Systems integration: a core capability of the ... - Semantic Scholar

Industrial and Corporate Change, Volume 14, Number 6, pp. 1109–1143 doi:10.1093/icc/dth080 Advance Access published November 7, 2005

Systems integration: a core capability of the modern corporation Michael Hobday, Andrew Davies, and Andrea Prencipe

Many of the world’s leading firms are developing a new model of industrial organization based on systems integration. Rather than performing all productive tasks in-house, companies are building the capabilities to design and integrate systems, while managing networks of component and subsystem suppliers. This article illustrates how systems integration evolved from its military, engineering-based, origins in the 1940s and 1950s to a modern-day strategic capability across a wide variety of sectors. Taking a resource-based view of the firm, the article shows how systems integration capabilities underpin the way high-technology companies compete by moving selectively up- and downstream in the marketplace through the simultaneous “twin” processes of vertical integration and disintegration. Systems integrators of capital goods move downstream into service-intensive offerings to expand revenue streams and increase profitability. By contrast, producers of high-volume components and consumer goods use systems integration capabilities to exploit upstream relationships with input suppliers. In both cases, strategic options and capabilities are shaped by the life cycle of each product. The article develops a clearer understanding of systems integration, arguing that it now represents a core capability of the modern high-technology corporation.

1. Introduction Systems integration has “two faces” similar to the two faces of research and development (R&D) identified by Cohen and Levinthal (1989).1 The first face concerns the internal activities of firms as they develop and integrate the inputs they need to produce new products. The second face, which has become more important in recent years, refers to the external activities of firms as they integrate components, skills, and knowledge from other organizations to produce ever more complex products and services. External organizations include suppliers, users, government agencies, regulators, production partners and, sometimes, competitors as firms work together and compete in projects.

1

A precise definition of systems integration is provided below. Section 2 shows how the systems integration of high-technology products, the focus of this article differs from and relates to other kinds of systems integration.

© The Author 2005. Published by Oxford University Press on behalf of Associazione ICC. All rights reserved.

1110

M. Hobday, A. Davies, and A. Prencipe

Recent research shows that both faces of systems integration have taken on a more strategic character and that systems integration is a capability central to the competitive advantages of prime contractors such as General Electric, ABB, Dell, Ford, IBM, Hewlett-Packard, BAE Systems, Cable & Wireless, Siemens, Nokia, Rolls-Royce, McDonell, and Thales. As industry leaders, one of their main tasks is to integrate together various kinds of technology, knowledge, and hardware supplied by other organizations involved in production. Although these arguments are often put forward by major firms themselves,2 there is little study of the nature and importance of systems integration in practice. The article therefore provides an assessment of the historical literature and an analysis of recent research on the subject.3 This evidence shows that systems integration rapidly developed in the 1940s and 1950s in the military arena, and then spread to other capital goods and high-volume industries. Evolving from an engineering practice (as part of the wider discipline of systems engineering) to a strategic business activity, systems integration has become increasingly important for organizing networks of production both within and across many high-technology firms. We argue that systems integration has evolved beyond its original technical and operational tasks to encompass a strategic business dimension becoming, therefore, a core capability of many high-technology corporations. In its broadest sense, systems integration can be defined as the capabilities which enable firms, government agencies, regulators, and a range of other actors to define and combine together all the necessary inputs for a system and agree on a path of future systems development. In the narrower sense of firm capability, systems integration is concerned with the way in which firms and other agents bring together hightechnology components, subsystems, software, skills, knowledge, engineers, managers, and technicians to produce a product in competition with other suppliers. The more complex, high technology, and high cost the product, the more significant systems integration becomes to the productive activity of the firm. In order to better understand systems integration from the perspective of the competing suppliers, the article adopts the resource-based view of the firm (Penrose, 1959; Richardson, 1972; Chandler, 1990) in its more recent incarnation proposed by Teece and Pisano (1994) and Teece et al. (1994), labeled dynamic capabilities.4 We argue that systems integration is an empirical instantiation of a firm’s dynamic capabilities, and therefore key to the broader competitive strategy of the firm and the particular position a firm takes within the value stream of an industry at any given time.5 Systems 2

See, for example, the recent annual reports of BAE Systems and Rolls-Royce in the UK.

3

For a collection of recent papers by the authors and others see Prencipe et al. (2003).

4

For consistency, the term capabilities is used throughout rather than the term competencies which is preferred by some management scholars (Barney, 1991; Hamel and Prahalad, 1994).

5

In this article, the term value stream refers to the industry wide value chain, rather than the intrafirm value chain. Porter (1990: 42) refers to this as value system.

Systems integration

1111

integration capabilities are inextricably linked to decisions on whether to make inhouse, outsource, or collaborate in production and competition. With the increase in outsourcing over the past decade or so, systems integration has become more directly linked to the operations, strategy, and competitive advantage of industry leaders in industries such as military systems, telecommunications, aerospace, software, computing, automotive, and hard disk drives (HDDs). At first sight, systems integration appears to be the “other side of the coin” of outsourcing and one of the capabilities that underpins successful outsourcing. As the article argues, however, systems integration capability is not merely the counterpart to outsourcing, but the capability needed to manage outsourcing as well as “joint sourcing” and “insourcing” to enable the systems integrator firm to gain the advantages of both outsourcing and vertical integration through different phases of the product life cycle. In other words, systems integration capabilities enable firms to move selectively up- and downstream in the marketplace through the simultaneous “twin” processes of vertical integration and disintegration.6 To illustrate how product systems integration can be bounded for analytical purposes, section 2 provides a simple typology of technological systems. Sections 3 and 4 then outline the American military origins of systems integration, identifying the factors and organizations which drove its initial development as a technical discipline within, and then beyond, the defense sector. Section 5 turns to non-military products arguing that, here too, systems integration has become a widespread strategic activity, rather than merely an operational or engineering task. Section 6 compares evidence from two highvolume industries, namely automobiles and HDDs, to show how firms use systems integration capabilities to upgrade technologically from one product to another. Section 7 examines recent trends from selected high-value capital goods industries, to reveal how large corporations have begun to use systems integration capabilities to bundle services together with hardware and software in order to gain competitive advantage. Section 8 points to future research directions and the conclusion summarizes the main findings.

2. A typology of technological systems Systems integration can be defined in many different ways according to the nature of the system being integrated, the integration processes involved, and the way in which the system is bounded for analytical purposes. Indeed, there are many different perspectives on the subject and a variety of different meanings in the literature (Prencipe et al., 2003). Therefore, it is useful to develop a simple typology of technological systems to show how different kinds of systems and systems integration relate to each other. The study of technological systems and their integration involves scholars of military systems (Sapolsky, 1972; Walker et al., 1988), project and technology management (Shenhar et al., 1994), geographical dynamics (Best, 2003), the measurement of complexity of 6

The authors are grateful to Paul Nightingale for this insight.

1112


System Scope 4 Large Technical system/System of Systems

3 ProductSystem

2 Component/Subsystem

1 Assembly

A4

B4

C4

D4

A3

B3

C3

D3

A2

B2

C2

D2

A1

B1

C1

D1

A Low-Tech

B Medium-tech

C High-tech

D Super high-tech.

Technological uncertainty/novelty

Figure 1 A simple typology of technological systems.

systems (Kline, 1990), large technical systems (Hughes, 1983), and evolutionary approaches to complex products such as aircraft (Mowery and Rosenberg, 1982; Nelson and Rosenberg, 1993). Historians of large technical systems such as Hughes (1983), Mayntz and Hughes (1988), and Summerton (1995) analyze the historical evolution of networks made up of complex interconnecting components which link together to form infrastructures such as rail, electricity, and, more recently, the internet. They also illustrate how these systems are shaped by social forces, examining the way network technologies are developed within particular social, historical, and institutional settings (Bijker et al., 1987). To understand the nature of these various systems and the systems integration processes involved, it is helpful to contrast the various types of system and map their boundaries for analytical purposes. Building on the work of Shenhar and others (1994), Figure 1 attempts to describe the scope of the system and its technological novelty. Scope refers to the physical nature and content of the system and, in particular, the extent and complexity of the hierarchy and interconnection contained within it. Shenhar (1994: 1312–1313) provides a three-part grouping (an assembly, a system, and an array). In Figure 1 a four-part categorization is developed, following Hughes (1983) who emphasizes the hierarchical scope of systems.7 7

In reality, there exists a spectrum of different kinds of system and considerable overlap between them (Walker et al., 1988). Consequently, any attempt at definition or typology tends to be somewhat arbitrary.

Systems integration

1113

Hughes (1983) defines the four different types of system for analytical purposes. An “assembly” is usually a mass-produced stand-alone product which performs a single function and does not form part of a wider system (e.g., a shaver, calculator, or personal computer), unless it is connected by a network. Systems integration usually takes place at the individual supplier firm level and is a fairly simple manufacturing process, often called assembly. Efficiency in systems integration can differ between firms and confer cost and price advantages. By contrast a “component” or “subsystem” always performs a role in a wider system (e.g., a telephone mobile base station or an avionics unit). Some components are relatively simple, low-technology devices (e.g., resistors, capacitors, or relays). Others may be extremely high technology in nature, involving thousands of hours of design input (e.g., semiconductor components such as microprocessors, HDDs for computers, and customized software packages for telecommunications systems).8 In the case of microprocessors, for example, systems integration takes place usually within a single firm and involves a highly complex process with many different steps, each of which affects the yield, cost, and quality of the final product. Design is closely integrated into manufacturing, and a wide range of capital goods dedicated to each generation of microprocessor product are central to the integration process. Efficiency in the integration of these types of components is usually critical to the competitive advantage of the supplier and different approaches to integration methods can distinguish one supplier from another.9 A “product system” is situated in Figure 1 between a component and a large technical system. A product system, which is invariably a capital good, can be defined in terms of its components, network structure, and mechanisms of control.10 Systems are made up of various types of components (e.g., semiconductor devices and software packages), hierarchically organized to perform a common goal (e.g., an aircraft, a business information system, or a weapon systems). Sometimes these are called complex product systems (CoPS) which represent the high-technology capital goods that underpin the production of goods and services (Hobday, 1998). Systems integration usually involves many firms and other actors, including government, regulatory agencies, users, small specialist suppliers, and other subcontractors. Unlike components and assemblies, systems integration tends to be project based, and because each product is,

8

Clearly, grey areas exist between assemblies and components. For example, a personal computer could be an assembly or component depending on its use. Note that the nature of the component in question will affect the tasks involved in systems integration.

