international manufacturing networks—to develop global ... · ... global manufacturing strategy;...

Ž .Journal of Operations Management 16 1998 195–214

International manufacturing networks—to develop globalcompetitive capabilities

Yongjiang Shi ), Mike GregoryManufacturing Engineering Group, The UniÕersity of Cambridge, Mill Lane, Cambridge CB2 1RX, UK

Abstract

This paper seeks to extend existing manufacturing system concepts and develop new structured knowledge aboutinternational manufacturing networks by analysing the networks, classifying the configurations and identifying thecapabilities. The design and operation of international manufacturing networks is an increasingly important issue fortransnational corporations faced with rapid changes in global market opportunity, competition and new managerialmechanisms. Four international manufacturing networks in mechanical and process industries are analysed and a number ofconclusions drawn: first, a novel configuration map is proposed; second, key strategic capability parameters are identified;third, networking trends and their implications for configuration are discussed. Finally, the paper explores strengths andweaknesses of the particular methodology adopted in this research. q 1998 Elsevier Science B.V. All rights reserved.

Keywords: International manufacturing networks; Global manufacturing strategy; Manufacturing strategic capability

1. Introduction

Business globalisation is leading to widespreadrestructuring of international manufacturing systems

Ž . Žin transnational corporations TNCs UNCTAD,.1993 . Little systematic research in manufacturing

engineering or productionroperations managementŽ .PrO M has focused on these new manufacturing

Žsystems Anderson et al., 1989; Lawrence andRosenblatt, 1992; Miller and Roth, 1994; Shi and

.Gregory, 1995 . Many critical issues, such as proper-ties of international manufacturing network systemsin terms of structural architecture, dynamic mecha-nisms, and related strategic capabilities and strategyprocesses, are poorly understood and are not covered

) Corresponding author. Fax: q44-1223-338076.

by current manufacturing research agendas. In indus-try, lack of global vision and appropriate strategiesduring the internationalisation have become majorbarrier to the effective management of international

Ž .operations Klassen and Whybark, 1994 .Manufacturing engineering and PrO M are expe-

riencing a rapid change in manufacturing systemconcepts from a factory focus towards a corporate

Ž .international factory network Ferdows, 1989 . Thereare strong parallels with computer developments fromcentralised mainframe systems to distributed net-works. Nowadays, in computing, nobody doubts thenetwork is the computer and has more capabilitiesthan standard machines alone. Similarly, the world-wide dispersed corporate factory network can beseen as the manufacturing system but with manydifferent characteristics from the classic model. Net-works, therefore, are not only a new type of manu-

0272-6963r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.Ž .PII S0272-6963 97 00038-7

( )Y. Shi, M. GregoryrJournal of Operations Management 16 1998 195–214196

facturing system deriving new strategic capabilitiesand requiring design tools but also posing new theo-retical questions about systems and decision pro-cesses.

The paper is organised in five parts. The ‘back-Ž .ground’ Section 2 explains why international man-

ufacturing needs more attention and what the mainŽ .issues are. The ‘research design’ Section 3 intro-

duces the research approach and some conceptualmodels of the networks, and details the relationshipbetween this paper’s results and a practical project.In Section 4, four international manufacturing net-works from 15 studied cases in engineering andprocess industries are then analysed to demonstratethe most recent transformation of the networks. Fol-lowing the cases, two revised conceptual models arepresented in Section 5—strategic capability parame-ters and the configuration map and discusses somenew findings from the main driving forces, trends ofnetworks, to optimised configurations. The final partof the paper discusses the need for further researchwork about networks.

2. Background

2.1. The changing industrial enÕironment

The international market is not only a sum ofdifferent national markets. Traditionally, interna-tional business strategy is based upon these individ-ual markets and sets up objectives and policies sepa-rately to satisfy the specific requirements of differentcountries. This type of international development isan outcome of ‘multidomestic’ oriented strategyŽ .Dicken, 1986; Porter, 1980 . As international com-munication and exchanges extend, especially underthe trends of political regionalisation and rapid eco-nomic growth in some developing countries, theglobal market—a new type of international market is

Ž .emerging Levitt, 1983; Yip, 1992 . This global mar-ket is based on the shared and common demands ofdifferent countries. It integrates different nationalpreferences into a core entity and presents this as afundamental and nondifferentiable market require-ment. To satisfy the growing global market, thetraditional products and related development strate-gies are clearly not enough to satisfy companies’

internationalisation. TNCs are therefore keen to learnabout and develop global or world products and torestructure their worldwide manufacturing systems.

Although domestic and regional markets will con-tinue to exist, the reality of global competition isvery clear. It means not only an increase in thenumber of competitors in the world but also a quali-tative change in the nature of competitive advantage.Many TNCs have sought to achieve lower cost,higher quality, faster product introduction, greaterflexibility, and shorter delivery time, all at the same

Ž .time Daniels and Daniels, 1993 . This has become anecessary requirement of global competition. The

Ž .‘trade-offs’ principle of Skinner 1969 and theŽ .generic competitive strategy of Porter 1980 obvi-

ously are not enough to satisfy the global competi-tive requirements. Companies therefore need to paymore attention to their internal capability buildingand new patterns of manufacturing system searchingto create a proactive competitive influence ratherthan a reactive responsiveness to the new industrial

Ž .environment Hayes et al., 1988 .The emergence of the global market and intensifi-

cation of global competition is matched by majordevelopments in technology. New generations ofcommunication and transportation technologies arecreating the possibility for TNCs to organise theirworldwide operations more effectively and effi-ciently. There are also major changes in the waypeople think about manufacturing. New system tech-nologies, for example, are guiding managers andengineers to have more integrated and dynamic vi-sion about manufacturing systems. Indeed manufac-turing systems are increasingly expected to fulfilnew requirements for learning and adaptationŽ .Leonard-Barton, 1992; Hayes and Pisano, 1994 .

In general, these new driving forces—global mar-ket opportunity, new patters of competition, andreorganising potential or possibility—require a newgeneration of networks beyond the classical pipelineof physical transformation. The networking charac-teristics of the new manufacturing system must in-volve a wide perspective covering geographic disper-sion and interdependent coordination rather than thetraditional focus on separated manufacturing sites.These new perspectives require new understandingsof the nature of manufacturing systems and the waysin which the required performance can be achieved.

( )Y. Shi, M. GregoryrJournal of Operations Management 16 1998 195–214 197

2.2. Literature reÕiew on international manufactur-ing networks

Over the last ten years, manufacturing has beentaken increasingly seriously by both industry andacademics. In contrast to studies at factory level,however, the research focusing on the internationalmanufacturing networks is quite scarce. There seemsto be a widespread belief that, if a company mastersmanufacturing strategy and factory design, the strat-egy can be applied no matter where the manufactur-

Ž .ing is located or distributed. Sprague 1990 repre-sented this principle as a series of decision grids bywhich the markets could be identified and served toconstruct a series of internationally distributed facto-ries. International manufacturing network design issimplified into location decision and good factorydesign. Following similar assumptions, DuBois et al.Ž .1993 studied the relationship between international

Ž .manufacturing configurations IMCs and manufac-turing strategy. They found four key manufacturing

Žperformance priorities efficiencyrcost, quality, de-.pendability, and flexibility which, in addition to

three other key variables, market orientation, experi-ence and product characteristics, play important rolesin the development of IMC strategies.

However, international manufacturing networksŽ .are integrated rather than aggregated. Flaherty 1986

studied one pattern of IMCs and provided a soundbasis for the observation and exploration about inter-national manufacturing systems. Based on empiricalresearch in microelectronic industry, she suggestedthat one type of IMC with particular patterns ofgeographic dispersed facilities and shared commoninfrastructurermechanism could lead to a synergyadvantage in the network. The work does not, how-ever address linkage to the corporate strategy ortackle comprehensively the range of possible IMCsand coordination to give companies more alterna-tives during their internationalisation.

