Software component architecture in supply chain management

Computers in Industry 53 (2004) 165–178Software component architecture in supply chain managementMartin Verwijmeren *MP Objects, P.O. Box 29126, 3001 GC Rotterdam, The NetherlandsReceived 17 January 2003; accepted 23 July 2003AbstractWe present a software component architecture for supply chain management across dynamic organisational networks. Thelocal management in the architecture is done by existing enterprise resource planning (ERP) systems, warehouse managementsystems (WMS) and transportation management systems (TMS). The integral management in the architecture is executed bysupply chain engines (SCEs). These software components make up a layer of systems for supply chain management running ontop of ERP, WMS and TMS. The SCEs provide both the intelligence and the flexibility as required for the integral managementof dynamic supply chains. The software component architecture can be implemented with technology for components,interfaces and services as available in Java Enterprise, CORBA and Web Services.# 2003 Elsevier B.V. All rights reserved.Keywords: Supply chain management; Networked organisations; Inventory management; Distribution management; System architecture;Software components; Distributed systems; Object-oriented systems; Supply chain engines; ERP; WMS; TMS; Java; CORBA; Web Services1. IntroductionIncreased global competition and demand forgreater product variety, result in the emergence ofdynamic supply chains, which need to be co-ordinatedwithout sacrificing their flexibility. Co-ordinationacross dynamic supply chains requires inter-organisationalinformation systems. However, rigid automationof supply chain co-ordination takes away theflexibility of the networked organisations.Traditional ERP systems and EDI interfaces are notgeared to supply chain management across dynamicorganisation networks [1]. Enterprise resource planning(ERP) systems provide integration for supplychain management. However, due to their central* Tel.: þ31-10-2900304; fax: þ31-10-2900305.E-mail address: martin.verwijmeren@mp-objects.com(M. Verwijmeren).server and procedural software, the ERP systems lackthe autonomy and flexibility required by networkedorganisations. Local systems linked by electronic datainterchange (EDI) support the flexibility for networkedorganisations. However, EDI interfaces justfocus on data exchange, and therefore miss the decisionrules required for supply chain management.Given the drawbacks of the traditional ERP systemsand EDI interfaces, we are in search for a new softwarearchitecture. The main question of our research is:what is a suitable software architecture for supplychain management across dynamic organisational networks?Our study has resulted in a software architecturebased on components, meant to support therequired co-ordination and flexibility simultaneously.First, we will outline dynamic supply chains, includingsupply chain management and networked organisations.We then position existing systems (ERP, WMSand TMS) and reveal complementary components for0166-3615/$ – see front matter # 2003 Elsevier B.V. All rights reserved.doi:10.1016/j.compind.2003.07.004

166 M. Verwijmeren / Computers in Industry 53 (2004) 165–178dynamic supply chain management. The softwarecomponent architecture for supply chain managementwill be illustrated for one practical area. The inventorymanagement architecture explains the design of inventorymanagement engines and their use in supplychains for consumer electronics.Furthermore, we have a look at available technology(Java Enterprise, CORBA and Web Services) forthe implementation of our software component architecture.We will conclude with the major findings ofthe study.2. Dynamic supply chains2.1. Supply chain managementIn business we face the trend of increased globalcompetition, which forces companies to improvetheir efficiency. One of the measures for efficiencyimprovement is supply chain management. Supplychain management focuses on the inter-organisationalmanagement of goods flows between independentorganisations in supply chains, such as raw materialwinners, component manufacturers, finished productmanufacturers, wholesalers and retailers. This integrativeapproach to planning, control and monitoringof product flows, from suppliers to end users, aims atimproved customer service at reduced overall costs[2,3].Supply chain management starts when the scopeof integration is extended from internal to externalco-ordination, a stage that takes into account customersand suppliers in the supply chains [4,5]. The coreprinciple behind supply chain management is thereduction of uncertainty in decision making processesof organisations in supply chains. The co-ordination ofthe management processes in a supply chain requiresinformation exchange between the organisations in thesupply chain. The extra availability of information indecision making units reduces uncertainty, resulting inbetter control and finally in improved performance [6].For the exchange of proprietary information, suchas sales and forecasts, a collaborative attitude of thesupply chain partners is needed [7].In most cases, supply chain management has beenrealised by one organisation dominating the otherorganisations in the supply chain. As presented inFig. 1. Supply chain management by domination of one organisation over others.

