IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 44, NO. 1, FEBRUARY 1997

67

The Historical Roots of Concurrent Engineering Fundamentals Robert P. Smith

Abstract—This paper explores the history of the ideas behind concurrent engineering from the end of the 19th century until the 1960’s. Concurrent engineering is the relatively recent term that is applied to the engineering design philosophy of crossfunctional cooperation in order to create products that are better, cheaper, and more quickly brought to market. The principles of concurrent engineering that are traced by this paper are: manufacturing and functional design constraints need to be considered simultaneously; combining of people with different functional backgrounds into a design team is a useful way to combine the different knowledge bases; engineering designers must bear in mind customer preferences during the design process; and time to market is an important determinant of eventual success in the market. None of these principles is by itself surprising; concurrent engineering has led to their propagation to many people and firms in the engineering world. The author has examined the engineering literature in order to locate the existence of similar themes in published engineering thought. All of these themes have recurred often in the literature. Concurrent engineering can be seen therefore as a summary of best practice in product development, rather than the adoption of a radically new set of ideas. The paper suggests some reasons that concurrent engineering ideas may not have been adopted more widely. Index Terms—Concurrent engineering, design for manufacturing, engineering design management, history of design.

I. INTRODUCTION

T

HIS paper examines the engineering literature in order to determine the degree to which the ideas contained within the concurrent engineering approach are novel. Concurrent engineering is a term that has been applied since the 1980’s to the product development process where, typically, a product design and its manufacturing process are developed simultaneously, cross-functional groups are used to accomplish integration, and the voice of the customer is included in the product development process. The ideas behind concurrent engineering are simple, but it has proven a powerful product development ideal, with many claims of significant product development successes available in trade publications. This paper is intended to explore several facets of the historical literature related to the concurrent engineering approach. There are several goals of this paper: 1) to determine the existence of the fundamental concurrent engineering ideas in Manuscript received December 15, 1994; revised April 1996. This work was supported by Program in Engineering and Manufacturing Management at the University of Washington. Review of this manuscript was arranged by Department Editor D. Gerwin. The author is with the Industrial Engineering Department, University of Washington, Seattle, WA 98195 USA (e-mail: [email protected]). Publisher Item Identifier S 0018-9391(97)00691-0.

the literature between the end of the 19th century and the 1960’s; 2) to examine the justifications the authors used in advocating the adoption of these principles; to observe the degree to which the various fundamentals were combined such that we can observe a more or less complete concurrent engineering package; 3) to suggest some reasons why concurrent engineering seams revolutionary; and 4) to examine why the adoption of concurrent engineering is causing a significant change in the way firms are accomplishing product development. We observe that there is much similarity between our concept of concurrent engineering and historical best practices in product development. It is worthwhile to note that the basic desires of customers (of value for money) and of producers (for profits) have not changed, and the need for coordination between complex, interacting technologies in order to achieve those goals has not changed. Therefore, the similarities between modern practices and preexisting practices are not surprising. Engineering educators need to be aware of the continuing importance of these issues in order not to allow them to fade into the background, and engineering practitioners can benefit from knowing that the seemingly difficult problems involved in engineering design management are indeed tough, but there is a long history of practices that have evolved to deal with this complexity. The structure of the paper is as follows. Section II describes the fundamentals of concurrent engineering as they are understood today. Section III describes the existence of the perceived need for integrating engineering design with other business functions. Section IV looks at the advocacy of specific mechanisms for accomplishing integration. Section V looks at the existence of specific design-for-manufacturing/design-forassembly (DFM/DFA) rules. Section VI looks at the role of the voice of the customer in the design process. Section VII discusses examples where the entire concurrent engineering philosophy is described in earlier times. Section VIII presents some conjectures about why concurrent engineering ideas were not always used earlier. Section IX discusses the significance of these observations, and Section X provides a summary. II. CONCURRENT ENGINEERING FUNDAMENTALS For the purposes of this paper, the fundamentals of concurrent engineering will be described as the following four. (Alternative definitions of concurrent engineering are available, but this list is a fairly typical one.) They are the increased role of manufacturing process design in product design decisions, the formation of cross-functional teams to

0018–9391/97$10.00  1997 IEEE

68

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 44, NO. 1, FEBRUARY 1997

accomplish the development process, a focus on the customer during the development process, and the use of lead time as a source of competitive advantage [53]. All products have a need to incorporate constraints imposed by the manufacturing process on details of the product design. Depending on which manufacturing process is considered, these effects may be encoded into formal or computer-based rules, or else may be conveyed through individual experience and expertise. Addressing these design concerns early in the development process creates the opportunity to reduce manufacturing costs and improve product quality. The failure to account for these concerns is often due to a functional barrier within an organization between design and manufacturing. Often the method of accomplishing the integration of design with other functions (and removing functional barriers) is the use of cross-functional teams. These teams may include people with expertise in production, marketing, finance, service, or other relevant areas, depending on the type of product. Beside the barrier between design and manufacturing, another important functional barrier is the separation between the engineering designer and the customer. Under the same philosophy as removing the design-manufacturing barrier, the designer can become more responsive to customer desires and thereby create a more successful product. This is known as design-marketing integration. Lead time has proved to be a significant facet of modern competition [10]. By reducing the development lead time a firm is able to respond more rapidly to market trends or to incorporate new technologies. A shortened lead time creates a market advantage for those firms who are able to produce products rapidly. All of these ideas, which are fundamental to concurrent engineering, have been discussed in the literature for many years before the emergence of the concurrent engineering movement. The following sections illustrate how these topics have been discussed by earlier generations of engineers.

III. NEED FOR INTEGRATING DESIGN WITH OTHER FUNCTIONS Product designs have existed for as long as mass production has existed. Early on, there arose a division of intellectual labor whereby the designer was responsible for producing the design and the manufacturer was responsible for making the physical product. As a result of this division there is the possibility of the product designer’s working in ignorance of manufacturing process considerations. The recognition of the problem of the designer’s ignorance of the manufacturer’s constraints is longstanding. The designer has been accused of “throwing the design over the wall” that separates design from manufacturing. A design that is thrown over the proverbial wall is generally difficult and costly to produce and does not necessarily conform to the desires of the market. This functional separation and its resulting adverse effect on the resulting product design may be repeated with other functions (such as marketing, maintenance, or others). The remedy for this situation is to have the designer become more aware of others’ concerns within and without the organization in which he or she works. Engineering writers have long implored

