Physical Distribution & Logistics Management

Volume 32 Number 7 2002 ISSN 0960-0035 International Journal of Physical Distribution & Logistics Management Agile logistics for the virtual organization: analysis and modeling Guest Editors: Laura Meade, Joseph Sarkis and Srinivas Talluri www.emeraldinsight.com

International Journal of ISSN 0960-0035 Physical Distribution & Volume 32 Number 7 2002 Logistics Management Agile logistics for the virtual organization: analysis and modeling Guest Editors Laura Meade, Joseph Sarkis and Srinivas Talluri Paper format Internet Online Publishing This issue is The International Journal of with Archive, Reference Linking, part of a Physical Distribution & Logistics Emerald WIRE, Key Readings, comprehensive Management includes ten issues in Research Register, Institution-wide multiple access traditional paper format. The contents of this Licence, E-mail Alerting Service and Usage information issue are detailed below. Statistics. service Access via the Emerald Web site: http://www.emeraldinsight.com/ft See p. 499 for full details of subscriber entitlements. Access to International Journal of Physical Distribution & Logistics Management online ________ 499 CONTENTS Editorial advisory board ___________________________ 500 Abstracts and keywords ___________________________ 501 French abstracts___________________________________ 503 Spanish abstracts __________________________________ 505 Japanese abstracts_________________________________ 507 Guest editorial ____________________________________ 510 Research agenda for e-business logistics based on professional opinions Jaana Auramo, Anna Aminoff and Mikko Punakivi ___________________ 513 A comparative study of three different SCM approaches R. Meenakshi Sundaram and Sameer G. Mehta _______________________ 532 Improving materials management effectiveness: a step towards agile enterprise M. Caridi and R. Cigolini _________________________________________ 556 Measuring supply chain agility in the virtual organization Mary Margaret Weber ___________________________________________ 577

A multi-dimensional empirical exploration of CONTENTS technology investment, coordination and firm continued performance Anthony Ross __________________________________________________ 591 Improving electronics manufacturing supply chain agility through outsourcing Scott J. Mason, Michael H. Cole, Brian T. Ulrey and Li Yan ____________ 610

International Journal of Physical IJPDLM online Distribution & Logistics Management online An advanced knowledge resource for the entire organization Access via the Emerald Web site – http://www.emeraldinsight.com/ft Subscribers to this journal benefit from access to a fully Research Register 499 searchable knowledge resource stretching far beyond the Research Registers are an Internet-based database where current volume and issue. International Journal of you can identify inside information on research activity Physical Distribution & Logistics Management online is worldwide. Whether you are seeking information or you enhanced with a wealth of features to meet the need for wish to provide details of your own research, you will find fast, effortless and instant access to the core body of this feature invaluable. knowledge. Furthermore, this user-friendly electronic library may be networked throughout the subscribing organization to maximize the use and value of the Emerald Alert subscription. This is augmented with advanced search The table of contents e-mail alerting service will e-mail you facilities and ‘‘choice of access’’ through a variety of the contents page of any chosen journal whenever the alternative journal gateways. latest issue becomes available online. For further information please go to http://www.emeraldinsight.com/ alerts Emerald online access includes: Support Resources Institution-wide Licence A comprehensive range of resources is available online Our liberal licence allows everyone within your institution that helps users learn how to use online information to access the journals via the Internet, making your resources and helps information professionals market subscription more cost-effective. It has been designed to effectively resources to their users. For further information provide a comprehensive, simple system with minimum please go to http://www.emeraldinsight.com/support administration. Access can be granted by IP range or username and password. ATHENS authentication is enabled. Choice of Access Electronic access to this journal is available via a number Current Issue + Journal Archive of channels, enabling end users’ libraries to reach the Internet access via Emerald Fulltext to information as it content through their preferred delivery system. becomes available online and to material from past The Emerald Web site – http://www.emeraldinsight. volumes. Users can choose to browse the database for com/ft – is the recommended means of electronic access, relevant articles or search using the powerful search as it provides fully searchable and high value-added engine provided. access to the complete content of the journal. Refer to the next section for ‘‘how to access’’ via the Emerald Web Active Reference Linking site. Subscribers can also access and search the article Direct links from the journal article references to abstracts content of this journal through the following services: of the most influential articles cited. Where possible, this Ebsco Online link is to the full text of the article. http://www.ebsco.com/online Emerald WIRE Huber E-Journals http://e-journals.hanshuber.com/english A fully searchable subject-specific database, brought to Minerva you by Emerald Management Reviews, providing article http://www.minerva.at/ abstracts from the world’s top management, computing OCLC Firstsearch ‘‘Electronic Collections Online’’ and engineering journals. Directly relating to the journal http://www.uk.oclc.org/oclc/menu/eco.htm subject area, the database is updated monthly and gives RoweCom’s ‘‘Information Quest’’ users details of how to obtain the full text of original http://www.informationquest.com articles. SilverPlatter http://www.silverplatter.com Usage Statistics SwetsBlackwell’s ‘‘SwetsNetNavigator’’ Online Journal Usage Statistics are now available. This http://www.swetsnetnavigator.nl feature allows Emerald Administrators to download their usage statistics with regard to their organization’s journal usage. Usage Statistics allow you to review the value of How to access this journal through Emerald electronic dissemination of your journal subscriptions Organizations must first register for online access throughout your organization. They can also help (instructions provided at http://www.emeraldinsight. determine the future trends for information within your com/register), after which the content is available to organization. For further information go to http:// everyone within the organization’s domain. To access this www.emeraldinsight.com/stats journal’s content, simply log on either from the journal Key Readings homepage or direct through the Emerald Web site. Abstracts of articles, relating to keywords, are selected to Emerald Customer Support Services provide readers with current awareness of interesting For customer service and technical help, contact: articles from other publications in the field. The abstracts E-mail: support@emeraldinsight.com are available online and are updated regularly upon Telephone: (44) 1274 785278 publication of the latest issue. Fax: (44) 1274 785204

IJPDLM EDITORIAL ADVISORY BOARD Dr Richard A. Lancioni Professor of Marketing & Logistics, Temple University, USA 32,7 Dr C. John Langley Jr Professor of Supply Chain Management, Georgia Institute of Dr Prabir Bagchi Technology, USA Professor of Logistics & Management, George Washington University, USA Dr Michael Levy Charles Clarke Reynolds Professor of Marketing, Dr Ronald H. Ballou Babson College, USA Professor of Operations, Case Western Reserve University, USA Dr Arvinder P.S. Loomba Associate Professor of Organization and Management, San Jose Rick D. Blasgen 500 Vice President Supply Chain, Nabisco Inc., USA State University, USA Clifford F. Lynch Dr Joseph L. Cavinato President, C.F. Lynch & Associates, USA Senior Vice President, National Association of Purchasing Management, USA John McCormick University of New South Wales, Australia Dr Garland Chow Associate Professor of Logistics, University of British Professor Alan McKinnon Columbia, Canada Logistics Research Centre, Heriot-Watt University, Edinburgh, UK Dr Martin Christopher Professor of Marketing and Logistics, Cranfield School of Norman E. Marr Management, UK Division of Marketing, University of Huddersfield, UK Dr David J. Closs Dr G.C. Meeuse Professor of Marketing and Logistics, Michigan State Rotterdam, The Netherlands University, USA Dr John Thomas Mentzer Dr Jacques Colin The Bruce Excellence Chair of Business Policy, University of Institut Universitaire Technologie, France Tennessee, USA Dr Rajiv P. Dant Dr Alan Mercer Associate Professor of Marketing, Boston University, USA Professor of Operations Research, Lancaster University, UK Dr Patricia Daugherty Dr Paul Murphy Siegfried Professor of Marketing, Division of Marketing, Professor of Marketing and Logistics, John Carroll University, University of Oklahoma, USA USA David A. Durtsche Dr Bruce Murtagh TranzAct Technologies, Inc., USA Professor of Management, Graduate School of Management, Macquarie University, Australia Dr Margaret A. Emmelhainz Associate Professor of Marketing, University of Georgia, USA Dr Pieter Nagel Partner, Burns Bridge Nagel Pty Ltd, Australia Graham A. Ewer Chief Executive, Institute of Logistics, UK Dr R. Mohan Pisharodi Associate Professor of Marketing, Oakland University, Patrick Forsyth USA Oklahoma State University-Tulsa, USA Cees J. Ruijgrok Frances Fowler Professor Logistics Section, INRO-TNO, The Netherlands Miami University, Ohio, USA Dr Jay Sankaran Thomas L. Freese Senior Lecturer, University of Auckland, New Zealand Principal, Freese & Associates, Inc., USA Dr Philip B. Schary Dr Jerry Goolsby Professor Emeritus, Oregon State University, USA Associate Professor of Marketing, University of South Florida, USA Dr Arun Sharma Associate Professor of Marketing, University of Miami, Dr Bernard J. Hale USA Logistics Consultant, USA Dr Tage Skjott-Larsen Dr Anthony F. Han Professor, Institute for Logistics and Transport, Copenhagen Professor of Transportation Management, National Chiao Tung Business School, Denmark University, Taiwan, Republic of China Alan Slater Dr Alan Harrison Director, Added Value Logistics Consulting Limited, Professor of Operations and Logistics, Cranfield School of Manchester, UK Management, UK Amrik Sohal Dr James L. Heskett Director, Monash University, Australia UPS Foundation Professor of Business Logistics, Harvard University, USA Dr Mark Speece Nanyang Technological University, Singapore Herbert Hodus Consultant, IFM Logistics, USA Dr Thomas W. Speh Professor of Marketing and Logistics, Miami University, Dr Daniel E. Innis USA Associate Dean, Ohio University, USA Dr Jay U. Sterling Dr Zahir Irani Associate Professor of Marketing and Logistics, University of Senior Lecturer of Information Systems, Brunel University, UK Alabama, USA Olof Johansson Dr Diana Twede University of Umea, Sweden Associate Professor, Michigan State University, USA Dr Andrew Kerr Hans van der Hoop Managing Director, Griffin Corporate Services, NSW, Logistics International, Rotterdam, The Netherlands Australia Dr Hugo T.Y. Yoshizaki Dr Bernard J. La Londe Assistant Professor of Production Engineering, University of International Journal of Physical Professor Emeritus, Ohio State University, USA Sao Paulo, Brazil ˜ Distribution & Logistics Dr Douglas M. Lambert Dr Paul H. Zinszer Management, Vol. 32 No. 7, 2002, Raymond E. Mason Professor of Transportation & Logistics, Associate Professor of Marketing, Syracuse University, p. 500. # MCB University Press, Ohio State University, USA USA 0960-0035

Research agenda for e-business approach. In the integrated approach, all Abstracts and logistics based on professional opinions decisions are assumed to be made at a Jaana Auramo, Anna Aminoff and single level. keywords Mikko Punakivi Keywords Logistics, Improving materials management Supply chain management, R&D, Supply, effectiveness: a step towards agile Networks, New product development enterprise This study investigated what should be the 501 M. Caridi and R. Cigolini major research and development areas Keywords Demand management, regarding the logistics of electronic business. Inventory control, MRP, The method of study consisted of five Manufacturing systems, Computer simulation elements: the creation of a preliminary e- logistics vision; focus interviews of e-business This research provides a literature review in and logistics experts, identification and the field of uncertainty dampening methods categorisation of the key R&D topics, a for manufacturing systems, and proposes a panel workshop to critically analyse the new model to improve materials management preliminary findings and prioritise the R&D effectiveness in materials requirements topics, and formulation of a research agenda planning environments. The literature to guide future research work in the field of e- review gives rise to a classification business logistics. The two-phased process, framework of the models along nine where the interviews were followed by the structural dimensions that refer to the safety workshop, enabled the evaluation and buffer treatment, the environmental prioritisation of the preliminary findings. characteristics and the type of approach. On The requirements of e-business on the the basis of the classification framework, the national logistics infrastructure were found proposed model provides guidelines for to be one of the focus research areas. approaching the problem of dimensioning, According to the study, integrated supply positioning and managing safety stocks network structure with suitable visibility and against demand uncertainty. The usage of real-time data transfer is another effectiveness of the proposed model has been area of great importance. Research and tested by comparing it to the traditional development of new logistics service approach, through a computer-based concepts should also be promoted as well as simulation. research on the effects and possibilities of using new product data management and product identification methods. Measuring supply chain agility in the virtual organization Mary Margaret Weber A comparative study of three different SCM approaches Keywords Agile production, Flexibility, Market share, Direct marketing R. Meenakshi Sundaram and Sameer G. Mehta The need for increased flexibility in responding to market demand is driving a Keywords Supply chain management, heightened interest in virtual, or agile Logistics, Distribution, Production, organizations. However, agile response in Decision making, Integration the supply chain may not always be A comparative study of three different necessary and may not always be a better approaches on a hypothetical supply chain alternative than more traditional organizational model is presented. The three approaches structures. The model proposed in this paper investigated are: independent; semi-integrated; provides a means of measuring both the need and integrated. In the independent approach, for agility and how agile an organization it is assumed that decisions are made actually is. This is accomplished through the International Journal of Physical independently at three different levels. use of a hierarchical model that details with Distribution & Logistics Management, Vol. 32 No. 7, 2002, Decisions are assumed to be made at two increasing specificity sources and levels of Abstracts and keywords. different levels in the semi-integrated variance in the supply chain. As the ability to # MCB UP Limited, 0960-0035

IJPDLM control specified variances increase, the need requirements. Given the diversity of firms for agility decreases. represented, we conclude that the way in 32,7 which these firms compete may also have a A multi-dimensional empirical direct influence on the extent of IT investment exploration of technology investment, and competencies. coordination and firm performance Anthony Ross Improving electronics manufacturing 502 Keywords Co-ordination, Performance, supply chain agility through outsourcing Technology Scott J. Mason, Michael H. Cole, As economic activities span the supply chain Brian T. Ulrey and Li Yan boundary, the effective use of technology as the medium for coordination (or integration) Keywords Supply chain, Agile production, among and within organizations has received Outsourcing, Electronics much attention. In the US manufacturing The highly competitive electronics sector, IT usage is increasingly becoming a manufacturing marketplace demands that source of sustained competitiveness and an suppliers provide low-cost, high-quality opportunity for improvement. And there is a products to their customers in a timely growing demand to achieve conflicting fashion. Shortened product life cycles and performance objectives (revenue versus increasingly global competition have caused profitability versus efficiency, for example). traditional manufacturers to focus on their This article explores the relationships company core competencies, such as product between information technology investment, design and development, choosing to performance, and productivity. While outsource the actual manufacturing of their management should continue to evaluate IT products to contract manufacturers. Although investments by any practical means that the decision to outsource can have both satisfies company needs, the development of positive and adverse effects on key areas of IT competencies and investment policies so as the manufacturing supply chain, one positive to optimize the firm’s performance seems to effect is that the manufacturer’s supply chain be a worthwhile goal. Our empirical findings agility is increased. Outsourcing has caused clearly suggest that IT investment has a an increase in the amount of information that positive impact on market performance as a is shared between supply chain partners. As a result of better coordination in the value result, a greater reliance on suppliers and chain, but that larger investments do not seem alliance partners has become essential for to lead to higher financial performance. company survival. We examine the ways in Additionally, coordination productivity which contract manufacturing has increased seems to benefit from increased investment the agility of the electronics manufacturing by reducing, say, working capital supply chain.