M. Verwijmeren / Computers in Industry 53 (2004) 165–178 167Fig. 1, a dominant organisation can make use ofhierarchical control to achieve integral logistics managementacross organisational boundaries. The subordinateorganisations in the supply chain are forced toobey to the instructions of the dominant organisation.The integral management enforced by one dominantorganisation is usually accomplished with one centralinformation system with fixed procedures, ruling theunderlying processes of the subordinate organisationsin the supply chain.This type of supply chain management, where oneorganisation dominates the others, could improveproductivity of the supply chain, including lower costsand better quality (customer service). This results in abetter price-quality ratio of products, as required bydemanding customers. However, supply chain managementimposed by one dominant organisation, doesnot contribute to the flexibility of the supply chain,which is needed to offer customers a more dynamicproduct assortment. Because subordinate organisationsare bound to one dominant organisation, theycan not easily change their portfolio in the supplychain or switch to other supply chains.2.2. Networked organisationsAnother business trend is the demand for greaterproduct variety, requiring companies to become moreflexible. A means to increase flexibility are networkedorganisations, which can continuously reform toaccommodate changing market demand. Typical characteristicsof networked organisations are autonomouscontrol, common goals, mutual trust, informationexchange, close co-operation and variable coupling[8–13]. Organisation types similar to networked organisationsare the value-adding partnership [14], thevirtual corporation [15], the lean enterprise [16] andthe extended enterprise [17].Networked organisations apply a co-ordinationmechanism which is an intermediate between marketco-ordination and hierarchical co-ordination. Hierarchiesand markets support productivity and flexibility,respectively, but can hardly support them simultaneously.In contrast to an open market, there is somejoint commitment among networked organisations toestablish and cultivate relationships [13]. Whereas in aclosed hierarchy there is unity of ownership, powerand loyalty, in a network of networked organisationsthere is no single trinity, but distributed ownership,power and loyalty instead [18]. Networked organisationsare able to couple and uncouple other organisationswith less cost and time than hierarchies [11].Hierarchies are bound to an installed base of dedicatedresources that can not be adapted immediately. It iseasier to start or end co-operation with a partner than itis to add or remove an organisational unit [18].To obtain the required flexibility, supply chainscould introduce supply chain management by cooperationacross a network of organisations. As isshown in Fig. 2, networked organisations apply lateralcontrol to achieve supply chain management beyondtheir organisational boundaries. When compared tosupply chain management by domination of one organisation,co-operation across networked organisationsFig. 2. Supply chain management by co-operation across networked organisations.

168 M. Verwijmeren / Computers in Industry 53 (2004) 165–178can also increase the flexibility of the supply chains.Networked organisations can frequently change theirpositions in supply chains to respond to changingmarket circumstances. In the context of networkedorganisations, new supply chains emerge to createpromising products, while supply chains for outdatedproducts disappear. The changing relationships of networkedorganisations result in dynamic supply chains.Traditional ERP systems and EDI interfaces arenot equipped for supply chain management acrossnetworked organisations. Enterprise Resource Planningsystems focus on the integral management ofprocesses within a company [19,20]. Because of itscentral architecture, an ERP system assumes onecentral organisation, but in dynamic networks thereis no central point of authority [21]. Moreover, theprocedural ERP software can not easily supportthe coupling and decoupling of organisations to thedynamic network. ERP systems have intelligencefor co-ordination, but miss the flexibility needed fornetworked organisations. Traditional company systemscan mutually be linked through electronic datainterchange (EDI) [22–24]. The systems and interfacescan resemble the network structure, but do notprovide additional functions for logistics controlacross organisations. They can align with the structureof networked organisations, but lack the intelligencefor co-ordination across the supply chain.3. Supply chain management architecture3.1. ERP, WMS and TMSThe fundament of a system architecture for supplychain management is constituted by ERP systems,WMS and TMS (see Fig. 3). These information systemscan be either standard software packages withparameter configuration or software programs tailormade to company specific needs. The basic systems ina supply chain provide specific functions for typicalusers: ERP, enterprise resource planning systems: functions: purchase, materials management andsales; users: manufacturers and trading companies. WMS, warehouse management systems: functions: receipts put-away, bin managementand order picking; users: logistics service providers and wholesalers. TMS, transportation management systems: functions: transport booking, planning and monitoring; users: forwarders and carriers.A system like ERP, TMS and WMS has its strengthin the consistent management of elementary businessFig. 3. ERP, WMS and TMS in the supply chain management architecture.