that these barriers be removed. The references below show a continual emphasis on the relationship between the design department and manufacturing and other functions being an essential function of engineering design. We see this imperative by, for example, a turn-of-thecentury designer at Brown and Sharpe [21, p. 593]: The relations between the drafting department and the shop, between the departments of the shop, and between individuals, from the highest to the lowest, is given far too little attention as a factor in economical shop management. A mechanical engineering professor at the Massachusetts Institute of Technology (MIT) in 1921 [19, p. 32]: It is plain that the manufacturing designer must take into consideration every circumstance involved in the production of the commodity. To be successful, he1 must work in close cooperation with all who will be engaged in the development and operation of the manufacturing equipment. This will include the tool designers and the superintendents and foremen of the various manufacturing and assembling departments. An industrial engineering professor at Columbia University in 1941 [81, pp. 158–159]: The design of the product should be based on both the conditions to be met and the characteristics of material, structure and mechanism available for embodiment in the design, all with respect to first cost, operating cost, and the life of the product. And a manufacturing manager at postwar Western Electric [78, p. 36]: Cooperation [between the product design and manufacturing engineering organizations] appears to me to be an essential for best results. It should begin in the early stages of product design and continue as long as the product is being produced. We see similar calls for cooperation by many other authors, such as some from the end of the last century (whose employers are, alas, lost) [16], [85]; European manufacturers [34], [55]; the editors of Forbes and Machine Design [48], [52]; business faculty from schools such as Chicago and Northwestern [67], [68]; engineering faculty at schools such as Penn State and Cornell [42], [45], [86]; manufacturing managers from companies such as Westinghouse and General Motors [37], [46]; and engineers from such companies as Kodak and Magnavox [54], [65]. It seems that every writer discovers the truth in the recognition of the relationship between design and manufacturing anew. In the modern concurrent engineering literature there have been several distinct reasons for claiming that the role of manufacturing concerns in the design process should be increased. The most frequently repeated among these reasons are an increasing level of competition, the role of new manufacturing processes, and the need to reduce development 1 The author recognizes that in this and other quotations there is used what would today be considered sexist language. Each quote is a function of its time and has been left as it was originally written.

SMITH: HISTORICAL ROOTS OF CONCURRENT ENGINEERING FUNDAMENTALS

lead time. All of these justifications for pushing concurrent engineering ideas have deep historical antecedents. These justifications are discussed below. A. Increased Competition One justification given for the need for increased cooperation in the product development process is an increased level of competition. This justification is incongruent with historical facts. There have been claims that the level of competition has increased “recently” at times that we no longer consider recent. Also, business historians have observed that the level and degree of competition were as severe, if not more severe, in previous eras than it is today. For example, in [35], the claim is made that the level of competition has increased since the beginning of the 20th century, and modern firms can not afford to ignore designmanufacturing interaction issues. In [88], the claim is made that competition is much more stringent than it was at the turn of the century. Similarly, in [58], the claim is made that the high level of competition in the 1950’s requires that design and manufacturing personnel cooperate on new product development. Economic competition is now, and has always been, fierce. This is not a new effect. Research on the relationship between innovation and strategy [92] describes the dynamic relationship between innovation and industrial competition. Each industry, as it matures, encounters a time when the number of firms increases, followed by a period when the number of firms is reduced through competitive pressure. These periods of competition vary by industry, but all industries have significant competition, and all periods have their highly competitive industries. The belief that modern competition is no worse than at earlier times is illustrated by other authors. For instance, an article on the history of marketing points out [41, p. 111]: [The era between the 1860’s and the 1930’s] was supposedly a time when steady growth in disposable income caused demand to increase faster than supply. If we could somehow send this description back through time to contemporary business people, they would find it absolutely ludicrous in light of their own trying experiences with the business cycle, overproduction, deadly serious competition, and a social and demographic environment that was changing bewilderingly and rapidly. The business world then was often as tough, and sometimes tougher, than today’s. Furthermore, Fullerton [41] suggests that economic historians generally believe that competition today is no worse than in historical times [24]. B. New Production Methods As new production methods come into service it becomes important for knowledge about the new production processes to affect the resulting product design. Knowledge about these processes must be made available to the product designer. This approach allows designers to take advantage of and respond to the limitations of the new processes. Such knowledge is often resident in the production engineer. Therefore, the situation

69

in which new production processes are used will often be important for ensuring that design engineers work closely with production engineers. One example of the relationship between new production processes and engineering design is described by a Columbia University mechanical engineering professor in 1931 [31, p. 259]: On account of the rapid and steady improvement in production machinery and the development of manufacturing processes, progressive manufacturers are constantly analyzing all production operations with the object of reducing costs. We observe the long-standing notion that new manufacturing technologies increase the need for design-manufacturing integration. We also observe that there have been many new manufacturing technologies developed during the course of this century. Among new manufacturing processes, the development of automatic assembly techniques has been frequently cited as requiring a higher level of integration between design and manufacturing. This requirement is due to the difficulty in developing fully automatic assembly systems, the integrative nature of the assembly process, and the role of the design of the product on the assembly system [71], [96]. The literature of the 1950’s and 1960’s illustrates several examples of the concurrent engineering philosophy in this context. For example, a Harvard professor who studied the adoption of automatic assembly processes noted [73, pp. 199–200]: Perhaps another way of looking at the matter is to say that another major engineering function has evolved, the task of joining products and processes to obtain economical production. The advent of automatic processing alone may not have been the sole cause or even the origin of this development, but it undoubtedly has hastened, or perhaps is only in the process of hastening, its evolution. As well as British engineering lecturers [79, p. 18]: Component design is an important factor if automatic methods of production are to be successful. A change to assembly by automatic or semi-automatic processes often demands redesign in order to facilitate the correct feeding and assembly of the components. Similar claims are stated by others in academia [15, pp. 95–97], [28, p. 53] and industry [27, p. 5], [37], [58]. New manufacturing processes are being developed continually. Each manufacturing process, when it is new, requires close cooperation between the designer and the manufacturing engineer. The automation of the assembly process focuses these needs, but the advent of automatic assembly equipment is not new. C. Lead Time One of the prime motivations for a concurrent engineering approach to product development is a desire to shorten the total time that it takes to bring a product to the marketplace. The notion that the length of the development cycle can provide

70

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 44, NO. 1, FEBRUARY 1997

an important competitive advantage, and that addressing all aspects of the design problem simultaneously might lead to a shortened development cycle is a long-standing precept. The claim is made by a group of MIT business faculty that the quickening pace of life early this century leads to a need for shorter lead times [35, p. 311]: Life quickened [around the turn of the century] and complexities developed at such a pace that the problem of making product design cope properly with changing conditions now involves the complete depiction and development of a product from the original idea to the point of use. as well as the observation of a British industrial survey that points to shorter lead times in the post-war period [51, pp. 19–20]: It is said that the design function has not changed. That is true. But the time factor has changed considerably. It is changing and shortening all the time. My first point concerns planning and co-ordination, which require that the whole design team and management should ensure that all designing and ancillary arrangements that can should go on in parallel rather than in series. It may be said that this is obvious; most of my comments are comparatively elementary, yet all too often these ideas are lacking in many firms. Similar claims are made by many authors [15, pp. 95–97], [58, pp. 406–408], [59, p. 88], [66, p. 83], [75, p. 79], [85]. In summary, there are several potential justifications for why a concurrent engineering approach might be of importance: economic competition, new production processes, and a shortened lead time. None of these phenomena are new. These are real, but long-standing, justifications for cross-functional cooperation. This section has documented the existing needs; the next section describes how such cross-functional cooperation has been implemented. IV. MECHANISMS FOR ACCOMPLISHING INTEGRATION Once the need for cross-functional integration is recognized, it remains a difficult managerial challenge to accomplish that integration. Several organizational methods that have been proposed for accomplishing the integration are: 1) requiring approval by other departments, 2) establishing a liaison department that is responsible for coordinating the activities of other departments, 3) forming all interested parties into one cross-functional team to ensure integration, and 4) using job rotation to ensure that functional cross pollination occurs. A. Requiring Approval by Other Functions One method for attempting to ensure that the produced design corresponds to the wishes of multiple functional departments is to require that each interested department show its approval. This is known as the “sign-off” type of coordination. It is a potentially weak form of cooperation since there is no mechanism to ensure that all departments will be consulted at an early stage. By the time a department is contacted for its approval many decisions may already be irrevocable