French abstracts French abstracts ´ Programme de recherche pour la logistique du commerce electronique, fonde sur ´ l’opinion professionnelle Jaana Auramo, Anna Aminoff et Mikko Punakivi ´ Mots-cles Logistique, Gestion de la chaıne d’approvisionnement, Recherche et developpement, ˆ ´ Approvisionnement, Reseaux, Mise au point de nouveaux produits ´ 503 L’etude que voici cherchait a determiner quels devraient etre les domaines principaux de recherche ´ ` ´ ˆ et de developpement, en ce qui concerne la logistique du commerce electronique. La methode ´ ´ ´ d’etude comprenait cinq elements: la creation d’une vision preliminaire de la logistique ´ ´´ ´ ´ electronique; des interviews aupres d’experts en matiere de commerce electronique et de ´ ` ` ´ logistique; l’identification et la categorisation des questions ayant trait a la recherche et au ´ ` developpement; un seminaire avec debats afin d’analyser, d’un oeil critique, les resultats ´ ´ ´ ´ preliminaires et d’etablir un ordre de priorite pour les questions de recherche et de developpment; et ´ ´ ´ ´ la formulation d’un programme de recherche permettant d’orienter les travaux de recherche futurs dans le domaine de la logistique du commerce electronique. Le procede qui se composait de deux ´ ´ ´ phases, des interviews suivis du seminaire, permit d’evaluer les resultats preliminaires et de leur ´ ´ ´ ´ accorder un ordre de priorite. Les besoins du commerce electronique sur l’infrastructure logistique ´ ´ nationale furent determines comme etant l’un des domaines de recherche principaux. Selon l’etude, ´ ´ ´ ´ une structure integree de reseaux d’approvisionnement, avec une visibilite et une utilisation du ´ ´ ´ ´ transfert des donnees en temps reel appropriees, constitue un autre domaine de grande importance. ´ ´ ´ Il conviendrait egalement d’encourager la recherche et le developpement de nouvelles notions de ´ ´ service logistique, ainsi que des recherches sur les effets et les possibilites de l’utilisation de la ´ gestion des donnees sur les nouveaux produits et des methodes d’identification des produits. ´ ´ ´ ´ ´ ´ Une etude comparee de trois methodes differentes de gestion de la chaıne ˆ d’approvisionnement R. Meenakshi Sundaram et Sameer G. Mehta ´ Mots-cles Gestion de la chaıne d’approvisionnement, Logistique, Distribution, ˆ Prise de decisions, Integration ´ ´ L’article donne une etude comparee de trois methodes differentes portant sur un modele ´ ´ ´ ´ ` hypothetique de chaıne d’approvisionnement. Les trois methodes examinees sont la methode ´ ˆ ´ ´ ´ independante, la methode semi-integree, et la methode integree. La methode independante suppose ´ ´ ´ ´ ´ ´ ´ ´ ´ que les decisions sont prises independamment, a trois niveaux differents. La methode semi-integree ´ ´ ` ´ ´ ´ ´ suppose que les decisions sont prises a deux niveaux differents. La methode integree suppose que ´ ` ´ ´ ´ ´ les decisions sont prises a un seul niveau. ´ ` ´ ´ ´ Ameliorer l’efficacite de la gestion des materiaux: un pas vers une entreprise agile M. Caridi et R. Cigolini ´ Mots-cles Gestion de la demande, Controle de l’inventaire, ˆ Planification des besoins en materiaux, Stocks de securite, Systemes de fabrication, ´ ´ ´ ` Simulation par ordinateur La recherche que voici donne un examen des publications existantes dans le domaine des methodes ´ de decouragement de l’incertitude pour les systemes de fabrication; elle propose un nouveau modele ´ ` ` permettant d’ameliorer l’efficacite de la gestion des materiaux dans les environnements de ´ ´ ´ planification des besoins en materiaux. L’examen des publications existantes produit une structure ´ de classification des modeles qui suit neuf dimensions structurelles se rapportant au traitement ` d’amortissement pour securite, aux caracteristiques presentes dans l’environnement et au type de ´ ´ ´ ´ International Journal of Physical methode. Le modele propose, fonde sur la structure de classification, offre des directives permettant ´ ` ´ ´ Distribution & Logistics d’aborder le probleme du dimensionnement, du positionnement et de la gestion des stocks de ` Management, Vol. 32 No. 7, 2002, securite face a l’incertitude de la demande. Nous avons mis l’efficacite du modele propose a l’essai ´ ´ ` ´ ` ´` French abstracts. en le comparant a la methode traditionnelle, en nous servant de la simulation par ordinateur. ` ´ # MCB UP Limited, 0960-0035

IJPDLM ´ ˆ Mesurer l’agilite de la chaıne d’approvisionnement dans l’organisation virtuelle Mary Margaret Weber 32,7 ´ Mots-cles Production agile, Souplesse, Variantes de la chaıne d’approvisionnement, ˆ Part du marche, Mercatique directe ´ Le besoin de donner une reponse de plus en plus souple a la demande du marche entraıne un interet ´ ` ´ ˆ ´ ˆ accru dans les organisations virtuelles ou agiles. Cependant, la reponse agile dans la chaıne ´ ˆ d’approvisionnement n’est peut-etre pas toujours necessaire et elle ne represente pas toujours une ˆ ´ ´ 504 meilleure solution de remplacement aux structures organisationnelles traditionnelles. Le modele ` propose dans l’article que voici est un moyen permettant de mesurer, non seulement le besoin en ´ agilite, mais aussi l’agilite reelle d’une organisation. Pour ce faire, on se sert d’un modele ´ ´ ´ ` hierarchique qui detaille de maniere de plus en plus specifique les sources et niveaux de variance a ´ ´ ` ´ ` l’interieur de la chaıne d’approvisionnement. Tandis que la capacite de controler des variances ´ ˆ ´ ˆ specifiques augmente, le besoin en agilite diminue. ´ ´ Une exploration empirique multidimensionnelle des investissements technologiques, de la coordination et de la performance de l’entreprise Anthony Ross ´ Mots-cles Coordination, Performance, Technologie Comme les activites economiques recouvrent l’ensemble de la chaıne d’approvisionnement, ´ ´ ˆ l’utilisation efficace de la technologie comme moyen de coordination (ou d’integration) aupres des ´ ` organisations et au sein de celles-ci s’est vu accorder beaucoup d’attention. Dans le secteur americain de la fabrication, l’utilisation de l’informatique represente de plus en plus une source de ´ ´ competitivite durable et une possibilite d’amelioration. Et le besoin de realiser des objectifs de ´ ´ ´ ´ ´ performance en conflit (les revenus face a la profitabilite face a l’efficacite, par exemple) se fait de ` ´ ` ´ plus en plus ressentir. L’article que voici explore les rapports qui existent entre les investissements en informatique, la performance et la productivite. Tandis que la direction devrait continuer a ´ ` evaluer les investissements en informatique par tous les moyens pratiques possibles qui permettent ´ de satisfaire les besoins de l’entreprise, le developpement des competences en informatique et des ´ ´ politiques d’investissement permettant d’optimaliser la performance de la firme semblent representer un but digne d’etre poursuivi. Nos decouvertes empiriques suggerent clairement que ´ ˆ ´ ` les investissements en informatique ont un impact positif sur la performance sur le marche, ´ provenant d’une meilleure coordination a l’interieur de la chaıne de valeur, mais que des ` ´ ˆ investissements accrus ne semblent pas entraıner une performance financiere superieure. De plus, ˆ ` ´ la productivite de coordination semble profiter d’une augmentation des investissements en ´ reduisant, par exemple, les besoins en capital d’exploitation. Etant donne la diversite des firmes ´ ´ ´ representees, nous arrivons a la conclusion que la maniere dont ces firmes rivalisent peut aussi ´ ´ ` ` avoir une influence directe sur l’ampleur des investissements et des competences en informatique. ´ ´ ´ ˆ Ameliorer l’agilite de la chaıne d’approvisionnement dans l’industrie de fabrication des pie `ces electroniques, au moyen de l’externalisation ´ Scott. J. Mason, Michael H. Cole et Brian T. Ulrey ´ Mots-cles Chaıne d’approvisionnement, Production agile, Travail a forfait, Electronique ˆ ` L’industrie de fabrication des pieces electroniques, qui est extremement competitive, requiert des ` ´ ˆ ´ fournisseurs qu’ils offrent a leurs clients des produits de qualite superieure, a un prix reduit et dans ` ´ ´ ` ´ les delais fixes. La reduction du cycle de vie des produits et l’augmentation de la concurrence au ´ ´ ´ niveau mondial ont force les fabricants traditionnels a se concentrer sur les competences essentielles ´ ` ´ de leur entreprise, comme par exemple la conception et la mise au point des produits et le choix d’externaliser la fabrication effective de leurs produits vers des fabricants travaillant a forfait. Bien ` que la decision d’externaliser puisse avoir des effets positifs aussi bien que nefastes sur les domaines- ´ ´ cles de la chaıne d’approvisionnement dans le secteur de la fabrication, l’un des effets positifs est que ´ ˆ l’agilite de la chaıne d’approvisionnement du fabricant s’en voit amelioree. L’externalisation a ´ ˆ ´ ´ entraıne une augmentation du nombre des informations qui sont partagees entre les differents ˆ ´ ´ ´ partenaires dans la chaıne d’approvisionnement. En consequence, une dependance accrue vis-a-vis ˆ ´ ´ ` des fournisseurs et des partenaires de l’alliance est devenue essentielle a la survie de l’entreprise. ` Nous examinons les diverses manieres dont la fabrication a forfait a permis d’augmenter l’agilite de ` ` ´ la chaıne d’approvisionnement pour le secteur de la fabrication electronique. ˆ ´

Spanish abstracts Spanish abstracts ´ ´ ´ Programa de investigacion para la logıstica del negocio electronico basado en opiniones profesionales Jaana Auramo, Anna Aminoff y Mikko Punakivi Palabras clave Logıstica, Gestion de la cadena de suministro, I y D, Suministro, Redes, ´ ´ Desarrollo de productos nuevos 505 Este estudio investiga cuales deberıan ser las principales areas de investigacion y desarrollo con ´ ´ ´ ´ respecto a la logıstica del negocio electronico. El metodo de estudio consistio en cinco elementos: ´ ´ ´ ´ la creacion de una vision preliminar de la logıstica electronica; entrevistas enfocadas con ´ ´ ´ ´ expertos en el negocio electronico y la logıstica; identificacion y categorizacion de los temas ´ ´ ´ ´ clave de I y D; un taller de panel para analizar crıticamente los descubrimientos preliminares y ´ establecer prioridades entre los temas de I y D; y, elaboracion de un programa de investigacion ´ ´ para guiar trabajos futuros de investigacion en el campo de la logıstica del negocio electronico. ´ ´ ´ El proceso de dos fases, donde las entrevistas se continuaron con el taller, facilitaron la evaluacion y establecimiento de prioridades entre los descubrimientos preliminares. Se ´ descubrio que los requisitos del negocio electronico en cuanto a infraestructura logıstica ´ ´ ´ nacional constituıan una de las areas basicas de investigacion. De acuerdo al estudio, una ´ ´ ´ ´ estructura en red integrada de suministro con visibilidad y uso adecuados de transferencia de datos en tiempo real es otra area de gran importancia. Tambien deberıa promocionarse la ´ ´ ´ investigacion y el desarrollo de nuevos conceptos de servicio logıstico, ası como investigacion ´ ´ ´ ´ sobre los efectos y posibilidades de utilizar la gestion de datos de nuevos productos y metodos ´ ´ de identificacion de productos. ´ ´ Un estudio comparativo de tres planteamientos de SCM (gestion de la cadena de suministro) diferentes R. Meenakshi Sundaram y Sameer G. Mehta Palabras clave Gestion de la cadena de suministro, Logıstica, Distribucion, ´ ´ ´ Toma de decisiones, Integracion´ Se presenta un estudio comparativo de tres planteamientos diferentes de un modelo hipotetico ´ de cadena de suministro. Los tres planteamientos investigados son: independiente, semi- integrado e integrado. En el planteamiento independiente se supone que las decisiones se toman independientemente en tres niveles distintos. En el semi-integrado se supone que las decisiones se toman en dos niveles distintos. En el nivel integrado se supone que todas las decisiones se toman en un solo nivel. ´ Mejora de la eficacia de la gestion de materiales: un paso hacia el negocio agil ´ M. Caridi y R. Cigolini Palabras clave Gestion de demanda, Control de inventario, ´ MRP (ingreso del producto marginal), Existencias de seguridad, Sistemas de fabricacion, ´ Simulacion por ordenador ´ Esta investigacion proporciona una revision de bibliografıa en el campo de los metodos para ´ ´ ´ ´ amortiguar la incertidumbre, para los sistemas de fabricacion, y propone un nuevo modelo para ´ mejorar la eficacia de la gestion de materiales dentro de los ambitos de planificacion de ´ ´ ´ requisitos de materiales. La revision de bibliografıa suscita un marco de clasificacion de los ´ ´ ´ modelos en base a nueve dimensiones estructurales que hacen referencia al tratamiento intermedio de la seguridad, las caracterısticas medioambientales y el tipo de planteamiento. En ´ International Journal of Physical base al marco de clasificacion, el modelo propuesto proporciona directrices para plantearse el ´ Distribution & Logistics problema de dimensiones, posicionamiento y gestion de las existencias de seguridad contra la ´ Management, Vol. 32 No. 7, 2002, incertidumbre de la demanda. La eficacia del modelo propuesto se ha ensayado comparandola ´ Spanish abstracts. con el planteamiento tradicional, a traves de una simulacion basada en ordenador. ´ ´ # MCB UP Limited, 0960-0035