9

This is similar to systems integration in hard-disk drives (section 6).

10

The term product system or complex product system is preferable to capital good, as the latter term includes low technology, routinely produced products, and also fails to capture the diversity of products, systems, networks, and functions of various kinds of product system.

1114


to some extent, tailor-made for each user, the tasks involved in systems integration will differ from product to product.11 Finally, “large technical systems” or “system of systems” are collections of distinct but interrelated systems, each performing independent tasks but which are organized together to achieve a common goal (e.g., an airport which consists of aircraft, terminals, runways, air traffic control, and baggage handling systems). Large technical systems represent the technological, energy, communications, and transportation infrastructures of the economy. They underpin progress at the wider economic and industrial levels and also the advance of each individual firm and household. The integration of large technical systems has been studied fairly extensively by historians of technology12 and by sociologists concerned with how these systems are shaped by human beings, the social desirability (or otherwise) of some systems, and how the policies affect (or should effect) their pattern and pace of advance (Bijker et al., 1987; Davies, 1996). In the case of large technical systems, systems integration involves interaction between different kinds of actors from business, government, non-governmental organizations (e.g., pressure groups), and regulatory bodies. Often, formal bodies emerge to define the systems requirements, ensure their environmental and social acceptability, and shape their future development. Through these bodies, leading firms are able to reach out and shape market institutions, and through them, the pace and pattern of demand and sometimes the structure of competition, as in the case of US naval weapons systems where firms participate in organizations such as the Center for Naval Analysis, the Institute for Defense Analysis, the Naval Research Laboratory, and the Software Engineering Institute of Carnegie Mellon University (Gholz, 2003). Government-business relationships have been shown to be extremely important in the weapons acquisition process in the US and elsewhere, where technological uncertainties preclude the development of a conventional market system (Cherington et al., 1962; Peck and Scherer, 1962). However, similar non-market characteristics have also been observed in the development and acquisition of complex civilian

11 While there has been considerable recent research on complex products and systems, this has only recently began to address the question of systems integration capabilities (Prencipe et al., 2003). 12

In his account of the early history of the electricity industry Hughes (1983) depicts electricity supply as a large technical system in which electricity is generated, transmitted over distance, and used in a variety of ways by industry and households. The electricity network is made up of many different, yet interrelated vintages of technology. Hughes shows that, as with other infrastructural systems, the development of electricity supply is uneven. Breakthroughs in one area (e.g., production) require corresponding innovations in transmission and utilization for their economic potential to be realized. Using the idea of “reverse salients,” Hughes shows that those areas of the technology that restrict the development of the system overall tend to become the focal points for innovative effort. Similarly, Rosenberg (1979) shows how uneven development of technology can produce incentive effects across different sectors of the economy. As Rosenberg (1979) argues “technological problems arising in industry A are eventually solved by bringing to bear technological skills and resources from industries B, C, or D.” Students of large technical systems reveal how technological imbalances can lead to a very uneven course of infrastructural progress.

Systems integration

1115

systems in areas such as nuclear power, off shore oil production, telecommunications, and aerospace where major firms take a prominent role in shaping the institutions of demand and rely on close political and bureaucratic interaction with government and regulatory agencies (Walker et al., 1988; Miller et al., 1995). These four types of system are closely interrelated. For example, components and assemblies form part of product systems, while various product systems are key elements of large technical systems. In fact, the four kinds of system can be seen as a hierarchy or production chain with components, materials, and assemblies feeding into product systems which then form the technological core of large technical systems. Technological advances in components and product systems to some extent shape the potential and performance of large technical systems, while the needs of the latter shape the pace and pattern of technical change in the former, providing a focussing device for technological efforts. The institutions set up to govern the development of large technical systems often take a direct interest in the development of product systems and their processes of integration and, sometimes, the key components which the large technical systems ultimately rely upon for their advance. Shenhar et al. (1994) and Shenhar (1994) show that technological uncertainty (defined by the degree to which new artefacts, knowledge, or techniques are embodied in a system) is a key dimension for understanding and producing any kind of system. Uncertainty is closely connected to the notion of high and low technology.13 For illustrative purposes, Figure 1 divides uncertainty into four types, which can be related to each kind of system described above.14 Type A (low-technology) systems rely on well-established technologies (e.g., roads and simple buildings). These can be large or small in value, but no new technology is required at any stage. Systems integration is a relatively straightforward task as firms and other actors will tend to have experience in integrating such systems and the level of uncertainty is low. Type B (medium-technology) systems incorporate some new features, but most technology is available as with new models of existing products. As with type A, systems integration is unlikely to pose major problems given the familiarity with most of the technology. Type C (high-technology) systems consist of, mostly, recently developed technology. Examples include internet super servers, intelligent buildings, and new passenger aircraft. Systems integration is likely to pose difficulties and require new types of capability both within the prime contractor or lead integrator and between the latter and other suppliers and users. Governments, standards, and regulatory bodies may also play a 13 Miller and Cote (1987: 11) operationalize the notion of high technology (in relation to industry) using R&D intensity and the proportions of scientists, engineers, and technicians in the workforce. However, most so-called high-technology industries and products (e.g., electronics) often have a substantial low-technology dimension to them (e.g., plastics, mechanical engineering, assembly, and manual testing) and vice versa. This also applies to each of the systems in Figure 1. 14 As in the case of system types, this categorization is arbitrary as in practice there will be a continuum with no clear boundaries and frequent overlaps.

1116


direct role in the systems development, including approaches to integration. Type D (super-high technology) systems, which depend on the development of new knowledge, artefacts, skills, and materials, are fairly rare and rely on emerging new technologies. They involve extremely high levels of uncertainty, risk, and new investment (e.g., new spacecraft and intelligent defense systems). In these cases, systems integration strategies, capabilities, and structures need to be developed alongside the technology. As noted above, while there has been important research on high-technology large technical systems, there has been comparatively little research on the suppliers of medium-, high-, and super-high technology product systems (categories B3, C3, and D3). As Miller et al. (1995) argue, the latter are an important category of capital goods which exhibit unusual innovation features (when compared with mass-produced assemblies and components) centered on project management and systems integration. Unlike consumer goods, for example, users are often intimately involved in the innovation and systems integration processes together with standards and regulatory bodies. The two dimensions above, system scope and technological novelty, provide a useful starting point for categorizing the nature and scope of different kinds of technological systems. In the case of product systems, there has been relatively little research on the competitive strategies of systems producers or the modern-day challenges posed by systems integration (Prencipe et al., 2003). Nevertheless, systems integration has occupied the minds of military and defense analysts, planners, and engineers for many years. In fact, the modern-day concept of systems integration has its origins in US defense technology.

3. The military origins of systems integration Sapolsky (2003) describes the military and cold war origins of systems integration in the US, the first country to develop and institutionalize formal systems integration processes. While systems integration in both large technical systems such as railways and electricity networks and the CoPS which underpinned them had a long history of development (Rosenberg 1979; Hughes 1983), as Sapolsky (2003) argues, events during and just after the Second World War led to new techniques for systems integration. The steeply rising costs and burgeoning complexity of weapons systems led to experimentation with new approaches to military systems development, including systems integration. During the Cold War period, the American Government came to believe that it needed more than financial investment and determination to prevail against the Soviet Union. It needed to create new institutions and tools to build and co-ordinate, over the long-term, the broadening range of complex new technologies needed for military purposes. During the Second World War, the pace of development of new weapons technologies had accelerated (e.g., nuclear weapons, ballistic missiles, and jet propulsion). Just after the war, as Sapolsky (2003) points out, already many of basic weapons technologies needed to wage the Cold War were developed or under development. However,

Systems integration

1117

the institutional systems required for their development, deployment, and renewal were inadequate for the needs of the Cold War.15 American military structures could control events when most in society were ready to accept military discipline and priorities because of War. However, these structures began to be severely tested in the partial mobilization of the Cold War, with the need to co-ordinate large numbers of business firms and civilian scientists and engineers. Weapons systems were becoming ever more complex and ambitious. The traditional single discipline, linear “over the wall” approach to systems development, which involved making one part of the system, then the next, then the next, came under increasing pressure. There was a need to mobilize multidisciplinary teams of scientists and engineers to work together on systems in a way which optimized design, engineering development, production, and operations and kept costs under control. In response, a variety of special organizations and skills were developed. Key among them were the systems integration skills and processes needed for developing, building, and operating the new weapons, as well as the new project-based organizational structures which were then developing.16 Up until the Cold War, these systems integration ideas and structures had not yet been formalized, and emerging notions of systems integration were primarily technical. During the late 1940s and early 1950s, in areas such as the atomic and hydrogen bombs, jet fighters, ballistic missiles, satellites and strategic defense command, and control systems, the US military created two related approaches. First, engineers created the technical discipline of systems engineering, including systems integration, for designing and developing systems. Second, managers and business leaders created new project management tools, techniques, and organizational structures. Both approaches emphasized meeting cost and schedule targets as well as technical requirements. One highly successful project was the Polaris missile system which invented management tools such as Programme Evaluation and Review Technique (PERT) which rapidly spread to other military projects and civil production (Sapolsky, 1972). Hughes (1998) describes several major US projects in the 1950s including Atlas [the first intercontinental ballistic missile (ICBM) system] led by General Bernard Schriever of the Air Force and Simon Ramo of TRW. Atlas was the first project which formally deployed systems engineering to co-ordinate the activities of large numbers of companies and thousands of scientists and engineers working across a diverse set of technologies. Because weapons systems were incorporating many new technologies and components (e.g., radar, nuclear weapons, rocket propulsion, and electronic controls), they were 15 Hughes (1998) describes major US projects in the 1950s such as The Atlas (the first intercontinental ballistic missile system) led by General Bernard Schriever of the Air Force and Simon Ramo of TRW. Atlas was the first project which formally deployed systems engineering in order to co-ordinate the activities of large numbers of private companies and thousands of scientists and engineers working across a diverse set of technologies. 16 Some of the origins of modern project management can also be traced back to the US military (Davies and Hobday, 2005).