Ž .Ferdows 1989 defines the international manufac-turing system as a network of factories. Each factoryplays different strategic roles in the network, e.g.,off-shore, source, server, contributor, outpost andlead. ‘‘Using it for analysis of factory networksprovides a fresh perspective and yields new insights.’’So the focus of Ferdows is on the relation betweennetwork and its factories. He tries to link the strate-

gic motivations to the role of each factory in anetwork and emphasises new opportunities for learn-ing in the networks. This model perhaps put a littletoo much emphasis on strategic role of separatefactories rather than functions of the integrated orcoordinated network.

Ž .Cohen et al. 1989 consider the network includ-ing vendors, plants and markets. But their interna-tional manufacturing strategy is mainly driven by theinternational financial factors, such as fluctuations ofcurrency exchange rate, global corporate taxes, tar-iffs and so on. These factors could influence thebehaviours of TNCs but may be not the drivingforces to the strategy.

The interdependent coordination in the networksis a determinant of network structure. Dicken, aBritish economic geographer, observing the tide oflarge-scale production geographic dispersion glob-ally, generalises two patterns of production coordina-tion strategy. He calls them horizontal expansionand Õertical integration. ‘‘Each of these strategieshas rather different implications for the internal rela-tionships between the individual parts of the enter-prise. In particular, units which are part of a verti-cally integrated operation tend to be locked into amuch tighter network of internal interdependencies

Ž .with other parts of enterprise’’ Dicken, 1986, 1992 .Ž .De Meyer and Vereecke 1994 recently also suggest

two types of manufacturing network architecture—product focus and process focus, the former is sameas Dicken’s horizontal expansion strategy and thelatter is as vertical integration strategy. They mention‘‘the choice between the process and product focushas strong implications for the range of tasks of thecorporate manufacturing staff and the factory man-agement. The factory in a process-focused companyin general operates less autonomously than the fac-tory in a product-focused company’’. These strategypatterns and network architecture at manufacturinglevel are the consequence of the globalised or mul-tidomestic focused strategy of the corporation.Therefore, the coordination between future marketrequirements and corporate strategies influences notonly the corporate organisation but also the manufac-turing network and factory internal characteristics.The pattern choices of coordination must be madebefore the international manufacturing network isphysically designed. The problem is that, even though


it is a very important parameter of the networks, thecoordination dimension alone is not sufficient topresent the whole picture of international manufac-turing networks.

In summary, there still appears to be little seriousresearch concerning integrated networks of factoriesas whole systems and little about the process ofstrategy development to build such networks. Eventhough reviewed research shows some people havestarted to observe the dispersed factories in an inte-grative and interdependent way and tried to put thefactories into some kinds of configurations, the gen-eral industrial requirements in manufacturing interna-tionalisation are far from satisfied. This representsanother serious missing link in corporate strategyŽ .Skinner, 1969 .

2.3. New agenda for international manufacturingsystems

A new concept of manufacturing systems needs tobe developed as manufacturing business developsinternationally. However, as TNCs’ internationalisa-tion is a very complex process, it is not possible toidentify one dominant pattern suitable for all futuredevelopment. Even while many TNCs believe themain stream will be global integration or coordina-tion, may others believe the multidomestic nation by

Žnation to be more appropriate Kanter, 1991; Fleenor,.1993 .

Faced by contingency options for internationalisa-tion, international manufacturing managers needknowledge about the options and network character-istics in their strategy analysis and network design.Work to develop the necessary new knowledge isdifficult to justify within individual companies but isan appropriate subject for research aimed at newtheory building and practical tool development. Re-search into the international manufacturing networkstherefore should answer the following fundamentalquestions.

Ø What are the international manufacturing net-works? Are existing manufacturing system conceptsadequate and if not what constructs should be?

Ø What should the international manufacturingnetworks do? What strategic requirements andrormissions are expected from networks and strategiccapabilities does this suggest?

Ø What could the international manufacturingnetworks do? How can different configurations ofnetworks and the relationship between capabilitiesand configurations be understood, e.g., by classifyingthe networks, mapping their transformations, andprofiling capabilities?

Ø How to match ‘should’ to ‘could’ to design anetwork? How might a strategy process be devel-oped to guide the transformation of internationalmanufacturing system towards global coordinatednetworks.

It is clear that such questions requires intensivelytheoretical and practical research. And it is alsoessential to design an integrated research approach totackle these issues.

2.4. International manufacturing network researchprogramme

Research adopted to tackle the above issues formsa wide research programme. The programme’s objec-

Ž .tives are: 1 To develop new theoretical knowledgethat better understands the international manufactur-ing networks in terms of their missions, configura-

Ž .tions and capabilities. 2 To develop a practicalprocess detailing the analysis and decision of theinternational manufacturing strategy and networktransformation towards globally coordination.

The programme thus includes two basic parts—knowledge-based research work and application-based research work—shown as the core of Fig. 1.

Fig. 1. Three stages of the international manufacturing networkresearch programme.


The research programme extends manufacturingsystem concept from factory focus to internationalfactory network, but at the same time is limitedwithin the factory network boundary. This type ofinternational factory network usually belongs to one

Ž .strategic business unit SBU in a transnational cor-poration, which is consistent with the boundary of

Žthe manufacturing strategy Skinner, 1969; Hayes.and Wheelwright, 1984 . As an international manu-

facturing system can be seen a factory network withmatrix connections, in contrast to a more linearsystem of a factory, the linkages with the environ-ment through sourcing, strategic alliances and R&Dnetworking are of course important for the networks.But for this research work, the focus is just theinternational manufacturing factory network in whicha TNC has direct investment, no matter how large orsmall percentage of the ownership, since this repre-sents the activities over which the company hasdirect management control contrasting with othertypes of collaboration. Another reason for this focusis that it should be the bottom line for a company tofirst optimise or better manage its own resources.

The international manufacturing network researchprogramme has been designed as three major stages—theoretical model building, workbook methodol-ogy pilot testing, and workbook testing through de-

Ž .tailed case studies Fig. 1 . The programme in gen-eral seeks to combine theory construction and practi-cal application through the ‘process research ap-

Ž .proach’ Platts and Gregory, 1990 . In its first andsecond stages, 15 case companies have been studiedbased on the conceptual models and pilot workbook.The second stage is a connection between theoryconstruction and its application, transferring the newknowledge into a practical workbook assisting inter-national manufacturing strategy formulation and net-work design. Currently the research programme is

entering the third stage to test and apply the work-book entitled as ‘Global Manufacturing Strategy—Developing New Manufacturing Capability’.

2.5. The focus of this paper

This paper mainly introduces the knowledge-baseresearch work in the research programme, by focus-ing on the first and second stages of the researchprogramme in Fig. 1, although the background to thefull programme is briefly introduced in the paper toprovide contest. Presenting four typical cases todemonstrate the emerging and transformation of in-ternational manufacturing network systems, the pa-per sets out to discuss typical options of the networkconfigurations and their related strategic capabilitiesand managerial behaviours. However, as this knowl-edge-based research is only a part of the widerprogramme, it might be helpful to remind that theresearch strategy, model constructions and methodol-ogy choice are designed from the point of view ofthe whole programme rather than just for findingknowledge or building theory.

3. Research design

3.1. The knowledge-based research

Fig. 2 sets out the approach to the knowledge-based research work, starting from pilot industrialcase studies and workshop discussions and relatedliterature survey to the identification of key indus-trial issues and research agendas. Drawing on indus-trial requirements for tools and processes for interna-tional manufacturing decisions in the first phase ofthe research, conceptual models are developed to

Fig. 2. The research procedure of the theory-building part in the knowledge-based research.


guide field observation and underpin workbook inthe second phase. In the third phase, supported witha pilot workbook application in companies, casestudies serve to detail and test the conceptual modelsleading to modifications where necessary. They alsoprovide an exploration of new issues not captured inthe conceptual models and a vehicle for maintainingclose links between research and practice.

ŽFifteen case companies or strategic business. Ž .units have been studied Fig. 2 , covering the phar-

maceutical, snack food and engineering industries. Ineach case, starting from access to its annual reportsandror Internet Web sites to find general vision,stage of internationalisation and proper people ordepartment, several unstructured interviews are or-ganised by introducing the collaborative project,searching configuration ‘data’, understanding evolu-tion process and strategic intentions at each turningpoints, and discussing recent developing issues andcooperative requirements for research work.

