheduling systems are all part of an information revolution in the factory. The general contours of this revolution point to a conver­ gence of administrative control, production, and design technologies towards what is now commonly called computer integrated manufacturing, or C I M . The competitive signifi­ cance of a C I M environment derives less from the power of the component technologies and more from their progressive integration. Independent functional systems are coming to­ gether into cross-functional systems, which will eventually be linked into a fully multifunctional C I M . The C A D / C A M linkage of design and production functions is perhaps the most advanced of the multifunctional systems today. In the C A D / C A M case, integration of computer-aided design, engineering, and drafting (grouped under the title C A D ) can be linked to various types of computer-driven manufacturing capability (equipment like machine tools and also manufacturing documentation systems). This linkage can be thought of as proceeding through a hierarchy of progres­ sively broader and deeper integration: 1) downloading of data directly from the CAD data base to the manufacturing environment; Manuscript received February 27, 1986; revised July 20, 1986. P. S. Adler is with the Industrial Engineering and Engineering Management Department, Stanford University, Stanford, CA 94305-4024. D. A. Helleloid is with the Hewlett-Packard Company, Lake Stevens Instrument Division, Everett, WA 98205-1298. IEEE Log Number 8612323.

2) inclusion in the CAD data base of manufacturability design rules, criteria, and models so as to assure the reliability of the data that is downloaded; 3) inclusion in the CAD data base of automatic manufactur­ ing process planning, broadening the ability of designers to incorporate manufacturing concerns; and 4) error recovery capabilities such that contingencies in manufacturing can be automatically identified, diag­ nosed, and rectified, or circumvented. Very few facilities have proceeded beyond stage 3) and the authors' recent field investigations in the electronics and aerospace industries suggest that most are still struggling with stages 1) and 2). Our fieldwork (to be reported in a subsequent article) and a reading of the more pragmatic literature ([22], [23], [28], [29]) suggests that writing the software that makes CAD and CAM systems able to communicate has proven to be only a small step towards functional integration; critical organiza­ tional issues remain to be addressed. The new technologies are a necessary but not a sufficient condition for *'getting the slash out of CAD/CAM" and realizing the associated benefits of improved quality, reduced cost, increased product perform­ ance, and shortened product development cycles. Our underlying model is shown in Exhibit I, where the effectiveness of product development projects (cost, quality, timing) is a function of the quality of project management, the technology being used, and key organizational characteristics. Moreover, we believe that the organizational factors dominate the model because they both influence the other two factors (by constraining or encouraging both project management approaches and technology development efforts) and mediate their impact on effectiveness. This paper elaborates on two key ideas. First, the various criteria of new product development project effectiveness need not 'trade o f f against each other. In many firms, interest in CAD/CAM is driven by the need to reduce product develop­ ment cycle time without adversely affecting product perform­ ance or cost; in other firms, priorities are differently ranked. Our proposition is that intelligent managerial strategies can often ensure that cost and performance improvements can be made even as development cycle times are reduced. Our second key idea flows directly from our model: if we measure the effectiveness of CAD/CAM integration efforts by improvements that these efforts generate in new product development project effectiveness indicators, then we should find that the effectiveness of CAD/CAM integration efforts depends more on the organizational context than on the level 4

0018-9391/87/0500-0101$01.00 © 1987 IEEE

102

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. EM-34, NO. 2, MAY 1987

new product development project eiiectiveness

Fig. 1.

Determinants of new product development project effectiveness.

of the CAD/CAM technology or project management ap­ proaches. The temptation is strong to look to CAD/CAM technology or to project management techniques as the answer to the persistent difficulties of coordinating engineering and manufacturing efforts. Our proposition is that the availability of the integration technology is at best an opportunity to confront these organizational difficulties—it is not itself a solution. Similarly, project management methods can make a real difference, but are necessarily dominated by the broader and deeper organizational pressures. This article presents a model of the factors that can facilitate the implementation of integrated CAD/CAM systems. To guide our effort to identify these factors, their relationships, and their impact on implementation success, we have surveyed a broad range of literature (Section II). We have distilled from this literature two background propositions (Section III), which in turn lead us to two central research hypotheses (Section IV). If research supports these hypotheses, our precepts for good project and technology management might be modified (Section V).