1118


becoming more complex and difficult to design, produce, and operate. As one indicator, the number of aircraft gas turbines parts increased from 9,000 in 1946 to 20,000 in 1957. The costs of the more complex weapons systems continued to increase steeply during the 1945–1955 decade. No individual or single engineering discipline could cope with the conception, design, and building of these systems. In response, engineers, physicists, and rocket scientists joined together in multidisciplinary teams to create an alternative to the traditional sequential approach (Johnson, 1997). Before and during the Second World War, organizations such as the American Army Air Forces procured aircraft from prime contractors. Afterwards, the Air Force added its own components and modifications (e.g., engines and armaments). By the late-1940s, formal systems engineering had begun to take hold. Under this approach, all core weapons systems components had to be designed and developed together from the start (including airframe, electronics, and armament) prior to integration. A very early example was the mobile ground radar units used in the Second World War. Ivan Getting, the engineer on the project, realized that radar and fire control components behaved differently when working together than they did individually. To account for the differences, he ordered that specifications on each unit be written with full consideration of the features and capabilities of other units. His early innovation was to allocate a systems integrator to each project. Close co-ordination between radar and gun subsystems was achieved by assigning MIT’s Radiation Laboratory as the formal systems integrator. The laboratory had access to all information, authority to test models, and to provide feedback on design through all stages of development. By the late-1940s, systems engineering began to set new standards for military projects. As a discipline, systems engineering was concerned with the whole system, rather than any single subsystem or technology. The idea was that an integrated system was a whole greater than the sums of its parts and that entire weapons systems, and their components had to be designed together concurrently (e.g., airframe, electronics, and armaments) so that the system could be integrated successfully. One landmark was the use of systems engineering in ICBM systems which emerged out of the German rocket programs. Bernard Schriever famously promoted the use of systems integration when he became head of ICBM programs in 1953. Another landmark was the Manhattan project which focussed on the development of nuclear warheads, where a single systems engineering organization (Ramo–Wooldridge) was assigned to co-ordinate the projects’ systems integration activities. The multidisciplinary approach enabled the team to push the technological frontier of nuclear developments much further than previously possible. For Simon Ramo, of Ramo–Wooldridge (and the “R” in TRW), the systems engineering-integration approach was a “cure for chaos” (Ramo, 1969). The advantages began to be widely appreciated and systems engineering and integration techniques spread beyond the military. For example, in 1950 Bell Labs deployed a version of systems engineering to maintain its telephone network. In 1952, MIT began to put on the first weapons systems engineering courses. The first text

Systems integration

1119

book on systems engineering (Goode and Machol, 1957) was soon followed by others in aerospace and telecommunications. Early breakthroughs in systems integration practices transferred from the military to related civilian sectors such as aerospace, where some of the same firms were engaged in both military and civil production. The new practice of project management also transferred from military to civil. While the complex relationship between the military and civilian industries cannot be discussed here, it is likely that systems learning and organizational practices and technologies transferred both from military to civil and vice versa, given the uneven pace of technological change. Later on, for example, as semiconductors and electronics began to be used more widely, military systems makers began to rely on commercial producers such as Texas Instruments and IBM for key components and subsystems. In the case of semiconductors (e.g., microprocessors and high speed logic circuits), military producers benefited from commercial production processes to achieve the levels of integration and performance required by military systems. Typically, key military components would be produced from mainstream commercial processes and then subjected to special performance tests for military suitability. Specialist military chip applications allowed firms such as Texas Instruments and Motorola to further capitalize on their rapidly advancing technologies. Today, most major military and civil systems producers refer to themselves as prime contractors. They see their primary task as one of systems integration. Figure 2, for example, presents the case of the UK aerospace and military producer, BAE Systems. Similar diagrams can be found in the annual reports of firms such as Thales (of

Systems Integration Capabilities WEAPONS SYSTEMS INTEGRATOR Provides the overall system…with processes and skills to: •translate customer requirements into a total systems solution •develop overallsystems specifications

PRIME CONTRACTOR

Prime contractor Weapons system integrator

•deliver fully tested and integratedweapons systems to meet customer requirements

Source: British Aerospace (now BAE Systems) Annual Report (1998)

Figure 2 The case of BAe systems.

Provides theproject and risk management of large scale complexprogrammes to deliver completevalue for money solutions to meet customer requirements

1120


France) and McDonell of the US. These firms see their main business task as occupying the prime contractor space in the production pyramid. They then outsource production, design and other activities to a wide range of other suppliers. In Figure 2, the prime contractor can be separated from the systems integrator which means that, in some cases, the systems integration task itself is outsourced to a supplier firm. In fact, the pyramid in Figure 2 is slightly misleading. In most cases, the customer (e.g., the Department of Defence or the Airforce) occupies the tip of the pyramid, providing instructions, guidance on customer requirements, and overall systems specifications. The customer will often specify which firms are below the prime contractor in the pyramid and lay down fairly tight guidelines on project processes, risk management arrangements, and scheduling. In their turn, the leading systems firms are able to influence the purchasing decisions and strategies of the buyers and users of the systems. Beneath the top two sections of the pyramid, not shown in the diagram, there is usually a layer of subsystem suppliers. These are usually very large firms themselves which act as systems integrators and prime contractors in other projects. Beneath this subsystem layer is another tier of firms made up of component suppliers, some of which are large producers of civil technology (e.g., semiconductors, software, and computer suppliers). Their inputs can also be extremely complex, specialized, and high-technology based (e.g., microprocessor components or super computers). Like the subsystem suppliers, these firms drive the base technology forward. Beneath this tier, lies another tier of basic parts and material suppliers who form the base of the pyramid or “food chain” of production.17 Figure 2 implies that, today, systems integration is not solely or even mainly a technical task. Instead, it is concerned with the apportioning of production and innovation tasks across the industry value stream. It involves the organization of major projects, the choice of business partners, and decisions over what to source internally and externally. Regarding the specific functions of systems engineering, as Figure 2 indicates, at the technical level the primary task is to draw up a set of overall specifications which map the performance of each subsystem and its interactions with every other. Other tasks include the evaluation of subsystems during development, the planning of their integration, the controlling and testing of subsystems, and the assessment of the operational environment in which systems are to be put to work. Systems engineering allows engineers to partition systems into smaller manageable subsystems, assemblies, and sub-assemblies, and, at the same time, to develop interface specifications for each component before they are designed and constructed. The purpose is to enable engineers to understand and predict the interactions among subsystems and so reduce emergent properties which might adversely affect the overall design and final functioning of the system. The approach allows engineers to freeze 17

As Walker et al. (1988) show, this production pyramid applies not only to military systems but also complex civilian products including telecommunications systems, energy control systems, transportation equipment, information technology products and networks, and civil engineering constructs. See Hobday (1998) for a list of 100 or so of these product systems.

Systems integration

1121

the systems design at the most appropriate, earliest, point in order to reduce the ripple effects of many design changes. In short, systems engineering-integration is an attempt to provide a systematic, multidisciplinary approach to systems development which is concurrent not sequential. In retrospect, the approach looks quite obvious and logical. Similar techniques are now widely used in the automobile and electronics industries and terms such as concurrent engineering are commonplace. However, in the 1950s the concept was a major breakthrough both in the way engineering was taught and in the way systems were produced. Using concurrent engineering, components were designed, produced, and tested simultaneously, reducing delays, and later technical difficulties. The advantages were quickly seen in superior performance (Sapolksy, 1972) as more systems were developed on time and to specification. The approach allowed engineers to understand the dynamic interaction between the system and its environment and to predict design changes which might be needed through the operational life cycle of the system. As Sapolsky (2003) points out, the American military forced the pace of technological change in a number of large complex weapon projects. Spurred on by its own internally competitive organizational structures, the military learned and helped others to learn the early lessons of formal systems integration.

4. The organizations involved in systems integration Today, systems integration has become increasingly sophisticated and no single organization can claim to have a monopoly on systems integration capabilities in any one project. Each subsector has its own profile of supplier and user organizations, all involved in systems integration. For example, in American defense systems, as Table 1 Table 1 Levels of systems integration capability in the defense industry

Distinguishing skills

Component systems

Platform systems

Architecture systems

integration

integration

integration

Technical capabilities in

Project/subcontractor

System definition

specific core areas

management

Key implementing tasks Engineering development, Production, system component production Example organizations

Subcontractors like

assembly

Trade-off studies, customer interface

Prime contractors like

Technical advisors like

Northrop Grumman

Lockheed Martin

MITRE and SAIC

Electronic Systems and

Aeronautics and

Raytheon Missile

General Dynamics

Systems

Bath Ironworks

Source: Gholz (2003: 282).