M. Verwijmeren / Computers in Industry 53 (2004) 165–178 169data, such as: customers and sales orders, items andprices, warehouses and bins, resources and workorders, suppliers and purchase orders.Each of the systems has its own database in whichthese data are stored. The software programs in anERP system, WMS or TMS provide functions for thetransformation of the stored data during the courseof the business process. The users can access theprograms they need for their tasks through a userinterface.The ERP, WMS and TMS focus on co-ordination ofthe processes within the organisation. Their databasesstore the internal data, the programs provide intelligencefor internal co-ordination and the user interfacesgive access to internal users. So, these traditionalsystems are fit for internal logistics management,but they miss the co-ordination capabilities requiredfor the management of dynamic supply chains. PossibleEDI interfaces between the local systems enablethe exchange of information, but they do not add therequired intelligence for supply chain management.3.2. Software componentsFor supply chain management across dynamic organisationnetwork, the basic ERP, WMS and TMS(including EDI interfaces) are not enough. Additionalsystems are needed in the supply chain managementarchitecture to support the co-ordination of logisticsprocesses which are distributed over different organisations.Therefore, in the supply chain management architecture(see Fig. 4), we introduce software componentson top of ERP, WMS and TMS to provide extra intelligencefor co-ordination as well as greater flexibilityto cope with dynamics. The required intelligence is builtin the software components, whereas the componentstructure enables the required flexibility.The software components in the architecture arecalled supply chain engines (SCEs), to express thepower they provide for supply chain management as anextension to the local ERP, WMS and TMS. The supplychain engines can run on the computers of the differentorganisations in the supply chain. Together, the softwarecomponents make up a system layer dedicated tosupply chain management. Whereas the ERP, WMSand TMS focus on internal management, the SCEs addfunctions and data for external management to thesupply chain architecture. The software componentsin the architecture can be classified in three categories,differentiated to the level of supply chain managementsupported by the engines: Communication engines: function: basic communication between thesystems (and users) in the supply chain; examples: data communication, message conversionand flow control engines. Information engines: function: transparent information over the systems(and users) in the supply chain; examples: stock visibility, track and trace andreport query engines. Management engines: function: advanced management across the systems(and users) in the supply chain; examples: inventory management, productionmanagement and distribution managementengines.In the communication layer, the message communicationengines, for example, support the exchangeof messages between systems using either SMTP(e-mail), FTP (file transfer) of HTTP (web protocol).A message communication engine is installed on eachof computers to be connected. Then, the local systemscan exchange messages through the local engines,which pack, send, transfer, receive and unpack themessages.At the information level, the stock visibilityengines, for example, can present stock data fromdifferent local systems. External users can specifytheir information request via a Web browser. Theengines retrieve the stock levels from the local systemsand integrate the information in a stock overview. Thisoverview can be displayed to an external user or can beimported in another system in the supply chain.The management layer includes engines foradvanced management of the supply chain. They haveintelligent rules for fully automatic decision makingand semi-automatic decision support. For example,distribution management engines can be used forthe integral management of distribution serviceswhich are provided by independent organisationsin a physical distribution network. Below, we willexplain the design and application of inventory managementengines in the software component architecturefor supply chain management.