and ability to change may be limited. Nevertheless, this is a potentially useful mechanism to create cross-functional information flow and cooperation [57, p. 206]: The logical method of resolving the various conflicts of interest is to have all interested divisions a party to the final decision. As a matter of fact, the design engineer may well consult the methods department and the manufacturing division as he progresses. Such coordination may easily result in modifications of design that do not interfere with the basic operation of the product yet make possible the use of present equipment, thus avoiding unnecessary expenditures for new equipment or later changes in design. The production department often may suggest modifications in design that result in marked savings in manufacturing costs. By working closely with the sales department the design engineer will have the benefit of the practical customer reaction as well as the enthusiastic support of the sales group in marketing a product for the design of which they feel some responsibility. The purchasing department may render valuable suggestions regarding economies in purchasing certain materials or parts that may be specified especially in terms of standards and dimensions that are used in the trade. Such an approach is described at RCA [40, p. 212], Western Electric [1], and Bausch and Lomb [6], as well as by academics [3, pp. 20–21]. In addition, obtaining approval from other departments, notably the purchasing department [94], sales, and service [52] has also been described. Published accounts of using the “sign-off” technique exist in the twenties and the thirties. Either it became standard practice subsequently and was not worthy of comment in the technical press, or it was supplanted by the growing techniques of cross-functional teams, or both. B. Using Liaison Personnel Another method of increasing the amount of coordination between design and manufacturing functions is to use liaison personnel. Liaison personnel are not members of any functional piece of an organization, but rather people who are capable and prepared to address issues that span functional organizational boundaries. Liaison personnel have as their fulltime job the coordination of the disparate functions. Under this approach, they become the primary modes of accomplishing information transfer between functional areas [35, pp. 318–319]: There always exists a chasm between shop and design groups into which goes much waste. The introduction of new products and changes in old products as they go into manufacture are problems for the highest type of special personnel which acts in the specific capacity of liaison between designers and shop supervisors. Similar descriptions are given by practitioners [34], [43], [50], [69] and by academics [22], [28], [60], [70], [73]. The use of liaison personnel in order to accomplish crossfunctional integration is not novel. The discussion of this technique occurred from the 1920’s through the 1960’s.

SMITH: HISTORICAL ROOTS OF CONCURRENT ENGINEERING FUNDAMENTALS

C. Meetings Between Representatives of Different Functions Another method that is used to ensure the cooperation of different functions is to have people representing each functional area meet regularly (typically weekly) to discuss items that are of boundary-spanning or general interest with regard to a development project. Meetings of this type enable an organization to maintain a functional structure while achieving a greater level of cross-functional interaction. These meetings have been advocated throughout this century for resolution of design-manufacturing and other coordination problems. Two distinct methods for applying the cross-functional meeting approach are apparent in the literature. In the first case the meetings are conducted by division heads, chief engineers, or other senior personnel, since they are able to make decisions affecting many people. In the second case the meetings are held among the personnel who are responsible for day-today progress on the project, since they are fully aware of the relevant issues as they relate directly to the project. Both of these philosophies are apparent in the literature. The approach of having interdepartmental meetings at a high level exists throughout the literature studied. This practice is not a typical approach to modern concurrent engineering practice, but rather is an attempt to solve organizationally a problem that is closely associated with concurrent engineering [79, p. 18], [62, p. 6]: New designs should be given the widest consideration and be carefully steered through the preapproval stages; one method of doing this is by the appointment of a Design Committee. The members of a typical committee would consist of senior representatives of the Research Department, Design Department, Production Engineering Department, Shop Superintendent, Quality Control Department and, possibly, the Production Control Department. The committee meets regularly until the design is successfully to production. If the product is for a specific industrial market, it may be advantageous to invite the customer to send a representative to these meetings. The two principle areas of technical creative activity are design and production. Design and production are closely interrelated. This fact has led to the practice, now common in many works, of bringing the responsible executives together from time to time for an interchange of experience with the design engineers. Other advocates have included practitioners [34, p. 104], [52, p. 28] and academics [58, p. 405], [97, p. 27]. Alternatively, the approach of having regular interdepartmental meetings of people who are lower on the organizational ladder is also an attempt to accomplish cross-functional integration. This approach is closer to current concurrent engineering practice, as even today it is debated whether it is appropriate to have a full-time product development team, or whether there should be a looser cross-functional committee with representatives from different functional areas. The committee approach is described by engineers in the 1920’s [5, p. 39]:

71

In all of the activities we have mentioned it is essential that the engineering department work closely with the shop and purchasing authorities. Sometimes informal contact when specific points come up will do, but this method offers chances for consultation to be overlooked. Regular meetings of representatives from each of these departments often work well and tend to systematize the contact. By an executive of the Dexter Folder Company [83, p. 8]: In our plant, we obtain close contact between the manufacturing and designing heads by a processing committee which meets from time to time to discuss processing for economical manufacture. In addition to the designing executive, there are represented on this committee the shop superintendent, the chief inspector, the tool supervisor, and the production manager. We have found the contacts obtained through these meetings to be invaluable in educating the designing department to recognize shop problems, and to design so as to minimize them. And in a survey of European engineering practices of the 1960’s [75, p. 73]: Coordinated co-operation in connection with the development of a new product to go into long run batch production is safeguarded by the utilization of committees. There is also discussion of the use of committees to accomplish cross-functional cooperation by industrial practitioners [55, p. 560], [82], management and engineering consultants [59, p. 88], [93, p. 4], and academics [7, p. 169–170], [31, p. 57], [76, pp. 126–128]. We see that the role of meetings to coordinate departments has been a feature of engineering management practice for many years. There was not agreement on whether the meetings should be at a high or low level of the organization. D. Using Cross-Functional Teams to Accomplish Integration The preeminent organizational technique for accomplishing concurrent engineering is the cross-functional design team. Under this approach, the design team is composed of experts from engineering, production, marketing, and any other functional area that has a vested interest in the development project. The team is formed to work on a specific project, and stays together throughout the development of the product. This approach seems more recent than high-level committees and liaison groups, as the bulk of its advocates wrote in the 1940’s–1960’s. The cross-functional team approach is described at Western Electric [46, p. 23]: Close working arrangements [between engineering and manufacturing] is done through the assignment of one or more manufacturing engineers to work with the design group in analyzing the alternative constructions that might be employed and to cooperate in cost analysis and in the building of models. On wartime aircraft production [90, p. 18]:

72

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 44, NO. 1, FEBRUARY 1997

Close cooperation with the Production Department during the development of a design is essential if the maximum degree of production efficiency is to be obtained. In fact, it is considered good practice to have one or more Production Planners stationed within the Engineering Department, to work with the designers and approve each drawing for adaptability to production and feasibility of manufacture. And on a research project examining industrial adoption of automation [73, p. 191]: In one relatively large project, a somewhat different type of committee, a task force group, was organized, including representatives of product design and production engineering. The task force members were relieved of their regular duties and moved physically to a new large office in order to undertake preparations to manufacture a new product. The group worked together continuously until the new product went into production, when it was disbanded and the individuals returned to their permanent job assignments. Since the product design had been frozen before this group was formed, the extent of the coordination was more limited in scope than it might have been, but the members of this task force were most enthusiastic about the degree of coordination achieved and the ease with which it was obtained. There are also discussions on the use of cross-functional teams at General Motors [37] and Johns–Manville [80, p. 53] as well as by engineering and business researchers [15, pp. 95–97], [56, p. 165], [60, p. 110], [61], [63], [70, p. 44]. One organizational structure that is associated with the advent of the cross-functional team is the matrix organization. A matrix organization allows functional specialization to be preserved while cross-functional teams are able to focus on finishing a focused project. There is a description of a matrix organization as practiced by Westinghouse in [46, pp. 26–27], including organizational charts and advantages and disadvantages of this organizational structure. The idea of using cross-functional teams extends back to the 1940’s. This idea has appeared in the literature regularly after its original appearance. We cannot therefore associate the cross-functional team with the rise of the concurrent engineering movement; its application extends to a significantly earlier period.

V. DFM/DFA RULES Designing a product so that it can be easily manufactured is an obvious advantage for a product. In recent years this approach has been referred to as DFM and a number of specific design rules have been suggested that, if followed, make the manufacturing processes straightforward. In particular, attention has been paid to the assembly process as a result of its presence in virtually every engineering industry and the degree to which every design decision is reflected in the assembly process. DFA is the name of this technique, which has been developed and spread by Boothroyd and Dewhurst [13] and others.

Examination of the literature shows us that not only was the philosophy of DFM/DFA long-standing, but specific design rules that mirror modern instances of DFM/DFA design rules can be found in older literature sources. First, consider the existence of DFA rules. The rules primarily involve reducing the number of parts, simplifying the part mating and securing processes, and creating symmetry or asymmetry so that it is difficult to put the parts together in any manner but the correct way. This simplification has the effect of reducing direct assembly costs, and often tends to reduce indirect costs such as incoming inspection and parts inventories. We see these design rules and justifications discussed in many sources [3, pp. 13–15]: One of the first ends which [the factory man] seeks in analyzing the product is to reduce the number of parts of which the automobile is constructed. [An] excellent example [is a] study made at the National Cash Register Company of a problem of this kind. Before the analysis of the product was made, 44 parts were required to make the key assembly in the illustration [not shown]. After the problem had been studied it was found possible to reduce the number of parts to three, a net saving of 41 parts. The saving from such a change in design come in several different ways. In the first place, there is the lessened material requirement, for less steel is actually consumed in making the parts. This means that smaller quantities of steel must be purchased and hence a smaller inventory is needed. Not so much space is required in the storeroom to store the raw material till it is required, and as a result storage costs are reduced. After the parts are made the effect of the storage problem is felt to a much greater extent, for there are only three parts to be stored instead of 44. Not only is less space required, but very much less labor on the part of the storekeeper and his assistants is necessary for handling the parts. The number of store records is also greatly reduced, and with the smaller number the chances for errors are minimized to a considerable extent. As the number of parts is cut many of the operations needed to make them are also eliminated, and in this way the cost of production in so far as direct labor and earned burden are concerned is reduced. More machine hours are made available for other tasks, and thus the capacity of the factory is virtually increased. Some of the machines which were formally required are no longer necessary at all, and, if not needed for other purposes, can be eliminated altogether. The whole problem of planning, scheduling, and dispatching the operations for making the subassembly is greatly simplified, and the number of records involved and labor entailed in the Planning Department are cut appreciably. All this means greatly increased ease in control and in manufacture in the shops and greatly decreased costs. There are similar descriptions of the importance of designing for assembly and rules for accomplishing assemblyoriented design in other works [23], [33], [39], [62], [72], [73], [75], [95].

SMITH: HISTORICAL ROOTS OF CONCURRENT ENGINEERING FUNDAMENTALS

Similarly, there have been design rules developed for many other manufacturing processes. Welding, casting, machining, and other common processes, because of their frequent use in engineering industries, have a large body of design rules. The older DFM design rules are in many cases similar to lists of such design rules today. Many authors have given manufacturing-oriented design rules [3, p. 18], [5], [11], [18, p. 444], [25], [26], [36, pp. 695–696], [38], [39], [64], [72], [73], [75, pp. 65–66], and [89]. These references target a variety of manufacturing technologies, but all of them specify ways that a design can be improved by paying attention to its manufacturing process early in product development. VI. INCLUDING THE VOICE OF THE CUSTOMER IN THE DESIGN PROCESS The ultimate goal of a development project is to create a product that makes a profit. The role of the customer who will pay for the product needs to be addressed during development (using, for example, the quality function deployment approach [44]). There is, however, the risk that development personnel become isolated from current market conditions and produce products that do not respond adequately to market desires. There have been continual statements to the effect that development personnel have become removed from market wishes. Several authors suggest that mechanisms be put in place to ensure that market wishes reach the development staff. These mechanisms run the gamut of organizational approaches described in Section IV. The important thing to point out here is that not responding to market needs is a continuing shortcoming in engineering design and development, and has long been recognized as such. As an expression of the need for marketing knowledge in the product development process, the editor of Product Engineering stated [14, p. 675]: Machine designers, and engineers responsible for the engineering development of many kinds of metal products, should have awakened to their responsibilities in putting sales appeal into their product long before they did. Similarly, a marketing professor from MIT observed [32, p. 544]: It is necessary to develop means for ascertaining with some degree of precision what consumers’ real wants are, in order that they may be correctly anticipated by manufacturers. Given ways to lean what consumers’ real wants are, the organization should be set up as to be able to translate them into production in an economical fashion. There are similar observations by academics [2] [3], [29], [70], [76], [84] and practitioners [4], [7], [40] over the last 80 years. We see many authors extolling the virtues of having engineers incorporate market wishes into the design of the product. VII. COMPLETE CONCURRENT ENGINEERING APPROACH It could be argued that what is new about concurrent engineering practice is not the adoption of any individual