3.2. Research methodology

The ‘process research approach’ was adopted inthe knowledge-based research. The process approachenables firms to capture strategic and operationaldecision-making into a relatively standardised imple-menting procedure which is easier for firms to apply

Žand control DTI, 1988; Platts and Gregory, 1990;.Platts, 1993; Maslen and Lewis, 1994 . Platts and

Ž .Gregory 1990 note:

The audit procedure has sought to ‘operationalise’the existing strategy frameworks . . . and make themaccessible to the management of operating compa-nies. This has been done by using worksheets at eachstage of the process to facilitate the recording andprogressing of the procedure. The use of worksheetsis most important as they require the users to interactformally with the process rather than simply review-ing a ‘checklist’. The worksheets also provide trace-ability; the logical and data of the analytical part ofthe process are recorded and can be revisited periodi-cally to check that the bases for the strategy are stillappreciate.

w xFor applying the approach , rather than act as exter-nal ‘auditors’, we have sought to act as ‘facilitators’.Personnel within the company have been intimately

involved in the process, our role being to provide therequired structure, and to advise and assist whenrequired. We have used multi-disciplinary workshopsinvolving, in addition to manufacturing personnel,representatives from marketing, product engineering,finance, personnel etc. Generally speaking the repre-sentatives have been at director level and the work-shops have been chaired by the Manager Director.We structure the workshops by using the worksheetsas an outline agenda and use them as the basis fordiscussion aiming to reaching a consensus view atthe completion of each stage. Between the mainworkshops there are data gathering activities andmini-workshops usually involving lower levels ofmanagement.

This approach in the knowledge-based researchspecifically integrated with the conceptual modeldevelopment by interview, understanding the histori-cal transformations of the manufacturing networksduring the pilot workbook application. One of themajor benefits of this research methodology is thatthe combination of theoretical exploration and practi-

Ž .cal process development Fig. 1 allows researchersaccess to industry to observe in detail the lateststrategic and operational practices. This providesnetwork knowledge and, at the same time, providesmanagers more practical tools.

3.3. OÕerÕiew of the conceptual models about thenetworks

As Fig. 2 details the research procedure, theconceptual models are basically developed throughpilot case studies, literature review, and wide discus-sion with academics and managers. The purpose ofconstructing the following conceptual models is tobetter steer the case studies in very complex interna-tional manufacturing systems. Another reason forusing the models, as mentioned before, is to stimu-late managers to use them and ‘test’ them.

3.3.1. Construction of international manufacturingnetworks

An international manufacturing system may beseen a factory network with matrix connections, incontrast to the linear system of a factory. Table 1lists some of the key system elements making up aninternational manufacturing factory network and

()

Y.Shi,M

.Gregoryr

JournalofO

perationsM

anagement16

1998195

–214

201

Table 1Comparison of the construction of two types of manufacturing systems

Characteristics of the system construction Difference of two types of manufacturing systemsaFactory manufacturing system International manufacturing

network system

Ž . Ž .Structural Elements: 1 Capacity: 1 Factory’s characteristics:Ž Ž .static levers controlling the architectural amount, timing, type. as whole left column .

Ž . Ž .configurations of corporate international 2 Facilities: 2 Geographic dispersion:.manufacturing system size, location, specialisation. distributed factory condition.

Ž . Ž .3 Technology: 3 Horizontal coordination:equipment, automation, linkage. coordinated mechanism.Ž . Ž .4 Vertical integration: 4 Vertical coordination:direction, extent, balance. international dispersion of the corporate

value-adding chains and their linkages.

Ž . Ž .Infrastructure elements: 5 Workforce: 5 Dynamic response mechanism:Ždynamic levers controlling the operational skill level, wage policies, employment security. opportunity identify, and manufacturing mobility.

Ž . Ž . Ž .mechanism of corporate international 6 Quality: 6 Product life cycle PLC and knowledge transfer in.manufacturing system defect prevention, monitoring, intervention. international manufacturing networks.

Ž . Ž .7 Production planningrmaterial control: 7 Operational mechanisms:sourcing policies, centralisation, decision rules. network daily co-ordination, management information system.Ž . Ž .8 Organisation structure: 8 Dynamic capability building and network evolution:structure, controlrreward system, learning by operations.role of staff groups.

a Hayes and Wheelwright, 1984.


compares them with those for a single factory. Thenetwork elements chosen in Table 1 are partly basedon existing research findings, such as the model of

Ž . Ž .Porter 1980 and the observation of Flaherty 1986 ,and partly based on our pilot case studies and indus-try workshop at Cambridge, especially for thosedynamic elements which make the network distin-guished from factories.

ŽAs for the elements of factory system Hayes and.Wheelwright, 1984 in Table 1, the levers controlling

the network are interdependent but have differentcharacteristics. The first four levers are system deter-minants settling the various patterns of structures,which have more radical and architectural influenceson the network construction. The following four aredynamic controllers from daily operations and accu-mulative improvement, to product transfer, and net-work evolution during internationalisation. The struc-tural elements determine the capabilities of the net-work and influence the patterns of infrastructure anddynamic mechanisms; but, it is also possible for thedynamic levers to act on the configuration in a longterm and accumulative way.

3.3.2. Classification of the international manufactur-ing networks

Based on the principal dimensions from the modelŽ .of Porter 1980 , Table 2 sets out two key dimen-

sions of the classification and presents seven typicalconfigurations of international manufacturing net-works and some of their characteristics.

Once a business has moved beyond a conven-tional single domestic operation it is obliged toestablish coordination mechanisms. The multidomes-tic approach, a mode of weak coordination, involvesthe development of more or less autonomous manu-facturing units geographically located close to targetmarkets. In the most extreme case such units mighthave complete design and manufacturing authoritywith only financial performance reported to the cen-tre. On the other hand, the globalised approachinvolves closely managed coordination of a dis-persed manufacturing system and integration of bothproduct design and development and production. Themanufacturing system is seen as a unified wholewith a mechanism of sharing knowledge, androrwith elements of the task being performed in themost advantageous areas.

Manufacturing dispersion covers the full range ofoptions from domestic manufacture to worldwidemanufacture. Domestic means that all production iscarried out in a single country serving both homeand export markets. Regional approaches set upfactories and networks located in a particular geo-graphical region, such as Europe, North America,Far East or South Pacific areas, sharing similarcultural value systems. Multinational approaches,with trans-regional dispersion, involve factories lo-cated in several economic sectors or countries. Thegeographic dispersion can cause large social andpsychological differences. Worldwide, an extremecase of the geographic dispersed approach, involvesthe establishment of many manufacturing operationsaround the world.

3.3.3. Strategic capabilities of the international man-ufacturing network

In the international manufacturing networks,strategic capabilities derived from the system config-uration can be categorised into resource accessibility,thriftiness ability, manufacturing mobility, and learn-

Ž . Žing ability Fig. 3 Dunning, 1994; Ferdows, 1989;.Yip, 1992 . The capability of a firm is to renew,

augment, and adapt its core competencies over time.Capabilities thus reflect the firm’s latent competen-

Ž .cies Teece et al., 1992 . As each different configura-tion has a particular structure and consistent opera-tional mechanisms, each configuration has differentcapabilities to satisfy the strategic requirements. Se-vere mismatch between configuration and require-ments is likely to make strong performance difficult.

The resource accessibility of the network meansthe ability to establish and serve a ‘presence to themarketplace’ and to reach the other corporate strate-gic resources. The thriftiness ability is derived fromcoordination andror integration of the network. Itreflects the ability to develop more efficiency throughnetworking. Combining accessibility and thriftiness,more competitiveness could be achieved from thewhole network system through the optimised geo-graphic dispersion of the corporate value-addingchain.

Unlike the above capabilities which could directlydecide the network’s effectiveness of performance,mobility and learning represent longer-term capabil-ities of network restructuring. Manufacturing mobil-

()

Y.Shi,M

.Gregoryr

JournalofO

perationsM

anagement16

1998195

–214

203

Table 2Classification of the international manufacturing networks

The degree of plants dispersion Coordination conditions in international manufacturing network

Multidomestic Orientation. Multidomestic-oriented strategies are Global Orientation. Global-oriented strategies are featuredcharacterised by manufacturing system tailoring to the local market with integration and coordination in networks. The interdepen-and having autonomy. Therefore the coordination is weak in network dence is developed from either designed system structures orand factories are independent. dynamic operational mechanisms.