principal conclusion from this literature is that for the functions to work effectively together, and hence achieve effective integration of systems, the functions must be on a more nearly equal footing than is commonly the case. But what sorts of action can be taken to effectively balance the functions when, as is often the case, they do make distinct and unequal contributions to the firm's competitive advantage? 3) The management and implementation of cross-functional activities is the topic of much research in both the project management and matrix management literatures ([5], [8], [10], [ 1 2 H 1 4 ] , [26], [30], [43], [48], [49]). The integration of CAD and CAM both depends on and reinforces the interaction of design and production, as well as marketing, finance, and vendors. The key lesson of this literature would appear to be that when environmental and technological "uncertainty" is low, the traditional functional organization will suffice, but that the greater the uncertainty of the project, the greater will be the need to coordinate the functions through project teams. But how can firms identify and reduce critical types of uncertainty? 4) Finally, the management of innovation literature ([1], [9], [25], [30], [32], [51]) points to a multiplicity of factors distinguishing successful from unsuccessful innovations and more from less innovative firms. One limitation of this literature is its relative neglect of process innovations like CAD/CAM (as distinct from product innovations). A second limitation, Rosenbloom [41] argues, is the lack of a theoretical framework which can make sense of the importance of both external environment and internal organization in explaining innovation performance. A strategy focus is needed, but what are the key components of an innovation strategy, and, in particular, of a process innovation strategy? I I I . Two

II. A VERY RAPID SURVEY OF THE RELATED LITERATURE

The relevant literature comes from four main sources: 1) the implementation of new technologies, 2) interdepartmental conflict, 3) project management, and 4) the management of innovation. In each area, certain general conclusions can be drawn, but key questions of critical importance to CAD/CAM implementation remain unresolved. 1) The difficulties of technology implementation have been explored in a variety of functions ([7], [13], [15], [16], [18], [27], [39], [42], [44]-[47]). Few of these writers, however, focus on the consequences of a given function's automation for other functions. The key idea we draw from this literature is that the successful implementation of advanced technologies depends at least as much on organizational adaptation as on technical adaptation. But what general propositions can one usefully make on the type of organizational adaptation needed when the technology is itself interfunctional, as in the case of CAD/CAM? 2) Research into the management of a multifunctional business unit has given extensive coverage to interdepart­ mental conflict, power, status, and politics in the firm (for example, [24], [38], [47], [52], [53]). In many companies, important differentials of status, compensation, and capabili­ ties distinguish the design and the production functions. Our

THEORETICAL BACKGROUND PROPOSITIONS

For our model of CAD/CAM integration to be useful in other areas of cross-functional integration, a rather broad theoretical perspective is needed. The subject we are studying is not new—it dates from the origins of the functionally organized firm—but there is a surprisingly limited theoretical basis on which to conduct research, other than that we can draw from the literature discussed above. The following propositions are directed primarily at the cross-functional implications of impending new technologies, but are more generally applicable for most situations involving cross-functional organizational changes. A. The Greater the Magnitude of Technical Change Sought, the Higher the Level of Learning Required in the Organization The higher level of automation implicit in the shift of technological change from intra- to interfunctional means that organizational learning is required at the strategic and cultural levels in addition to the skills and procedural levels. With an intrafunctional technological change, such as CAD, organiza­ tional learning is required at the lower, more localized levels 1

This follows Pava 137] except that we have replaced his "structure" level of learning with "strategy" for reasons that are explained in a later section. 1