IJPDLM ´ Medicion de la agilidad de la cadena de suministro en la organizacion virtual ´ 32,7 Mary Margaret Weber Palabras clave Produccion agil, Flexibilidad, Variantes de la cadena de suministro, ´ ´ Cuota de mercado, Marketing directo La necesidad de una mayor flexibilidad para responder a la demanda del mercado esta ´ encabezando un interes intensificado en las organizaciones virtuales o agiles. No obstante, ´ ´ puede que una respuesta agil en la cadena de suministro no siempre sea necesaria y que no ´ 506 represente siempre una mejor alternativa que otras estructuras organizacionales mas ´ tradicionales. El modelo propuesto en este trabajo ofrece un medio para medir la necesidad de agilidad y hasta que punto una organizacion es realmente agil. Esto se logra a traves del uso ´ ´ ´ ´ de un modelo jerarquico que ofrezca detalle con fuentes de especificacion creciente y niveles de ´ ´ varianza en la cadena de suministro. A medida que aumenta la habilidad para controlar varianzas especıficas, disminuye la necesidad de agilidad. ´ ´ ´ Una exploracion empırica multidimensional de la inversion en tecnologıa, ´ ´ ´ coordinacion y rendimiento de empresa Anthony Ross Palabras clave Coordinacion, Rendimiento, Tecnologıa ´ ´ A medida que las actividades economicas sobrepasan las barreras de la cadena de suministro, el ´ uso eficaz de la tecnologıa como medio de coordinacion (o integracion) entre y dentro de las ´ ´ ´ organizaciones, ha recibido mucha atencion. En el sector de fabricacion de los EE UU, el empleo ´ ´ de la informatica se esta convirtiendo cada vez mas en una fuente de competitividad sostenida, y ´ ´ ´ en una oportunidad para la mejora. Y, existe una demanda creciente para lograr objetivos conflictivos de rendimiento (por ejemplo, ingresos contra rentabilidad contra eficiencia). Este artıculo explora las relaciones entre la inversion en informatica, el rendimiento y la productividad. ´ ´ ´ Mientras que la direccion deberıa seguir evaluando las inversiones en informatica mediante ´ ´ ´ cualquier medio practico que satisfaga las necesidades de la empresa, el desarrollo de ´ competencias en informatica y de polıticas de inversion para optimizar el rendimiento de una ´ ´ ´ empresa parecen ser un objetivo merecedor de atencion. Nuestros descubrimientos empıricos ´ ´ sugieren claramente que la inversion en informatica tiene una influencia positiva sobre el ´ ´ rendimiento del mercado, como resultado de una mejor coordinacion en la cadena de valor, pero ´ que las inversiones mas grandes no parecen conducir a un mayor rendimiento financiero. ´ Asimismo, la productividad en coordinacion parece beneficiarse de una mayor inversion mediante ´ ´ la reduccion, por ejemplo, de los requisitos de capital circulante. Dada la diversidad de las ´ empresas representadas, concluimos que la forma en que estas empresas compiten tambien podrıa ´ ´ tener una influencia directa sobre la extension de las competencias y la inversion en informatica. ´ ´ ´ ´ Mejora de la agilidad de la cadena de suministro en la fabricacion de productos ´ ´ electronicos a traves del empleo de recursos externos Scott J. Mason, Michael H. Cole y Brian T. Ulrey Palabras clave Cadena de suministro, Produccion agil, Trabajo por contrato, Electronica ´ ´ ´ El altamente competitivo mercado de la fabricacion de productos electronicos exige que los ´ ´ proveedores ofrezcan productos de bajo coste y alta calidad a sus clientes de manera oportuna. Ciclos de vida del producto mas cortos y una competencia mundial creciente han hecho que los ´ fabricantes tradicionales se centren en sus competencias empresariales basicas, tales como el ´ diseno y desarrollo de productos, decidiendo encargar la fabricacion real de sus productos a ˜ ´ fabricantes externos contratados. Aunque la decision de utilizar fuentes externas puede tener ´ efectos tanto positivos como negativos sobre areas clave de la cadena de suministro de la ´ fabricacion, un efecto positivo es que se incrementa la agilidad de la cadena de suministro del ´ fabricante. El empleo de fuentes externas ha provocado un aumento de la cantidad de informacion´ que se comparte entre socios de la cadena de suministro. Como resultado, se ha hecho esencial una mayor dependencia en los proveedores y los socios en alianza para la supervivencia de las empresas. Examinamos las formas en que la fabricacion por contrato ha incrementado la agilidad ´ de la cadena de suministro dentro de la fabricacion de productos electronicos. ´ ´

Japanese abstracts Japanese abstracts 507 International Journal of Physical Distribution & Logistics Management, Vol. 32 No. 7, 2002, Japanese abstracts. # MCB UP Limited, 0960-0035

IJPDLM 32,7 508

Japanese abstracts 509

IJPDLM 32,7 Guest editorial Guest Editors Laura M. Meade graduated with a BSME from Valparaiso University. She later attended University of Texas at Arlington where she received both an MBA and a PhD in Industrial Engineering. Her work experience includes General Dynamics and the Automation and Robotics Research Institute. At present she is an Assistant Professor at the University of Dallas, 510 Graduate School of Management. She has published several articles in the area of enterprise modeling, performance management, supply chain management and logistics. She is a member of the American Production and Inventory Control Society (APICS), Council of Logistics (CLM) and Institute of Industrial Engineers (IEE). Her current research interests include reverse logistics, and sustainability in supply chain management. Joseph Sarkis is currently a Professor in The Graduate School of Management at Clark University. He earned his PhD from the State University of New York at Buffalo. His research interests include supply chain management and management of technology with a specific emphasis on performance management, justification issues, enterprise modeling and environmentally conscious operations and logistics. He has published over 140 articles in a number of peer reviewed academic journals, conferences and edited books. He is a member of the Decision Sciences Institute, APICS, the International Society for Industrial Ecology and INFORMS. Srinivas Talluri received the BS degree in mechanical engineering in India in 1989 and the MS degree in industrial and manufacturing systems engineering and the PhD degree in production and operations management from the University of Texas at Arlington in 1992 and 1996, respectively. He is an Associate Professor of Operations Management at Michigan State University, East Lansing, Michigan. His research interests are in the areas of purchasing/supply chain management, technology management, business process improvement, and multi-criteria decision modeling. His research is published in a variety of academic journals including the International Journal of Production Research, European Journal of Operational Research, IEEE Transactions on Engineering Management, International Journal of Production Economics, International Journal of Flexible Manufacturing Systems, International Journal of Operations and Production Management, Computers and Industrial Engineering: An International Journal among other journals, and in various national and international conferences. Dr Talluri is a member of the Decision Sciences Institute and the Institute for Operations Research and Management Sciences. Today’s busness and technical environment is characterized by rapid, unanticipated, and often dramatic change. In this dynamic and uncertain environment, the continued competitiveness of the international logistics community depends on the ability to respond effectively and efficiently, i.e. to be agile. Agile enterprises are focused organizations that combine visionary management, highly motivated workforces, effective processes, and agile technology into an integrated system capable of rapidly responding to market demands. Among the more important managerial and organizational issues facing companies in the new competitive environment is the formation of the virtual enterprise; a temporary alliance of companies, each providing a core competency, to take advantage of a market opportunity. Participation in a virtual enterprise requires the utmost in agility. Participation in a world where quick response, e-commerce, time-based competition, mass customization, and International Journal of Physical Distribution & Logistics shortened product lifecycles all call for organizational agility. Management, Vol. 32 No. 7, 2002, Logistics plays an increasingly important strategic role for organizations pp. 510-512. # MCB UP Limited, 0960-0035 that strive to keep pace with market changes and supply chain integration. The

logistics environment has undergone several transitions, from regulatory Guest editorial deregulation, third party outsourcing, supply chain emphasis, and now the newest paradigm of electronic logistics. The Internet has provided immense opportunities for the existence of virtual organizations. An efficient logistics system is a necessity for companies to respond to unanticipated change and the supply chain and its management, itself can be considered a core competency and thus make a contribution to the virtual enterprise. This field of knowledge 511 needs to be expanded and a facilitation of this expanding knowledge is through our special issue of the International Journal of Physical Distribution and Logistics Management. We feel you are in for a treat as you read the papers included in this issue. The papers include a variety of conceptual, practical, analytical, and empirical methodologies, with each of them helping to improve the design and development of agile supply chains for participation in a virtual organization. The following is a brief preview of what appears in this special issue volume. We begin with Jaana Auramo, Anna Aminoff and Mikko Punakivi: ‘‘Research agenda for e-business logistics based on professional opinions’’. This paper is an informative introductory article for the special issue as it focuses on determining a research agenda for logistics in a virtual enterprise. Using the viewpoint that e-business logistics is a wide-ranging topic related to supply chain integration the authors conducted focus interviews and panel workshops in order to evaluate and prioritize popular e-business considerations. Requirements of e-business on the national logistics infrastructure, integrated supply network structure, new logistics service concepts and new product data management and product identification methods were some of the primary areas that need to be considered for future research. Sundaram and Mehta in their article entitled ‘‘A comparative study of three different SCM approaches’’ highlight the importance of integration across various supply chain processes. They utilized mathematical programming models in evaluating three different supply chain decision-making scenarios that involved independent, semi-integrated, and integrated approaches. In the independent approach, it is assumed that decisions are made independently at three different levels. Decisions are assumed to be made at two different levels in the semi-integrated approach, and all decisions are assumed to be made at a single level in the integrated approach. Their optimization model results suggest that supply chain costs decrease as level of integration increases, thus stressing the need for integrated decisions across the supply chain. The use of simulation to test a novel approach to managing MRP nervousness is the central approach of Maria Caridi and Robert Cigolini’s ‘‘Improving materials management effectiveness: a step towards agile enterprise’’. MRP nervousness adds to the uncertainty of managing the logistics function. To help ability to function in this uncertain environment, as is required by agile systems, they help to dampen it through their recommended approach. Testing of their technique helps to show its advantageous characteristics.

IJPDLM Mary Margaret Weber, ‘‘Measuring supply chain agility in the virtual 32,7 organization’’, addresses the overwhelming question of determining the degree of agility an enterprise needs to perform in order to be successful. This paper proposes a means of measuring the organization’s need for and ability to develop an agile business strategy within the context of the virtual organization. Logistics inventory variances and marketing variances are 512 incorporated into the hierarchical model and the specific sources of variance from the planning to the execution phase are brought to light. The model also addresses the degree of responsiveness to unanticipated change in the marketplace; in other words, how agile was the company’s supply chain? The effective management of supply chain and virtual enterprises is heavily dependent on technology, and especially information technology. Whether or not this intuitive thought is real and practical is investigated by Anthony Ross in his paper: ‘‘A multi-dimensional empirical exploration of technology investment, coordination and firm performance’’. Exploring the multi- dimensional nature of managing agility through various performance metrics, Ross determines whether investments in information technology truly contribute to the bottom line by investigating corporate market and financial performance. In some cases larger investments do lead to better market performance (coordination productivity), yet the financial performance linkage is much more tentative. These measures and the various relationships are evaluated with actual empirical data and the relatively innovative analytical approach of data envelopment analysis. The final paper, entitled ‘‘Improving electronics manufacturing supply chain agility through outsourcing’’ by Mason, Cole, Ulrey and Yan, stresses the importance of outsourcing in improving supply chain agility through a case study relating to the electronics industry. According to them, due to the constantly changing needs of the customer in the electronic industry, companies are responding by primarily focusing on their core competencies and outsourcing other operations. While this approach has both positive and adverse effects on key areas of the manufacturing supply chain, one positive effect is that the manufacturer’s supply chain agility is increased. They examined the ways in which contract manufacturing has increased the agility of the electronic manufacturing supply chain. Specifically, they discuss the impact of outsourcing on facility location, customer service and product distribution. Laura Meade Joseph Sarkis Srinivas Talluri

The research register for this journal is available at The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/researchregisters http://www.emeraldinsight.com/0960-0035.htm Research agenda for Research agenda for e-business e-business logistics based on logistics professional opinions Jaana Auramo 513 Department of Industrial Engineering and Management, Helsinki Received May 2001 University of Technology, Finland Revised February 2002 Accepted April 2002 Anna Aminoff Industrial Systems, VTT Technical Research Centre of Finland, Finland, and Mikko Punakivi Department of Industrial Engineering and Management, Helsinki University of Technology, Finland Keywords Logistics, Supply chain management, R&D, Supply, Networks, New product development Abstract This study investigated what should be the major research and development areas regarding the logistics of electronic business. The method of study consisted of five elements: the creation of a preliminary e-logistics vision; focus interviews of e-business and logistics experts, identification and categorisation of the key R&D topics, a panel workshop to critically analyse the preliminary findings and prioritise the R&D topics, and formulation of a research agenda to guide future research work in the field of e-business logistics. The two-phased process, where the interviews were followed by the workshop, enabled the evaluation and prioritisation of the preliminary findings. The requirements of e-business on the national logistics infrastructure were found to be one of the focus research areas. According to the study, integrated supply network structure with suitable visibility and usage of real-time data transfer is another area of great importance. Research and development of new logistics service concepts should also be promoted as well as research on the effects and possibilities of using new product data management and product identification methods. Introduction With the growth in new information and communication technologies, managing the flow of information has become as vital as managing the flow of material (Demkes et al., 1999). It is now possible to separate the information flow from the physical material flow if necessary. Furthermore the Internet makes it possible to connect all of those working within a single supply chain, which creates new challenges for enterprises. New technologies are used to try to optimise the supply chain, with the final goal of planning all the enterprise’s activities in as close a relationship to the demands of clients as possible. The outcome of the process should be profitable growth (Ranagan and Adner, 2001). International Journal of Physical Appreciation is extended to the National Technology Agency of Finland for support of this Distribution & Logistics project. The authors wish to thank Jan Holmstrom, Eila Jarvenpaa, Juha Luoma and John ¨ ¨ ¨¨ Management, Vol. 32 No. 7, 2002, pp. 513-531. # MCB UP Limited, Sillincer for their helpful suggestions on the earlier drafts of this paper and Hanna Pajunen- 0960-0035 Muhonen, Jarkko Lehtinen and Hannu Yrjola for active participation during the project. ¨¨ DOI 10.1108/09600030210442568

IJPDLM The Trilog-Europe project (Demkes et al., 1999) states that the Internet can 32,7 be seen as a supply chain management concept as well as a trading platform. Most concepts for improving supply chain efficiency are based on sharing information throughout the supply chain in order to improve transparency. It is clear from a functional perspective that there are benefits to the supply chain as a whole but the distribution of the benefits remains a difficult task. 514 It is difficult to define e-business logistics comprehensively because the potential impact of e-business on logistics and supply chain management is not yet fully understood. One possible definition is that e-business logistics simply means processes necessary to transfer the goods sold over the Internet to the customers. On the other hand it is also possible to define e-business logistics as a wide-ranging topic related to supply chain integration, which is the viewpoint taken in this paper. Many companies are struggling with the question of how e-business impacts the supply chain. It could have the effect of eliminating some intermediaries (such as the wholesaler or retailers), but it also fosters the emergence of new players like logisticians, whose role is to adapt traditional logistics chains to take into account the requirements of e-business. Many companies are also struggling with internal change management processes when trying to make employees adapt to electronic business. Two recent studies (CEST, 2000; Ferrari, 2000) have listed issues that should be addressed regarding e-business logistics: . Where is the value proposition of e-business relative to the supply chain? Can e-business make the supply chain experience more valuable? . What will e-business do to the supply chains? . What are the uses for e-business in transportation and logistics specifically? . Is there benchmarking information available on how to re-engineer for e- business? . How does a company match its current operations to the e-future? . How to harness information and link it to the company’s operations? . How will the dynamics of B2B (business to business) and B2C (business to consumer) supply change? . What new, valued supply chain ‘‘e’’-services are possible? . Who are the right partners? . How will different industry sectors look when e-business is the norm? As we can see based on these selected references, the questions regarding the e-business logistics and supply chain management cover many overlapping themes. It is not always easy to see what is the interrelation of these different topics. The questions can be divided based on how concrete or abstract they are by nature. There are some very concrete (but still not simple) issues, like the

home delivery concepts related to e-commerce operations (Punakivi and Research agenda Saranen, 2001) and a great variety of very abstract and large issues like the for e-business potential changes in the value chains of different industries that need to be logistics studied further. Some of the questions can be considered as prerequisites for any future development and it is very important to realise what they are and quickly start initiatives to solve them. The overall picture of e-business logistics needs categorising into different 515 research and development themes. There also seem to be different opinions whether the IC technology is ready to redeem the vision of the e-business (Supply Chain Decisions, 2001). Guidelines are needed so that the resources can be allocated to the areas in the field of logistics and supply chain management, which mostly support development of profitable electronic business. Figure 1 illustrates how grouping and prioritisation of different R&D questions could be organised so that high quality research agendas can be formulated to guide the work (Auramo et al., 2001). This paper identifies, categorises and prioritises the most crucial research and development topics that should be dealt with in order to help enterprises to face the new logistics challenges of e-business. The study included an iterative process, where focus interviews of professionals were followed by the workshop. This iterative process enabled respondents’ consensus views to be summarised and evaluated. The categorisation and prioritisation of different but often interrelated e-business issues help people to see the overall picture of e-business logistics more clearly. However, there was a need to somehow pinpoint a few larger research themes to guide the future R&D work in research institutes and companies. Thus, as the final result of this study, we came forward with a proposal for the research agenda with four themes to support the development of e-business logistics. Figure 1. Categorisation and prioritisation of R&D topics to formulate research agendas