1122


points out, various organizations share the overall task of systems integration. They include the prime contractors and major subcontractors that build the weapons systems, non-profit technical advisors (including university groups), specialist government laboratories, and organizations that manage weapons acquisition (e.g., the DoD) as well as the military users of the weapons (e.g., the army or navy). Typically, these organizations are engaged in systems development and integration at both the level of large technical system (e.g., integrated weapons systems) and the product systems which underpin the larger system. However, each of these groups has a slightly different perspective on the meaning of the systems integration, reflecting its own particular interest in the system under development. For the private sector weapons producers (e.g., BAE or Lockheed Martin), systems integration concerns the ability to generate the conceptual design, control the supplier network, and to produce to specification. For acquisition planners that award contracts (e.g., the MoD or DoD), systems integration is the expertise to set the initial technical requirements of the system and evaluate competing bids from prime contractors. For military “user” organization (e.g., the army or navy), systems integration is the skill to understand the capabilities and limitations of the weapons platforms, and their suppliers, and to predict the use of the system in an operational environment. Typically, the user (e.g., the navy) will engage with the producers, even if they are organizationally separate from the purchaser (e.g., the DoD). In the defense sector Gholz (2003) divides the different levels and types of systems integration capability into three major types, component systems integration, platform systems integration, and architecture systems integration.18 As Table 2 summarizes, different organizations maintain and develop these capabilities in each area. In components, firms such as Northrop Grumman Electronic Systems and Raytheon Missile Systems require the ability to design and integrate major components in relation to others in any particular system. Systems integration at the weapons platform level is carried out by prime contractors such as Lockheed Martin Aeronautics and General Dynamics Bath Ironworks. In terms of overall concept or architecture technical advisors such as MITRE and SAIC, as well as the DoD or in the UK, the MoD are engaged. The skills of the organizations listed in Table 2 are not exclusive to them. For example, platform systems integration involves many technical capabilities which overlap with those of subcontractors, while component systems integration often involves some assembly tasks as well as subcontractor management skills. Gholz describes the various organizations involved in systems integration in US naval defense (Table 2). Each specific system will involve different combinations of these organizations and a different profile of systems integration capabilities. Among the many organizations that contribute to systems integration for the US Navy are the SPAWAR Systems Center, the Naval Air Warfare Center, and the Naval Surface Warfare 18

Recalling Figure 1, the latter two correspond to large technical system, while the former are concerned with product systems and core component integration.

Systems integration

1123

Table 2 The organizations involved in systems integration in US naval defense

Policy analysis

Government

Private, non-profit

System Commands

Center for Naval Analysis, ANSER, TASC, Booz-Allen

(SPAWAR, NAVSEA,

Institute for Defense

NAVAIR)

Analysis, Rand

Private, for-profit

Scientific research Naval Research Laboratory, APL, Lincoln Laboratory, SPAWAR Systems Center—San Diego Technical support SPAWAR Systems Center—San Diego

Software Engineering Institute APL, MITRE, Aerospace

SAIC, SYNTEK

Corporation

Production

Lockheed Martin—Naval Electronics and Surveillance Systems, Raytheon Command Control Communications and Information Systems

Testing and fleet support

SPAWAR Systems Center—San Diego

Source: Gholz (2003).

Center. Some of these organizations have small-scale activities that overlap with the capabilities of other firms in the matrix in Table 2. For example, the SPAWAR Systems Center in San Diego is not only concerned with policy but also manufactures Link 16 antennas for surface combatants.

5. Systems integration in non-military sectors 5.1 Suppliers of high cost civilian product systems As noted above, in the area of high-cost systems civilian suppliers firms need a deep systems integration capability to compete (e.g., in the area of aircraft engines, flight simulators, air traffic control networks, railway engines, civil engineering, telecommunication systems, and internet equipment). As in the military, the major users of these systems, and associated regulatory and government bodies, need an understanding of systems integration and the pattern of development of the large technical system in which the product system is embedded. In some cases, product system users, suppliers, and other groups may need to establish a formal system integration function or body where it does not already exist (e.g., in the case of European air traffic control). In the latter case, one of the roles of the systems integration body is to define and plan the

1124


future evolution of the large technical system and resolve any conflicting priorities and pressures which may exist (e.g., over the environment versus capacity growth in the case of air traffic control). These plans have a direct bearing on the product systems being developed by suppliers. In order to plan for the product technology, the systems integrator actors need a deep understanding of the changing supply and demand structure of the large technical system and the environment in which it evolves. Large technical systems are invariably dependent on high-technology product systems (e.g., in railways, air transport, information and communications systems, electricity, and gas networks). The drivers for more elaborate product systems integration capability by firms are similar to those witnessed in military systems, namely the increasing complexity of the technology, the rapid pace of market and technical change, and the broadening range of knowledge and skills required to produce the product system in question. As in the case of military systems, integrators are extensively outsourcing to lower-tier suppliers as they themselves move downstream to service their customers (see Section 7). System users (e.g., airlines and the utilities) are demanding more functions and higher performance and lower costs as liberalization, increased competition, and the de-regulation of markets proceeds. These competitive pressures are forcing many product system users to move away from systems development and towards the supply of differentiated, higher quality services to their final customers.

5.2 Firms as systems integrators—an historical perspective While the Second World War may have ushered in a new, more frantic pace of systems integration led by the military, Pavitt (2003) argues that systems integration, as a business function, is simply an unfolding of industrial specialization—another example of Adam Smith’s division of labor. Pavitt describes the role of firms which specialize in systems integration, arguing that recent examples are the result of two intensifying features of technical change. First, is the continuous increase in specialization in both the production of artefacts and of knowledge. Second is the advance of information and communications technology (ICT). These two forces have, Pavitt argues, increased the opportunities for “disintegration,” both within product development activities themselves and between product development and manufacturing. In one sense, disintegration (or from a managerial perspective, outsourcing) can be viewed as the “other side of the coin” of systems integration. Firms can only outsource if they acquire the capability to integrate the components, knowledge, or software then produced by their specialist suppliers and subcontractors. As Pavitt (2003) argues, some firms specializing in systems design and integration have grown to challenge large-scale manufacturing firms, although in many areas there are limits to complete outsourcing, because arm’s-length relationships can be an inefficient means for exchanging and integrating fast-changing fields of knowledge. However, even in those firms which have made heavy investments in product

Systems integration

1125

development, we see some vertical disintegration in manufacturing processes and design. This process has been occurring since the nineteenth century, for example in machine tool manufacture, as pointed out by Rosenberg (1963). Specialized machine-tool makers (early examples of product systems) emerged then as a result of advances in metal cutting and metal forming technologies. This led to technological convergence in operations that were common to a number of manufacturing processes. For example, boring accurate circular holes in metal was common to the making of both small arms and sewing machines. Although the skills associated with such machining operations were craft-based, their output could be codified and standardized. Once the size of the market for these common operations became large enough, small specialized firms designing and making the machines emerged. The larger manufacturing customers could therefore benefit by purchasing these machines which incorporated the latest improvements fed back from many users in many different industries. Because these machines were superior to the machines produced internally by manufacturers, designing and making such machines in-house no longer gave manufacturing firms a competitive advantage. The machine tool user firms described by Rosenberg became the systems integrators of other producers’ goods. As Pavitt (2003) shows, these historical processes based on technological convergence and outsourcing occur regularly (Table 3) in many sectors as new opportunities for convergence arise because of technological breakthroughs. These advances have occurred in material shaping and forming, the properties of materials, continuous chemical processes, and the storage and manipulation of information for controlling various business functions, including manufacturing operations and design. Firms have emerged to specialize in materials analysis and testing (Mowery and Rosenberg, 1989) and measurement and control instruments for continuous processes. Table 3 Modern examples of increasing outsourcing and systems integration Underlying technological breakthrough

Technological convergence

Vertical disintegration

Metal cutting and forming

Production operations

Machine tools makers

Chemistry and metallurgy

Materials analysis and

Contract research

testing Chemical engineering

Process control

Instruments makers, plant

Computing

Design, repeat operations

CAD makers, robots makers

New materials

Building prototypes

Rapid prototyping firms

ICT (information and communications

Application software,

KIBS, contract manufacture

contractors

technology)

production systems

KIBS, Knowledge-Intensive Business Services. Source: Pavitt (2003: 83).