103

ADLER AND HELLELOID: IMPLEMENTATION OF INTEGRATED CAD/CAM LEVEL OF LEARNIN6 REQUIRED

of skills and procedures, but little change is required at the higher strategic and cultural levels. The integration of CAD and CAM provides an opportunity, and in many cases imposes a requirement, for learning at these higher levels. B. The Higher the Level of Learning, Time to Adjust

the Greater the

These four levels of learning have progressively higher levels of "viscosity." From this follows three expectations. First, the more of the appropriate skills, procedures, strategies, and culture that are in place before a given phase of the CAD/CAM integration program begins, the greater will be its effectiveness, measured as its impact on the effectiveness of product development projects that will use the new tools. Even well-designed project management is, we hypothesize, rela­ tively ineffective in poorly prepared organizations. Second, and perhaps less obvious, CAD/CAM's effective­ ness will depend more on the preexisting culture and strategies than on the preexisting procedures and skills. Not only are these higher levels more viscous, but the appropriate culture and strategies are needed to ensure that the right procedures and skills are rapidly identified and developed. Third, and a consequence of the preceding two factors, the pace of the CAD/CAM integration process will, in all probability, be dictated less by its purely technical component than by the time necessary to establish the skills, procedures, strategies, and culture required to effectively implement those technical capabilities. Our model thus focuses on the organization's characteristics (skills, procedures, strategies, and culture) prior to the implementation and integration of CAD/CAM systems. These two propositions can be easily represented on a single graph (Exhibit I I ) : the greater the magnitude of the change one wants to make to the level or span of automation, the higher the level of organizational learning required and therefore the greater the time required to realize this potential. I V . THE HYPOTHESES

The first hypothesis of our model is basically methodologi­ cal, referring to the relationship between different dimensions (or measures) of effectiveness of changes in the level of integration technology. The second identifies the factors that can improve that effectiveness. HI:

A

Τ

CULTURAL

The Effectiveness of CAD/CAM Integration Efforts Can be Measured by Changes Induced by Those Efforts in the Effectiveness of New Product Develop­ ment Projects

One of the difficulties of both managerial practice and research in this area is to determine a measure by which we can judge the effectiveness of steps towards CAD/CAM integration. Among other measures, one might consider improvements in product cost, in product quality, product performance, in degree of interdepartmental cooperation, and in the time required to bring new products from idea to production ("development cycle time"). One is, however, immediately confronted with the possibility that improvements in one of these dimensions may be obtained at the cost of

YEARS

MONTHS

nA6NITU0E OF TECHNOLOGICAL CHAN6E SOUGHT

TIME TO ADJUST

Fig. 2.

Organizational learning.

losses in another dimension, making the evaluation task almost impossibly complex. Our working hypothesis is that this dilemma will not typically arise: as the design and manufactur­ ing technologies advance, improvements in these objectives are not related to each other as trade-offs, but rather are positively correlated. Even in the absence of technological change, trade-offs are not inexorable. Crosby [11], for example, has made the argument that "quality is free." The underlying premise is sound: firms are virtually never on a production possibilities frontier, and therefore the things that are done to improve the quality dimension of effectiveness can often stimulate im­ provement in the cost dimension. Crosby's argument is generalizable. The idea of an inexora­ ble trade-off between effectiveness dimensions is only relevant in the hypothetical limiting case where performance capabili­ ties are exhaustively known and fully exploited. But the real world is more plausibly modeled as one of cumulative and idiosyncratic discovery of new production capabilities [33], and the relevant capabilities are therefore often complemen­ tary rather than substitutes. Besides cost and quality, other complementarities can easily be identified: managers explor­ ing ways of reducing design-to-market cycle time by more careful delimitation of the project mission often discover that higher quality upfront preliminary design is a prerequisite. The notion of trade-off can, in this dynamic perspective where all other things are not held constant, be very misleading. There are clearly limits to this complementarity; beyond them, we return to the conventional world of trade-offs. But it is critical to see that the domain over which the complementar­ ity holds will expand relatively if, at the same time as we make changes to cost, quality or timing policies, we can also change the technology in use. The " n o trade-off argument holds even more strongly, therefore, as a firm progresses into CAD/ CAM integration. We therefore expect to find that in the better-managed firms, CAD/CAM integration brings an improvement in all new product development project effectiveness dimensions. Whatever the strategic orientation of the firm, whether emphasizing product sophistication at a relatively high cost and a relatively long product development cycle or emphasiz­ ing low cost for simpler product designs, the effectiveness of the implementation of integrated CAD/CAM will normally be manifest in some improvement in all the performance dimen­ sions. We should therefore be able to form an indicator of CAD/CAM effectiveness by comparing the average effective-