IJPDLM Method 32,7 Selected method The method included aspects of the Delphi method (Linstone, 1975), which employs an iterative process of summarising and evaluating the respondents’ views on a consensus view (McKinnon and Forster, 2000). The method consists of focus interviews, to be followed by a workshop. Ideally, the people 516 interviewed should also attend the workshop. This kind of two-phased process enables the evaluation and prioritisation of the preliminary findings. Face-to-face conversations foster interactive communication, which is a precondition for knowledge creation and new innovations (Nonaka and Takeuchi, 1995). Previous research has identified the task-orientated, interaction-centred focus groups (workshops) as an ideal methodology for exploring professionals’ experiences and for describing that expedience. These professionals are often in a position to influence future events and, thus, make their forecasts come true (Belzowski, 2000). Application of the method The study consisted of five stages as illustrated in Figure 2. First, the preliminary E-Logistics Vision was created based on the literature study to be used as a discussion guideline during the interviews (Stage 1). A total of 50 focus interview sessions were organised, with 65 people interviewed in total (Stage 2). The interviews can be divided into four categories: the leading edge companies (trade and manufacturing companies that have a reputation of being early adapters of new business trends), logistics service companies, consultants and IT companies and researchers at universities and research centres (Table I). The objective of the interviews was to determine the key research and development topics in the field of e-business logistics. The preliminary vision was used as a discussion guideline, which became more defined during the interview process. However, the interview sessions were kept unstructured and Figure 2. The stages of the study

open to promote the free flow of ideas. The interviewees from industry, trade Research agenda and service sectors were responsible for the development of e-business in their for e-business companies. Researchers and other professionals in the fields of logistics, supply logistics chain management, e-business and future technologies were also interviewed. In Stage 3, the data from the interviews was analysed. The key R&D topics identified during the interview phase were listed. Similar topics were divided into eight preliminary categories, which were formulated during the process. 517 Preliminary categories and respective R&D topics were incorporated into three final categories. In Stage 4, a workshop panel was organised and the findings from the interviews were prioritised, further evaluated and specified. The workshop was attended by a total of 45 people of which 30 were the same previously interviewed. The question: ‘‘What should be the key research and development topics in the field of e-business logistics?’’ was thoroughly discussed in five parallel groups. The group discussions were semi-structured based on categorised and listed topics. First, the categorisation was presented to the workshop and participants’ approval was sought. Second, the participants were encouraged to identify the most important research and development topics and prioritise them within each category in a consensus view. During the interview and workshop process a total of 80 professional opinions were analysed. The objective of Stage 5 was to formulate a research and development agenda to guide future research work in the field of e-business logistics. The presented themes of the research agenda are combinations of prioritised R&D topics. The key research and development areas The framework When analysing the results of the interviews and workshop discussions, we realised that the company’s level of integration in the supply network influenced how well the complexity of different interrelated e-business issues were understood. This formed the framework that helped us in the analysis processes (Figure 3). Companies Experts Industry 15 21 Trade 6 10 Logistics service companies 9 13 Consultants, IT companies 6 6 Table I. The background Universities, research centres 10 15 information of Total 46 65 interview data

IJPDLM 32,7 518 Figure 3. The level of integration in the supply network Figure 3 illustrates the development steps when moving from the traditional business practices towards the integrated supply and demand network. Most of the interviewed companies were located between the first and the second step or on the second step of the development curve. The companies have problems with the internal integration of information, as well as with integration with their suppliers and customers. Only very few companies were able to see what a fully developed e-collaboration could mean to their business. Many interviewees paid a lot of attention to the readiness of individual companies to collaborate in the electronic supply network. Issues related to information management and internal capabilities of connecting to the other participants of the network were seen as critical prerequisites. However, it was concluded in the workshop discussions that they should be omitted from this study because the companies themselves are responsible for developing them. As a result of the interview process it was concluded that the research and development topics in the field of e-business can be divided into three levels: strategic level; business level; and technology level (Tang, 2000). The prerequisites for the development of e-business and related logistics are created on the strategic level. The business level issues relate to the supply chain management processes that support the implementation of the selected strategies. The technology level deals with the actual order-delivery process. Categorisation and prioritisation of R&D topics After a thorough analysis of the interview results the R&D topics were divided into three categories as described in stage 3 in the methodology section. This division was presented to the workshop participants and their approval was sought. The three categories were: (1) Information flow. The research and development issues deal with the transparency of the information and its real-time distribution to the necessary parties in the supply-demand network.

(2) Supply network. Research and development topics related to potential Research agenda changes in value chains and new collaboration possibilities within for e-business networked companies. logistics (3) Physical material flow. Utilisation of new technologies and their impact on distribution and logistical infrastructure, consolidation possibilities and emerging home delivery concepts. 519 The R&D topics were prioritised within each of the three categories in the workshop. The issues were discussed in five parallel sessions. The participants of each session were encouraged to identify the most important topics and to achieve a consensus view. The prioritisation of topics is presented in Tables II- IV. The numbers 0-5 in the columns headed ‘‘Importance’’ describe how many of the groups identified the topic as ‘‘very important’’. Each of three categories will be covered in more detail. Information flow Information and communication technology have played a key role in transforming logistics and supply chain planning (Demkes et al., 1999). Increased visibility and the transfer of real-time information bring new effectiveness to supply chain management. For e-business to grow as fast as many expect it to, The R&D topics Importance (Æ) Development of standards and interpreters 5 Integrated ERP (enterprise resource planning) – systems in the supply networks 4 PDM (product data management) 2 Development of POS (point of sales) – and forecasting data Table II. management 2 R&D topics related to the information flow Information management in the supply network: responsibilities, and their importance correctness and rights 2 according to the Centralised vs decentralised data management 1 workshop The R&D topics Importance (Æ) Changes in value chain 4 Evaluation and development of different e-business models 3 Modelling and optimising of the supply network 3 Table III. Management of outsourced resources 3 R&D topics related to Logistics networks, 4 plTM 2 the supply network and their importance Management of channel conflict 2 according to the Responsibilities and role in the e-market places 2 workshop

IJPDLM The R&D topics Importance (Æ) 32,7 National logistics infrastructure and requirements of e-business to it 4 Home delivery 3 Management of regional differences 3 520 Identification technologies 4 Mobile technologies 3 Consolidation of material flows, cross-docking 2 Improvement of order fulfilment accuracy 4 Service level variation 1 Table IV. R&D topics related to Physical logistics of e-market places 1 physical material flow Material handling technologies 1 and their importance according to the Positioning technologies 1 workshop Reverse logistics, direct deliveries 0 trading partners must be able to more easily exchange data between their back- end databases and core business applications (Webster, 2001). The most important R&D topics regarding the information flow within the supply network according to the workshop are presented in Table II. The numbers in the column headed ‘‘Important’’ describe how many of the five groups identified the topic as ‘‘very important’’. Five groups found development of standards and interpreters as a very important R&D topic. During the interviews and workshop there was a lot of discussion about the transfer of information between company interfaces. There are different standards and interpreters being developed at the moment to overcome this problem (Cox, 2001; Webster, 2001). Standards-based Web services have the potential to solve the issue, but for most companies this is still a technology of the future. There were also differing opinions about the role of electronic data interchange (EDI) in the development of Internet based integration. It should, however, be kept in mind that the processes are being integrated across company boundaries and the IT development has just a supporting role. It is very challenging to try to integrate two different companies. Only after the integration can be optimised over the entire supply net can the full potential of new value-added integration be reached (Demkes et al., 1999). During the iterative process of interviews and workshops the information transfer within the supply chain or in the future within the supply net was seen as one of the key research and development areas. Here are a few comments from the workshop regarding the information transfer across the company interfaces: . ‘‘Standards are seen as the key to solve the problem’’. . ‘‘Modular information systems need to be able to communicate with each other’’.

. ‘‘The business processes need standards, not only the information Research agenda transfer’’. for e-business . ‘‘Service providers can take the role of the standards’’. logistics . ‘‘EDI systems should form a basis for the development’’. . ‘‘Attention should be paid also to the actual utilisation of the information; better to transfer less and ‘focused’ information than 521 provide visibility that cannot be utilised’’. In relation to the development of standards for information transfer, companies need to study and understand the new role of enterprise resource planning (ERP) systems in the networked environment. ERP is viewed as the root from which data is pulled into a complex IT organism that links with customers’ and suppliers’ systems, or with clusters of trading partners in Internet hubs. It has been argued that in the new economy having ERP is not enough (Mullin, 2001). ARM Research (Boston) has coined a new term, enterprise commerce management (ECM), to describe multi-vendor IT systems (Mullin, 2001). Simultaneously there are solutions being developed where Web-based browsers allow companies without ERP systems (SME companies mostly) to complete transactions with other companies with ERP systems (D’Amico, 2001). This whole development process, related to information sharing across company interfaces, is a very complicated issue from a single company’s point of view. As discussed earlier, a company’s level of integration in the supply network influenced how well the complexities of different e-business related issues were understood. Thus more research and development resources should be allocated to various fields in this area to help the enterprises fully benefit from emerging solutions, and of course to develop solutions that can foster integration and new innovative ways to collaborate. Product data management (PDM), is one of the key elements when integrating the information flow across the entire supply net (Philpotts, 1996). Two groups identified it as a very important research and development topic. The product and the information related to it must be unambiguously described so that messages can be distributed to the network. Another potential benefit of PDM is the better management of the product portfolio throughout the life cycle of different products. There should be more research and development activities regarding the opportunities of product data management and different technologies available. The development of e-business will create more tightly integrated supply nets where collaboration among different parties will play a key role (CPFR, 2001). According to the visions of e-business, forecasting and demand information is available to all the partners in the supply net in real time and without the Forrester or bullwhip effect, i.e. without the enlargement of demand variability as orders move up the supply chain (Forrester, 1958; Lee et al., 1997). Sharing information and re-thinking responsibilities between companies are

IJPDLM topics receiving a lot of attention in supply chain management at the moment. 32,7 Sharing demand information electronically more frequently or in real-time has become possible and, more recently, less costly (Kiely, 1998). According to the company interviews there is still a lot of work to be done before this vision can be realised. Collaborative planning and forecasting methods need to be developed and one should not forget the changes needed in the management 522 processes when real-time information flows will be visible throughout the supply net. There is also a need for indicators that are capable of measuring the degree of transparency within the supply network. When transferring information within the supply net the questions of who manages and owns the information and who guarantees the correctness of the information become vital issues. During the company interviews a lot of time was spent discussing whether the information management should be centralised or decentralised. A strong opinion was in favour of decentralised information management practices, but there is currently not enough knowledge about what this really means in practice. The results of the prioritisation work done during the workshop focuses more on technological issues related to information transfer. The overall findings of the study suggest that it is at least as important to foster research and development initiatives that help to understand what information should be shared and how to use the shared information to add value to the supply network. Supply network Internet technology has dramatically altered the trade-off between vertical and virtual integration (Shah, 2001). The question, ‘‘Do the advantages of focusing on core competencies and outsourcing the rest outweigh the cost of managing added complexity’’, is very relevant. The R&D topics that help to understand the integration processes and their importance according to the workshop are presented in Table III. The numbers 0-5 describe how many of the groups in the workshop identified the topic as ‘‘very important’’. One of the major challenges of e-business is the potential change in the value chains of different industries. Four groups found it to be a very important research and development focus area. The Internet overturns the old rules about competition and strategy (Werbach, 2000). The changes will vary in different fields and will also depend on the role the companies play in the value chain. However, the value chain will only be changed if there is added value for the supply net. How to identify and measure the added value for the whole supply net or how to optimise the entire value chain were the key research and development areas according to the interviews. The analysis should be done separately for the different industries: electronics industry, forest industry, food industry, consumer goods industry, etc. It is, however, most likely that some general trends will also be revealed. It was further noted that changes in value chains should also be analysed from an individual company’s point of view: how one single company can understand

the changes it will be facing and how it can prepare itself for the new rules of the Research agenda game. Knowledge about different e-business models was seen as important for e-business during the transfer process from the traditional supply chains to the networked logistics business environment. What type of partners will have the key roles in the new value chains and whether there will be differences in the amount of decision- making authority when deciding how the future supply networks operate were also identified as topics of interest. A lot of attention was also paid to the 523 requirements for SMEs to be able to integrate with the new value chains. The channel conflict, which occurs when the traditional sales channels and the new electronic channels are servicing the same client base (Reda, 1999; Hanover, 2000), was identified as an important research topic. Managing outsourced logistics resources was another field to which new research and development initiatives should be directed. The major challenges that affect the logistics infrastructure and the potential needs for new service providers in the networked business environment were also considered as significant topics to be studied further (Bauknight, 2001; Bade and Mueller, 1999; Timmers, 1999). As B2B commerce shifts to the Internet, companies that control the on-line markets will exert enormous influence over the way transactions are carried out, relationships are formed, and how profits flow within the supply net (Kaplan and Sawhney, 2000). The roles and the responsibilities of the electronic marketplaces or electronic hubs as they are also called are not yet fully understood according to the workshop discussions. They were merely seen as trading platforms and not as integrated parts of supply networks. There should be more studies done regarding electronic marketplaces and their potential role as an information intermediary within the supply Web. The main conclusion was that development of IC technology enables business process reengineering and fosters potential changes in the value chains. This creates requirements for new types of logistics services that are supporting new value chains. Research and development initiatives in this area should be focused so that they support the development of integrated and transparent supply network structures. Physical material flow The development of e-business will and has already set increased quality requirements on logistics services and on the logistics infrastructure as a whole. It will be important to be able to provide a different level of service to different customer segments or to individual clients at home. These services must create added value for the clients and at the same time optimise the overall cost structure of the supply net. The management of the physical material flow of the supply net has become one of the key development topics (Thomas and Griffin, 1996) especially in the e-business. Table IV summarises the key research and development topics and their importance according to the workshop. The numbers 0-5 describe how many of the groups in the workshop identified the topic as ‘‘very important’’.