1126


Computer-aided design and manufacturing systems, originally developed in the transport sectors, are now widely used, as are robots which originally developed in metal manufacturing. Specialized applications software and rapid prototyping are today deployed in a wide range of industries. Many specialized chemical engineering firms now contract out the designing and building of complete large-scale continuous production facilities (Landau and Rosenberg, 1992; Arora and Gambardella, 1999). More recently, as Pavitt (2003) argues, there is evidence of intensification in the outsourcing of product design by manufacturers. Sturgeon (2002), for example, has pointed to the rise of outsourcing in electronics. New specialized suppliers not only take over product design from manufacturers but also carry out the detailed engineering and manufacture. Although this trend is well known in East Asia (Hobday, 1995), Sturgeon argues that the outsourcing of manufacturing is growing in industries such as apparel, footwear, toys, data processing, offshore oil drilling, home furnishings and lighting, semiconductor fabrication, food processing, automotive parts, brewing, enterprise networking, and pharmaceuticals. In addition, Prencipe (1997) has documented increases in the outsourcing of production of aircraft engine components. In all these cases, the final manufacturer takes on the role of product integrator whether it be an assembly, a component, or a product system. As Pavitt also points out, scholars (Zuboff, 1988; D’Adderio, 2001; Balconi, 2002) have argued that recent advances in ICT suggest that ICT has increased the potential for technological convergence and outsourcing for several reasons. First, by reducing the costs of searching for standard components and sub-subsystems. Second, by increasing the degree of standardization through automation and the adoption of standard software tools (e.g., integrated enterprise software systems like PDM and ERP, D’Adderio, 2002). Third, ICT-based simulation technology and modelling have increased the scope of “learning before making” (Pisano, 1997), thereby reducing the risks of “bugs” and technical difficulties in subsequent production (D’Adderio, 2001). Fourth, ICT can increase the ease with which digitized information about new products can be transferred from product designer to producer. Fifth, in some cases, ICT enables product designers to monitor subsequent production instantaneously and gain valuable feedback, helping to resolve both manufacturing difficulties and improving design for production. However, despite clear advances in ICT (often hyped up by supplier firms and consultants),19 Pavitt (2003) goes on to argue that ICT has yet to fully achieve the conditions for “modular” products and production systems with complete outsourcing. On the contrary, links between product design and manufacture are not, and cannot, be 19

For example, in the auto sector: “Our global vision is that by 2005, every production factory will be planned, built, launched, and operated first using full simulation, before going to bricks and mortar. Every digital vehicle must pass the digital factory quality gate—meeting cost, quality, and timing targets—before approval will be given for the actual factory.”—Sue Unger, Chief Technology Officer, DaimlerChrysler AG. (Manufacturing Daily, August 28, 2002), cited in Pavitt (2003: 89).

Systems integration

1127

based entirely on codified information. Indeed, the design and integration of CoPS are extremely difficult to formalize. Also, as research by D’Adderio (2001) shows, the “digitizing” of a product’s characteristics by designers involves simplification, and digitized models must subsequently be “re-actualized” by the human teams responsible for production. Therefore, many production processes still require close personal contacts involving the transfer of tacit knowledge. Also, as Brusoni et al. (2001) and Dosi et al. (2003) point out, the division of labor in knowledge and technology is not necessarily mirrored by a division of labor in production. Product system makers which outsource design often have to retain substantial knowledge about production process and product design in-house. Therefore, human engineers and designers remain involved in relational transactions and cannot yet rely on purely arm’s-length market transactions, limiting the extent and scope of outsourcing and systems integration.

5.3 Systems integration as a strategic business activity It is clear from the evidence that systems integration is much more than an operational or technical task. At the strategic level, systems integration capability can confer competitive advantage for at least two sets of reasons. First, systems integration is the technological capability which underpins new product development and introduction. As Prencipe (1997) shows for the case of aircraft engines, systems integration embodies the static (intrageneration) capabilities a firm requires to establish a product concept design, decompose it, and co-ordinate the network of suppliers needed to produce a new product within a given technological family. Equally importantly, systems integration also refers to the dynamic (intergeneration) capabilities required to envisage and produce new product architectures and novel product families. Because the evolution of new products depends on a variety of technological fields and key components, one of the most difficult tasks facing firms is how to establish dominion over these fields which cross organizational boundaries. Second, at the level of the industry value stream, systems integration is the capability by which a firm decides where and how to situate itself, influencing how a firm competes, who it collaborates with, and who it competes with. These, arguably, are among vitally important boardroom decisions because they decide the market positioning of the firm and its evolution through time. Best (2003) argues that one of the key roles of the systems integrator firm is to exploit the technological capabilities which reside in other firms, sometimes located in regional clusters. Within regional clusters, systems integration represents the technological and organizational capability by which firms collectively foster technological and market change. Best cites the cases of Silicon Valley, Boston’s Route 128, and the UK Cambridge Science Park to illustrate the importance of this external face of systems integration. Sometimes, product design is decentralized amongst networked enterprises. In these cases, small teams, individual entrepreneurs, and large high-technology firms draw upon these pools of talent which exist above and beyond any individual firm.

1128


6. Insights from automobiles and HDDs 6.1 Automobiles—limits to modularity It is interesting to compare systems integration trends in CoPS with trends in hightechnology, high-volume products such as automobiles and HDDs as these reveal different kinds of insights. The automobile example is interesting because it shows that, despite the rhetoric of outsourcing, the lead systems integrator companies (i.e. the major car manufacturers) still retain a great deal of control over the design and production of components. As Sako (2003) shows, in automobiles, systems integrators have turned to strategies of “modularity” to cope with the demands of technological complexity and operational efficiency. Essentially, modularity is a method of outsourcing major subsystems and components to first- and second-tier suppliers, allowing the systems integrator firm to move downstream to service the final customer more directly (e.g., in styling, distribution, after sales service, and finance).20 Pressures on car producers have led them to adopt a variety of decision paths toward outsourcing modules leading to new power balances between final car makers [or original equipment manufacturers (OEMs)] and supply chain manufacturers. As Sako (2003) shows, outsourcing represents the reallocation of tasks from within one organizational unit to another unit, normally separated by ownership. Some OEMs claim to have moved toward the outsourcing of the entire design and development of fairly high cost, complex modules. Others only outsource some basic production and assembly, while still others outsource both sets of activities. Regarding design and development, a module supplier may assume full responsibility for the module. Alternatively, the OEM may co-develop the module with the contractor in a co-located design team and perform the systems integration task with the supplier.21 Figure 3 provides a typical example of the sequence of an outsourcing strategy. In this case, the OEM has decided first to outsource the logistics and assembly of modules, second quality assurance and purchasing, and, third, the development and sourcing of major components, giving the contractor greater control over second-tier suppliers. Eventually, the contractor is given full engineering and development responsibility. In each phase, the systems integration capability of the OEM is transformed as the lead firm moves away from an in-depth control over component design and manufacture to the systems integration knowledge and skills needed to integrate the modules produced by others in the supply chain. In this kind of example, gradual outsourcing 20

Modularity can therefore be seen as a particular form of systems integration strategy, to be compared, for example, with integrated strategies where product architectures are designed with interfaces and components tailored for each product (e.g., in complex software packages and ICT networks).

21 These types of arrangements are also very common in high-volume consumer electronics and semiconductors (Hobday, 1995).

Systems integration

LOGISTICS and MODULE Assembly OEM

Supp lier

Purchasing and Quality Management OEM

Supp lier

1129

Development and Sourcing

OEM

Supp lier

Logistics Assembly Quality Purchasing Sourcing Engineering Development Source Sako (2003) p238

Figure 3 Sequencing of outsourcing tasks in automobiles.

by the OEM in theory allows a step-by-step increase in the buyer–supplier confidence which, in turn, is supposed to prevent the OEM being “captured” by the contractors. In principle, the development of large intermediate markets allows contractors to gain economies of scale and specialization, leading to an improved technological focus and greater efficiency. Contractors, it is argued, can capture more value-added by learning to design modules and through systems integration activities at the module level. This step-by-step sequence provides an incentive for contract suppliers to engage in the initial low-margin, assembly-only business. In some cases (e.g., in developing country locations such as Brazil), the OEM may never intend to allow locally based suppliers to progress much beyond the assembly of modules designed by the OEM or other first-tier suppliers. By contrast, some large module suppliers, such as Intier (who produce the car wing of the Magna) have invested heavily in systems knowledge about the whole car, so that they compete by designing and developing whole modules from the outset. However, as Sako (2003) shows, despite much rhetoric there is less modularity in the car industry than might be imagined. Figure 4 provides detailed evidence of which particular firm, the OEM or the module supplier, has control over selecting the second-tier suppliers of the components which form cockpit modules produced in Europe. As Figure 4 shows, in most cases the OEM still nominates component suppliers despite the view that OEMs should relinquish control over such components. Indeed, when interviewed, many contractors complained about the OEMs’ “reluctance to let go” and the practice of “shadow engineering.” As Sako (2003) goes on to show, this seemingly wasteful duplication of design and supplier selection tasks is, in fact, an attempt by OEMs to retain and increase their systems integration capabilities in order to maintain their control not only over current but future vehicle designs, as occurs in the case of aircraft engines (Prencipe, 2003). The publicly expressed strategies of OEMs (e.g., that they wish to focus purely on styling, financing, and marketing and withdraw from manufacturing and assembly) is in fact

1130


A IP /d ash b o ard C ro ss car b eam In stru m en t clu ster C en tral d isp lays/d ials S w itch es C en tre co n so le R ad io /I.C .E S at n av ig atio n etc. G lo v eb o x Aird u cts B ezels/v en t co n tro l H V AC system S teerin g co lu m n S teerin g w h eel D riv er airb ag P assen g er airb ag P ed al b o x W irin g h arn ess O th er w irin g F irew all K eyless en try S teerin g co lu m n sh ro u d S o u n d in su latio n

F X X

G X X

H X O

I X X

J X O

K X O

L

M X X X

O X

X

X

X XO O

X X X X O

X X X

O O O X X

X X X

X

X

X

X

X = Produced in-house by module supplier O = bought from suppliers selected by module supplier = bought from suppliers nominated by OEM

Source: Sako (2003) p238 (original data from IMVP European Module Supplier Survey)

Figure 4 Control over components for car cockpits in Europe.

not the actual practice. Few OEMs risk delegating systems integration responsibilities to powerful first-tier suppliers that could take over the design of the whole car.