Worldwide MMC3: Glocalised Manufacturing Configuration. Glocalisation means GMC4: Global-Coordinated Manufacturing Configuration. Itsglobal localisation, which is the maximum of multinationalisation. manufacturing is dispersed worldwide with homogeneousThis strategy is strong market and local management resource strategy-separated facilities and shared product, technologyoriented and autonomy focused. But, some Japanese companies, like and operation mechanism. Generally, the configuration in-

Ž . Ž .Honda Mair, 1994 and Mitsubishi company file , are adopting this cludes global product and standardised process and managerialstrategy. The competitive advantages could be generated from taking mechanisms. The network disperses its nodes globally tofull advantage of local resources, especially, national characteristics to access the markets. McDonald and KFC could be goodhave real adaptability, in terms of quick responsiveness and special examples. Many automobile component firms also try toservice, and network synergy, in terms of product R&D, capability follow this path. When company focuses more on its coreand culture fusion, from the autonomy. competence and adopts advanced manufacturing technology,

the factory could be more integrated and easier to bedistributed like KFC restaurants.

Multinational MMC2: Multidomestic Manufacturing Configuration. Its plants dis- GMC3: Global-Integrated Manufacturing Configuration. TheŽ .perse in some or a lot of countries with no or weak linkage. The corporate value-adding chains VACs or supplying chains are

network is designed for accessing corporate strategic points including dispersed in many countries to access to the most optimisedmarkets andror production factors. Basically the plants have more resources, markets and strategic capabilities according to theautonomy in product, process and management. In many cases of corporate strategic intentions. Contrast to GMC4 integratingmerge and acquisition, the companies have this type of configuration its VAC in a factory, this configuration distributes VACif they don’t adopt a coordinated strategy to transform the networks. vertically and centralises each stage of process to reduce the

duplication of manufacturing facilities. Coke Cola productionnetwork can be a good example.

Regional MMC1: Regional Uncoordinated Manufacturing Configuration. Its GMC2: Regional Exporting Manufacturing Configuration. Itsinternational manufacturing disperses centrally in only one region international manufacturing is focused in a region but itsŽ .Europe, Far East, Northern America . . . and the plants are tailored to products could reach the global market based on its regionalthe local country or regional market. There are no coordination coordinated manufacturing network in the region and globalbetween the plants. As the culture is very similar in the region, there product development.is no serious problem for manufacturing transfer.

Domestic It does not belong to the international manufacturing networks GMC1: Home Exporting Manufacturing Configuration. Itbecause it has no transnational manufacturing operations. centralises manufacturing in home country but usually has a

global logistic system. Its product lines could cover the globalmarket. Boeing jet and Rolls-Royce car are examples.


Fig. 3. Key capabilities derived from international manufacturing networks.

ity leading to the system’s swift transfer and quickresponse of strategic dispersion could be directlydetermined by the international manufacturing con-figurations.

4. Case studies

4.1. Case 1— new Õision of manufacturing systems

The company is one of the largest industrial andmunicipal boiler manufacturers in the world, whichup to the 1980s adopted an export policy worldwidebased on its Northern America manufacturing centre.Boiler production typically requires 500 000 to600 000 manufacturing hours and so very expensivein labour. Since the middle of 1980s, the companyhas changed its policy to locate its manufacturingoperations more in foreign countries, especially forbidding developing countries’ projects.

Ž .The company set up its first joint venture JVfactories in Jakarta, Indonesia in 1984 and Beijing,China in 1986. After assessing the performances ofthe JVs, the company found the benefits from mar-keting, in terms of penetrating the national or re-gional markets and better service to local customers,to be much larger than the benefits from cost reduc-tion. Especially, in China, the company’s JV soontook about 30% of Chinese market. The success inmarket area by manufacturing dispersion stimulatedthe company to establish more manufacturing basesin strategic regions in following years, includingIndia in 1988, Turkey in 1989, and Egypt in 1994.

Such widely dispersed manufacturing systems andstrong autonomy JVs transformed the company into

a real multinational corporation with unique competi-tive position, especially in developing countries, andlead to very fast growth in the second half of 1980s.As a business manager mentioned, the company withsuch wider international operations has the followingadvantages.

Ø Competitively priced products, mainly by re-ducing transportation cost and time, and taking ad-vantage of local labour and materials, but with thesame quality assurance standards regardless of man-ufacturing location.

Ø Better local service in terms of quicker re-sponse time, knowledge of customers and their facto-ries, and local supply of parts.

Ø Strong worldwide export product coverage in-cluding Canada, China, Cyprus, Egypt, Indonesia,Inner Mongolia, Philippines, Saudi Arabia, Syria,Thailand, United States.

Ø Enhanced eligibility for government fundingwith local manufacturing activities and localisationlinkages.

Ø Flexibility to apply expertise and resources ofvarious joint ventures to commit to customer’s dead-line.

With the successful presence in the strategic re-gions and capacity growing in each joint venturefactory, however, the company also faced seriousduplication issues. Two years ago, therefore, thecompany started adapting its global manufacturingstrategy by networking the JVs in different nationsbased on specialising its product lines and centralis-ing key manufacturing processes. For example, itsChinese JV is playing more and more important rolein the field of main components manufacturing, final


assembly, and engineering design service of indus-trial boilers, especially in high pressure sector, withinthe networks. Since 1986, the Beijing JV has devel-oped its manufacturing capacity more than four timesand upgraded its manufacturing capability to the 600MW boiler from only 50 MW boiler.

JVs in the network now have more specialisedproduct focus according to national strategic re-sources and existing or developing competencies, aswell as undertaking routine manufacturing operationsto optimise the manufacturing and transportationcosts. Of course, operating such a network is muchcomplex to design and coordinate, but the companybelieves this is the key success factor for its futuresurvival.

In summary, the company has experienced threemain transformation, from regional manufacturingŽ .MMC1 in Table 2 to wider distributed and au-

Ž .tonomised system MMC2 and then towards globalŽ .coordinated network GMC3 , during recent twelve

years. The network structural elements in terms ofgeographic dispersion, horizontal and vertical coordi-nation and factory characteristics determine the ar-chitectural pattern of the networks in the transforma-tion. As the configurations were changed, the strate-gic capabilities were transformed from more accessi-bility domination towards more thriftiness, efficiencyand resource integration focus.

4.2. Case 2— new motiÕation of deÕelopment

The company has three business groups, house-hold and toiletry, pharmaceutical, and food. Its phar-maceutical business is focused on ‘over the counter’Ž .OTC sales. As the businesses had very close rela-tion with local customers and national taste, tradi-tionally the company adopted very sharp multido-mestic strategy to pursue market growth through itsmanufacturing presence. Based on its aggressive for-eign direct investment, mergers and acquisitions, thecompany positioned itself to be a truly worldwideplayer ‘at home in over 120 countries’ in the early1990s.

‘‘The national market driven strategy made usvery successful in foreign markets and a leadingposition in the industry,’’ an international manufac-turing manager reflected. ‘‘At the same time, how-ever, the duplication of our manufacturing operation

cost us a lot. In some extreme cases, to penetrate aprotected national market, we had to put some facto-ries in the country first, even though we were notquite sure whether sufficient market existed. Forexample, several factories in South America nevergot chance to operate, simply because there were noactual customers.’’

The company decided to rationalise its worldwidemanufacturing system from national based manufac-turing towards global coordinated networks in 1995.What the company did in its most important businesssector—household and toiletry—was the following.

Ø To restructure the world manufacturing opera-tions into seven regions based on geographical andcultural characteristics.

Ø To centralise manufacturing sites based on therequirement of economies of scale to support wholeregional market rather than traditionally nationalmarkets.

Ø At global manufacturing level, to standardisethe manufacturing and managerial processes andshare them within seven regions.

Ø At product development level, to promote thegeneralisation of regional products from national onesand further global products from regional ones; butall of them share the standardised manufacturingsystem.