104

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. EM-34, NO. 2, MAY 1987

ness of new product development projects conducted before CAD/CAM integration with the average effectiveness of those conducted after integration. H2: Ex Ante Organizational Conditions are More Impor­ tant for CAD/CAM Integration Effectiveness than Either the CAD/CAM Technology Itself or Project Management Techniques This second hypothesis reflects the relative viscosity of the four levels of learning. CAD/CAM technology, we hypothe­ size, is not an answer to organizational problems at the engineering/manufacturing interface. If it is to be a help rather than a hindrance, organizations must already have tackled, or at least have committed themselves to tackle, these problems. The CAD/CAM integration process, like specific new product development projects, can support (can be the opportunity for) some learning, especially of the " l o w e r " kinds. But the "higher" the level of learning required, the more difficult will be the "concurrent" management of the new product development project and its conditions of success. The ex ante conditions that characterize the context of the CAD/CAM integration effort should therefore deter­ mine the final outcome (preintegration to post-integration improvement m new product development project effective ness) to a greater extent than the quality of the technology or of project management itself. This is not to suggest that all organizational learning can or should take place prior to integration. The implementation and use of CAD/CAM will certainly provide the opportunity for new organizational approaches that were not possible nor even conceivable prior to the integration. But because these sorts of changes are difficult to make in most organizations, it is important to prepare the organization for the new technology before it is to be implemented. The following four subhypotheses are designed to explicate this proposition. They highlight some possible paradoxes in the management of CAD/CAM integration efforts. -

H2.A:

Preintegration

experience and skills:

Computerized information systems must be well-developed individually in both engineering and production functions before integration can occur. In other words, system effective­ ness will be greater when all potential users of the system have already developed their own computer capabilities before the integrated system is developed. At first glance, this proposition seems to promise the creation of a software Tower of Babel nightmare. We suspect, however, that the most difficult task of CAD/CAM implemen­ tation is not the integration of software, but the acceptance of these new tools. Acceptance is all the more likely when the tools are well-designed—which can only be guaranteed when the tool developers can benefit from the experience gained by the users in starting up their own systems and finding the applications that are particularly appropriate for them. Com­ puter literacy, both in general and in relation to specific applications, is a prerequisite for implementation success ([2], [4], [15], [18], [22]). This hypothesis, if confirmed, would justify the recommen­ dation often heard in regard to advanced technologies that the

appropriate approach is a) top-down basic parameter setting for the system to insure at least a minimal level of compatibil­ ity, but b) bottom-up responsibility for specific system choice and implementation. The hypothesis leads also to the sobering implication that there is no easy solution to the "islands of automation" problem. Grandiose detailed plans imposed by a corporate MIS or CIM "czar" cannot but ignore the real diversity of functional needs and the nature of the learning process required to specify these needs and to cooperatively commit to the integration process. H2.B: Preintegration

procedures:

Procedures for the coordination of the design and production functions must be working well before the systems are integrated in order for the use of the CAD/CAM system to be effective in the development of new products. Whereas many CAD/CAM integration enthusiasts are inclined to see it as the solution to persistent coordination problems, our hypothesis suggests that it is at most a late opportunity to attack these problems. Much recent discussion in management circles has focused on whether design and manufacturing phases should be sequential or overlapping. There may not, of course, be one best way: depending on the activity, the form of the most effective procedure might differ. Our proposition is therefore somewhat different from the now-popular endorsement of "parallel" development. It is simply that CAD/CAM should not be thought of as a substitute for cooperative coordination. Effective new product development will with CAD/CAM still require the input of manufacturing at the early stages of design. The more "participative" the process, the higher the chance of effectiveness. There is, however, one important caveat to this proposition: if this coordination could somehow take place before any particular product development project began, then time spent in coordination on the specific project might be reduced and quality and cost improved. The development of design rules for producibility as one such tool of ex ante coordination—it can serve as a powerful lever with which to take the coordination load off the project [54]. It is important to note, however, that the elaboration of such design rules requires extensive proceduralized cooperation between design and manufacturing engineers. Furthermore, it is not because the CAD data base incorporates good design rules that ongoing discussions with manufacturing people relative to producibility will be redundant. Although design rules and other formal modes of communication are important, informal procedures, such as phone calls and "management by wandering around," are critical. The prior existence of effective procedures (formal and informal) for close coordination should therefore prove to be a major factor in implementation success. H2.C: Preintegration functional