IJPDLM More sophisticated customer demand chains and electronic business pose 32,7 new challenges to supply chain management. Customers are demanding products and deliveries customised to their specific needs and flexible outsourcing of supply chain operations is a necessity in volatile business environments (Karkkainen and Holmstrom, 2001). ¨ ¨ ¨ When moving towards electronic business the material flows are presumed 524 to become narrower in the B2B environment due to increased visibility and the possibility to optimise the inventory levels in the different parts of the supply net. The material flows to the individual consumers in the B2C sector follow the same tendency when more and more goods will be transported directly to households and other delivery points (Witt, 1999). Forrester estimates that along with the growth of the market share, the number of residential deliveries will exceed 2.1 billion by 2003 (Brooksher, 1999). R&D issues related to the supply network in the B2B environment are more globally valid; companies within the same industry sector or parallel players in the supply networks are struggling with similar issues. In the B2C sector, however, more country-specific elements have been reported (Demkes et al., 1999). The logistics challenges when creating the home delivery concepts in the B2C environment are dependent on local infrastructure, the geographical distribution of the end users and the consumer habits rather than the line of business. Thus R&D initiatives need to be allocated to study the national logistics infrastructure and its suitability to meet the needs of e-business, both from B2B and B2C points of view. It is necessary to offer different service levels and flexibility throughout the supply net. New evolving concepts like cross-docking terminals (Daugherty, 1994), direct deliveries, and office and home delivery solutions (Punakivi and Saranen, 2001) are processes that need to be researched and developed. There are also emerging consolidation possibilities, which should be further studied (Hall, 1987; Goolley, 2000). Reverse logistics, the reverse flow of products from customers back to suppliers, is another issue that faces new challenges. There is a trade-off between investing in the process of supporting reverse logistics activities versus investing in the process of prevention to ascertain why goods are coming back. The development of electronic business will increase the importance of delivery accuracy, delivery frequency and delivery time in emerging new supply networks. This applies to logistics both in B2B and B2C segments. Research and development programmes are necessary in order to study how the required delivery patterns can be reached and how to measure their potential benefits. New technologies, like product identification technologies, mobile technologies and applications that utilise satellite location technologies (GPS) are seen as enablers for the new logistical systems that will be necessary to fully benefit from e-business (Jedd, 2000; Shulman, 1999; Radding, 1994). According to the workshop opinion mobile data transfer together with RFID (radio frequency identification) may lead to great efficiency improvements in the logistics network. Use of wireless identification technology could be one of the keys to significantly speed up and increase accuracy in sorting and

distribution (Jones, 1999; Boxall, 2000; Lindstrom, 2000). Wireless product Research agenda ¨ identification is already used with great benefits in the functional areas of for e-business innovative companies and there is potential for its use in supply chain wide logistics solutions, i.e. item level supply chain management (Karkkainen and ¨ ¨ Holmstrom, 2001). But the current knowledge is still insufficient and further ¨ research should be supported in this field. As a conclusion, research and development topics related to physical 525 material flow should be divided into two major themes: technology related R&D to support the development of new logistics services to enable the necessary changes in the value chains; and analysis of potential changes in the national logistics infrastructure to support e-business development. Further research agenda As discussed earlier in this article, many companies are struggling with the question of how e-business impacts their logistics operations and supply chain efficiency. The development processes related to information sharing and visibility across the company interfaces are very complicated issues. Also according to the interviews the emergence of new players in the field of logistics and information services and the potential changes in value chains are great challenges to overcome. The categorisation and prioritisation of these different but often interrelated e-business issues discussed earlier helped us to see the overall picture of e-business logistics more clearly. But there was still a need to somehow pinpoint a few larger research themes to guide the future R&D work in research institutes and companies. As the final result of this study we came forward with a proposal for the research agenda with four themes to support the development of e-business logistics. These research and development themes are combinations of different R&D topics across the three categories supported with the findings from the international body of knowledge in the field. The proposed themes are: (1) integrated supply network structure with suitable visibility and usage of real-time data transfer possibilities, e-collaboration and CPFR (collaborative planning, forecasting and replenishment); (2) research and development of new logistics service concepts and their effect on the whole supply network; (3) research on the effects and possibilities of using new identification methods such as RFID (e.g. product, parcel or batch identity) in the supply network; and (4) the requirements of e-business for the national logistic infrastructure. Although the R&D themes seem to be independent entities, there is a very strong interrelationship between them in the context of e-business logistics as described in Figure 4. Technology innovations are enablers to business process re-engineering aiming towards supply network integration.

IJPDLM 32,7 526 Figure 4. Interrelations of research agenda themes Integrated supply network structure with suitable visibility Creating integrated supply networks with suitable visibility is a very complicated exercise. There are at least three major elements that need to be solved almost simultaneously as illustrated in Figure 5. Solutions are needed to enable information flow across company interfaces. There must also be know- how of what information should be shared and how to utilise the shared information. And last but not least there are potential changes in the structures of the value chains. During the iterative process of interviews and workshops the information transfer within the supply chain or in the future within the supply net was seen as one of the key R&D areas. E-collaboration is one of the most promising areas related to information sharing. According to e-collaboration experts it is important to understand what information should be shared and how to add value to the information shared (CPFR, 2001; Cameron and Gormley, 1998; Holmstrom et al., 2000). ¨ Figure 5. Building blocks of integrated supply network with suitable visibility

There are also more strategic problems to be solved: new integration of the Research agenda processes within the supply network and orchestration of the operations for e-business including the management of outsourced resources (Sawhney and Parikh, 2001). logistics R&D of new logistics services in the supply network More sophisticated customer demand chains and e-business pose new challenges to logistics service providers. R&D of new logistics service concepts 527 are required regardless of the line of business both in B2B and B2C environments. Unless the products are non-material, physical distribution logistics is one of the key challenges for the supply networks or e-marketplaces. According to the interviews the development of e-business will increase the importance of, especially, delivery accuracy, delivery frequency and delivery time (see Table IV). Thus a major challenge is to identify and develop new logistics service concepts, like 4PLTM (Bauknight, 2001; Bade and Mueller, 1999) or information intermediaries (Timmers, 1999), which could more efficiently manage, consolidate and optimise both information and material flows. This is especially important when different product group (Fisher, 1997) requirements, such as service levels, are to be created using the same resources. These new logistics services would enable companies to focus on their core competencies (Prahalad and Hamel, 1990) and added value could be created for both the network participants and customers. Research on identification methods in the supply network Development of e-business creates a need for new solutions for more efficient handling of lean material flows, efficient customisation of products and logistic services and information sharing across company limits. Research and development related to these topics can be approached from two perspectives: (1) Technology perspective. What are the possibilities that new technologies bring? (2) Business need perspective. Are there new technology demands when developing supply chain or network management? Technological innovations and huge increases in efficiency are often connected together (Brynjolfsson and Hitt, 1998). According to the workshop opinion mobile data transfer together with RFID (radio frequency identification) may lead to great efficiency improvements in logistics networks. Applications using wireless product identification technologies have until now been mostly at the individual company level. However, making item level supply chain management solutions work across company limits could reap enormous benefits (Karkkainen and Holmstrom, 2001). ¨ ¨ ¨ Requirements of e-business to the national logistic infrastructure According to the prioritisation done in the workshop, one of the most important R&D initiatives is concerned with the national logistics infrastructure responding to the developing e-business. When trying to identify the requirements of

IJPDLM e-business on national and regional levels, current material flows and current 32,7 logistics infrastructure need to be analysed. Additionally, various possible development directions ought to be analysed in respect of the requirements and challenges described, for example, in this article. One of the most interesting research areas would be to analyse the logistics service concepts needed in B2B and B2C supply chains to enable profitable 528 growth in e-business. Another challenge is to identify and define the responsible parties in the development and maintenance work for the physical logistics network infrastructure. In the development and maintenance of the Internet, the responsibilities are basically divided among all the actors who have Web sites. When creating the physical logistics network, where management of the information and material flows affects several parties, similar kinds of responsibility problems are faced. Discussion and conclusions In this study we have identified, categorised and prioritised the most crucial research and development topics that should be dealt with in order to help enterprises to face the new logistics challenges of electronic business. The analysis was conducted through an iterative process, where focus interviews of professionals were followed by the workshop. One of the problems of the selected method is the difficulty in analysing the data obtained during the focus interviews when open questions are used. The researchers’ role as an objective interpreter is very challenging. Another disadvantage of the method is that there is a possibility that, during the interview process, today’s problems may hinder interviewees in seeing the future possibilities, problems and broader development issues. Despite the potential disadvantages, the selected method was found to be suitable to this study. Inventing and developing new service concepts or business models to support the development of profitable e-business calls for an innovative and flexible attitude in the R&D and business operations. The rapid rate of change includes great possibilities but also threats. When implementing new technology to old processes there is a danger of creating only new, expensive and inefficient performance. To utilise the value of e-business efficiently, innovative new operating models have to be developed. However, development of e-business skills requires that the potential changes or modifications to existing business models and logistic processes can be identified and defined at the single company level. In developing new service or business concepts for e-business, the logistics service infrastructure covering information flow, co-ordination of supply network and material flows is a crucial issue. Several of the most important R&D topics that were identified are related to logistics services, service providers or the overall logistics service structure. The R&D initiatives in these fields ought to be interdisciplinary and cover a wide range of business-related issues. Additionally, there ought to be space for development initiatives with both a shorter and longer time span, as shown in Figure 6.

Research agenda for e-business logistics 529 Figure 6. Time span in the research and development Although new technologies are enablers to network integration, successful integration requires confidence and excellent professional skills from the various supply network partners. Research and development resources should also be allocated to the areas that support the ‘‘soft’’ side of the integration process. References Auramo, J., Aminoff, A. and Punakivi, M. (2001), What Are the Key Research and Development Topics in the Field of E-business Logistics?, Nofoma 2001 Conference, Reykjavik. Bade, D. and Mueller, J. (1999), ‘‘New for the millennium: 4PL’’, Transportation & Distribution, Vol. 40 No. 2, February, pp. 78-80. Bauknight, D. (2001), Fourth Party Logistics – Breakthrough Performance in Supply Chain Outsourcing, available at: www.accenture.com Belzowski, B., Flynn, M., Londal, G., DiBernardo, M., Cole, D.E., Smith, B. and Jimenez, T. (2000), Forecast and Analysis of the North American Automotive Industry, Delphi X, For 2004 and 2009, Office for the Study of Automotive Transportation, University of Michigan Transportation Research Institute, available at: www.osat.umich.edu/delphi.html Boxall, G. (2000), ‘‘The use of RFID for retail supply chain logistics’’, Tag 2000, 24 May. Brehmer, P.-O., Bus, L., Demkes, R., Hultkrantz, O., Ladonet, A., Sjostedt, M. and Waidringer, J. ¨ (1999), ‘‘Trilog Europe, Indicator report’’, Department of Transportation and Logistics, Chalmers University of Technology, Gothenburg. Brooksher, K. (1999), ‘‘E-commerce and logistics’’, Traffic World, Vol. 260 No. 7, 15 November, pp. 31-4. Brynjolfsson, E. and Hitt, L.M. (1998), ‘‘Beyond productivity paradox: computers are the catalyst for bigger changes’’, Communications of the ACM, Vol. 41 No. 8, August. Cameron, B. and Gormley, T. (1998), ‘‘Extend for collaboration’’, Manufacturing Systems, Vol. 16 No. 7. p. 20. CEST (2000), e-Supply for the European Rim – Securing our Role in the e-Commerce Revolution, CEST, London, available at: www.cest.org.uk Cox, J. (2001), ‘‘E-business specification passes a key milestone’’, Network World, 26 March. CPFR (2001), available at: www.cpfr.org/ Daugherty, P.L. (1994), ‘‘Strategic information linkage’’, The Logistics Handbook, The Free Press, New York, NY.

IJPDLM D’Amico, E. (2001), ‘‘Envera Web browser links companies without ERP systems’’, Chemical Week, 4 April. 32,7 DeJong, C.A. (1998), ‘‘Material handling tunes in’’, Automotive Manufacturing & Production, Vol. 110 No. 7, pp. 66-9. Demkes, R., Brugge, R. and Verduin, T. (1999), ‘‘Trilog – Europe Summary Report’’, TNO-report, Inro/Logistiek 2002 – 25, Delft. 530 Ferrari, R. (2000), ‘‘White paper frontline supply chain managers struggle with e-business’’, a Richmond Events and AMR Research White Paper, September, New York, NY. Fisher, M.L. (1997), ‘‘What is the right supply chain for your product’’, Harvard Business Review, Vol. 75 No. 2, pp. 105-16. Forrester, J. (1958), ‘‘Industrial dynamics, a major breakthrough for decision makers’’, Harward Business Review, July-August. Goolley, T. (2000), ‘‘Growth spurt’’, Logistics Management and Distribution Report, Vol. 39 No. 11, pp. 77-84. Hall, R. (1987), ‘‘Consolidation strategy: inventory, vehicles and terminals’’, Journal of Business Logistics, Vol. 8 No. 2, pp. 57-74. Hanover, D. (2000), ‘‘Channel conflict? Put a lid on it’’, Sales and Marketing Management, Vol. 152 No. 3, March. Holmstrom, J., Framling, K., Kaipia, R. and Saranen, J. (2000), ‘‘Collaborative planning forecasting ¨ ¨ and replenishment: new solutions needed for mass collaboration’’, working paper, available at: www.tuta.hut.fi/ecomlog/ Jedd, M. (2000), ‘‘Sizing up home delivery’’, Logistics Management and Distribution Report, Vol. 39 No. 2, February, pp. 51-6. Jones, H. (1999), ‘‘Asset management easier with RFID’’, Automatic ID News, Vol. 15 No. 9, p. 52. Kaplan, S. and Sawhney, M. (2000), ‘‘E-Hubs: the new B2B marketplaces’’, Harward Business Review, May-June, pp. 97-103. Karkkainen, M. and Holmstrom, J. (2001), ‘‘Wireless product identification: enabler for handling ¨ ¨ ¨ efficiency, customisation and flexible outsourcing’’, International Journal of Electronic Business. Kiely, D. (1998), ‘‘Synchronizing supply chain operations with consumer demand using customer data’’, The Journal of Business Forecasting Methods & Systems Flushing, Vol. 17 No. 4, pp. 3-9. Lee, H., Padmanaghan, V. and Whang, S. (1997), ‘‘The bullwhip effect in supply chains’’, Sloan Management Review, Vol. 38 No. 3, pp. 93-102. Lindstrom, T. (2000), Personal contact, Rafsec Oy, 17 July. ¨ Linstone, H. (1975), The Delphi Method : Techniques and Applications, Reading, MA. McKinnon, A. and Forster, M. (2000), European Logistical and Supply Chain Trends 1999-2005: The Results of a Delphi Survey, Logistics Research Network 2000, conference proceedings. Mullin, R. (2001), ‘‘ECM: where ERP meets the Web’’, Chemical Week, 25 April. Nonaka and Takeuchi (1995), Knowledge Creating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford University Press, New York, NY. Philpotts, M. (1996), ‘‘An introduction to the concepts, benefits and terminology of product data management’’, Industrial Management & Data Systems, Vol. 96 No. 4., pp. 11-17. Prahalad, C.K. and Hamel, G. (1990), ‘‘The core competence of the corporation’’, Harvard Business Review, Vol. 68 No. 3, pp. 79-92. Punakivi, M. and Saranen, J. (2001), ‘‘Identifying the success factors in e-grocery home delivery’’, International Journal of Retail & Distribution Management, Vol. 29 No. 4.

Radding, A. (1994), ‘‘With satellites, Boyle keeps trucking all night long’’, InfoWorld, Vol. 16 Research agenda No. 42. Ranagan, S. and Adner, R. (2001), ‘‘Profitable growth in Internet-related business: strategy tales for e-business and truths’’, working paper. logistics Reda, S. (1999), ‘‘Internet channel conflicts’’, Stores, Vol. 81 No. 12, pp. 24-8. Sawhney, M. and Parikh, D. (2001), ‘‘Where value lives in a networked world’’, Harward Business Review, January, pp. 79-86. 531 Shah, P. (2001), Interprice Platforms Ease Collaboration – Help Connect ‘‘Inticate Ecosystem’’ of Comlex High-Tech Supply Chains, Ebn, Manhasset, 7 May. Shulman, R. (1999), ‘‘Trucking in real time’’, Supermarket Business, Vol. 54 No. 2, pp. 14-22. Supply Chain Decisions (2001), Supply Chain Decisions, Managing the Integrated eB2B, challenge conference on 6-7 March, London. TAI (2000), Ecomlog-project, available at: www.tai.hut.fi/ecomlog/ Tang, V. (2000), ‘‘E-business and technologies of the Web: Shumpeter’s fifth-wave and Kondratieff’s next long-cycle’’, paper presented at the National Technology Agency of Finland, Helsinki, 19 October. Thomas, D. and Griffin, P. (1996), ‘‘Coordinated supply chain management’’, European Journal of Operations Research, Vol. 94 No. 1, pp. 1-15. Timmers, P. (1999), Electronic Commerce – Strategies and Models for Business-to-Business Trading, John Wiley & Sons, Chichester. Webster, J. (2001), ‘‘B2B transactions – will Web services do the trick? Business partners need a better way to share back-end application data. Standards-based services promise fast deployment and lower costs’’, Internetweek, 16 April. Werbach, K. (2000), ‘‘Syndication, the emerging model for business in the Internet era’’, Harward Business Review, May-June, pp. 85-93. Witt, C. (1999), ‘‘Update: material handling in the food industry’’, Material Handling Engineering, Vol. 54 No. 11, pp. 38-50. Further reading Wilson, R. and Delaney, R.V. (2000), 11th Annual State of Logistics Report – Logistics and the Internet: In the Frantic Search for Space, It is Still About Relationships, ProLogis & Cass Information Systems, Washington, DC, June.