Ø At group coordination level, to develop, intro-duce and implement sharable managerial processeslike new product development and manufacturingoperation benchmarking globally.

Such strategy is clearly different from themarket-drive approach. And it is also distinct fromtraditional cost-cutting policy. As its group manufac-turing manager emphasised, ‘‘Efficiency or cost re-duction, of course, is a target. But, to foster newstrategic capabilities perhaps is more important issuefor the new manufacturing network design. Thiscould include more responsiveness through our re-gionalised dispersion and intensive coordination be-tween regions, learning mechanisms between re-gional factories through expertise sharing and inter-nal competition. This might be basic difference be-tween factory and factory network design.’’

Since early 1996, the regionalised global strategyhas been transferred into its pharmaceutical sector.The company believes, if the global strategy can besuccessful in the household market, it should be


appropriate in pharmaceuticals—‘a truly global in-Ž .dustry’ Ballance et al., 1992 . In the pharmaceutical

sector, the company at first rationalised its Europeanmanufacturing sites from 16 into only three—oneintegrated and two specialised—bases. A very simi-lar strategy was adopted to focus on core technologyand standardise manufacturing and managerial pro-cesses. In the near future, there will be only 10manufacturing sites serving seven regional markets.

As each manufacturing site has been standardisedand focused on core technology and operations, theinternal operations are much easier to manage andsome learning mechanisms, such as knowledge shar-ing, problem solution, and various kinds of bench-marking, are adopted within the network. The emerg-ing problem, however, is outsourcing to support thiscoordinated regional network. The company is facingnew challenges dealing with interfirm networks. An-other issue is to keep the original strength in thedevelopment of customer-tailored products and at thesame time transfer them into regional or even globalproducts effectively. It is not easy at factory level toproduce different levels of products while simultane-ously introducing new products.

The transformation of this company highlightseffective response to current major changes in world-wide trade environment. Configuring its internationalmanufacturing network into GMC4, the companyintegrated its regional products to global products,standardised the manufacturing processes and cen-tralised the manufacturing sites based on its strategicregional markets. The company hopes that the globalcoordinated configuration can accelerate sharing ofresources and knowledge and learning in the groupto strengthen the economies of scope and scale andother capabilities for the future.

4.3. Case 3— new architecture of the network

A top pharmaceutical company in the world hadmanufacturing operations at 47 sites in 32 countries,utilising assets with a net value of £1.2 billion andemploying 17 000 staff. Although its internationalmanufacturing operations were more widely spreadthan many of its competitors, the company’s manu-facturing systems were highly globally integratedalong the value-adding chain. Its international manu-

facturing strategy embodied the following principles:Ž .1 highly centralised strategic sites for global supplyof bulk drug substances and new product develop-

Ž .ment; 2 specialised and relatively widely dispersedŽ .regional sites for more complex formulations; 3

more widely dispersed and market driven local for-mulationrpackaging sites for close relation with na-tional or local markets.

In contrast with its other competitors’ strategies,such as the efficiency driven strategy of Merck andEli Lilley, the dispersed packaging strategy of Astra,and the export base dependent strategy of mostJapanese pharmaceutical companies, Case 3 com-pany had a more optimised or balanced network tomanufacture its products efficiently and cover itslocal market effectively.

ZTK is the company’s most important productaccounting for more than 40% of its total sales in1993r1994, and is also the most successful medicinein the world. ZTK’s international manufacturing net-work is a very good example of the company’sinternational manufacturing strategy and the networkdesign principles.

ZTK’s bulk drug substances were manufacturedmainly in the UK and Singapore. In fact, the Singa-pore factory supplied most of the regional formula-tion sites apart from the UK. The formulation pro-cesses for ZTK were dispersed to some of its 24strategic and regional secondary sites, which coveredevery continent. As the manufacturing process ofZTK tablet is robust and relatively simple, the sec-ondary sites generally integrated the final packagingprocess to supply local markets directly. Besides thestrategic and regional secondary sites, more than 20local formulationrpackaging sites were used to ac-cess other local markets.

Before ZTK’s success, the company was a UK-based pharmaceutical manufacturer ranked number25 in the world. Great global market opportunitiesfrom ZTK pushed it into being a global reach com-pany. But it is worth noting that, unlike other marketdriven companies, ZTK’s manufacturing network,based on wider geographic dispersion, was alsohighly integrated rather than purely multidomestic.Experiencing fast geographic expansion in the 1980s,the company rationalised its world wide manufactur-ing operations since the early 1990s. It changed itsloose linked sites into more integrated and better


coordinated networks especially along the valueadding chains. For ZTK’s networks, more complexformulation processes were more centralised.

There were a few differences between the mainproduct lines. For many newer products, they appearto have more centralised and closely coordinatednetworks, compared with ZTK’s networks. This wasmainly because of the better international trade envi-ronment, especially in some free trade regions, butalso partly because of the earlier product life cycle,and partly because they were more potent com-pounds requiring lower manufacturing volume. Themain trends in the company, therefore, were to de-velop more integrated networks along the valueadding chain based on its existing dispersed sites.

In summary, such balanced or optimised globalintegrated manufacturing networks, more GMC3Ž .Table 2 oriented, have clear advantages in higherefficiency, a high level of quality and supply controlsecurity, exploitation of both manufacturing sites andlocal market opportunities, and higher responsive-ness to various kinds of changes. But on the otherhand, some weaknesses include difficulty of min-imising cost of production and supply, some under-utilised bases and fragmented supply chains, diffusedcontrol of global manufacturing, and little immediatecontribution to the competitive advantage from highlycoordinated and complex network operations.

4.4. Case 4— new mechanisms of manufacturing sys-tem

The company is a leading manufacture of bakeryfood processing equipment. During the 1980s,through a merger with another liquid food processingmachinery company and wide range of acquisition,the company grew very fast and developed a strongpresence in different geographic markets. Two facto-ries in the States serve the whole American conti-nent, an Australian factory supplies the south Pacificregion, and two factories in UK cover the markets ofthe rest of the world.

One of the main characteristics of the company’sproducts is its made to order, one off or very smallbatch, and ‘highly customer engineered content’,although there is a product catalogue guiding theorder. The company has a large range of products,

approaching more than 500, and has been trying tomodularise its designs for a long time to balancecustomer’s featured requirements and internal manu-facturing efficiency. Up till 1993 when the companyidentified new directions of market and product, littleprogress had been made.

The company has similar manufacturing facilitiesat its different locations around the world. This islargely because its products are tailored to the cus-tomers, its acquisition of similar companies, heavyweight and expensive transportation of the products,and different standards of product and engineering inthe areas in which it operates. This multidomesticproduct strategy and manufacturing duplication his-torically helped it to access the different countriesand compete effectively.

But many markets in traditional regions have seensustained period of growth, with rising demand forhigher quality products. On the other hand, for serv-ing new growing markets, the company faced theproblem of penetrating the developing countries’markets where its products are quite price sensitive.In order to meet the new challenges the company

Ž .was faced with a number of options: 1 to reinforcethe current approach with more highly focused man-ufacturing and duplicated operations to satisfy the

Ž .local market; or 2 to develop its competence tochallenge the global market with dynamic capabilitywithout trade-offs between higher quality and lower

Ž .cost; or 3 to diversify the business by merging oracquiring new business to develop a totally differentnew competence.

The company decided it would develop new capa-bilities to face the global competition. It adopted aset of new business global policies including devel-oping global products, restructuring the businessvalue-adding chain, and implementing the strategythrough pilot testing.

The company has developed its first generation ofworld product through the coordination of the engi-neers in three continents, thanks to the modern com-munication technology. The development was nevereasy, the R&D cost of the new world product was60% higher than the normal one. ‘‘But it is neces-sary learning fee.’’ Besides the world product, thecompany also set up a product module system tocope with global homogeneous demand and nationalspecific demands. It developed the engineering plat-


form for new product development in the future bystandardising the international and imperial technicalstandards.