73

element of this package, but rather the adoption of all of this approach, and its synthesis into a novel method for product development. The alternative to concurrent engineering would be to have strong organizational separation between design, manufacturing and marketing, and/or to separate the functional design of the product from production design and manufacturing process design. However, we do observe several writers from the 1950’s and 1960’s integrating the full range of concurrent engineering practices. These works describe the importance of manufacturing processes on product design, recognize the need to formally incorporate marketing analysis in the product development process, and suggest organizational approaches to accomplishing integration (such as cross-functional teams). As described by a manager of new product development at Magnavox, the necessity of and the techniques for concurrent engineering practices were recognized in industry by the 1950’s [54, pp. 112–114]: The principal [sic] reasons, then, for integrating industrial and production engineering into the development program are (1) to cut costs and (2) to help get production started on schedule. The second reason can be just as important—and, in some cases, even more important—than the first. The market research function should be integrated into the development program in its initial stages and should continue to function throughout the life of the resulting product. The integration of all these diverse functions cannot be achieved by waving a magic wand; each of the various organizational elements involved must be carefully combined with all the rest to form an effectively functioning team. Serious consideration should be given to assigning this task to a new-product development group, so placed organizationally that it can operate effectively throughout the company. In addition, we see similar complete packages in the British Ministry of Supply for aircraft development [58], research done at Harvard Business School on industrial practice [60], [61], [73], surveys of best practice in industry [51], [75], and industrial engineering faculty members at Pennsylvania State University [72]. The complete package of concurrent engineering existed at that time. It was a recognized technique, even if the title concurrent engineering did not exist. VIII. POSSIBLE CAUSES OF LACK OF ADOPTION It is possible to offer a number of conjectures as to why concurrent engineering did not become the dominant product development strategy earlier, but it is difficult to determine which of these conjectures are valid. This section introduces some conjectures and presents evidence concerning their credibility. Conjecture One: New manufacturing processes make cooperation between manufacturing process development and product development more necessary than under previous manufacturing technology. There have been changes in technology that may, in recent years, have made a concurrent engineering approach both more

74

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 44, NO. 1, FEBRUARY 1997

possible and more necessary. New manufacturing processes have been developed (for example, in electronics, flexible machining, and materials processing). It is conceivable that these changes in the technology have increased the need for concurrent engineering. If this conjecture is to hold, we must explain why current manufacturing processes make concurrent engineering more necessary than earlier manufacturing processes. Differences between modern manufacturing processes and older processes are not apparent. There is no reason a relatively new manufacturing process like composite structures would require more design-manufacturing cooperation than an older technology such as welding needed in the early years of its application. As pointed out in Section III-B, assembly is a particularly difficult operation to automate, requiring a high level of cooperation during the design phase between many participants. It has been suggested that the advent of flexible assembly methods has heightened the need for integration between design and manufacturing [71]. Robots and other automatic assembly equipment require more forethought into the assembly process, which focuses attention on the underlying product design. I would suggest that the argument that assembly is singularly important is valid, but focusing on recent developments in flexible assembly equipment does not acknowledge other longstanding aspects of design-manufacturing integration related to assembly. As the literature on assembly points out, benefit from integration of design and manufacturing exists for both automatic and manual assembly methods. The references in Section VII that discuss many facets of concurrent engineering practice in the 1950’s and 1960’s are concentrated on automatic assembly. Assembly does have a central role in the need for cross-functional cooperation, but neither the need nor its solution is novel. There are always new manufacturing processes. It may be true that newer manufacturing processes require greater design-manufacturing interaction during the first years of their application, but as a manufacturing process becomes well understood the design needs may stabilize and no longer require extensive cooperation. Conjecture one seems unlikely to provide a justification for increasing the need for crossfunctional cooperation only in recent times. Conjecture Two: Current communication and other enabling technologies have lowered the costs of cross-functional cooperation, thereby encouraging such behavior. This conjecture suggests that new technologies have affected firms’ behavior in adopting a concurrent engineering approach by making cross-functional cooperation easier. Better methods for electronic communication and better databases are available. Easier exchange of electronic data is possible. These changes make the actual work in doing concurrent engineering easier, which might overcome some of the difficulties with adopting concurrent engineering. The existence of new communication and computational technologies is undoubted. These technologies include CAD/CAM systems, integrated databases, electronic communication tools, and others [77]. These technologies enable the rapid exchange of information, and the application of a range

of analytical frameworks that are necessary in cross-functional design work. To determine whether these technologies are able to explain the novelty of concurrent engineering practices, we must examine whether it would be possible to implement those practices in the absence of the technologies. In Section VII, it is argued that a number of firms had applied concurrent engineering practice before the advent of modern technologies. Communication and computational technologies are not necessary for the adoption of concurrent engineering practices. If the manufacturing plant and the designers were near to each other, and suppliers were also within a short distance, then the telephone and other technologies available earlier this century could be sufficient to coordinate plans, have cross-functional groups, establish and implement DFM rules, and accomplish much of the structure of concurrent engineering practice. Conjecture Two nevertheless seems to have some merit. I would suggest that the new technologies are neither necessary nor sufficient for cross-functional cooperation, but do serve to lower the barriers against the adoption of such cooperation and have fostered the types of communication and analysis that occur in the product development process. Conjecture Three: A functionally separated organization makes it difficult to implement concurrent engineering mechanisms. There are several considerations that might have caused hierarchical, functional organizations to discourage the types of cross-functional cooperation associated with concurrent engineering. First, in an organization that encourages crossfunctional work the lines of control may become blurred. Second, more communication and coordination work becomes necessary, which has the potential to distract from the fundamental (technical) work of the employees. These difficulties may have led to the lack of adoption, perhaps by default, of a concurrent engineering approach. Evidence for this conjecture comes from some writers who explicitly advocated using purely functional organizations because of the potential confusion that would ensue if a less rigid organizational structure were employed. An educator from LaSalle University advised [30, p. 7]: Do you ever stop to wonder why so many manufacturing organizations are divided into engineering and production groups, each reporting to and the work coordinated, by the general manager? Is there any real value in the written plans and schedules by which production programs are guided? Is there any truth in the assertion that some men are thinkers and others doers; that each has his place in industry and that it is difficult for most men to cross the invisible line that separates their respective spheres of activity? The truth is that the scheme of such organization for production works, and so the answer to the last two questions must be affirmative. Another writer commented in American Machinist [17, pp. 234–235]: Lately, we have come to hear a great deal about cooperation as a possible cure-all for organization troubles.

SMITH: HISTORICAL ROOTS OF CONCURRENT ENGINEERING FUNDAMENTALS

Co-operation in principle is all right, but as a work and a force it is being promiscuously used, and is becoming a menace through its misuse. The dangerous feature of co-operation is the tendency to use it as an easy way out of difficulties that require managerial ability to master. Another feature of organization that is much in vogue is the conference Like cooperation, the conference principle is a dangerous factor to deal with and unless tactfully handled will set up more ills than it will cure. A conference would lead to discussions and arguments of the merits of each individual’s thought as to how he would have done the work if it had been assigned to him. Some benefit might be derived but the chances are strongly toward disruption, jealousies, and delays, all of which are fatal to the well-being of any organization. Similarly, there is the belief expressed by a mechanical engineering professor at MIT that a simultaneous approach to engineering design might fail to address some engineering issues, and a slower, more methodical approach is likely to be more successful [20]: Effective machine design involves two distinct tasks. The first is the functional design, the second the production design Today, few plants recognize this two-fold nature of design. Too often the attempt is made to solve simultaneously the two types of problems involved. It is seldom that this is done adequately. The suspicion of these authors is that integrating functions will confuse the organization, and it will lose its singular focus and become unproductive. The statements quoted above are the only such statements found in the literature; the bulk of the references that discuss the product development process do not share the outlook of these authors. These pieces of evidence lead to the possibility that the choice of a more hierarchical approach over concurrent engineering was a conscious choice. It is not clear if these were minority or mainstream ideas strictly from the sources. Apart from the suggested ease associated with managing functional organizations, there is another reason that this type of organization might be preferred. Functional organizations are better able to develop and retain advanced expertise in specific technical areas [91]. This expertise typically comes at the expense of cross-functional coordination and product development lead time. If this mechanism were able to explain the predominance of functional organizations we would have to accept that most firms were technologically oriented and did not focus on lead time as a competitive factor. Another consideration is that changing to concurrent engineering involves changing the organizational culture, which is an inherently difficult process. For example, changes that reduce the power base of existing functional managers are likely to be resisted by those managers. While true, this observation does not explain how sequential engineering arose in the first place to become the dominant philosophy, but may explain why it remained entrenched, even in the face of growing evidence of the superiority of concurrent engineering ideas.