The research register for this journal is available at The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/researchregisters http://www.emeraldinsight.com/0960-0035.htm IJPDLM 32,7 A comparative study of three different SCM approaches R. Meenakshi Sundaram and Sameer G. Mehta 532 Department of Industrial and Manufacturing Engineering, Tennessee Technological University, Cookeville, Tennessee, USA Received June 2001 Revised March 2002 Keywords Supply chain management, Logistics, Distribution, Production, Decision making, Integration Abstract A comparative study of three different approaches on a hypothetical supply chain model is presented. The three approaches investigated are: independent; semi-integrated; and integrated. In the independent approach, it is assumed that decisions are made independently at three different levels. Decisions are assumed to be made at two different levels in the semi-integrated approach. In the integrated approach, all decisions are assumed to be made at a single level. Introduction The advent of economic globalization in the past 20 years has had two major impacts in manufacturing. A global market has evolved very rapidly as more and more companies are embracing e-commerce. The competition has become fierce as companies can go beyond national borders. Manufacturing companies are under intense pressure to produce quality products faster and inexpensively. Innovative new technologies and techniques are evolving to meet the demand for cost reduction and quality improvements. Techniques such as lean manufacturing, Kanban systems, total quality management, just-in-time, and kaizen have been proven to be extremely effective on the shop floor. Manufacturers are continuously striving to improve the internal operations and are beginning to focus on the external operations as well. The unrelenting task of cutting costs to improve the profit margin has resulted in the management of all upstream operations as well as downstream operations external to manufacturing. These downstream operations are responsible for delivery of the products to the market. This has led to connecting both upstream and downstream companies culminating in the concept of supply chain management (SCM). Thus a supply chain may be defined as a network of facilities and distribution operations to perform the functions of procurement of materials, transformation of these materials into intermediate and finished products, and the distribution of these finished products to customers (Ganeshan, 1999). SCM constitutes a set of approaches utilized to efficiently integrate suppliers, manufacturers, warehouses, and retailers, so that merchandise is produced and distributed in the right quantities, to the right locations, and at the right times, in order to minimize system-wide costs while satisfying service level International Journal of Physical requirements (Simchi-Levi et al., 2000). Logistics is defined as ‘‘the process of Distribution & Logistics Management, planning, implementing, and controlling the efficient, effective flow and storage Vol. 32 No. 7, 2002, pp. 532-555. # MCB UP Limited, 0960-0035 of materials, finished goods, services, and related information from origin to the DOI 10.1108/09600030210442577 location where they are used or consumed’’ (Fawcett and Clinton, 1997).

A typical supply chain has a linear structure with links connecting one Three different constituent to the next. Raw materials are procured and products are made at SCM approaches one or more plant locations. The products are then either stored locally or shipped to warehouses/distribution centers for intermediate storage. This description essentially represents a traditional brick-and-mortar supply chain. Depending on the market demand, the products are shipped either to distributors or retailers directly. Thus a network of suppliers, manufacturers, 533 distribution centers, and retailers communicate for the manufacture and distribution of products in a supply chain. Each member of the supply chain operates independently without very many interactions with members farthest in the hierarchy. The interaction between two immediate members is limited to just transfer of information and materials. In a traditional supply chain, members are concerned with decisions that directly affect their bottom line. Many manufacturing operations were designed in the past to maximize throughput and lower costs with little consideration for the impact on inventory levels and distribution capabilities. The result of these factors is that there was not a single, integrated plan for the entire supply chain. Independent and conflicting plans were made by each of the supply chain members to achieve individual goals. Therefore, there exists a need to integrate these different functions. SCM is a strategy through which such integration can be achieved. The primary focus is on manufacturing and logistics so that manufactured products can be delivered as required on a timely basis and cost-effectively. Collaborative efforts among supply chain members are leading to improvement in supply chain efficiency. The end-user benefits since this results in an attractive price, improved quality and better service. Review of literature Early studies related to SCM were fragmented, with focus on improving functions such as purchasing, distribution, materials management, and logistics. The role of planning in the supply chain was totally independent between production and distribution. The independent approach results in higher cost due to inventory at different levels. Glover et al. (1979) developed a network model for production scheduling and distribution operations at Agrico Chemicals. In the model, a decision support system was embedded to analyze both short-term and long-term planning decisions, which involved determining the size and location of distribution centers and the volume and mode of shipping. Lawrence and Burbridge (1976) developed a multi-objective linear programming approach to determine alternative production schedules for groups of products in a two-echelon supply chain. The model helped determine what products to make, how much to produce, and where to produce these products. The model attempted to achieve coordinated production and distribution schedules subject to stated objectives of the firm.

IJPDLM The objectives considered in the model presented by Lawrence and 32,7 Burbridge were: . maximization of the total sales revenue for specific locations; . minimization of the total cost of production and distribution; . maximization of the production volume of a particular item at a specific 534 location. These conflicting objectives radically affect the operations of a company. Therefore, instead of optimizing each of the objectives independently, the goal programming technique was used to simultaneously satisfy the three objectives by assigning priorities. The goal programming technique tries to satisfy the first goal with the highest priority first, then the second and so on until the last goal is satisfied. This provides the decision maker some flexibility in compromising the goals. Cohen and Lee (1988) developed a decision model that linked management control and system performance in the supply chain using several submodels. The submodels were linked by inventory and scheduling decisions. Each submodel was optimized subject to a defined level of customer service. With the help of heavy traffic queuing theory, Arreola-Risa (1989) demonstrates that by assuming an exponential distribution for the manufacturing lead time, the accuracy of the operating characteristics of the model by Cohen and Lee (1988) could be improved. Blumenfield et al. (1986) considered the problem of scheduling the production and distribution for a manufacturing company supplying parts to assembly plants. A model was proposed that included the following limitations: . a single destination for each part type; . identical production cycles for each part; . fixed transportation costs for each shipment. It was concluded that simultaneous consideration of the production and distribution schedules would result in reduced costs. It was also found that cost saving for a given production-distribution schedule was maximized when the demand, item value, and variable costs were the same for each product. The model, reportedly (Blumenfield et al., 1987) implemented at General Motors Delco electronics division, resulted in a savings of $2.9 million per year in logistics cost. Chen et al. (1994) presented a production-planning model for a system composed of a central factory and several satellite factories. Each satellite factory specialized in the production of specific part types and supplied them to other plants in the network as required. Only the central factory dealt with outside customers. Material flow within the network was to satisfy the customer demands at the central factory. The production-planning problem was to minimize total cost. This included transportation, processing, holding

raw materials, and manufacturing products in a network of factories. Two Three different nonlinear mathematical programming models were developed to determine the SCM approaches best production and procurement policies. Heuristic algorithms were developed for solving the models. Ozdamar and Yazgac (1999) defined a production system as a chain of subsystems from the supplier subsystem to the distribution subsystem. They concluded that it was impossible to separate the subsystems and achieve the 535 goal of minimizing total costs. Therefore, an integrated production-distribution model was proposed to solve the production and distribution problems. The model was based on the operating system of a multinational company producing detergents in a central factory. Products were distributed to geographically distant warehouses. The overall system costs included factory and warehouse inventory costs as well as transportation costs. A hierarchical approach was used to schedule the weekly fluctuating demand. Cachon and Zipkin (1999) investigated a supply chain that consisted of a supplier and a retailer. Stationary stochastic demand and fixed transportation times were assumed. Inventory and backorder costs were considered at both stages. Two types of games were developed for the solution. The games minimized the total costs but differed in the way each operated. In the first game, the firms tracked inventory levels along echelons; in the other game, each firm tracked only local inventory. In the first case, there was an effort to reduce the total supply chain costs while the other focused on reducing the local costs, thus inducing competition. Solutions to the test problems showed a difference between the two approaches. The competition introduced by the second approach reduced the efficiency of the operation. For some specific cases, competition increased total cost only by a fraction of 1 percent; in others, the increase was much greater. Hence it was concluded that the supply chain approach was better. From the above discussions, it can be concluded that the investigations demonstrated the importance of cooperation among different functions of the supply chain. While some studies focused on combining the inventory and transportation functions, others combined the production and distribution decisions to reduce costs. Methods such as Benders decomposition, goal programming, non-linear programming, linear programming, and heuristics have been used. Different configurations of supply chains were used in these studies. While some used the single warehouse-multiple retailer configuration, others used the multiple manufacturers-retailers configuration. In most of these studies, only two stages were considered. Though proven to be cost-effective, the production-distribution models lacked the element of complete integration of the supply chain. Therefore it is worth investigating the integration of a supply chain with more than two stages. Currently, SCM is receiving significant attention. Newer models and methodologies are being developed for integrating many functions in the entire supply chain. A review of some recent studies is presented in the next few paragraphs.

IJPDLM Ganeshan (1999) presented a near optimal type of inventory-logistics cost 32,7 minimizing model. The model considered a production distribution network with a single distribution center receiving consignments from several suppliers. The consignments were distributed to a number of lower echelon retailers. The model addressed the following issues in the system: . the inventory analysis at the retailers; 536 . the demand process at the warehouse; . the inventory analysis at the warehouse. Cost components in the model were inventory and transportation costs. Solutions for the test problems were verified using simulation. The following inferences were made from the study: . The model was accurate in estimating the service levels at the retailer as well as the warehouse level. It was recommended that the model could be used to determine stock levels at each center. . The model was flexible enough to include the changes in supply chain configuration. The author recognized the following shortcomings of the model: . The model assumed that the supplier always had sufficient capacity to satisfy demand. The model did not extend beyond two echelons. . The model assumed identical suppliers and retailers for simplicity. Matta and Sinha (1995) investigated a two-echelon system, which included a warehouse, a system of suppliers, and a set of downstream retailers. The model considered the ordering costs only at the warehouses, not at the retailers. A cost model was developed to derive an optimal ordering-distribution policy for the distribution system. However, the model had its shortcomings. Although it extended beyond a manufacturing echelon into multiple echelons, it failed to integrate production-scheduling decisions in the model. Simchi-Levi and Chan (1998) considered the problem of integrating inventory control and vehicle routing to develop an optimal cost strategy for a distribution system. The system consisted of a single outside vendor, a fixed number of warehouses, and a set of geographically dispersed retailers. Demands at the retailers were assumed to be constant. In addition, inventory- holding charges were applicable at the warehouses and retailers. The authors emphasized the strategy of cross-docking for distribution planning in the system. Cross-docking is a strategy in which warehouses receive fully loaded trucks from suppliers. Warehouses coordinate the delivery to retailers but never hold inventory. Using this strategy, a model was developed to coordinate the distribution process. Another study related to integrated decision making in a supply chain is the inventory planning at Libbey-Owens-Ford. Martin et al. (1993) developed a model called FLAGPOL. This large-scale mathematical model simultaneously

considered the production and distribution, as well as inventory operations in Three different the flat glass business of Libbey-Owens-Ford. The model was used to make SCM approaches decisions in a supply chain covering four plants, over 200 products, over 40 demand centers, and a planning period of 12 months. Fumero and Vercellis (1999) aimed to coordinate important and interrelated logistics decisions such as capacity management, inventory allocation, and vehicle routing. An optimization model that accounted for the various 537 production and distribution capabilities of the system was developed. The model was solved using the three approaches; Lagrangian relaxation, Lower bound solutions, and Heuristic solution. The computational results on test problems demonstrated the effectiveness of the model. The model was solved for a single plant, multi-period, and multi-product configuration. The capacities were assumed to be infinite. Back orders were not allowed. The model ignored the effects of suppliers and warehouses in the system. Recent literature reviews demonstrate the increasing emphasis on synchronization of decision processes throughout the supply chain. Inventory and routing problems have been investigated under deterministic as well as stochastic assumptions, but most studies have ignored production decisions. Models that considered the production processes at different facilities failed to include the distribution plans. Ideal conditions such as single product, single plant, single period, and no back orders are the assumptions in most of these models. There is a need for developing more realistic models relaxing these assumptions over a wide range of conditions with total integration. Proposed method The major objective of the research reported here is to develop mathematical approaches that would aid decision making at various levels in the supply chain. Three approaches are presented incorporating varying degrees of integration of the decision processes. Integration in a supply chain can be defined as an association of customers, retailers, distribution centers/warehouses, and manufacturers using techniques enabling them to work together to optimize their collective performance in the creation, distribution, and support of the end product (NRC, 2000). There are basically two types of integration processes: namely, internal integration, involving coordinated management of a company’s internal operational activities like production scheduling, labor allocation, inventory holding, job sequencing, shipping, etc., and external integration, which refers to integration of activities external to the company across the supply chain. Less than 5 percent of companies have achieved total integration with others (Copacino, 1997). The importance of supply chain integration for developing successful and competitive strategies has been stressed. Independent approach In this approach, it is assumed that links of the supply chain operate independently as detailed in Figure 1. Decision-making occurs through a series

IJPDLM 32,7 538 Figure 1. Independent approach

of modules operated by various members of the supply chain. There are separate Three different modules for decision making such as production planning and shipping at every SCM approaches echelon of the supply chain. Modules focus on local optimization. This type of supply chain represents the segregated association of businesses where decisions are localized in the form of ‘‘silos’’. The forecast demands from retailers are consolidated into the distribution-planning module maintained by the distributor. The module indicates replenishment is shipped from each 539 distribution center/warehouse to each retailer. The module considers stocks on hand at each distribution center/warehouse and the capacity of the trucks. The module also determines the replacement volume of products needed at each distribution center/warehouse. The manufacturer’s plant-production-planning module receives the quantity requirements from the distribution centers/ warehouses. The module develops production plans for the plants and the master production schedule required by the upstream tier-1 supplier. The plant-shipping module develops the shipping plan from the production plan. The tier-1 supplier-production-planning module receives the master production schedule from the manufacturer’s plants. The module develops a desegregated production plan for supplier plants. In addition, it develops a master production schedule for the material-requirements-planning module of tier-1 suppliers. It also determines the quantities of material to be supplied by each tier-2 supplier to the tier-1 supplier plants. The tier-1 supplier-shipping module develops a logistics plan for shipment of components from the tier-1 suppliers to the manufacturer’s plants. Semi-integrated approach Unlike the independent approach, the semi-integrated approach involves some degree of coordination among the constituents of the supply chain. The decision processes are illustrated in Figure 2. Like the independent approach, the demand forecasts from each retailer are consolidated and sent to the distributor. The distributor passes on these requirements to the integrated plant-supplier-planning (IPSP) module residing at the manufacturer and accessible to the tier-1 suppliers. This module is used to develop production plans at the manufacturing plants and the tier-1 supplier plants. It is also used to develop the shipping plan for the tier-1 supplier plants. When the distribution-planning module receives the plant production plans, the module determines the product-shipping schedule from the manufacturing plants to the distribution centers and to the retailers. The module also considers the stock levels of the distribution centers/warehouses and the capacities of the trucks available. The tier-1 supplier, upon receiving the master production schedule, develops the material requirement plan. The requirements are sent to the tier-2 supplier for material procurement. Integrated approach In the integrated approach shown in Figure 3, total collaboration among the various links of the supply chain is achieved. Unlike the previous two

IJPDLM 32,7 540 Figure 2. Semi-integrated approach approaches, the integrated-supply-chain-planning (ISCP) module makes all the decisions using a single module. Each facility is connected to the decision- making process through the integrated-supply-chain-planning module or the ISCP module. In this process, once the retailers place their orders with the ISCP module, the planning across the whole supply chain occurs. The ISCP module helps in making the following decisions in the supply chain: . Quantity of products to supply to each retailer from each distribution center/warehouse.