Paralleling with the development of the worldproduct, the company restructured its UK manufac-turing factories as a pilot testing for the manufactur-ing globalisation. Before the change, the duplicationof manufacturing not only happened in differentcountries but in the UK and USA. Based on theirdifferent manufacturing competence, the companycentralises the manufacturing operations in differentfactories, one focuses on the two-dimensional com-ponents fabrication for tins and cans; another focuseson the metal process, control system and final equip-ment assembly. Such change, as a pilot test, winsmore economies of scale without compromisingquality and delivery. The company is currently trans-ferring this mechanism to its factories in the States.

In the future, the company plans to set up a newfactory in China to cover the Far East market, whichmeans the company has a global product and rela-tively standardised manufacturing process shared bysix factories in four continents, which configures itsnetwork more like GMC4 in Table 2. This willpromote efficiency by coordination and learning. Atthe same time, it also increases the accessibility tothe markets and different kinds of resources. Theduplication of manufacturing operations based onsimilar systems in different continents, as mentionedby the managers, is largely because of the hugetransportation costs, but it could increase the dy-namic responsibility for the global change of thedemands. Some more common parts, such as thecontrol box, may be centralised further in the furtherfuture.

5. Findings and revised models

5.1. DriÕing forces towards globalisation

Globalisation of manufacturing has two clearcharacteristics—geographic dispersion and interde-

Žpendent coordination between factories Porter, 1980;.Ferdows, 1989; Bartmess and Cerny, 1993 . As the

cases in Section 4 show, however, the driving forcesbehind the dispersion and coordination are complexand different. Understand of these forces is essential

to the understanding the missions, capabilities andbehaviours of manufacturing globalisation.

Geographic dispersion of the manufacturing net-works is drawn by external forces, especially newmarket opportunities. Emerging global marketsŽ .Levitt, 1983; Yip, 1992 have fuelled developmentof global product and related strategies in classicalglobal industries such as pharmaceuticals and elec-tronics and also customer dominated industries suchas household products and engineering, as demon-strated in the cases. Fast growth of some developingcountries is another key engine of dispersion. Recent

Ž .annual reports of UNCTAD 1992, 1993 show thatthe dominant countries getting foreign direct invest-

Ž .ment FDI are developing countries. Many manufac-turing managers share the same opinion as Gomez ofThomson—‘‘you do not choose to become global.The market chooses for you; it forces your hand’’Ž .McCormick and Stone, 1990 . Besides market at-tractiveness and local presence strategy, global com-petition is another driving force dispersing manufac-turing to access strategic resources along the value-

Ž .adding chain Porter, 1980 . In our case studies,however, the global competition pressure appears tocause more concentration or integration of manufac-turing rather than the dispersion to access theeconomies of scale. In general, however, dispersionof manufacturing networks is a very powerful, com-petitive weapon to access market and fuel companygrowth. When we study the restructuring of networkstowards more integrated and specialised patterns, forexample case 1 company was losing its EasternEuropean market to ABB, its main competitor inworldwide, simply because there was no manufactur-ing presence in the area.

Coordination between dispersed factories is an-other key characteristic of manufacturing networkingin the case studies. Theoretically, the coordination isa more critical feature of globalisation than geo-

Ž .graphic dispersion Fleenor, 1993 . From the cases,regional free-trade and economic deregulation in FDIare clear drivers for Case 2 company to adopt itsregionally based global strategy. For the other com-panies, global competition, especially the require-ment for higher efficiency and lower cost withoutquality and service compromises, demands manufac-turing coordination. Many studies have emphasiseddifferent coordination mechanisms to balance or op-


timise efficiency and tried to explore strategic moti-Ž .vations Porter, 1980; Flaherty, 1986 . The cases

indicate that for many companies, strategic capabilityis much more important than the traditional driversof responsiveness to local market requirements.Globalisation of manufacturing networks is typicallythe result of an internal push for extended capabili-ties rather than externally driver.

This kind of capability-building strategy such asthe case 1 and 2 companies challenges the traditionalmanufacturing ‘trade-offs’ mindset. Some companieslike case 3 and 2 did depend heavily on their opti-mised network architecture to provide competitiveedge; but increasingly companies found that it isimportant to have a capability development processand learning mechanisms within the networks.

5.2. New strategic capabilities of manufacturing

What are the strategic capabilities that many globalcompanies are pursuing and the networks are tryingto generate? Fig. 3 suggested a framework fromsystem structure and function aspects. Using this

framework, managers in the case companies, fromgroup manufacturing coordinators and business CEOsto project and factory managers, were interview.They identified and detailed the required strategiccapabilities of international manufacturing networksfrom framework to a operational system design.

The new strategic capabilities of the networks canbe categorised in Table 3. Like computer networksfocused on communication instead of processing, themissions of the network are significantly differentfrom those of a single factory. As a whole system,the nodes still play a transformation role in thenetwork, but the total configuration can generatemore additional functionality for both corporationand its factories.

In Table 3, accessibility and thriftiness are moredirectly determined by or derived from the architec-ture of the networks. Mobility and learning abilityare dynamic, process oriented and more determinedby operational stages rather than mainly designstages. While the former two have some trade-offrelationship separated by the networks’ dispersion

Ž .and coordination Prahalad and Doz, 1987 , the latter

Table 3Strategic capabilities of the international manufacturing networks

Ž .1 Strategic targets accessibility. Mainly derived from dispersion of the networkØ Strategic markets: beating trade barriers; closing to the customers; quick response.Ø Production factors: labour; materials; energy; product and process technology etc. to tap national resources and advantages.Ø Managerial skills: managerial knowledge, organisational skill, administration heritage, and corporate value and culture.Ø More sensitive to global changes: understand wide range and different customer requirements; more sensitive to future trends,information, technology, competition.

Ø Other special benefits: nontax; policy benefits; partner’s business and social connections.

Ž .2 Thriftiness ability. Mainly derived from coordination of the networkØ Economy of scale: special for the dispersed value-added chain configuration;Ø Economy of scope: wide product lines with shared R&D, engineering, manufacturing, marketing and distribution facilities . . .Ø Reducing duplication of activities: for all business activities

Ž .3 Manufacturing mobility. Mainly required and derived from dispersion of the networkØ Productrprocess mobility: transferred technology and system for donors and receivers and robust process for such transfer.Ø Managerial skill mobility: learning process for the skill, knowledge, culture, value . . .Ø Factory manufacturing flexibility: wider product lines and economy of scope for global changes and competition; more flexiblesystem for the product life cycle.

Ø Network manufacturing flexibility: network is more flexible for change its location, node linkages, value-chain relationship.

Ž .4 Learning ability. Mainly required and derived from coordination of the networkØ Special learning opportunity: wider internal and external comparison, exchange, and benchmarking.Ø National capability integration: culture fusion, learning and tapping the special national strengths.Ø Global product integration: learn from the worldwide market demands and abstract core requirements fordevelopment of world product.


Fig. 4. Manufacturing system’s new missions and capabilities.

two capabilities have strong synergistic power andsuggest integration or optimisation.

From the cases, the indications are that the fourstrategic capabilities can be built up by developingthe global products, dispersing and coordinating be-tween factories. Fig. 4 tries to generalise the newmissions of manufacturing systems from the casesand link them to required strategic capabilities ofinternational manufacturing networks. It illustratesthat manufacturing network systems need new struc-tures and mechanisms to deliver the capabilities andsatisfy the missions.

5.3. International manufacturing network configura-tions

Lack of systematic support to strategy formulationseverely limits the effectiveness and efficiency of thesystem design. Configuration provides a better wayof concisely representing an complex organisationand an approach forming an integrated view. Theissues of the network design, such as geographicdispersion, horizontal and vertical coordination andthe various dynamic operational mechanisms, cannot be picked and chosen independently in the way‘a shopper picks vegetables at the market’. Rather,these and other elements of the network designshould logically configure into internally consistent

Ž .groupings Mintzberg, 1979; Miller, 1993 .The configurations of international manufacturing

networks not only represent the structural groupingsin terms of dispersion and coordination but also,more importantly, demonstrate capability patternsderived from the structures. The configuration there-fore can bridge very effectively and efficiently be-tween the strategic requirements from corporate or

business level and the network capabilities duringthe system design. It is essential to develop a system-atic way to present the networks, to provide a con-cise visualisation to the complex network systems ofinternational business.