75

It is possible that firms chose not to adopt concurrent engineering practices based on rational behavior. Conjecture three suggests that concurrent engineering practices were not easily applied in existing functional organizations. This conclusion does seem at odds with the evidence presented in Sections IV–VII that demonstrates the application of concurrent engineering tactics in a variety of environments. I believe that functional organizations are not incompatible with concurrent engineering practices, and that best practice in engineering design has always worked to structure organizational mechanisms to account for cross-functional interactions. Conjecture Four: Engineers have been undertrained in product development, and therefore did not learn about best practice, so best practice did not spread even though it had been developed. There are multiple ways that the educational and training systems of this century may have foiled the earlier development and spread of concurrent engineering ideas. First, emphasis in the last 50 years within engineering education has been strongly on engineering science rather than applied fields such as product development and manufacturing [87]. The emphasis on engineering science may mean that several generations of engineers were undertrained in product development and therefore less likely to use sophisticated product development processes. The existence of the engineering science emphasis in the postwar era does not explain the lack of integrated product development attitudes prior to the 1940’s. During the first half of the century engineering education focused on engineering practice, so we might expect greater training on design in this period, which should have focused students’ attitudes on product development issues. In order to investigate Conjecture four further it would be necessary to study the history of engineering education, which is outside the scope of this paper. Conjecture Five: Competition increasingly focuses on lead time, which leads to increased benefits for concurrent engineering organizations. There may have been changes in the strategic environment that have led to an increased need for and acceptance of concurrent engineering ideas in recent years. The nature of competition between firms may have changed. There may be a decreasing willingness on the part of customers to accept below par goods. There may be an increasing focus on using lead time as a source of competitive advantage. As described in Section III-C, there is little reason to believe that present times are more competitive than previous eras. Competition between firms is not a new idea; competition based on value to the customer and the incorporation of the latest technologies is fundamental. We have to be careful before we attribute the rise of concurrent engineering purely to the increasing role of competition.

IX. DISCUSSION There is relatively little written about the history of the interaction between product development and manufacturing in this century, and how organizations have solved those sorts of problems. There is only one other paper discussing the

76

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 44, NO. 1, FEBRUARY 1997

history of concurrent engineering of which the author is aware [98]. That paper discusses how concurrent engineering ideas were applied during World War II, and describes some of the products that were developed under those methods. The authors point out that some of the ideas were used in prewar projects at the same companies. They also claim that with the end of the war the application of concurrent engineering ideas became significantly less prevalent until rediscovered during the late 1970’s–early 1980’s. Another author has looked at current engineering methodology and its relationship to previous engineering design practice [8], [9]. That work concentrates on the idea that modern developments in engineering design (such as CAD/CAM systems and the related cooperation between manufacturing and product development) are closely related to ideas that were developed in the 19th century. The close cooperation between artisans that was natural in the 19th century is being rediscovered in modern DFM practice. According to that work, this century has been characterized by the development of functional separation that has inhibited close cross-functional cooperation. In contrast, I have argued that the ideas are traceable continuously from the beginning of the century until the 1960’s. My belief is that the ideas have existed all along, and have always been applied by companies that were using a sophisticated approach to product development. In comparison to the relative lack of information about the history of engineering design practices, several researchers have documented the history of manufacturing technology development [49], [74]. In these histories it is acknowledged that the technologies being used have changed significantly, but the authors do not discuss the interaction between process development and product development; nor do they discuss how changes in manufacturing technology have affected managerial approaches (cross-functional interaction and so forth). A similar study has pointed out that much of the success in applying mass production techniques in the 19th century relied on designing the product in accordance with necessary manufacturing constraints, but that study did not look at organizational structures used to accomplish that goal [47]. Although the literature review undertaken attempted to be exhaustive, I must acknowledge the shortcomings of this approach. This paper examines primarily the U.S. literature on the management of the mechanical engineering design process. The review may have missed important revelations in other fields or other countries. Also, we cannot know if the ideas published by these authors were minority opinions or were for some other reason disregarded. We can observe that the publications cited are predominantly mainstream research and trade publications, the books are published by major publishing houses, and the authors represent leading companies and universities. We can, therefore, suspect that these were widespread ideas. Furthermore, there is no way to determine the reason that a particular theme was not discussed at any particular time or in any given work. All that we can be certain of is that the publication of a particular idea at a particular time ensures that the idea was recognized at that time as being of importance by that author.

The historical patterns that we observe are clear. There are well-expressed and long-standing needs for cross-functional cooperation, and many organizational mechanisms for accomplishing those ends. As it came to the fore in the 1980’s, the concurrent engineering movement nevertheless appeared to be a significant change in the ways of doing business. How those seemingly incongruous facts can be reconciled is discussed in the conjectures of Section VIII. I believe that the evidence suggests that Conjecture Two (on technology lowering the barriers for cross-functional cooperation) and Conjecture Four (on lack of training in best practice in product development) are valid, Conjecture five (on lead time as a novel focus of competitive pressure) and Conjecture one (on whether novel manufacturing processes require more crossfunctional cooperation) can be rejected, while Conjecture three (on whether functional organizations discourage crossfunctional cooperation) provides a likely truth without being able to explain the lack of adoption of concurrent engineering practices. I suggest that the concurrent engineering ideas have existed for a long time, but were not put into practice both because older methods seemed easier and because the educational system did not advocate sufficiently a change to preexisting practices. Educators and researchers have the duty to ensure that this does not happen again. X. SUMMARY We observe that many of the needs for and methods of accomplishing cross-functional integration have deep historical roots. Although the unification and popularization of these ideas under the label of concurrent engineering has only occurred in recent years, we can see that the important pieces of this puzzle have been in place for a long time. Good engineering design practice has always been, and will always be, the inclusion of the customer, the manufacturing process, and all other relevant functions into the product development process. ACKNOWLEDGMENT The author wishes to thank P. Niland, the two anonymous reviewers, and the department editor for their helpful comments. He also wishes to thank W. von Hansen and J. Naor for their help in obtaining items for the bibliography. REFERENCES [1] J. L. Alden, “Machine-design management,” Trans. Amer. Soc. Mech. Eng., vol. 54, no. MSP-54-9, pp. 105–118, 1932. [2] L. P. Alford, Cost and Production Handbook. New York: Ronald, 1934. [3] P. M. Atkins, Factory Management. New York: Prentice-Hall, 1926. [4] A. J. Baker, “The sales engineer and his relation to production and machine design,” J. Amer. Soc. Mech. Eng., vol. 38, no. 11, pp. 855–858, 1916. [5] W. R. Basset and J. Heywood, Production Engineering and Cost Keeping. New York: McGraw-Hill, 1922. [6] C. L. Bausch, “Coordination of research and engineering with production and sales,” Trans. Amer. Soc. Mech. Eng., vol. 54, no. MAN-54-4a, pp. 21–24, 1932. [7] L. L. Bethel, F. S. Atwater, G. H. E. Smith, and H. A. Stackman, Industrial Organization and Management. New York: McGraw-Hill, 1945.