Three different SCM approaches 541 Figure 3. Integrated approach . Levels of stocks to be maintained at the distribution centers/warehouses. . Production plans for the manufacturing plants. . Shipping plans from plants to the distribution centers/warehouses. . Quantity of components required at the plants.

IJPDLM . Production plans for the tier-1 suppliers. 32,7 . Shipping plans from the tier-1 suppliers to the plants. . Material requirements at the tier-1 suppliers. . Quantity of material to be supplied to each tier-1 supplier by each tier-2 supplier. 542 Discussion Mathematical models were developed for each of the three approaches. For each of the supply chain approaches using the demand and resource limitations, the developed models were solved. The mathematical models developed are included in the Appendix. The following assumptions were made in developing the mathematical models for all three approaches: . A fixed number of echelon levels exists in the supply chain. A supply chain can involve any number of entities at the same level. . The mathematical models are multi-product and multi-period type. . Manufacturing operations are assumed at the manufacturer and tier-1 supplier level. . Tier-2 suppliers are assumed to have fixed capacity to supply materials. No production or inventory decisions are involved at tier-2 supplier level. . All products in a category have the same shipping costs. . There is no material flow between entities at the same level. . Less than partial truckload shipments are allowed in the supply chain. . There are no location decisions involved in the supply chain. . Lead times are ignored in the formulation of the supply chain models. . Back orders are allowed at the retailers. No back orders are allowed at other levels. . The truckload decisions in the supply chain are weight-constrained. This is contrary to the real world situation where both weight as well as volume constraints are considered in making shipping decisions. These solutions provided a detailed plan for each of the following functions: . Quantity of each product manufactured at each plant. . Quantity of product shipments from manufacturers to distribution centers/warehouses. . Quantity of product shipments from distribution centers/warehouses to the retailers. . Component requirements at the manufacturer.

. Component production plans for each tier-1 supplier facility. Three different . Quantity of components shipped from tier-1 suppliers to manufacturers. SCM approaches . Material requirements at tier-1 supplier facilities. . Material procurement quantities from each tier-2 supplier. Model performance 543 Several test problems were developed for the supply chain described. Further, the results obtained were tested statistically to assess the effect of supply chain integration. The hypothetical supply chain configuration used in the tests is described in Table I. The supply chain consists of an organization that owns and operates two manufacturing plants. Both plants are assumed to manufacture two products. It is also assumed that the two plants have different production capacities and, consequently, different production costs. The manufacturers ship products to the distribution centers/warehouses based on the demand schedule. The distributors place purchase orders with the manufacturers, who in turn develop the production plans. Each distribution center/warehouse has a specific capacity to handle each product and ships units to retailers upon receipt of orders. Three retailer outlets are assumed in the supply chain investigated and all three are assumed to be operated by the same company. It is assumed that the company has demand forecasts to place orders with the distributor. The manufacturers use two component parts in production. Component orders are placed with the tier-1 suppliers. The tier-1 suppliers level includes two production facilities to meet the component demands. Both the facilities are assumed to be operated by a single company. The tier-1 suppliers develop their production plans based on the master production schedule provided by the manufacturer. The suppliers are assumed to use two different raw materials for making the two components. The material orders for the tier-1 supplier are placed with two second-tier suppliers for replenishment. Supply chain constituent Configuration Number of horizon periods 3 Number of manufacturing plants 2 Number of DC/warehouses 2 Number of retailers 3 Number of tier-1 suppliers 2 Number of tier-2 suppliers 2 Types of products 2 Table I. Types of components 2 Supply chain Types of materials 2 configuration

IJPDLM Experimental design 32,7 Having formulated the mathematical models for the three different approaches, it was necessary to assess the benefits of an integrated decision-making process. Also it was necessary to investigate the effect of various supply chain parameters on the benefits of supply chain integration. An analysis of variance (ANOVA) was performed to test the statistical significance of the factors. An 544 experiment with two levels of each parameter was designed. The first step in the design was to identify factors that influence cost. The cost reduction that can be achieved was of interest in this experimental analysis. The following factors were chosen: . the distribution center/warehouse capacity; . the manufacturing plant capacity; . the tier-1 supplier production capacity; . total shipping cost in the supply chain; . model type. The factor levels selected for the experiment are tabulated in Table II. The models were tested for different combinations of the factors at high and low levels. A high level indicates a capacity utilization of 80 percent for all three capacity factors. In other words, 80 percent of the total capacities available at the distribution centers/warehouses, plants, and tier-1 suppliers are utilized to meet the retailer demands. The low level is 60 percent capacity utilization. Shipping cost was also tested at high and low levels. The high level represents shipping costs as 10 percent of the total costs. The high level for the model type indicates cost reduction achieved by using the integrated approach as compared to the independent approach; the low level indicates the cost reduction achieved when the semi-integrated approach is used as compared to the independent approach. Comparison of test runs for each of the approaches The objective of conducting the test runs was to compare the supply chain costs for each of the three approaches and to assess the benefits of integrating the decision-making processes in a supply chain. Factor High Low Distributor capacity utilization (%) 80 60 Plant capacity utilization (%) 80 60 Supplier capacity utilization (%) 80 60 Table II. Shipping costs as percentage of total costs (%) 30 10 Factor levels Model type Integrated Semi-integrated

Total supply chain cost was obtained by solving the mathematical models Three different developed for each of the following three approaches: SCM approaches (1) Independent approach – six decision levels. (2) Semi-integrated approach – three decision levels. (3) Integrated model – one decision level. 545 The total supply chain cost for each approach is shown in Table III. The mathematical models were utilized to determine the following: . volume of each product manufactured at each plant; . volume of product shipments from manufacturers to distribution centers/warehouses; . volume of product shipments from distribution centers/warehouses to the retailers; . component requirements at the manufacturer; . component production plans for each tier-1 supplier facility; . quantity of components shipped from tier-1 suppliers to manufacturers; . material requirements at tier-1 supplier facilities; . material procurement quantities from each tier-2 supplier; Type of supply chain approach Run Integrated ($) Semi-integrated ($) Independent ($) 1 33,038,000 33,327,200 34,587,900 2 33,641,000 33,867,600 35,105,900 3 33,650,800 33,867,600 35,084,900 4 33,619,000 33,841,250 34,936,900 5 33,628,800 33,841,250 34,915,900 6 34,222,000 34,378,600 35,454,900 7 34,231,800 34,378,600 35,433,900 8 33,047,800 33,332,800 34,202,900 9 27,688,092 27,952,357 28,415,530 10 27,746,639 27,988,111 28,408,558 11 27,400,727 27,641,871 28,040,040 12 27,419,144 27,674,245 28,033,068 13 27,299,112 27,508,056 27,859,080 14 27,356,340 27,509,527 27,852,108 Table III. 15 27,010,695 27,159,106 27,483,410 Comparison of total 16 27,028,589 27,163,182 27,468,078 supply chain costs

IJPDLM . number of truckload trips between manufacturers and distribution 32,7 centers/warehouses; . number of truckload trips between distribution centers/warehouses and retailers; . beginning and ending inventories at each location. 546 From results of the test runs, supply chain costs for decision making using the independent approach are the highest followed by the semi-integrated approach. The integrated supply chain approach had the least total cost. This supports the contention that increased collaboration within the supply chain ensures better supply chain performance. Experimental design characteristics Having confirmed that the integrated approach produced better results than the independent approach, an investigation needed to be carried out to assess the effect of various supply chain parameters on the benefits of supply chain integration. Thus, the response variable in the experiment was the percentage cost reduction that was achieved using the semi-integrated approach and integrated approach as compared to the independent approach. The experiment was a 25 factorial design with five factors at two levels, as shown in Table IV. With each run a unique combination of only one replicate for a given treatment set, this experiment was an unreplicated factorial for a 25 design. The data for the percentage cost reduction achieved by supply chain integration were obtained from all the test runs tabulated in Table III. The experimental design data are not included in the paper. Analysis The factors found to have significant influence on the percentage of cost reduction were the following: . distributor capacity utilization; . plant capacity utilization; . supplier capacity utilization; . shipping cost; . model type; . interaction between distributor capacity utilization and shipping cost; . interaction between distributor capacity utilization and model type. Parameter Number Number of factors 5 Number of levels 2 Table IV. Experimental design Number of treatment runs 32 characteristics Number of replicates 1

The following inferences were made from the analysis of variance. These Three different conclusions apply only to the data tested. No inferences can be made about SCM approaches supply chains in general. (1) It can be seen from the results that as the shipping cost in the supply chain changes from high to low, the mean percentage cost reduction drops from an average of 3.5 percent to a value of 1.6 percent. Thus, the 547 higher the shipping costs, the more beneficial it is to integrate the decision-making process in the supply chain. (2) It can be seen from the results that as the shipping cost in the supply chain changes from high to low, the mean percentage cost reduction drops from an average of 3.5 percent to a value of 1.6 percent. Thus, the higher the shipping costs, the more beneficial it is to integrate the decision-making process in the supply chain. (3) The integrated supply chain approach results in greater cost reduction than the semi-integrated approach as compared to independent approach. This result supports the contention that the model costs decrease as the degree of supply chain integration increases. (4) It can be inferred from Table V that supply chain integration benefits are much higher when the supply chain is less constrained in resources available at each level. The semi-integrated approach and the integrated approach take a system-wide view of the decision-making process in the supply chain and provide globally optimized solutions for the given assumptions. Since decisions in the independent approach are made at several levels, they tend to generate localized optimum solutions at each level. The greater the available capacities at each level of supply chain, the more localized are the decisions generated by the independent approach, and thus the greater the supply chain cost resulting from the independent approach as compared to the globally optimizing integrated supply chain approaches. (5) There is a significant effect of shipping costs-distributor capacity utilization interaction and model type-distributor capacity utilization interaction on the benefits of supply chain integration. It can be inferred from the cost reduction achieved by integrating the supply chain is high for the following conditions: Cost reduction at 60% Cost reduction at 80% Resources capacity utilization capacity utilization DC/warehouses 2.90 2.30 Plants 2.71 2.40 Table V. Response to variation Suppliers 2.70 2.50 in capacity utilization

IJPDLM . shipping costs in the supply chain are high; 32,7 . capacity utilization at the distributor level is low; . an integrated supply chain approach is used. The manufacturers use two component parts in production. Component orders are placed with the tier-1 suppliers. The tier-1 level includes two production 548 facilities to meet the component demands. Both the facilities are assumed to be operated by a single company. The tier-1 suppliers develop their production plans based on the master production schedule provided by the manufacturer. The suppliers are assumed to use two different raw materials for making the two components. The tier-1 material orders are placed with the two second-tier suppliers for replenishment. The objective for the test runs is to compare the supply chain operation costs for each of the three approaches and validate the benefits of integrating the decision-making process for different processes across the whole supply chain. As reported earlier, the integrated approach involves integration of all the decision-making processes across the supply chain. The semi-integrated supply chain approach includes integration of decision processes at the plant- supplier level and plant-retailer logistics level thus providing for some degree of integration in the supply chain. The independent approach involves independent decision making at each level of the supply chain as an example of minimum collaboration. From the results obtained for the test runs, it is evident that the supply chain cost for decision making using the independent approach are the highest followed by the semi-integrated supply chain approach. The integrated supply chain approach resulted in a plan with the least cost along the supply chain as compared to the other two approaches. This supports our contention of better supply chain performance with increased collaboration across the supply chain. Conclusion For each of the three supply chain approaches, linear programming models were formulated and the models for one of the approaches are shown in the Appendix. Linear programming is a well-known mathematical technique to optimize a linear function subjected to several linear constraints. Also the decision to formulate the above models as linear programming models was largely influenced by the availability of a linear programming software package. The models formulated are all five echelons, multi-facility, multi-period, multi-product type. The models can only be applied to a supply chain representing five levels, namely, the retailers, the warehouses, the plants, the suppliers, and the second-tier suppliers. However the model can support any number of facilities at each level. Actual production occurs only at the plants and the suppliers.

Models for each of the three formulated supply chain approaches were Three different solved for a set of 16 test problems. For a given supply chain configuration, the SCM approaches formulated models were solved and the total cost across the supply chain was computed. The Linear Interactive Discrete Optimizer (LINDO) optimization software tool was used to solve the models. The mainframe version was used due to the magnitude of the supply chain formulation at hand. Each of the formulated models incorporated varying degrees of integration of the decision- 549 making process for the various functions across the supply chain. From the results obtained for the test runs, it is evident that the supply chain costs for decision making using the independent supply chain approach are the highest followed by the semi-integrated supply chain approach. The integrated supply chain approach resulted in a plan with least costs along the supply chain as compared to the other two approaches. The integrated supply chain approach represents a system, which involves integration of all the decision-making processes across the supply chain. It represents a synchronized supply chain where decisions at each level of plants, warehouses, suppliers and retailers are taken simultaneously under a single decision system. The semi-integrated supply chain approach includes integration of decision processes at the plant- supplier level and plant-retailer logistics level thus providing for some degree of integration in the supply chain. It represents collaboration between the plants and the suppliers on one hand and the plants and the warehouses on the other. The independent approach involves independent decision making at each level of the supply chain and thus displays minimum collaboration. Each module herein can be thought of as a decision system that optimizes the specific processes for which it is developed such as production planning, logistics, and others. Thus it can be concluded from the results that increased collaboration across the supply chain results in improved performance of the supply chain. In general, optimizing each process of the supply chain in isolation from other process does not guarantee optimization for the whole supply chain. Considering the advancements in technology, manufacturers should try to achieve optimality in the supply chain by integrating the decision making processes across the supply chain under a single system. References: Arreola-Risa, A. (1989), ‘‘Integrated production distribution systems with capacitated manufacturing facilities’’, unpublished PhD dissertation, Department of Industrial Engineering and Engineering Management, Stanford University, Palo Alto, CA. Blumenfield, D.E., Burns, L.D. and Daganzo, C. (1986), Synchronizing Production and Transporation Schedules, Research Publication GMR 5519,1986, General Motors Research Laboratories. Blumenfield, D.E., Burns, L.D., Daganzo, C., Frick, M.C. and Hall, R.W. (1987), ‘‘Reducing logistics costs at General Motors’’, Interfaces, Vol. 17, pp. 26-37. Cachon, G.P. and Zipkin, P.H. (1999), ‘‘Competitive and cooperative inventory policies in a two-stage supply chain’’, Management Science, Vol. 35 No. 7, July, pp. 936-53.