Fig. 5 presents a map for international manufac-turing network configurations which groups sevenconfigurations into four blocks, in terms of RegionalFocused Networks, Global Exporting Networks,Multidomestic Autonomy Networks, and Global Co-ordination Networks, along verticalrdispersion andhorizontalrcoordination dimensions. It is populatedwith examples from other literature referencesŽFlaherty, 1986; Mair, 1994; McCormick and Stone,

.1990; Taylor, 1991; Maruca, 1994 and our own casework. This map shows the relationship between theconfigurations and characteristics of the networksand their transformation. There is quite clear trend ina number of companies towards more global config-urations generating global competitive capabilitiesfrom Fig. 5.

ŽIn the global coordinated networks top-right block.in the map , there are two typical global network

configurations.Ø The geographically dispersed and vertical inte-

Ž .grated value-adding chain GMC3 reflects a drive toaccess the most favourable resources through appro-

Ž .priate location of nodes Porter, 1980 . because ofeach node of its VAC can access most favourableresources. Case 1 Company provides an good exam-ple of this configuration trying. Practically, pureGMC3s are quite rare. Very like the global exporting

Ž .networks bottom-right block losing their global-covering power, GMC3 is also challenged by itslimited presence and sensors in local markets.


Fig. 5. Map of international manufacturing network configurations.

Ø Geographically dispersed and horizontal coor-Ž .dinated factories GMC4 provides the ability to

replicate and deliver a synergy advantage from ashared common infrastructurermechanism in the

Ž .network Flaherty, 1986 . Currently favoured by ser-vice industries such as McDonald’s and KFC, Mac-

Ž .Cormack et al. 1994 argue that this model willproliferate and help companies focus more on com-petence. The Case 2 shows this trend but the prob-lem of increased duplication of manufacturing facili-ties and potential to damage efficiency remains.

Case 3 Company combines advantages fromGMC3 and GMC4 by centralising andror specialis-

Žing upstream manufacturing operations bulk of.medicine and special formulation processes and lo-Žcalising or tailoring down-stream operation packag-

.ing based on local market demands. It is a balanceor optimisation between higher efficiency, control

security, market accessibility, and higher responsive-ness to various kinds of changes.

Besides optimising the structure of the networks,most companies in the case study are more keen onfostering a dynamic mechanism to promote learningand sharing between networking factories. This canprovide unique and protectable future competitive-ness. But such system may be much more difficult tomanage.

In summary, the configuration map provides apractical tool underpinning the workbook for interna-

Žtional manufacturing strategy formulation Shi and.Gregory, 1996 , because it provides a better way of

concisely representing an complex organisation andforming an integrated view. The issues of the net-work design, such as geographic dispersion, horizon-tal and vertical coordination and the various dynamicoperational mechanisms, can be logically configured


into internally consistent groupings. More impor-tantly, the configuration map can bridge effectivelyand efficiently between the typical configurationsand the network capabilities. It can be used to bestadvantage in the early stage of network design suchas developing global vision for new global capabili-ties, self-assessing the network performance, config-uration selecting and outside requirement matching.

6. Trends and methods of international manufac-turing research

6.1. Globalisation: different perspectiÕes and pre-scriptions

The term globalisation is used rather freely andŽthere are no widely accepted definitions Fleenor,

.1993 . Marketing people argue about reality of theŽmarket globalisation Levitt, 1983; Yip et al., 1988;

.Baden-Fuller and Stopford, 1991 . Strategists iden-tify emerging trends in global competition and

Žchanging rules of business Hout et al., 1982; Hamel.and Prahalad, 1985; Daniels and Daniels, 1993 .

Manufacturing staffs tend to pursue global sourcing,seeding factories close to markets and then network-

Žing Ferdows, 1989; DuBois et al., 1993; MacCor-.mack et al., 1994 . As there is a lack of common

language and shared vision about globalisation amongthese people, ‘‘their prescription on how to manageglobally have also been very different, and often

Ž .contradictory’’ Ghoshal, 1987 .Based on the case studies presented here, manu-

facturing globalisation might be defined as the pro-cess of moving from an independently managedbusinesses serving local markets to networks of busi-nesses serving the businesses’ chosen markets in acoordinated and optimised way. Manufacturing net-works are the main outcome of globalisation andcapability development is one of the key motivationsof the process. Just as global strategy cannot bereally operated without manufacturing networking,the manufacturing system and whole business re-quire globalised coordination and cooperation.

In the future, the term globalisation might be lost,but manufacturing networking with vertical integra-tion or horizontal coordination or other synergymechanisms between factories will last in long term.

From this perspective, the manufacturing competi-tiveness will heavily depend upon network construc-tion and mechanisms rather than separated factorynodes. The definition of a manufacturing systemtherefore needs to be extended beyond the scope offactory towards the network, just as in the earlyyears of this century manufacturing people extendedtheir scope from separate machine towards a productmanufacturing line.

6.2. Research focus: international manufacturingnetworks

Manufacturing globalisation, in our definition, isone kind of internationalisation involving internalcoordination and external cooperation in manufactur-ing networks. This emphasis on globalisation seeksto overcome some of the limitations of internationalmanufacturing strategy research.

At the highest level it is clearly important for astrategist to select the appropriate configuration tomatch its corporate strategic goals. However, thesuccessful implementation of the desired interna-tional manufacturing configuration requires a greatdeal of detailed input, selection between options, andthe process of matching and transforming. Configu-ration models can give some support to the decision.They can never be definitive. Managers still needsome craft process and tools to make effective deci-sions for the network design even though the config-uration could help the efficient decision-making. Be-sides intra-network relations emphasised in this pa-per, the inter-network relationship, especially withstrategic alliances and other functional departments,is becoming more and more important in internation-alisation. Therefore, surrounding the networks, the

Ž .following issues need to be tackled in the future: 1To develop transform strategy processes betweenconfigurations to facilitate adopting an international

Ž .strategy. 2 To understand more about internationalstrategic alliances based on manufacturing networks.Ž .3 To explore wider network relations with otherfunction departments, especially for global product

Ž .development and capability building. 4 To explorethe detailed relationship between network capabili-ties and factory capabilities to understand the contri-

Ž .bution of a network to each of its nodes. 5 Toidentify the detailed distinctive capabilities of each


network configuration to make international strategicdecision more efficiently.

6.3. Research methodology tackling the network is-sues

The research methodology adopted in this re-search project is quite unique, rather engineeringoriented than social science oriented. The majorreason for using the process approach is to balanceboth equally important issues—develop new net-work knowledge and design the network system.Traditional academic observation methodology canbe very efficient to explore the network knowledgebut not sufficient for practical design. On the otherhand, the classical system engineering approach cansatisfy customer requirements more efficiently butinhibits new exploration. From the cases presented,the process approach can combine both facets andintegrate them in the research process. The knowl-edge based research part in the project can heavilyinvolve various kinds of practitioners and gener-alised knowledge can be tested and modified in arelatively short cycle. This is difficult for any otherresearch methodology.

It is clear that this methodology is not all-purpose,even though it can combine several approaches to-gether in a process. Alternative approaches includelarge scale questionnaire surveys which are highefficient in developing new knowledge and illustrat-ing mega trends. But there is a concern that thereliability of knowledge generalisation from suchmethodology may be inhibited by low feed-backrate. Pure observation has the fundamental weaknessof developing no constructive relation with businessworld. From this perspective, the process methodol-ogy pursuing tangible benefits to companies makes iteasier to engage companies in a productive way.

A mutual trust and benefit relationship is ex-tremely critical in the process methodology. Therelationship is of course not easy to set up based onsingle research project. It depends on long-term co-operation and network building.

7. Conclusions

In summary, manufacturing businesses are in-creasingly faced with decisions involving global con-

siderations and capability development. The litera-ture of manufacturing strategy and of internationalbusiness are reasonably well developed but the linksbetween them are weak. The research set out in thispaper suggests the following.

Ø Manufacturing globalisation, based on internalcoordination or rationalisation rather than pure geo-graphic expansion, clearly dominates current manu-

Ž .facturing business internationalisation Fig. 5 .Ø International manufacturing networks are new

manufacturing systems in terms of mission, struc-ture, infrastructure, capability, and design process,which needs more detailed observation and theorybuilding.