SMITH: HISTORICAL ROOTS OF CONCURRENT ENGINEERING FUNDAMENTALS

[8] I. Black, “Back to the future with CAD: Its impact on product design and development,” Design Studies, vol. 11, no. 4, pp. 207–211, 1990. [9] , “Product design in British manufacturing—New processes from old practices,” Proc. Inst. Mech. Eng., vol. 208, no. B2, pp. 81–87, 1994. [10] J. D. Blackburn, Time-Based Competition: The Next Battleground in American Manufacturing. Homewood, IL: Business One Irwin, 1991. [11] R. W. Bolz, Production Processes: Their Influence on Design. Cleveland: Penton, 1949. [12] W. H. Booth, “The relation of the draughtsman to the workshop,” Eng. Mag., vol. 24, no. 1, pp. 98–104, 1902. [13] G. Boothroyd, P. Dewhurst, and W. Knight, Product Design for Manufacture and Assembly. New York: Dekker, 1994. [14] G. S. Brady, “Product design for increased utility and improved marketability,” Mech. Eng., vol. 53, no. 9, pp. 675–676, 1931. [15] J. R. Bright, Automation and Management. Boston: Harvard Bus. Sch., 1958. [16] W. H. Bryan, “The relations between the purchaser, the engineer, and the manufacturer,” Trans. Amer. Soc. Mech. Eng., vol. 19, no. 774, pp. 686–694, 1898. [17] P. H. Bryant, “Some comments on engineering department organization,” Amer. Machinist, vol. 63, no. 8, pp. 233–235, 1925. [18] E. Buckingham, “Influence of design on production,” Mech. Eng., vol. 48, no. 5, pp. 442–444. 1926. [19] , Principles of Interchangeable Manufacturing. New York: Industrial, 1921. [20] , Production Engineering. New York: Wiley, 1942. [21] L. D. Burlingame, “The drafting department as a factor in economical shop management,” Eng. Mag., vol. 27, no. 4, pp. 589–604, 1904. [22] T. Burns, “Research, development and production: Problems of conflict and cooperation,” IRE Trans. Eng. Manage. vol. EM-8, no. 1, pp. 15–23, 1961. [23] J. Calder, “The assembly of small interchangeable parts,” Trans. Amer. Soc. Mech. Eng., vol. 33, no. 1312a, pp. 195–209, 1911. [24] A. D. Chandler, “The emergence of managerial capitalism,” Bus. History Rev., vol. 58, pp. 473–503, 1984. [25] G. O. Clifford, “Planning and coordination,” Cutting Tomorrow’s Costs. New York: Amer. Manage. Assoc., production series no. 168, pp. 3–13, 1947. [26] H. G. Conway, “Design and productivity,” in Institution Mech. Eng. Proc., 1963, vol. 177, pp. 1181–1200. [27] R. E. Cross, “Automation tomorrow,” Amer. Soc. Tool Eng. Collected Papers, no. 24T34, 1956. [28] J. Diebold, Automation. New York: Van Nostrand, 1952. [29] H. Diemer, “Management ABCs: 3. Engineering and research,” Factory Indust. Manage., vol. 81, pp. 991–993, 1931. [30] , “Relation of the engineer to management,” Prof. Eng., pp. 5–7, Jan. 1926. [31] F. L. Eidmann, Economic Control of Engineering and Manufacturing. New York: McGraw-Hill, 1931. [32] R. F. Elder, “Product design for the market,” Mech. Eng., vol. 54, no. 8, pp. 543–546 and 565, 1932. [33] H. Ellis, “The process of assembling a small and intricate machine,” Trans. Amer. Soc. Mech. Eng., vol. 33, no. 1312b, pp. 211–231, 1911. [34] H. Fayol, General and Industrial Management. New York: Pitman, 1949 (transl. Administration Industrielle et G´en´erale. Paris: Dunod, 1916). [35] K. D. Fernstrom et al., Organization and Management of a Business Enterprise. New York: Harper, 1935. [36] R. E. Flanders, “Design, manufacture and production control of a standard machine,” Trans. Amer. Soc. Mech. Eng., vol. 46, no. 1933, pp. 691–713, 1924. [37] W. A. Fletcher and W. C. Edmundson, “Product and production engineers join forces to make new manufacturing techniques succeed,” General Motors Eng. J., vol. 1, no. 9, pp. 2–5, 1954. [38] F. E. Folts, Introduction to Industrial Management. New York: McGraw-Hill, 1938. [39] G. Frenz, Kritik des Taylor-Systems. Berlin: Springer-Verlag, 1920. [40] A. W. Frey, Manufacturers’ Product, Package and Price Policies. New York: Ronald, 1940. [41] R. A. Fullerton, “How modern is modern marketing? Marketing’s evolution and the myth of the ‘production era’,” J. Mktng., vol. 52, pp. 108–125, 1988. [42] C. C. Furnas, Research in Industry: Its Organization and Management. New York: Van Nostrand, 1948. [43] J. F. Hardecker, “The functions of a project engineer in a technical organization,” Amer. Machinist, vol. 64, no. 16, pp. 641–642, 1926. [44] J. R. Hauser and D. Clausing, “The house of quality,” Harvard Bus. Rev., pp. 63–73, May/June 1988. [45] D. B. Hertz, The Theory and Practice of Industrial Research. New York: McGraw–Hill, 1950.