170 M. Verwijmeren / Computers in Industry 53 (2004) 165–178Fig. 4. Software component architecture in supply chain management.4. Inventory management architecture4.1. Inventory management enginesInventory management engines enable integralinventory management across networked organisations(see Fig. 5). These software components together withexisting systems in the supply chain represent an inventorymanagement architecture. An inventory managementengine (IME) is an extremely elementary system,managing the stock level of just one single SKU stockpoint.The software components are loosely coupled toone another in networks, in order to support integralinventory management in dynamic supply chains [25].The local management systems (ERP, TMS or WMS)can be used by the IMEs to get and set stock information.The IMEs can be distributed over several locationsand organisations in the supply chains.The inventory management engines support integralinventory management according to base stock control(BSC), material/distribution requirements planning(MRP/DRP) and line requirements planning (LRP)[26–30]. With the help of the system variables andsystem equations in a network of IMEs, integralinventory management can be supported accordingto one of these algorithms. The software componentsintegrate stock levels in the time dimension as well asin the place dimension, by using extra informationon time-phased demand (order plans) and integralinventory (local plus downstream stock).

M. Verwijmeren / Computers in Industry 53 (2004) 165–178 171Fig. 5. Inventory management architecture.For the support of networked organisations somesupplementary system variables and equations areincluded in the IMEs. The software components supportconfiguration flexibility, timing flexibility andalgorithm flexibility. Configuration flexibility is theability to couple and decouple of IMEs in a dynamicnetwork. Timing flexibility refers to coping withdifferent review moments and plan moments. Algorithmflexibility includes algorithm selection in anIME and algorithm transition across the engines.These properties allow autonomy of networked organisationswith respect to the place, the timing and thetype of management.The software components for integral inventorymanagement across networked organisations havean object-oriented design (see Fig. 6). An informationsystem for supply chain management can be builtof objects that represent real-world entities in thesupply chain [31,32]. Forthcoming objects are durable,and have data and functions that naturallybelong together. An inventory management engineconsists of five object classes, each responsible forthe functions and data related to an entity type inthe system environment: customers, suppliers, operationalprocess, inventory (stock-point) and strategist(supervisor) [33].The object classes in an IME are equipped withattributes and operations which enable them to performin accordance with their targets [34,35]. Theoperations of an object class give the ability to provideits functions, while the attributes give the ability tomaintain its data. The objects work together by sendingmessages, representing a request for a service andthe response to that service request. Customers andsuppliers in the environment can again be equippedwith IMEs, but they may also use other informationsystems or no system at all. If a customer or a supplieralso makes use of an IME, the same system associationsapply from the viewpoint of the customer or thesupplier.4.2. Supply chains for cordless phonesThe inventory management architecture has beentested against a network of supply chains for manufacturingand distribution of cordless phones [36,37].

172 M. Verwijmeren / Computers in Industry 53 (2004) 165–178Fig. 6. Object classes in the inventory management engine.A dozen of manufacturers in various countries supplya telecommunications retailer with cordless telephones(see Fig. 7). The retailer sells the productsthrough more than hundred outlets to customers in theconsumer and business market. Each of the organisationshas its own information systems in place for localmanagement. Due to market and technology changes,the network of organisations in the supply chain isexpanding and changing more frequently. At the sametime, the supply chains need to be managed in a moreintegral way to remain competitive.In the supply chain for cordless telephones there isthe need for integral inventory management across thenetworked organisations. However, it is practicallyunfeasible to develop and maintain the networkedinventory management with the existing informationsystems in the supply chain. The heterogeneityand instability of the local systems would requirea prohibitive number of dedicated interfaces. Networkedinventory management can be achieved byapplication of IMEs for every single SKU stock-point.Loose coupling enables the IMEs to combine integrationand flexibility. The software components form asystem layer for supply chain management, abstractedfrom the existing systems for local management.5. Implementation technology5.1. Components, interfaces and servicesImplementation technology is needed for the developmentand deployment of the software componentsin the supply chain management architecture.The technology used for the implementation of thesupply chain engines includes: application components,

M. Verwijmeren / Computers in Industry 53 (2004) 165–178 173Fig. 7. Inventory management engines in a supply chain for cordless telephones.