strategies:

The procedures discussed in the previous hypothesis need to be reflected in and supported by organizational strategies that give coordination an appropriate institutional significance. CAD/ CAM is not a solution to the perennial problem of product

105

ADLER AND HELLELOID: IMPLEMENTATION OF INTEGRATED CAD/CAM

versus functional structures. The organizations that can best deploy C A D / C A M are those which have developed a set of long-term functional strategies. Signifying a higher level of interfunctional integration, a CAD/CAM environment will heighten the need for ongoing coordination of engineering and manufacturing. Such ongoing coordination will require more than procedural devices to sustain it—it will require organizational structures and strate­ gic commitment. Structure is important above and beyond procedure if only because procedures themselves will need to be adapted as the process evolves; a means of managing that adaptation in real time is essential. Only a well-designed organizational structure can process the requisite information in the requisite time [17] and mobilize the requisite resources [38]. We are not, however, advocating a matrix structure for all contexts. The relative importance of the product and the functional dimensions of organization depend, with or without CAD/CAM integration, on the factors that have been well identified in the organizational literature (complexity, uncer­ tainty, resource availability, etc.) ([6], [9], [17], [30], [38]). There is no magic solution to the choice between the ease of integration provided by product-oriented structures and the depth of expertise provided by functionally-specialized struc­ tures. The danger with the current infatuation with the former is that the project team will find itself having to solve too many technological problems. The "dilemma" is only resolvable by adding a third dimension to the problem: the degree to which the functions' pre-project activity has prepared the terrain by establishing all the basic technologies before the project begins [54]. But such ex ante coordination requires strategic thought and action at a new level of detail in the functions. It is through such detailed, functional strategic analysis that a firm can first identify and then manage the critical environmental uncertainties. We hypothesize that this development of strategic thinking cannot be an exclusively general management concern, but on the contrary must be rooted in strategic planning in each of the functions. When both engineering and manufacturing capabili­ ties are undergoing such profound transformations with new CAD and CAM systems coming on stream, general manage­ ment should be cognizant of the new business opportunities being created. Moreover, the mission of "supporting" and "implementing" the business strategy is no longer sufficient guidance to functional managers when the rate of change of their capabilities accelerates as it has: technology choices (e.g., system compatibility) made today will commit the whole business for years to come. Under these circumstances, the individual functions must develop their own functional strate­ gies [21] and these functional strategies must be congruent with each other and with the overall business unit strategy. Strategic planning must, under the new conditions of acceler2

Strategy can be seen as mediating environmental uncertainty and organizational structure. Galbraith 117] shows a correspondence between low uncertainty and functional structure, and between high uncertainty and product project teams. These coordinated functional strategies are a way of reducing uncertainty and shifting back toward the functional end of the spectrum. 2