77

[46] J. M. Hipple, “Coordination of research and development with other specialized departments in manufacturing,” Research and Development Departments. Amer. Manage. Assoc.: Production Executives’ Series, no. 78, pp. 17–31, 1929. [47] D. R. Hoke, Ingenious Yankees: The Rise of the American System of Manufactures in the Private Sector. New York: Columbia Univ., 1990. [48] C. Hoskins, “There is no bottom to costs,” Speeding Up Inventory Turnover: Meeting the Market Price. Amer. Manage. Assoc.: Production Series no. 6, pp. 27–35, 1933. [49] D. A. Hounshell, From the American System to Mass Production 1800–1932: The Development of Manufacturing Technology in the United States. Baltimore: Johns Hopkins Univ., 1983. [50] H. E. Howard, “Coordination between research and engineering departments,” Industrial Eng. Better Production. Amer. Manage. Assoc.: production series no. 153, pp. 11–13, 1944. [51] Inst. Mechan. Engineers, The Practice of and Education for Engineering Design. London: Inst. Mech. Eng., 1964. [52] L. E. Jermy, “Co-ordinating engineering with other company activities,” Machine Design, vol. 3, no. 6, pp. 27–29, 1931. [53] H. H. Jo, H. R. Parsaei, and W. G. Sullivan, “Principles of concurrent engineering,” in Concurrent Engineering: Contemporary Issues and Modern Design Tools. New York: Chapman and Hall, 1993, pp. 3–23. [54] D. W. Karger, “Integrating the over-all development program,” Organizing for Product Development. Amer. Manage. Assoc.: Manage. Rep. no. 31, pp. 109–115, 1959. [55] E. A. Kraft, “Konstruktion und Fertigung in Wechselwirkung,” Maschinenbau, vol. 10, no. 17, pp. 559–562, 1931. [56] E. Laitala, Engineering and Organization. Homewood, IL: Irwin, 1959. [57] R. H. Lansburgh and W. R. Spriegel, Industrial Management. New York: Wiley, 1940. [58] L. H. Leedham, “Impact of development and research on production and design,” in Proc. Institution Mech. Eng., 1954, vol. 168, pp. 403–408. [59] W. O. Lichtner, Planned Control in Manufacturing. New York: Ronald, 1924. [60] J. W. Lorsch and P. R. Lawrence, “Organizing for product innovation,” Harvard Bus. Rev. pp. 109–122, Jan. 1965. [61] J. W. Lorsch, Product Innovation and Organization. New York: Macmillan, 1965. [62] R. Matousek, Engineering Design: A Systematic Approach. London: Blackie, 1963. [63] H. B. Maynard, Industrial Engineering Handbook. New York: McGraw-Hill, 1956. [64] F. K. McCune, “Product design,” Cutting Tomorrow’s Costs. Amer. Manage. Assoc.: Production Series no. 168, pp. 21–35, 1947. [65] C. E. K. Mees, “Industrial research laboratory organization,” Trans. Amer. Soc. Mech. Eng., vol. 41, no. 1690, pp. 83–90, 1919. [66] G. F. Metcalf, “Engineering management challenge in weapons development,” IRE Trans. Eng. Manage., vol. 4, pp. 82–84, 1957. [67] W. N. Mitchell, Organization and Management of Production. New York: McGraw-Hill, 1939. [68] F. G. Moore, Manufacturing Management. Homewood, IL: Irwin, 1952. [69] A. V. H. Mory and L. V. Quigley, “How the Bakelite Corporation effectively manages organized research: A striking example of the modern trend in industry,” Mgmt. and Admin. in Manufac. Industries, vol. 11, no. 1, pp. 13–16, 1926. [70] R. Muther, Production-Line Technique. New York: McGraw-Hill, 1944. [71] J. L. Nevins et al., Concurrent Design of Products and Processes: A Strategy for the Next Generation in Manufacturing. New York: McGraw-Hill, 1989. [72] B. W. Niebel and E. N. Baldwin, Designing for Production. Homewood, IL: Irwin, 1963. [73] P. Niland, Management Problems in the Acquisition of Special Automatic Equipment. Boston: Harvard Bus. Sch., 1961. [74] D. F. Noble, Forces of Production: A Social History of Industrial Automation. New York: Knopf, 1984. [75] Organization for Economic Co-operation and Development, Design Departments. Paris: OECD, 1967. [76] C. F. Phillips, Marketing by Manufacturers. Chicago: Irwin, 1948. [77] B. Prasad, Concurrent Engineering Fundamentals: Integrated Product and Process Organization. Upper Saddle River, NJ: Prentice-Hall, 1996. [78] C. A. Purdy, “Process research,” Cutting Tomorrow’s Costs, Amer. Manage. Assoc.: Production Series, no. 168, pp. 36–51, 1947. [79] J. D. Radford and D. B. Richardson, The Management of Production. London: Cleaver-Hume Press, 1963.

78

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 44, NO. 1, FEBRUARY 1997

[80] C. F. Rassweiler, “Product research,” Cutting Tomorrow’s Costs. Amer. Manage. Assoc.: Production Series no. 168, pp. 13–20, 1947. [81] W. Rautenstrauch, The Design of Manufacturing Enterprises. New York: Pitman, 1941. [82] C. S. Redding, “Research and development departments: Their functions, activities, and relations to other departments,” Research and Development Departments. Amer. Manage. Assoc.: Production Executives’ Series no. 78, pp. 3–16, 1929. [83] O. D. Reich, The Effect of Design on Overhead. Amer. Manage. Assoc.: Production Series no. 93, 1931. [84] S. A. Robertson, Engineering Management. New York: Philosoph. Library, 1961. [85] A. W. Robinson, “The relation of the drawing office to the shop in manufacturing,” Trans. Amer. Soc. Mech. Eng., vol. 15, pp. 965–976, 1894. [86] E. S. Roscoe, Organization for Production. Homewood, IL: Irwin, 1955. [87] B. Seely, “Research, engineering and science in American engineering colleges: 1900–1960,” Tech. Culture, vol. 34, no. 2, pp. 344–386, 1993. [88] R. Seybold, Controlling the Cost of Research, Design and Development. Amer. Manage. Assoc.: Production Series no. 86, 1930. [89] R. E. Smith, Machining of Metal. Bloomington, IL: McKnight and McKnight, 1949. [90] J. E. Thompson, Aircraft Production Design. San Carlos, CA: Aviation, 1945. [91] K. T. Ulrich and S. D. Eppinger, Product Design and Development. New York: McGraw-Hill, 1995. [92] J. M. Utterback, Mastering the Dynamics of Innovation: How Companies Can Seize Opportunities in the Face of Technological Change. Boston: Harvard Bus. Sch., 1994.

[93] E. E. Vender, Controlling Avoidable Manufacturing Expenditures During a Change of Design. Amer. Manage. Assoc.: Production Executives Series no. 60, 1927. [94] A. C. Ward, “The purchasing department of a manufacturing organization,” Eng. Mag., vol. 46, no. 3, pp. 349–355, 1913. [95] T. N. Whitehead, “Planning standardized components to secure variety in products,” Harvard Bus. Rev., vol. 10, no. 3, pp. 257–268, 1932. [96] D. E. Whitney, “Manufacturing by design,” Harvard Bus. Rev. July/Aug. 1988. [97] R. F. Wilder, “Coordination of research, sales, and production,” Trans. Amer. Soc. Mech. Eng., vol. 54, no. MAN-54-4b, pp. 25–39, 1932. [98] M. C. Ziemke and M. S. Spann, “Concurrent engineering’s roots in the World War II era,” in Concurrent Engineering: Contemporary Issues and Modern Design Tools. New York: Chapman and Hall, 1993, pp. 24–41. Robert P. Smith received the B.S. and M.S. degrees in mechanical engineering from Stanford University, Stanford, CA, and the Massachusetts Institute of Technology (MIT), Cambridge, MA, respectively. He received the Ph.D. degree in management from MIT. He is presently an Assistant Professor of Industrial Engineering at the University of Washington, Seattle. His research and teaching focus on management of the product development process. His papers have appeared in Management Science, Journal of Engineering and Technology Management, and Research in Engineering Design.