174 M. Verwijmeren / Computers in Industry 53 (2004) 165–178Fig. 8. Components, interfaces and services in the supply chain architecture.interfaces of the components and common services forthe components (see Fig. 8). The supply chain enginesare the application components in the architecture,which can be distributed over different servers, locationsand organisations.The SCEs have an object-oriented design and arealso built as a collection of object classes and instances[38]. The objects have attributes to store data and havemethods to execute functions. Only objects with aremote interface can be accessed by other components.The components provide services to othercomponents through these interfaces [39]. A clientcomponent (object) can send a request for a service toa server component (object), which can send back aresponse to the client component.The software components have interfaces for communicationin the organisation and across the supplychain. The supply chain engines can be equipped withthe following interfaces: System interfaces to other supply chain engines,for example, interfaces between distribution managementengines to support integral distributionmanagement. System interfaces to local information systems,for example, interfaces between the inventorymanagement engines and ERP systems to retrievethe stock levels. User interfaces to users, for example, the supervisor(strategist) of the inventory management enginewho sets the parameters for integral inventorymanagement. Database interfaces, for example, to store the detailsof a customer order in a distribution managementengine persistently in a database.Common services are available in the supply chainmanagement architecture to facilitate the interactionbetween the distributed components. Some commonservices in the software component architecture are[38–41]: Naming service: the naming service contains adirectory of logical names and technical addressesof components. With the help of a naming service, aclient component can send a service request toa server component by using its logical name.The naming service provides the technical addressof the server component. Trading service: components can use a tradingservice to publish their service interfaces withnames, attributes and types. Components can searchfor available services in other components. The

M. Verwijmeren / Computers in Industry 53 (2004) 165–178 175trading service provides references to the discoveredservices, so that a client component can send aservice request to the server component. Messaging service: a messaging service uses intermediatequeues for the exchange of messages(requests/responses) between components to guaranteemessage delivery. Asynchronous communicationmakes that a client component does not have towait for a response after a request has been sent.Publish-subscribe features facilitate distribution ofmessages from a publisher to all subscribers. Transaction service: a transaction is a sequence ofoperations in which several components can beinvolved. The transaction service makes sure thata transaction complies with the requirements ofatomicity, consistency, isolation and durability. Atwo-phase commit protocol ensures that all componentscommit to transaction completion or rollback to an original state in the event of a failure.5.2. Java Enterprise, CORBA and Web ServicesAvailable technologies for the implementation ofthe software component architecture for supply chainmanagement are Java Enterprise, CORBA and WebServices (see Fig. 9). In the Java Enterprise architecture,the objects in a supply chain engine are written inthe Java programming language [39,41–44].Some Java objects of an application componenthave a RMI interface, which makes the interfaceavailable to Java objects in other components. Remotemethod invocation (RMI) is a Java technology forinterfacing between distributed objects written in Java.A RMI client object can send a service request to aRMI server object. The remote invocation is handledin the Java platform by an RMI stub at the client sideand a RMI skeleton at the server side. The protocol forRMI communication is Java Remote Method Protocol(JRMP) or Internet Inter-ORB Protocol (IIOP), bothrunning on top of the TCP/IP Internet protocols.The Java Enterprise platform consists of J2EE (Java2 Enterprise Edition) servers, which offer a range ofcommon services to the application components. TheJava Naming and Directory Interface (JNDI) service isa naming service, for translation between logicalnames and technical addresses. In addition, Jini providesa trading service in Java for registration anddiscovery of distributed services.The supply chain engines could also be implementedas CORBA components [40,45–48]. The objectsin CORBA components can be written in any programminglanguage, like C, Cþþ or Java. CORBAcomponents interact via IDL interfaces, specified inthe neutral interface definition language (IDL) andindependent from the programming language of theobjects.Fig. 9. Java Enterprise, CORBA and Web Services in the component architecture.