6.2 HDDs—the dynamics of systems integration Research into the HDD industry throws new light on the dynamics of systems integration. As shown earlier by Prencipe (2003), systems integration is not simply a static capability concerned with current product generations. It is also a dynamic capability essential for moving successfully from one product generation to another. This dynamic feature of systems integration capability is borne out in the case of HDDs, where research by Chesbrough (2003) vividly illustrates what happens to this capability when firms move from one generation to another. Chesbrough (2003) argues that systems integration is not an end stage in the evolution of a technology or a product. At regular, recurring points existing product architectures must be transcended if performance limits are to be overcome and technological progress is to continue. The case of HDD components shows that new product designs can oscillate between modular and “integrative” states as progress from one design to another occurs. In the modular state there is far more scope for outsourcing supported by in-house modular systems integration. However, in the integrative state, there is far less scope for component standardization, the basis of modularity, and a greater proportion of component design and production occurs

Systems integration

1131

in-house (the systems integrator’s) compared with the relatively steady state where component interfaces are predictable or known. In these new design phases, the systems integrator plays an intense part in defining key components and their interfaces. Chesbrough (2003) shows that a failure to sustain and nurture different kinds of systems integration capabilities in all phases of the product life cycle can result in a “modularity trap” where lead manufacturers no longer possess the ability to incorporate novel components based on new technologies into the product. In other sectors, firms such as Microsoft and Intel, aware of this fact, have made substantial investments in systems integration (including product design and architecture, testing equipment, and design tools) to avoid these difficulties. In fact, this is probably also the case in automobiles as suggested by Sako (2003). In HDDs, Chesbrough (2003) shows that the depth of integrative knowledge in components such as thin film heads and magnitoresistive (MR) heads22 evolves over time. In the 1980–1986 period, there was a performance advantage to using thin film head, and especially the components produced in-house. However, by the 1994–1995 period, when thin film heads became widely available and MR heads began to be used, thin film heads were negatively associated with competitive performance. By contrast, the MR head was becoming crucial to superior performance, with leading firms gaining advantages by supplying MR heads (compared with businesses which outsourced these components). What Chesbrough shows was that during times where modular conditions became feasible and a competent supply chain emerged, competitive advantages were gained from outsourcing in the marketplace. By contrast, when new technology led to new types of components (e.g., MRs) then internal integration was a superior strategy. After a period (in this case circa 1995), the performance of firms producing MR drives internally fell when the detailed knowledge associated with MR became better understood. In the late 1990s MR technology became widely available, independent head makers could then offer products in large volumes at low prices, and the intermediate market for MR heads began to expand and mature. Still later, a new head technology [giant magneto-resistive (GMR)] entered the market, based on IBM’s research, and advantages were again seized by the integrated drive manufactures which had internal capabilities not only in R&D but also materials, and associated electronics. In each phase drive manufacturers with deep systems integration capabilities were better positioned to incorporate new head technology into their HDD product designs. This particular research shows not only that internal and external systems integration knowledge is a key factor in competitive advantage but also that, in fast growing mass-produced goods, an understanding of the integration-modular cycle is an essential part of forward-looking competitive strategy. Even during the modular outsourcing 22 MR heads offered a ten-fold improvement over the performance of earlier thin-film technology (Chesbrough, 2003: 183).

1132


phase, the HDD example shows that suppliers needed to retain an in-house systems integration capability covering a wide range of technologies and disciplines. This may also apply to other products such as PCs, TVs, semiconductors, cameras, and microwave ovens where product architectures are unlikely to remain in a steady stage indefinitely. Failure by firms to sustain and upgrade these integration capabilities may result in a poor competitive performance as new product designs enter the marketplace.

7. Trends in CoPS In deploying systems integration capabilities to achieve competitive advantage, we find major differences between CoPS and high-volume products. In the case of CoPS, the scope for final product standardization and modularization is less than in massproduced goods, whether assemblies (e.g., automobiles or camcorders) or components (such as HDDs or semiconductors). Because each product system tends to be tailor-made for a particular user and because volumes are far lower, the scope for standardization and the development of high-volume intermediary markets is more limited. CoPS tend to remain in the fluid design stage and never reach the mass production stage (Hobday, 1998). Consequently, for each unit of production the scope standardization, modularity, and outsourcing is less than in high-volume consumer goods. However, the importance of systems integration in manufacturing is greater than in the case of high-volume goods because of the project-based nature of production. In complex products, systems integration is always core to production, whereas in mass-produced goods it becomes a routine part of manufacturing during the highvolume stage of the product life cycle. By contrast, the scope for integrating high-value services into each product system is greater than in the case of most volume produced goods. For example Davies (2003) illustrates the “integrated solutions” strategies by which system makers deploy new systems integration skills to integrate a wide range of services and software to produce a package attractive to individual users, in the hope of gaining competitive advantage. Some of the world’s leading capital goods producers (e.g., GE and IBM) have begun selling whole solutions and research indicates that there is far greater revenue and profit from integrated service solutions than from the sales of individual product systems (e.g., in rail or aircraft engines) (Wise and Baumgartner, 1999). This trend in product systems implies a broadening of the scope of systems integration capability as firms move away from their core to provide a range of services in partnership with other suppliers or based on new in-house divisions. This strategy goes against conventional notions of core competence which imply “sticking to your knitting.”23 In fact, as Davies (2003) shows, for integrated solutions providers the ability to integrate related services, support, and sometimes externally produced hardware 23

See, for example, Hamel and Prahalad (1994).

Systems integration

1133

becomes critical. Within an integrated solutions business model, firms attempt to meet the growing range of requirements of, typically, large business or government customers as they outsource other non-core activities. To do this, suppliers need to re-assess their positions in the industry value stream and develop the systems integration capabilities to bring together these novel combinations. This trend toward solutions does not mean, as some authors suggest (e.g., Wise and Baumgartner, 1999) that firms are uniformly marching downstream from manufacturing to services. In fact, as Davies (2004) shows, product system suppliers are moving from both down- and upstream positions to try and capture the higher value territory situated between manufacturing and services. This involves systems integrators learning how to specify, deliver, finance, maintain, support, and operate a system throughout its life cycle. Davies provides detailed evidence on the solutions strategies of five major solutions suppliers in railways, mobile communications systems, corporate information networks, flight simulators, and civil engineering markets. To understand recent trends in high cost capital goods, it is helpful to show how firms position themselves against the main areas of value activity, as depicted in Figure 5, where the outputs of one value-adding activity are the inputs of the next, making up the overall value stream. Each of these activities becomes progressively closer to the final consumer, such as the railway passenger or the telephone users. The four core activities (in bold in Figure 5) include:

• Manufacture: taking raw materials and sub-assemblies, and transforming them into physical components and subsystems which are then used to manufacture a system; • Systems integration: adding value through the design and integration of physical components, subsystems, software, and embedded services; • Operational services: supporting and maintaining the system to provide services, such as a corporate telecom network, baggage handling, flight simulation training, and rail services; • Final service provision: providing to the final consumer with the service (e.g., telecom or air transport), and engaging in brand, marketing, distribution, and customer care activities. Mobile communications during the mid-1990s provides an interesting example of new value stream positioning. Initially, suppliers such as Ericsson and Nokia were primarily manufacturers of equipment, supplying mobile phone operators with the equipment (for example, base stations, transmission equipment, and switches) they needed to build a mobile network, while the mobile operators maintained systems integration expertise in-house. By the late-1990s mobile operators began outsourcing the systems integration task and a new type of service provider entered the market, namely the so-called “mobile virtual network provider.” Virgin Mobile, for example, developed its subscriber base through brand image, advertising, and customer care

Design and physically make components & subsystems

(Backwards)

Manufacture

Figure 5 The value stream in high-technology capital goods.

Source: adapted from Davies (2004)

Raw materials, intermediate goods, primary product manufactures

Earlier stages

UPSTREAM

Services

Design & integrate products & systems

Systems integration

Manufacturing

Maintain and operate products & systems

Vertical moves

Operational services

Buy in maintenance & operational capacity to provide services to final consumers

Final Service provision

Consumption of service by final consumer or end user (e.g. train passenger)

(Forwards)

Final consumer

DOWNSTREAM

Added value

1134


Systems integration

1135

while buying in network capacity from another operator to carry its telecommunications traffic. In reality, rather than the simple linear stages of Figure 5, the value-adding activities involve a series of dynamic feedback loops between later and earlier stages of product development (Hobday, 1998: 694). Systems integrators ensure that the manufacturers in earlier stages of production produce their components as “integrateable” packages that conform to an overall design. Through “learning by using,” operators and service providers jointly identify opportunities to improve the overall system performance and feedback lessons into the design of future products. Research shows that other product system suppliers are moving into maintenance, renovation, training, and operation of product systems. As Davies (2003) shows, firms such as Alstom and Ericsson now offer services both to their business customers and to final consumers. Alstom, for example, has developed a train management system to monitor and resolve faults on the fleet of Pendolino trains operated by Virgin Trains. Similarly, Ericsson provides a 24-hour software-controlled network management service to support their customers’ mobile phone network. Some firms, such as Alstom, GE, and Ericsson, have set up new divisions to provide technical support and final services to consumers. For example, Alstom’s Service Business unit offers “Total Train-Life Management” to its buyers. GE in the US has a similar division. These strategies are designed to generate revenues through all stages in the operating life cycle of a train, from services such as maintenance, renovation, spare parts sales, and asset management. The typical life cycle of a train extends over 30 years, including 2 years to design, build, and manufacture followed by a 28-year in-service period. For example, although the cost of building a fleet of 70 diesel trains can be around £65 million, the service life generates revenues worth around £200 million, making the service market extremely attractive for systems integrators. As some systems integrators begin to provide operational services they gain a direct incentive to design systems from the start that are more reliable, efficient, and easily maintainable. Also, by participating in services, systems integrators are able to learn about design and integration from in-service problems and can use learning to improve overall system re-design and performance. There is also a strong incentive to do this as in the case of Rolls-Royce “power by the hour” strategy where profit is made from service contracts rather than spare parts, as in the past. Because manufacturers such as Alstom, Ericsson, and Thales develop technology and perform operational services, they are able to create new and broader feedback channels within different parts of the same firm. Using this closed loop, firms hope to initiate a virtuous cycle of innovative improvements between systems integration and service activities, leading to more reliable and efficient systems (Geyer and Davies, 2000). By contrast, pure systems integrator firms such as WS Atkins and C&W which have traditionally relied on external manufacturers for equipment and technology are unable to take advantage of these dynamic feedback loops. Revenue streams and profitability are far larger in-service

1136


operations and often more predictable and defendable as buyers engage in long-term contracts for service-intensive solutions.