ating technological change, be multilevel (general and func­ tional management) and multifunctional; it must be a coopera­ tive, iterative effort rather than a top-down unilateral process [20]. The set of functional strategic maps provides that frame­ work within which the CAD/CAM system can develop and evolve, as both the needs of the business change and new options for the configuration of the system become apparent. The specifics of the system need not be cast too rigidly in advance if all functions' goals are in sufficient congruence to assure that the eventual outcome will be consistent with the uses envisioned (and therefore consistent with overall business unit strategy). H2.D: Preintegration culture: The effectiveness of the C A D / C A M integration will depend upon the preintegration culture of cooperation and learning— and its symptoms in egalitarian norms of status, pay, educa­ tion, benefits, mobility between the functions and encourage­ ment of experimentation. The most successful integration will occur in firms in which the functions view each >ther as peers in a learning process. The structures and strategies of coordination are a necessary but insufficient condition of success. For these to echieve tbp desired effect, a culture of cooperation and learning is required. It is not simply a matter of "goal congruence." In the more ambitious projects, there will necessarily be some incongruence and conflict as each function sees opportunities for capitalizing on and building its own capabilities. A certain degree of such conflict is healthy, encouraging the parties to further research and clearer debate, and thereby stimulating the innovation process itself. But for this process to have beneficial rather than destructive effects, the organization needs an overarching culture of cooperation and experimenta­ tion. Such a culture takes time to emerge. A particular project might be the occasion for some dramatic turn-around in the attitudes and values of the project team members, but such radical change will not have much impact on the broader organization. The earlier the establishment of the right cultural context, the greater the payoff. CAD/CAM integration is not a way of avoiding the need for such cultural change. On the contrary, if, for example, manufacturing is not seen as a critical competitive resource for the firm, not only will Design Engineering fail to seriously consider producibility criteria, but Finance will block the funds needed by Manufacturing to capitalize on the potential of CAD/CAM since such improvements are often difficult to justify by traditional quantitative approaches [21]. Marketing too must be on board if, for example, the MRP systems are to reflect sales goals without putting Manufacturing under impossible time constraints. V . DISCUSSION

If these hypotheses are supported, some important conse­ quences will follow for project management, and indeed for the management of the CAP/CAM technology development process itself. An ex ante rather than the concurrent or ex post approach to the management of the CAD/CAM integration process

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. EM-34, NO. 2, MAY 1987

106

implies some changes to the management of individual projects and phases. Traditional project management ap­ proaches must be adapted to support new levels of automation and new levels of learning. On the basis of the preceding propositions, one would expect top management support to be beneficial, but top management involvement in intricate details to be a distrac­ tion, if only because of the faster pace of technological change and the cross-functional nature of many of the new technolo­ gies. For similar reasons, the existence of an interfunctional network of support would be more important to success than the presence of a single champion. Under these conditions of dynamic technological change, relations with the vendor organizations will need their ex ante conditions also: a ' 'marriage''' configuration will be more effective than arm'slength transactions. And finally, the focus on a culture of continual adaptation implies that training should be directed at developing longer term learning capabilities, rather than being limited to immediate operational proficiency. The preceding hypotheses concerning primarily the rela­ tions between engineering and production in their use of CAD/CAM should, furthermore, apply also to the relations between designers and users in the development of the CAD/ CAM system. As is common in a broad range of advanced technologies, if the designers of the CAD/CAM system do not take the user's needs seriously, the system will not be recognized nor accepted as an effective tool and will be discredited by underutilization ([37], [45]). But for designers to become cognizant of user needs, the latter need a certain level of familiarity with the technology's potential, and the two parties need procedures, strategies, and cultures that allow cooperation. These cannot be improvised. V I . CONCLUSION

In our focus on préexistent organizational conditions, we have probably overstated our case against the primacy often accorded technology and project management methods. There is clearly much in common between our propositions and those developed by or implicit in the best work on project management (as reviewed by Rubenstein and Ginn [42], for example). And the new technologies open up opportunities, not only to do the old things in new ways, but to do radically different things. But hopefully research along the lines suggested here can elaborate on the "conditions of validity" of the more established project management propositions, and the "conditions of effectiveness" of the new technologies. ACKNOWLEDGMENT

The general approach adopted here owes much to S. C. Wheelwright. This draft has also benefited from comments by R. S. Rosenbloom and discussions with M. Cutkosky. REFERENCES [1]

[2] [3] [4]

B. Achiadelis, P. Jervis, and A. Robertson, Project SAPPHO: A Study of Success and Failure in Industrial Innovations, Science Research Council, Univ. of Sussex, Sussex, England, 1976. P. Adler, "New technologies, new skills," Harvard Bus. School, working pap. 9-794-076, 1984. R. Beckhard and R. Harris, Organizational Transitions: Managing Complex Change. Reading, MA: Addison-Wesley, 1977. W. D. Beeby, "Integrating engineering and manufacturing," in