176 M. Verwijmeren / Computers in Industry 53 (2004) 165–178When a client component invokes a service on aservice component, the CORBA platform internallyuses stubs and skeletons to forward the request andto feed back the response. The IIOP is used over TCP/IPto communicate the messages between the components.The CORBA platform has its own CORBA Namingservice for name resolution and CORBA Trading forservice discovery. By IDL interfaces in addition toRMI interfaces, Java components can also worktogether with components in a CORBA platform.Web Services refer to XML, WSDL, SOAP andUDDI as a set of emerging web technologies for theimplementation of a software component architecture[49]. The supply chain engines in the Web Servicesplatform can be built in any programming language aslong as they provide the services as specified in theWSDL interfaces.The Web Services Definition Language (WSDL)uses eXtensible Markup Language (XML) to specifywhat service is offered by a component and how thisservice can be requested by other components [50].The components in the Web Services platform interactby sending messages in XML format that comply tothe WDSL interfaces.The communication protocol for the messages isSimple Object Acces Protocol (SOAP) [51]. SOAPisan XML-based protocol that defines the messageenvelop, the encoding rules and the transport options.The SOAP protocol makes use of the web protocolHyperText Transfer Protocol (HTTP) to communicatemessages between widely available Web servers.Alternatively, SOAP can run over the e-mail protocolSimple Mail Transfer Protocol (SMTP).The low level naming service for the Web Servicesplatform is Domain Name Service (DNS) whichtranslates URL-names into IP-addresses. The tradingservice in the Web Services architecture is based onUniversal Description, Discovery and Integration(UDDI) [52]. Server components can publish theirservices in the UDDI service. Client components canget extensive information from the UDDI serviceabout server components, including names, addresses,services and WDSL descriptions.6. ConclusionSupply chain management can increase productivityas a result of improved co-ordination across organisations.Networked organisations can provide greaterproduct variety as a result of higher flexibility of thesupply chain. In dynamic supply chains there is the needfor information systems which support co-ordinationand flexibility at the same time.The software component architecture for supplychain management across dynamic organisational networksconsists of existing systems supplemented withnew software components. The local management inthe architecture is done by existing enterprise resourceplanning systems, warehouse management systemsand transportation management systems.The integral management in the architecture isexecuted by supply chain engines. These softwarecomponents make up a layer of systems for supplychain management running on top of ERP, WMS andTMS. The SCEs provide both the intelligence and theflexibility as required for the integral management indynamic supply chains. The supply chain engines canbe classified in communication, information and managementengines.Inventory management engines can make up asystem layer for integral inventory management,using the algorithms BSC, MRP or LRP and providingflexibility for algorithm selection, network configurationand review timing. The IMEs have an objectorienteddesign, with the classes based on entities inthe environment managed by the system. The IMEscan be applied for the integral inventory managementacross networked organisations in the supply chain ofcordless phones.The software component architecture can be implementedwith technology for the application components,their interfaces and common services. Theapplication components can be built from the objectsin the object-oriented design. The interfaces are usedto send requests and receive responses between theclient and server components. Examples of facilitatingservices are naming, trading, messaging and transactionservices.Java Enterprise, CORBA and Web Services aretechnologies for the implementation of the softwarecomponent architecture. The Java platform makes useof RMI-interfaces, the JRMP or IIOP protocol, andprovides the JNDI naming service and Jini tradingservice. The CORBA solution is based on IDL interfaces,the IIOP protocol and the CORBA namingand trading services. The Web Services architecture