8. Research implications Research shows that systems integration is no longer “merely” a technical task (if it ever was) but is now a core strategic business capability which enables high-technology firms to shape their boundaries and their position in an industry value stream, enabling them to decide who to compete with, who to collaborate with, what to make in-house, and what to outsource. Systems integration includes the principal technical, strategic, and organizational capabilities involved in the main stages of the production of each product system. Recently, systems integration capabilities have enabled productsystem suppliers to bundle together their traditional products and software with services designed to provide customers with long-term solutions to their problems. According to recent interpretations, lead firms as systems integrators represent the visible hand of the marketplace, playing the mediator between the “Smithian” and “Chandlerian” firm, enabling the market to gain the benefits of both specialization and integration (Chesbrough, 2003; Dosi et al., 2003; Pavitt, 2003). According to this view, systems integrators deploy purposeful strategies to create networks of production, using both market competition and vertical integration to deliver products and systems. On the one hand, increasing specialization and the division of labor improves productivity (Pavitt, 2003). Vertical disintegration can lead to the growth of intermediary markets and the emergence of specialist firms who are able to produce inputs more creatively and efficiently than large manufacturers. On the other hand, increasing and more effective integration can also improve productivity.24 System integration capabilities are the mechanism by which firms are able to gain the benefits of both specialization and integration through the life cycle of new product as shown in the case of HDDs (Chesbrough, 2003). What the evidence implies is that systems integration is the primary capability by which lead firms simultaneously manage (and gain the benefits from) the “twin” processes of vertical integration and disintegration, as these change through time for each product and system in question. How these processes are managed technically and organizationally is central to the competitiveness of the firm as indicated by studies of aircraft engines (Prencipe, 1997), automobiles (Sako, 2003), and HDDs (Chesbrough, 2003). Rather than assuming one is always better than the other research (i.e. specialization versus integration), future research might help us understand how system integration capabilities allow firms to decide whether to move up- or downstream,

24

As Nightingale et al. (2003) argue, the benefits from integration at the firm level include capacity utilization which can lead to falling costs and prices. Improved capacity utilization itself can flow from the controlled allocation of resources to specialized services

Systems integration

1137

integrating some activities and disintegrating others, and which particular systems integration skills are needed to do this. Nightingale et al. (2003) and Davies, (2004) point to other Chandlerian benefits to be gained from firms with strong systems integration capability. For example, as the complexity of the product system increases, so does the number of feedback loops in the design and production process from later to earlier stages of production. A deep systems integration capability is needed to generate an understanding of how to avoid the risks of uncontrolled feedback as “systemness” increases. Forward integration into services provides a direct and wide channel by which learning about system functioning can feed back into the design process. Further research might show precisely how these feedback loops are developed and managed in practice. The external environment has a strong shaping effect on the development of systems integration capabilities. In countries such as the UK, users of systems increasingly expect suppliers to engage in the financing of the production of new product systems and sometimes the operation of the system (e.g., in flight simulation and telecommunications). This trend means that product system suppliers need to gain new capabilities to compete. Another external trend, privatization, has encouraged previously national monopolies, with integrated R&D departments, to outsource the design and development and, sometimes, the operations of major systems to systems integrators, again leading to the acquisition of new capabilities. In the UK, this has occurred in the electricity, telecommunications, rail, airlines, banking, water, and gas supply industries. Further research into Government programs such as the private finance initiative and public–private partnership (e.g., in underground rail systems, military, and hospital construction) could show how far systems integrators have been successful or not in moving further downstream into operations, finance, training, and consulting. Finally, emerging corporate strategies toward systems integration pose interesting research questions. For example, research on automobiles and electronic components shows how integration strategies and capabilities must change through the cycle of new product design and introduction, especially when new kinds of technology are incorporated. From a strategic management perspective, understanding and acting on the dynamics of the cycle is the key to competitive performance as the nature and intensity of internal and external systems integration changes radically over the cycle. However, these trends only appear to apply to high-volume consumer and components industries. In low-volume product systems, there is more advantage to be gained from by increasing service revenues per unit of production, which explains why systems integrators are shifting downstream into services such as maintenance, finance, consultancy, and operations (Davies, 2004). Future research could show if those firms which fail to build up the capabilities for integrating new services into their systems offerings fall behind competitively or whether the traditional “stick to your knitting” approach remains valid in some industries.

1138


9. Conclusions In complex products and systems and other high-technology goods, systems integration is a core technical, strategic, and organizational capability. Systems integration is no longer solely an engineering or operations task and perhaps never was. As a capability, it permits a supply chain leader to develop and deliver new products, co-ordinate supply networks, and continually optimize its position within the value stream of an industry. The historical path, depth, and trajectory of a system integrator’s capability helps determine the firm’s distinctive position in the marketplace and shapes its decisions over who to partner with and whether to buy, make, or collaborate in design and production. The distinctive strengths and dimensions of this capability enable a particular firm to choose whether or not to outsource specific elements of design and production to other firms. As system scope and novelty increases, and as outsourcing becomes increasingly commonplace, systems integration capability becomes ever more important. Viewed in this way, systems integration is a capability at the heart of the strategic management of the modern high-technology corporation. Recent evidence shows that suppliers of high-value capital goods and systems are forming long-term service-intensive partnerships with downstream customers who use the systems to provide services to consumers. This trend is in response to new customer demands and, in some cases, privatization and de-regulation. Demands for ever more complex products have also led systems integrators to form long-term outsourcing relationships with upstream suppliers of components, technologies, skills, and knowledge. In countries such as the UK, the outsourcing of system development activities previously carried out by national monopolies has increased the importance of systems integration capability to the private sector firms engaged in supplying these systems. In both military and civilian sectors, systems integration capabilities are widely diffused, underpinning outsourcing and “insourcing” practices and, in some cases, strategies of modularity in design and production. Research reveals major differences in the strategies of systems integrators depending on the product in question. In high-volume products such as automobiles and HDDs, firms use their capabilities to achieve competitive advantage by exploiting upstream component supplier relationships in ways which differ according to the particular phase of each product life cycle. By contrast, in low-volume, high-cost capital goods, manufacturing firms are focussing more on exploiting downstream relationships with system users by integrating services such as maintenance, finance, consultancy, and operations within their product offerings. In both cases, systems integration capability enables firms to move selectively, and simultaneously, up- and downstream to gain advantages in the marketplace. The article outlines some new research directions. By bringing together existing approaches to understanding systems, the typology introduced provides a starting point for systematically examining and comparing the capabilities required to integrate different kinds of technological system. In high-volume products, research is

Systems integration

1139

needed to compare the systems integration strategies and outcomes in different industries to look for common patterns and sectoral differences. How firms actually deploy systems integration capabilities to optimize the twin processes of vertical integration and disintegration is also an important research question. In high-value capital goods, it would be interesting to see whether those firms which fail to build up capabilities for integrating services to provide customer centric “solutions” fall behind competitively. Research from the system user perspective could investigate the risks and disadvantages of lock in, as prime contractors offer bundled solutions on a long-term contractual basis. While many research questions remain, there can be little doubt that systems integration is a fascinating perspective for helping to understand the changing landscape of high-technology industries.

Acknowledgements This research is part of the Complex product systems Innovation Centre, funded by the UK Economic and Social Research Council (ESRC). The authors are grateful to Tim Brady and Paul Nightingale and to an anonymous referee for insightful, extensive comments on earlier drafts. The normal disclaimers apply.

Address for correspondence Michael Hobday and Andrea Prencipe, Complex Product Systems Innovation Centre, Science and Technology Policy Research (SPRU), Freeman Centre, University of Sussex, Falmer, Brighton BN1 9QE. e-mail: [email protected] Andrew Davies, Innovation Studies Centre, Tanaka Business School, Imperial College London, South Kensington Campus, London, SW7 2AZ. e-mail: a.c.davies@ imperial.ac.uk

References Arora, A. and A. Gambardella (1999), ‘Chemicals,’ in D. Mowery (ed.), US Industry in 2000: Studies in Competitive Performance. National Academy Press: Washington. Balconi, M. (2002), ‘Tacitness, codification of technological knowledge and the organisation of industry,’ Research Policy, 31, 357–379. Barney, J. (1991), ‘Firm resources and sustained competitive advantage,’ Journal of Management, 17(1), 99–120. Best, M. H. (2003), ‘The geography of systems integration,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford. Bijker, W. E., T. P. Hughes and T. J. Pinch (1987), The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology. MIT Press: Cambridge, MA.