Advances in CAD/CAM: Case Studies, P. C. C. Wang, Ed. Boston, MA: Kluwar Academic Pub., 1984, ch. 2, pp. 37-55. S. A. Bergen, "The new product matrix," R&D Management\ vol. 5, [5 no. 2, pp. 149-152, 1975. [6] L. J. Bourgeois, III, "Strategic goals, perceived uncertainty, and economic performance in volatile environments," Acad. Manag. J., vol. 28, no. 3, 1985. H. Brody, "Overcoming barriers to automation," High TechnoL, pp. 17 41-46, May 1985. is: L. D. Brown, Managing Conflict at Organizational Interfaces. Reading, MA: Addison-Wesley, 1983. [9 T. Burns and G. Stalker, The Management of Innovation. London, England: Tavistock, 1961. [10] J. Child, "New technology and developments in management organiza­ tions," Omega, vol. 12, no. 3 , pp. 211-223, 1984. 11 P. B. Crosby, Quality is Free. New York: McGraw-Hill, 1979. B. G. Dale, "The management of the engineering change procedure," 12; Eng. Manag. Int., vol. 1, p p . 201-208, 1982. D. B. Davis, "Renaissance on the factory floor," High TechnoL, pp. 13; 24-47, May 1985. P. F. Drucker, "Playing in the information-based ' o r c h e s t r a , ' " Wall 14; Street J., p. 28, June 4, 1985. J. E. Ettlie, "The implementation of programmable manufacturing innovations," Center for Social and Economic Issues, Industrial 15 Technology Institute, Ann Arbor, MI, Mar. 1985, unpublished. J. E. Ettlie and A. H. Rubenstein, "Social learning theory and the implementation of production innovation," Decision Sci., vol. 11, no. 16 4, Oct. 1980. J. Galbraith, Designing Complex Organizations. Reading, MA: Addison-Wesley, 1973. D. Gerwin, " D o ' s and don'ts of computerized manufacturing," n ; Harvard Bus. Rev., pp. 107-116, Mar.-Apr. 1982. J. D. Goldhar and M . Jelinek, "Computer integrated flexible manufac­ turing: Organizational, economic, and strategic implications," Inter­ faces, vol. 15, no. 3, May-June 1985. 19; R. R. Hayes, "Strategic planning—Forward in reverse?" Harvard Bus. Rev., Nov.-Dec. 1985. Competitive 20] R. R. Hayes and S. C. Wheelwright, Restoring Our Edge: Competing Through Manufacturing. New York: Wiley, 1984. 21 G. J. Hess, "Computer integrated manufacturing—How to get started," in Proc. Auiofact 5, 1983. 22; G. J. Hess, "Computer integrated flexible manufacturing," in Proc. Autofact 6, 1984. 23; D. J. Hickson et al., " A strategic contingencies theory of intraorgani­ 24] zational power," Admin. Sci. Quart., vol. 16, n o . 2, June 1971. P. Kelly and M. Kranzberg, Eds., Technological Innovation: A San Francisco, CA: San 25; Critical Review of Current Knowledge. Francisco Press, 1978. H. F. Kolodny, "Matrix organization designs and new product 26; success," Res. Manag., pp. 2 9 - 3 3 , Sept. 1980. 2 7 G. B. Korte, " C A D D requires radical organizational changes," Comp. Graph. Today, vol. 4 , no. 12, Aug. 1984. 28; J. F. Lardner, "Integration and information in an automated factory," in Proc. Autofact 6, 1984. 29; J. F. Lardner, " A new manufacturing strategy: Reintegration," Operations Manag. Rev., p p . 3 - 8 , Spring 1985. 30] P. A. Lawrence and J. W . Lorsch, Developing Organizations: Diagnosis and Action. Reading, MA: Addison-Wesley, 1969. McKinsey & Company, "Forging CAD/CAM into a strategic weapon," in A Review of Preliminary Survey Findings: Report to Participants, Jan. 1984. 5. Myers and D. Marquis, "Successful industrial innovations," National Science Foundation, Washington, DC, Rep. NSF69-17, 1969. R. R. Nelson, "Research on productivity growth and productivity differences: Dead ends and new departures," J. Econ. Lit., pp. 10291064, Sept. 1981. Office of Technology Assessment, "Computerized manufacturing automation: Employment, education, and the workplace," Washing­ ton, DC, 1984. J. N. Orr, " C A D and C A M far from being integrated," Comp. Graphics Today, pp. 8-9, Sept. 1984. R. F. Ouimette and L. A. Wickstrom, "Improved responsiveness through C A D / C A M , " in Proc. Autofact 6, 1984. C. Pava, Managing New Office Technology. New York: Free Press, 1983. J. Pfeffer, Power in Organizations. Marshfield, MA: Pitman, 1981.