M. Verwijmeren / Computers in Industry 53 (2004) 165–178 177applies WSDL (XML) interfaces, communicates withthe SOAP protocol, and use DNS and UDDI as namingand trading service.References[1] M.A.A.P. Verwijmeren, Networked inventory managementby distributed object technology, Ph.D. thesis, EindhovenUniversity of Technology, Eindhoven, The Netherlands, 1998.[2] L.M. Ellram, Supply chain management: the industrialorganisation perspective, International Journal of PhysicalDistribution & Logistics Management 21 (1) (1991) 13–22.[3] T.C. Jones, D.W. Riley, Using inventory for competitiveadvantage through supply chain management, InternationalJournal of Physical Distribution & Materials Management 15(5) (1985) 16–26.[4] J. Stevens, Integrating the supply chain, International Journalof Physical Distribution & Materials Management 19 (8)(1989) 3–8.[5] M. Christopher, Logistics and Supply Chain Management,Financial Times, London, UK, 1994.[6] H.S. Sheombar, Understanding logistics co-ordination: afoundation for using EDI in operational (re)design of dyadicalvalue adding partnerships, Ph.D. thesis, Katholieke UniversiteitBrabant, Tilburg, The Netherlands, 1995.[7] M. Walker, Supplier–retailer collaboration in europeangrocery distribution, Logistics Information Management 7(6) (1994) 23–27.[8] H.B. Thorelli, Networks: between markets and hierarchies,Strategic Management Journal 7 (1) (1986) 37–51.[9] W.W. Powell, Neither market nor hierarchy: network formsof organisation, Research in Organisational Behaviour 12(1990) 295–336.[10] M.S. Scott Morton (Ed.), The Corporation of the 1990s:Information Technology and Organisational Transformation,Oxford University Press, New York, USA, 1991.[11] R.E. Miles, C.C. Snow, Causes of Failure in NetworkOrganisations, California Management Review (1992) 53–72.[12] F.E. Webster, The changing role of marketing in thecorporation, Journal of Marketing 56 (1992) 1–17.[13] C. Ching, C.W. Holsapple, A.B. Whinston, Toward ITsupport for co-ordination in network organisations, Information& Management 30 (1996) 179–199.[14] R. Johnston, P.R. Lawrence, Beyond Vertical Integration: TheRise of the Value-Adding Partnership, Harvard BusinessReview (1988) 94–101.[15] W.H. Davidow, S.M. Malone, The Virtual Corporation:Structuring and Revitalizing the Corporation for the 21stCentury, Harper Business, New York, USA, 1992.[16] J.P. Womack, D.J. Jones, From Lean Production to the LeanEnterprise, Harvard Business Review (1994) 93–103.[17] J. Browne, P.J. Sackett, J.C. Wortmann, Future manufacturingsystems: towards the extended enterprise, Computers inIndustry 25 (1995) 235–254.[18] J.E. Aken, L. van Hop, G.J.J. Post, The virtual organisation: aspecial mode of strong inter-organisational co-operation,in: M.A. Hitt, J.E. Ricart, R.D. Nixon (Eds.), ManagingStrategically in an Interconnected World, Wiley, Chichester,UK, 1998.[19] F.L.G. van Lierop, H. Vanmaele (Eds.), Multi-site BusinessInformation Systems, DELWEL, Den Haag, The Netherlands,1996.[20] Ovum, ERP for Manufacturers, Ovum, London, UK, 1997.[21] M.J. Euwe, Decentralized control systems for logisticalcoordination, Ph.D. thesis, Eindhoven University of Technology,Eindhoven, The Netherlands, 1999.[22] P. van der Vlist, W.J. de Jong, A.E. Kolff, D.J. van der Net, A.van Overbeek, A.T.C. Siebbeles (Eds.), EDI in de Handel(EDI in Trade), Samsom, Alphen aan den Rijn, TheNetherlands, 1991 (in Dutch).[23] P. van der Vlist, W.J. de Jong, A.E. Kolff, D.J. van der Net, A.van Overbeek, A.T.C. Siebbeles (Eds.), EDI in de Industrie(EDI in Industry), Samsom, Alphen aan den Rijn, TheNetherlands, 1992 (in Dutch).[24] P. van der Vlist, W.J. de Jong, A.E. Kolff, D.J. van der Net, A.van Overbeek, A.T.C. Siebbeles (Eds.), EDI in de Transportsector(EDI in the Transport Sector), Samsom, Alphen aanden Rijn, The Netherlands, 1994 (in Dutch).[25] M.A.A.P. Verwijmeren, P. van der Vlist, K. van Donselaar,Networked inventory management information systems:materialising supply chain management, International Journalof Physical Distribution & Logistics Management 26 (6)(1996) 16–31.[26] J.F. Magee, Production Planning and Inventory Control,McGraw-Hill, New York, USA, 1958.[27] A.J. Clark, H. Scarf, Optimal policies for a multi-echeloninventory problem, Management Science 6 (1960) 475–490.[28] J. Orlicky, Material Requirements Planning, McGraw-Hill,New York, USA, 1975.[29] A.J. Martin, DRP: Distribution Resources Planning, OliverWight, Essex Junction, USA, 1983.[30] K.H. van Donselaar, Material coordination under uncertainty,Ph.D. thesis, Eindhoven University of Technology, Eindhoven,The Netherlands, 1989.[31] J.H.A. Harink, K.M. Duursma, P. van Raad, M.A.A.P.Verwijmeren, Architectuur Voor Besturing van LogistiekeKetens (Architecture for Management of Supply Chains),PTT Research, Groningen, The Netherlands, 1993 (in Dutch).[32] M.A.A.P. Verwijmeren, Object-oriented design of integraldistribution services, KPN Research, Leidschendam, 1996.[33] M.A.A.P. Verwijmeren, Exploiting distributed object technologyto achieve networked inventory management, Computersin Industry 2000 (41) (2000) 239–250.[34] L.F. Toia, Prototyping networked inventory managementinformation systems, KPN Research, Groningen, The Netherlands,1997.[35] E. van het Hof, Networked inventory management informationsystems: ontwerp en implementatie (networked inventorymanagement information systems: design and implementation),KPN Research, Leidschendam, The Netherlands, 1997(in Dutch).