1140


British Aerospace Annual Report (1998), Farnborough Hants. Brusoni, S., A. Prencipe and K. Pavitt (2001), ‘Knowledge specialization and the boundaries of the firm: why do firms know more than they make?,’ Administrative Science Quarterly, 46, 597–621. Chandler, A.D. Jr (1990), Scale and Scope: The Dynamics of Industrial Competition, Harvard University Press: Cambridge, MA. Cherington, P. W., M. J. Peck and F. M. Scherer (1962), ‘Organization of research and development decision making within a government department,’ in R. R. Nelson (ed.), The Rate and Direction of Inventive Activity: Economic and Social Factors, A Report of the National Bureau of Economic Research, New York, Princeton University Press: Princeton. Chesbrough, H. (2003), ‘Towards a dynamics of modularity: a cyclical model of technical advance,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: . Cohen, W. and D. A. Levinthal (1989), ‘Innovation and learning: the two faces of R&D,’ The Economic Journal, 99, 569–596. Davies, A. (1996), ‘Innovation in large technical systems: the case of telecommunications’, Industrial and Corporate Change, 5(4), 1143–1180. Davies, A. (2003), ‘Integrated solutions: the changing business of systems integration,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford. Davies, A. (2004), ‘Moving base into high-value integrated solutions: a value stream approach,’ Industrial and Corporate Change, 13(5), 727–756. Davies, A., T. Brady, P. Tang, M. Hobday, H. Rush and D. Gann (2003), ‘Delivering Integrated Solutions,’ mimeo, SPRU: Brighton, p. 35. Davies, A. and Hobday, M. (2005), The Business of Projects: Managing Innovation in Complex Products and Systems. Cambridge University Press: Cambridge. D’Adderio, L. (2001), ‘Crafting the virtual prototype: how firms integrate knowledge and capabilities across organisational boundaries,’ Research Policy, 30(9), 1409–1424. D’Adderio, L. (2002), Bridging Formal Tools with Informal Practices: How Organisations Balance Flexibility and Control. Research Centre for Social Sciences (RCSS): Edinburgh. Dosi, G., M. Hobday, L. Marengo and A. Prencipe (2003), ‘The economics of systems integration: towards an evolutionary interpretation,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford. Geyer, A. and A. Davies (2000), ‘Managing project-system interfaces: case studies of railway projects in restructured UK and German markets,’ Research Policy, 29, 991–1013. Gholz, E. (2003), ‘Systems integration in the US defence industry: who does it and why is it important?’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford. Goode, H. H. and R. E. Machol (1957), System Engineering: An Introduction to the Design of Large-Scale Systems. McGraw-Hill: New York.

Systems integration

1141

Hamel, G. and C. K. Prahalad (1994), Competing for the Future. Harvard Business School Press: Boston, MA. Hobday, M. (1995), Innovation in East Asia. Edward Elgar: Aldershot. Hobday, M. (1998), ‘Product complexity, innovation and industrial organisation,’ Research Policy, 26(6), 689–710. Hobday, M., A. Prencipe and A. Davies (2003), ‘Introduction,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford. Hughes, T. (1983), Networks of Power: Electrification in Western Society, 1880-1930. Johns Hopkins University Press: Baltimore. Hughes, T. P. (1998), Rescuing Prometheus. Pantheon Books: New York. Johnson, S. B. (1997), ‘Three approaches to big technology: operations research, systems engineering, and project management,’ Society for the History of Technology, pp. 891–916. Johnson, S. B. (2000), ‘From concurrency to phased planning: an episode in the history of system management,’ in A. C. Hughes and T. P. Hughes (eds), Systems, Experts, and Computers: the Systems Approach in Management and Engineering, World War 11 and After. MIT Press: Cambridge, MA. Kline, S. J. (1990), ‘A numerical measure for the complexity of systems: the concept and some implications,’ Report INN-5, Thermosciences Division, Department of Mechanical Engineering, Stanford University: California. Landau, R. and Rosenberg, N. (1992), ‘Successful commercialization in the chemical process industries,’ in N. Rosenberg, R. Landau and D. Mowery (eds), Technology and the Wealth of Nations. Stanford University Press: Stanford. Mayntz, R. and T. P. Hughes (eds) (1988), The Development of Large Technical Systems. Campus Verlag: Frankfurt am Main. Miller, R. and M. Cote (1987), Growing the Next Silicon Valley: a Guide for Successful Regional Planning. Lexington Books: Massachussetts. Miller, R., M. Hobday, Leroux-Demers and X. Olleros (1995), ‘Innovation in complex systems industries: the case of flight simulation,’ Industrial and Corporate Change, 4(2),. Mowery, D. and N. Rosenberg (1982), ‘The commercial aircraft industry,’ in R. Nelson (ed.), Government and Technical Progress: a Cross-Industry Analysis. Pergamon Press: New York. Mowery, D. and N. Rosenberg (1989), Technology and the Pursuit of Economic Growth. Cambridge University Press: Cambridge. Nelson, R.R. and N. Rosenberg (1993), ‘Technical innovations and national systems,’ in R. R. Nelson (ed.), National Innovation Systems: A Comparative Analysis. Oxford University Press, New York. Nightingale, P., Brady, T., A. Davies and J. Hall (2003), ‘Capacity utilisation revisited: software, control and large technical systems,’ Industrial and Corporate Change, 12(3), 477–517. Pavitt, K. (2003), ‘Specialization and systems integration: where manufacture and services still meet,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford.

1142


Peck, M. J. and F. M. Scherer (1962), The Weapons Acquisition Process: an Economic Analysis, Division of Research, Graduate School of Business Administration, Harvard University: Boston, MA. Penrose, E. T. (1959), The Theory of the Growth of the Firm. Basil Blackwell: Oxford. Pisano, G. (1997), The Development Factory. Harvard University Press: Boston, MA. Porter M. E. (1990), The Competitive Advantage of Nations. Macmillan: London. Prencipe, A. (1997), ‘Technological competencies and product’s evolutionary dynamics: a case study from the aero-engine industry,’ Research Policy, 25, 1261–1276. Prencipe, A. (2003), ‘Corporate strategy and systems integration capabilities: managing networks in complex systems industries,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford. Prencipe, A., A. Davies and M. Hobday (eds) (2003), The Business of Systems Integration. Oxford University Press: Oxford. Ramo, S. (1969), The Cure for Chaos: Fresh Solutions to Social Problems Through the Systems Approach. David McKay Company: New York. Richardson, G. B. (1972), ‘The organisation of industry,’ Economic Journal, 82, 883–896. Rosenberg, N. (1963), ‘Technological change in the machine tool industry, 1840–1910,’ The Journal of Economic History, 23(4), 413–443. Rosenberg, N. (1979), ‘Technological interdependence in the American economy,’ Technology and Culture, 20, 25–50. Rosenberg, N. (1994), Exploring the Black Box: Technology, Economics, and History. Cambridge University Press: Cambridge. Sako, M. (2003), ‘Modularity and outsourcing: the nature of co-evolution of product architecture and organization architecture in the global automotive industry,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford. Sapolsky, H. M. (1972), The Polaris System Development: Programmatic and Bureaucratic Success in Government. Harvard University Press: Cambridge, MA. Sapolsky, H. M. (2003), ‘Inventing systems integration,’ in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford University Press: Oxford. Shenhar, A. J. (1994), ‘A new conceptual framework for modern project management, in T. M. Khalil and B. A. Bayraktar (eds), Management of Technology, IV. Institute of Industrial Engineers: . Shenhar, A. J., D. Dvir and O. Levi (1994), ‘Mapping the dimensions of project’s success,’ in T. M. Khalil and B. A. Bayraktar (eds), Management of Technology, IV, Institute of Industrial Engineers: . Sturgeon, T. (2002), ‘Modular production networks: a new American model of industrial organization,’ Industrial and Corporate Change, 11(3), 451–496. Summerton, J (ed.) (1995), Changing Large Technical Systems. Westview Press: Oxford.

Systems integration

1143

Teece, D. J. and G. Pisano (1994), ‘The dynamic capabilities of firms: an introduction,’ Industrial and Corporate Change, 3, 537–556. Teece, D. J., G. Pisano and A. Shuen (1994), ‘Dynamic capabilities and strategic management,’ CCC Working Paper No. 94–99, Centre for Research in Management, University of California: Berkeley. Walker, W., M. Graham and B. Harbor (1988), ‘From components to integrated systems: technological diversity and integration between the military and civilian sectors,’ in P. Gummett and J. Reppy (eds), The Relations Between Defence and Civil Technologies. Kluwer Academic Publishers: London. Wise, R. and P. Baumgartner (1999), ‘Go downstream: the new profit imperative in manufacturing,’ Harvard Business Review, September-October, 133–141. Zuboff, S. (1988), In the Age of the Smart Machine: The Future of Work and Power. Heinemann: Oxford.

Systems integration: a core capability of the ... - Semantic Scholar

Systems integration: a core capability of the ... - Semantic Scholar

Suggest Documents

Protocol Systems Integration, Interface, and ... - Semantic Scholar

A Cortical Core for Dynamic Integration of ... - Semantic Scholar

functional integration and systems implementation ... - Semantic Scholar

functional integration and systems implementation ... - Semantic Scholar

the usage of capability maturity model integration

Integration of e-Health Systems with the Open ... - Semantic Scholar

Operationalization of the Capability Approach ... - Semantic Scholar

Implementing a Scientific Visualisation Capability ... - Semantic Scholar

Unpacking Dynamic Capability: A Design ... - Semantic Scholar

Evaluating a Usability Capability Assessment - Semantic Scholar

from a dynamic capability perspective - Semantic Scholar

A Data Integration Broker for Healthcare Systems - Semantic Scholar

Quantifying the Relationship between Capability ... - Semantic Scholar

Surge capacity and capability. A review of the ... - Semantic Scholar

(VORD) With The Capability Model Integration (CMMI

The AMIS Approach to Systems Integration: An ... - Semantic Scholar

Integration of positioning capability into an existing

Predictive Capability of Anorectal Physiologic ... - Semantic Scholar

Building organisational capability - Core

Integration of Bt Cotton in IPM systems: An ... - Semantic Scholar

Integration of Descending Command Systems for ... - Semantic Scholar

Systems biology of myasthenia gravis, integration ... - Semantic Scholar

The Integration of Cumulative Environmental ... - Semantic Scholar

Process Capability Analysis of a Centrifugal ... - Semantic Scholar