107

ADLER A N D HELLELOID: IMPLEMENTΑΉΟΝ OF INTEGRATED CAD/CAM

[39] [40] [41]

[42] [43]

[44]

[45] [46]

J- Potts, "Computer aided design: A revolution of the m i n d , " Government Data Systems, pp. 3 2 - 3 3 , Nov.-Dec. 1982. N. Rosenberg, Inside the Black Box. New York: Cambridge Univ. Press, 1982. R. S. Rosenbloom, "Technological innovation in firms and industries: An assessment of the state of the a r t , " in Technological innovation: A Critical Review of Current Knowledge, P . Kelly and M. Kranzberg, Eds. San Francisco, CA: San Francisco Press, 1978. R. Rothwell and Paul Gardner, " T h e role of design in product and process change," Design Studies, 1983. A. H. Rubenstein and M . E. Ginn, "Project management at significant interfaces in the R^D/innovation process.'* in Project Management, Β. V. Dean, Ed. Amsterdam, The Netherlands: North-Holland, 1985. H. Salzman, " T h e new Merlins or Taylor's automatons?—The impact of computer technology on skills and workplace organization," draft pap., Department of Sociology, Brandeis University and Center for Applied Social Science, Boston University, Boston, MA, 1985. L. Schneider, R. Howard, and F . Emspak, "Office automation in a manufacturing setting: A case study prepared for the Office of Technology Assessment," Apr. 10, 1985. P. Senker, "Implications of CAD/CAM for management," Omega, vol. 12, no. 3 , 1984, pp. 225-231.

[471 [48] [49] [50] [51] [52] [53] [54]

[55]

H. Shaiken, Work Transformed: Automation and Labor in the Computer Age. New York: Holt, Rinehart, and Winston, 1984. W. E. Souder and Α. Κ. Chakrabarti, " T h e R&D/marketing interface: Results from an empirical study of innovation projects," IEEE Trans. Eng. Manag., Nov. 1978. W. E. Souder, "Managing the coordination of marketing and R&D in the innovation process," TIMS Studies in the Management Sciences, vol. 15, pp. 135-150, 1980. L. Suchman and E. Wynn, "Procedures and problems in the office," Office: Technol. and People, vol. 2, pp. 133-154, 1984. J. M. Utterback, "Innovation in industry and diffusion of technology," Sci., pp. 620-626, p . 183. R. E. Walton and J. M. Dutton, " T h e management of interdepart­ mental conflict: A model and review," Admin. Sci. Quart., vol. 14, 1969, pp. 73-84. R. E . Walton, J. M . Dutton, and T. P. Cafferty, "Organizational context and interdepartmental conflict," Admin. Sci. Quart., vol. 14, no. 4 , pp. 522-542, 1969. S. C. Wheelwright, "Product development and manufacturing start­ u p , " in Manufacturing Issues. New York: Booz-Allen & Hamilton, 1985. S. C. Wheelwright and R. H. Hayes, "Competing through manufactur­ i n g , " Harvard Bus. Rev., Jan.-Feb. 1985.