178 M. Verwijmeren / Computers in Industry 53 (2004) 165–178[36] L.B.P. Rijen, Herontwerp van een logistieke keten met behulpvan simulatie (redesign of a supply chain with the help ofsimulation), PTT Research, Groningen, The Netherlands,1994 (in Dutch).[37] R.J.W. van Grinsven, Case-studie onderzoek bij PTT telecomnaar networked inventory management information systems(case study research at ptt telecom into networked inventorymanagement information systems), KPN Research, Groningen,The Netherlands, 1997 (in Dutch).[38] Object Management Group (OMG), OMG Unified ModelingLanguage Specification, Version 1.4, Object ManagementGroup, Framingham, USA, September 2001.[39] P.J. Perrone, V.S.R.R. Chaganti, Building Java EnterpriseSystems with J2EE, SAMS Publishing, Indianapolis, USA,2000.[40] J. Siegel, CORBA: Fundamentals and Programming, Wiley,New York, USA, 1996.[41] C.J. Berg, Advanced Java Development for EnterpriseApplications, Prentice-Hall, Upper Saddle River, USA, 2000.[42] B. Shannon, M. Hapner, V. Matena, J. Davidson, E. Pelegri-Llopart, L. Cable, Java 2 Platform Enterprise Edition,Addison-Wesley, Boston, USA, 2000.[43] N. Kassem, The Enterprise Team Designing EnterpriseApplications with the Java 2 Platform Enterprise Edition,Addison-Wesley, Boston, USA, 2000.[44] Sun Microsystems, Java TM 2 Platform, Enterprise EditionSpecification Version 1.4 (Public Draft), Sun Microsystems,Santa Clara, USA, July 2002.[45] R. Ben-Natan, CORBA: A Guide to Common Object RequestBroker Architecture, McGraw-Hill, New York, USA, 1995.[46] T.J. Mowbray, R. Zahavi, The Essential CORBA: SystemsIntegration Using Distributed Objects, Wiley, New York,USA, 1995.[47] R. Orfali, D. Harkey, J. Edwards, The Essential DistributedObjects Survival Guide, Wiley, New York, USA, 1996.[48] Object Management Group (OMG), The Common ObjectRequest Broker: Architecture and Specification, Version2.6.1, Object Management Group, May 2002.[49] R.J. Brunner, I. Cohen, F. Curbera, D. Govini, S. Haises, M.Kloppmann, B. Marchal, K. Scott Morrison, A. Ryman, J.Weber, M. Wutka, Java Web Services, SAMS Publishing,Indianapolis, USA, 2000.[50] World Wide Web Consortium (W3C), Web ServicesDescription Language (WSDL), Version 1.2 (Working Draft),World Wide Web Consortium (W3C), July 2002.[51] World Wide Web Consortium (W3C), SOAP, Version 1.2,Part 0, Primer (Working Draft), World Wide Web Consortium(W3C), June 2002.[52] UDDI.org, UDDI, Version 3.0, Specification (Open Draft),UDDI.org, July 2002.Martin Verwijmeren graduated cumlaude in Industrial Engineering andManagement Science. He obtained aPhD from the Eindhoven University ofTechnology for his research into systemsfor integral inventory managementacross networked organisations, usingdistributed and object-oriented technology.He worked as project manager forKPN Research, the R&D department ofa telecommunications company. Martinwas senior project manager at TNT, responsible for engineeringand IT projects at the logistics service provider. Currently, he isdirector of MP Objects, a company that develops softwarecomponents for supply chain management. He has published inthe European Journal of Operations Research, The InternationalJournal of Physical Distribution & Logistics Management,Computers in Industry and several logistics journals in TheNetherlands.