Ø A classification of manufacturing configura-tions based on geographic reach and differentiatingbetween multidomestic and global orientations canprovide a useful characterisation of internationalmanufacturing systems.

Ø The particular capabilities which internationalmanufacturing systems can provide to a networkinclude accessibility and thriftiness ability, but thesuccess of the network depends upon the mobility

Ž .and learning ability in the networks Table 3 .Ø Businesses need tools and techniques both to

identify appropriate manufacturing configurationsand to navigate between them; configuration ‘map’Ž .Fig. 5 can provide complementary support to inter-national decision-making.

Ø Further study should focus on both internal andexternal network issues, and the process of globalmanufacturing strategy formulation.

References

Anderson, J.C., Cleveland, C., Schroeder, R., 1989. OperationsŽ .strategy: a literature review. J. Operat. Manage. 8 2 .

Baden-Fuller, C., Stopford, J., 1991. Globalisation frustrated: thecase of white goods. Strategic Manage. J. 12.

Ballance, R., Pogany, J., Forstner, H., 1992. The World’s Pharma-ceutical Industries: An Introduction Perspective on Innovation,Competition and Policy. Edward Elgar, Hants, England.

Bartmess, A., Cerny, K., 1993. Seeding plants for a globalŽ .harvest. McKinsey Q. 2 107–132.

Cohen, M.A., Fisher, M., Jaikumar, R., 1989. International manu-facturing and distribution network: a normative model frame-

Ž .work. In: Ferdows, K. Ed. , Managing International Manufac-turing. Elsevier, Amsterdam.


Daniels, Daniels, 1993. Global Vision: Building New Model forthe Corporation of the Future. McGraw-Hill.

De Meyer, A., Vereecke, A., 1994. Strategies for internationalmanufacturing. Working paper of INSEAD: 94r25rSMrTM.

Dicken, P., 1986. Global Shift: The Internationalisation of Eco-nomics Activities, 1st edn. Paul Chapman.

Dicken, P., 1992. Global Shift: The Internationalisation of Eco-nomics Activities, 2nd edn. Paul Chapman.

DTI, 1988. Competitive Manufacturing: A Practical Approach tothe Development of a Manufacturing Strategy. IFS Publica-tions.

DuBois, F.L., Toyne, B., Oliff, M.D., 1993. International manu-facturing strategy of U.S. multinationals: a concept frameworkbased on a four-industry study. J. Int. Business Stud., 2ndQuarter.

Dunning, 1994.Ferdows, K., 1989. Mapping international factory networks. In:

Ž .Ferdows, K. Ed. , Managing International Manufacturing.Elsevier, Amsterdam.

Flaherty, M.T., 1989. Coordinating international manufacturingŽ .and technology. In: Porter, M.E. Ed. , Competition in Global

Industries. Harvard Business School Press.Fleenor, D., 1993. The coming and going of the global corpora-

Ž .tion. Columbia J. World Business Winter .Ghoshal, S., 1987. Global strategy: an organising framework.

Strategic Manage. J. 8, 425–440.Hamel, G., Prahalad, C.K., 1985. Do you really have a global

Ž .strategy? Harvard Business Rev. Jul–Aug 139–148.Hayes, R.H., Pisano, J.P., 1994. Beyond world-class: the new

Ž .manufacturing strategy. Harvard Business Rev. Jan–Feb .Hayes, R.H., Wheelwright, S.C., 1984. Restoring Our Competitive

Edge: Competing Through Manufacturing. Wiley.Hayes, R.H., Wheelwright, S.C., Clark, K.B., 1988. Dynamic

Manufacturing: Creating the Learning Organisation. The FreePress, New York.

Hout, T., Porter, M.E., Rudden, E., 1982. How global companiesŽ .win out. Harvard Business Rev. Sep–Oct 98–108.

Kanter, R.M., 1991. Transcending business boundaries: 12,000Žworld managers view change. Harvard Business Rev. May–

.Jun .Klassen, R.D., Whybark, D.C., 1994. Barriers to the management

of international operations. J. Operat. Manage. 11.Lawrence, S.R., Rosenblatt, M.J., 1992. Introducing international

issues into operations management curricula. Prod. Operat.Ž .Manage. 1 1 .

Leonard-Barton, D., 1992. The factory as a learning laboratory.Ž .Sloan Manage. Rev. Fall .

Levitt, T., 1983. The globalisation of markets. Harvard BusinessŽ .Rev. May–Jun .

MacCormack, A.D., Newman, L.J., Rosenfield, D.B., 1994. Thenew dynamics of global manufacturing site location. Sloan

Ž .Manage. Rev. Summer .Mair, A., 1994. Honda’s Global Local Corporation. St. Martin’s

Press.Maruca, P.F., 1994. The right way to go global: an interview with

Whirlpool CEO David Whitwam. Harvard Business Rev.Ž .Mar–Apr .

Maslen, R., Lewis, M.A., 1994. Procedural action research in thestrategy process. British Academy of Management 1994 An-nual Conference. Lancaster, UK.

McCormick, Stone, 1990. From national champion to globalcompetitor: an interview with Thomson’s Alain Gomez. Har-

Ž .vard Business Rev. May–Jun .Miller, D., 1993. The architecture of simplicity. Acad. Manage.

Ž .Rev. 18 1 .Miller, J.G., Roth, A.V., 1994. A taxonomy of manufacturing

Ž .strategies. Manage. Sci. 40 3 .Mintzberg, H., 1979. The Structuring of Organisations. Prentice-

Hall, Englewood Cliffs, NJ.Platts, K.W., 1993. Characteristics of methodologies for manufac-

turing strategy formulation. Comput. Integr. Manufact. Syst. 7Ž .2 , 93–99.

Platts, K.W., Gregory, M.J., 1990. Manufacturing audit in theprocess of strategy formulation. Int. J. Operat. Prod. Manage.

Ž .10 9 .Porter, M.E., 1980. Competitive Strategy. Free Press, New York.Prahalad, C.K., Doz, Y., 1987. The Multinational Mission: Bal-

ancing Local Demands and Global Vision. Free Press.Shi, Y., Gregory, M.J., 1995. International manufacturing config-

urations: contingency maps for manufacturing internationalisa-Ž .tion. In: Draaijer, D., et al. Eds. , Management and New

Production Systems. Univ. of Twente, The Netherlands.Shi, Y., Gregory, M.J., 1996. International manufacturing config-

uration map: a practical tool for global manufacturing strategyŽ .formulation. In: Voss, C. Ed. , Manufacturing Strategy: Oper-

ations Strategy in a Global Context. EUROMA ConferenceProceedings. London Business School.

Skinner, W., 1969. Manufacturing: missing link in corporateŽ .strategy. Harvard Business Rev. May–Jun .

Sprague, L.G., 1990. Strategic analysis for global manufacturing.Ž .In: Moody, P.E. Ed. , Strategic Manufacturing: Dynamic

New Direction For the 1990s. Business One Irwin.Taylor, W., 1991. The logic of global business: an interview with

Ž .ABB’s Percy Barnevik. Harvard Business Rev. Mar–Apr .Teece, D.J., Pisano, G., Shuen, A., Feb 1992. Dynamic capabili-

ties and strategic management. Working Paper, Univ. of Cali-fornia at Berkeley.

Ž .UNCTAD United Nations , 1992. World investment report 1993:transnational corporations as engines for growth. TransnationalCorporations and Management Division, Sales No.E.92.II.A.19.

Ž .UNCTAD United Nations , 1993. World investment report 1993:transnational corporations and integrated international produc-tion. Transnational Corporations and Management Division,Sales No. E.93.II.A.14.

Yip, G., 1992. Total Global Strategy: Managing for WorldwideCompetitive Advantage. Prentice-Hall.

Yip, G.S., Loewe, P.M., Yoshino, M.Y., 1988. How to take yourcompany to the global market. Columbia J. World BusinessŽ .Winter 37–48.

international manufacturing networks—to develop global ... · ... global manufacturing strategy;...

Documents