IJPDLM Chen, C.-F., Egbelu, P.J. and Wu, C.-T. (1994), ‘‘Production planning models for a central factory with multiple satellite factories’’, International Journal of Production Research, Vol. 32 32,7 No. 6, pp. 1431-50. Cohen, M.A. and Lee, H.L. (1988), ‘‘Strategic analysis of integrated production-distribution systems: models and methods’’, Operations Research, Vol. 36, pp. 216-28. Copacino, W.C. (1997), Supply Chain Management – The Basics and Beyond, The St. Lucie Press/ APICS Series on Resource Management, Delray Beach, FL. 550 Fawcett, S.E. and Clinton, S.R. (1997), ‘‘Enhancing logistics to improve the competitiveness of manufacturing organizations: a triad perspective, logistics in manufacturing’’, Transportation Journal, Fall, pp. 18-28. Fumero, F. and Vercellis, C. (1999), ‘‘Synchronized development of production, inventory, and distribution schedules’’, Transportation Science, Vol. 33 No. 3, August, pp. 330-40. Ganeshan, R. (1999), ‘‘Managing supply chain inventories: a multiple retailer, one warehouse, multiple supplier model’’, International Journal of Production Economics, Vol. 59, pp. 341-54. Glover, G., Jones, G., Karney, D., Klingman, D. and Mote, G. (1979), ‘‘An integrated production, distribution and inventory planning system’’, Interfaces, Vol. 9 No. 5, pp. 21-35. Lawrence, K.D. and Burbridge, J.J. (1976), ‘‘A multiple goal linear programming model for coordinated production and logistics planning’’, International Journal of Production Research, Vol. 14 No. 2, pp. 215-22. Martin, C.H., Dent, D.C. and Eckhart, J.C. (1993), ‘‘Inventory planning at Libbey-Owens-Ford’’, Interfaces, Vol. 23, June, pp. 68-78. Matta, K.F. and Sinha, D. (1995), ‘‘Policy and cost approximations of two-echelon distribution systems with a procurement cost at the higher echelon’’, IIE Transactions, Vol. 27, pp. 638-45. National Research Council (2000), Surviving Supply Chain Integration – Strategies for Small Manufacturers, National Academy Press, Washington, DC. Ozdamar, L. and Yazgac, T. (1999), ‘‘A hierarchical planning approach for a production- distribution system’’, International Journal of Production Research, Vol. 37 No. 16, pp. 3759-72. Simchi-Levi, D. and Chan, L.M.A. (1998), ‘‘Probabilistic analyses and algorithms for three-level distribution systems’’, Management Science, Vol. 44 No. 11, Part 1 of 2, November, pp. 1562-75. Simchi-Levi, D., Kaminsky, P. and Simchi-Levi, E. (2000), Designing and Managing the Supply Chain: Concepts, Strategies, and Case Studies, Irwin McGraw-Hill, Boston, MA. Appendix. Formulation of mathematical models The notations used for the various parameters, cost components and decision variables are as follows: Parameters i – Plant. j – Distribution center/warehouse. k – Retailer. p – Product type. q – Material type. r – Component type.

s – Tier-1 supplier. Three different t – Planning period in a planning horizon. SCM approaches v – Tier-2 supplier. CDpkt – Demand for product ‘‘p’’ at retailer ‘‘k’’ in planning period ‘‘t’’. MPpi – Maximum production capacity of plant ‘‘i’’ for product ‘‘p’’ in any given period. MOpi – Maximum overtime production capacity of the plant ‘‘i’’ for product ‘‘p’’ in any given period. 551 Nrp – Component determination factor. It is the number of components ‘‘r’’ required to make product ‘‘p’’. NRqr – Material determination factor. Quantity of material ‘‘q’’ required for component ‘‘r’’. QVqr – Maximum capacity of tier-2 supplier ‘‘v’’ to supply component ‘‘q’’. SPrs – Maximum production capacity of tier-1 supplier ‘‘s’’ for component ‘‘r’’. SOrs – Maximum overtime production capacity of tier-1 supplier ‘‘s’’ for component ‘‘r’’. VWpj – Maximum quantity of product ‘‘p’’ which can be handled at distribution center/ warehouse ‘‘j’’ in any given period. Wp – Weight of the product ‘‘p’’. WXij – Weight capacity of trucks operating between plan ‘‘i’’ and distribution center/ warehouse ‘‘j’’. WZjk – Weight capacity of trucks operating between distribution center/warehouse ‘‘j’’ and retailer ‘‘k’’. Decision variables Decision variables at manufacturing plants Drit – Quantity of component ‘‘r’’ needed for production at plant ‘‘i’’ in the planning period ‘‘t’’. Ipit – Inventory of product ‘‘p’’ to be held at plant ‘‘i’’ in planning period ‘‘t’’. Ppit – Production in units for product ‘‘p’’ at plant ‘‘i’’ in planning period ‘‘t’’. PWijt – Number of truckload trips required from plant ‘‘i’’ to distribution center/warehouse ‘‘j’’ in planning period ‘‘t’’. Pit – Overtime production in units for product ‘‘p’’ at plant ‘‘i’’ in planning period ‘‘t’’. Rarity – Inventory of component ‘‘r’’ to be held at plant ‘‘i’’ in planning period ‘‘t’’. Sapient – quantity of product ‘‘p’’ to be shipped from plant ‘‘i’’ to distribution center/warehouse ’j’ in planning period ’t’. Decision variables at tier-1 suppliers DQqst – Quantity of material ‘‘q’’ required for production at tier-1 supplier ‘‘s’’ in the planning period ‘‘t’’. IQqst – Number of units of ending inventory of material ‘‘q’’ at tier-1 supplier ‘‘s’’ in planning period ‘‘t’’. Rrsit – Quantity of component ‘‘t’’ shipped from tier-1 supplier ‘‘s’’ to plant ‘‘i’’ in the planning period ‘‘t’’. RPrst – Production in units for component ‘‘r’’ at tier-1 supplier ‘‘s’’ in planning period ‘‘t’’. ROrst – Overtime production in units for component ‘‘r’’ at tier-1 supplier ‘‘s’’ in period ‘‘t’’. SIrst – Number of units of ending inventory of component ‘‘r’’ at tier-1 supplier ‘‘s’’ in planning period ‘‘t’’.

IJPDLM Decision variable at tier-2 suppliers 32,7 Qqvst – Quantity of material ‘‘q’’ to be supplied by tier-2 supplier ‘‘v’’ to supplier ‘‘s’’ in planning period ‘‘t’’. Decision variable at distribution centers/warehouses SCpjkt – Quantity of product ‘‘p’’ shipped from distribution center/warehouse ‘‘j’’ to retailer outlet ‘‘k’’ in the planning period ‘‘t’’. 552 WCjkt – Number of truckload trips from distribution center/warehouse ‘‘j’’ to retailer outlet ‘‘k’’ in the planning period ‘‘t’’. WDpjt – Demand for product ‘‘p’’ at distribution center/warehouse ‘‘j’’ in planning period ‘‘t’’. WIpjt – Inventory of product ‘‘p’’ to be held at distribution center/warehouse ‘‘j’’ in planning period ‘‘t’’. Decision variable at retailers Bpkt – Units of product ‘‘p’’ backordered at retailer ‘‘k’’ in planning period ‘‘t’’. Cost components CBpk – Backorder cost for a unit of product ‘‘p’’ at retailer ‘‘k’’. CIpi – Inventory holding cost per period for a unit of product ‘‘p’’ at plant ‘‘i’’. CIQqs – Inventory holding cost per period for a unit of material ‘‘q’’ at tier-1 supplier ‘‘s’’. CPpi – Production cost for a unit of product ‘‘p’’ at plant ‘‘i’’. CPWij – Shipping cost for a truckload from plant ‘‘i’’ to distribution center/warehouse ‘‘j’’. COpi – Overtime production cost for a unit of product ‘‘p’’ at plant ‘‘i’’. CQqvs – Cost of a unit of material ‘‘q’’ supplied by tier-2 supplier ‘‘v’’ to tier-1 supplier ‘‘s’’. CRrsi – Shipping cost of component ‘‘r’’ from tier-1 supplier ‘‘s’’ to plant ‘‘i’’. CRIrsi – Inventory holding cost per period for a unit of component ‘‘r’’ at plant ‘‘i’’. CRPrs – Production cost for a unit of component ‘‘r’’ and tier-1 supplier ‘‘s’’. CROrs – Overtime production cost for a unit of component ‘‘r’’ at tier-1 supplier ‘‘s’’. CSIrs – Inventory holding cost per period for a unit of component ‘‘r’’ at tier-1 supplier ‘‘s’’. CWpj – Inventory holding cost per period for a unit of product ‘‘p’’ at distribution center/ warehouse ‘‘r’’. CWCjk – Shipping cost for a truckload from distribution center/warehouse ‘‘j’’ to retailer ‘‘k’’. Formulation of mathematical model for integrated approach In the integrated approach as shown in Figure 3, all supply chain links are connected through the integrated supply chain planning (ISCP) module. It utilizes supply chain constraints to determine the following: . Distribution centers/warehouses’ shipping plans. . Distribution centers/warehouses’ stock requirements. . Production planning at the manufacturers. . Manufacturers’ shipping plans. . Manufacturers’ component requirements. . Tier-1 supplier production planning. . Tier-1 supplier shipping plans. . Tier-1 supplier material shipping plans.

. Tier-1 supplier material requirements. Three different . Tier-1 supplier shipping plans. SCM approaches Objective function The objective function for the ISCP module is to minimize the following costs: . Manufacturer production costs. . Manufacturer overtime production costs. 553 . Manufacturer inventory costs. . Tier-1 supplier shipping costs. . Tier-1 component production costs. . Tier-1 overtime production costs. . Tier-1 supplier inventory costs. . Backorder costs. . Manufacturer-distribution center/warehouse shipping costs. . Distribution center/warehouse-retailer shipping costs. . Distribution center/warehouse inventory costs. . Tier-1 supplier material procurement cost. . Tier-1 supplier inventory costs for materials. Minimize Æt Æp Æi CPpi Ppit þ Æt Æp Æi COpi Opit þ Æt Æp Æi CIpi Ipi þ Æt Ær Æs Æi CRrsi Rrsit þ Æt Æs Ær CRPrs RPrst þ Æt Æs Ær CROrs ROrst þ Æt Æs Ær CSIrs SIrst þ Æt Æp Æk CBpk Bpkt þ Æi Æj Æt CPWij PWijt þ Æt Æj Æk CWCjk WCjkt þ Æt Æp Æj CWpj WIpjt þ Æt Æs Æq CIQqs þ Æt Æq Æv Æs CQqv Qqvst Constraints (1) Product consumption. For each period, beginning inventory at all DCs/warehouses + beginning manufacturing plant inventory + regular production + overtime production – retailer demand – previous period backorders + current period backorders = ending inventory at all DCs/warehouses + ending manufacturing inventory Æj WIpjðtÀ1Þ þ Æj IpiðtÀ1Þ þ Æj Rpit þ Æj Opit À Æk CDpkt À Æk BpkðtÀ1Þ þ Æj Bpk ¼ Æj WIpjt þ Æj Ipit ð8 p; tÞ (2) Manufacturer production capacity – regular time. For each period, regular production maximum production capacity. Ppit MPpi ð8; p:tÞ (3) Manufacturer production capacity – overtime. For each period, overtime production maximum overtime capacity. Opit MOpi ð8i; p:tÞ (4) Component requirements at manufacturing. For each period, component demand = component determination factor* (regular + overtime production scheduled) Drit ¼ Æp Nrp ðPpit þ Opit Þ ð8i; r; tÞ

IJPDLM (5) Manufacturing demand. For each period, quantity of components shipped by tier-1 suppliers = manufacturing component demand 32,7 Æs Rrit ð8i; r; tÞ (6) Tier-1 shipping quantities. For each period, beginning inventory + regular production + overtime production – quantity shipped to the manufacturing plants = ending inventory 554 SIrsðtÀ1Þ þ RPrst þ ROrst À Æi Rrst ¼ SIrst ð8 r; s; tÞ (7) Tier-1 production capacity – regular time. For each period, regular component production maximum component production RMrst SPrs ð8 r; s; tÞ (8) Tier-1 production capacity – overtime. For each period, overtime component production maximum overtime production ROrst SOrs ð8 r; s; tÞ (9) Manufacturing-DC/warehouse shipping quantities. For each period, beginning inventory + manufacturing plant regular production + manufacturing overtime production = quantity shipped to all DCs/warehouses = ending inventory IpiðtÀ1Þ þ Ppit þ Opit À Æj SApijt ¼ Ipit ð8 i; p; tÞ (10) Distribution centers/warehouse inventory. For each period, beginning inventory + quantity received by DC/warehouse – quantity to be shipped by DC/warehouse = ending inventory WIpjðtÀ1Þ þ Æj SApijt À SCpjkt ¼ WIpjt ð8 j; p; tÞ (11) Retailer demand. For each period, retailer demand + previous period backorder = quantity received from DC’s/warehouses + current period backorder CDpkt þ BpkðtÀ1Þ¼Æj SCpjkt þBpkt ð8 k;p;tÞ (12) Truckload trips between manufacturing plants-DC/warehouses. For each period, weight of the product * quantity supplied from manufacturing plants weight capacity of truck * number of truckload trips Æp Wp SApijt PWijt WXij ð8 i; j; tÞ (13) Number of truckload trips between DC/warehouse-retailers. For each period, weight of the product * quantity supplied from DC/warehouse weight capacity of truck * number of truckload trips Æp Wp SCpjkt WCjkt WZjk ð8 j; k; tÞ (14) Distribution center/warehouse capacity. For each period, quantity received by DC/ warehouse + ending inventory DC/warehouse capacity Æi SApij þ WIpjðTÀ1Þ VWpj ð8 j; p; tÞ (15) Tier-1 material requirements. For each period, tier-1 supplier demand = material determination factor * (tier-1 regular + overtime production scheduled) DQqst þ Ær NRqr ðRPrst þ ROrst Þ ð8 q; s; tÞ

(16) Tier-1 material replenishment. For each period, tier-1 supplier beginning inventory + Three different quantity of material received from tier-2 suppliers – material demand at tier-1 supplier = ending inventory at a tier-1 supplier SCM approaches IQqsðtÀ1Þ þ Æv Qqrst À DQqst ¼ IQqst ð8 q; s; tÞ (17) Tier-2 supplier capacity. For each period, quantity of material shipped by tier-2 supplier maximum tier-2 supplier capacity 555 Æs Qqvst QVqv ð8 q; t; vÞ

The research register for this journal is available at The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/researchregisters http://www.emeraldinsight.com/0960-0035.htm IJPDLM 32,7 Improving materials management effectiveness A step towards agile enterprise 556 M. Caridi and R. Cigolini Received June 2001 Department of Management, Economics and Industrial Engineering, Revised May 2002 Politecnico di Milano, Milano, Italy Keywords Demand management, Inventory control, MRP, Manufacturing systems, Computer simulation Abstract This research provides a literature review in the field of uncertainty dampening methods for manufacturing systems, and proposes a new model to improve materials management effectiveness in materials requirements planning environments. The literature review gives rise to a classification framework of the models along nine structural dimensions that refer to the safety buffer treatment, the environmental characteristics and the type of approach. On the basis of the classification framework, the proposed model provides guidelines for approaching the problem of dimensioning, positioning and managing safety stocks against demand uncertainty. The effectiveness of the proposed model has been tested by comparing it to the traditional approach, through a computer-based simulation. 1. Introduction The pressure of time-based competition, the relentless spread of information and communication technologies within organizations, and the speed of response required by the Internet-era orders fulfillment procedure, forced leading edge companies to recognize the need to be agile, by effectively planning manufacturing activities, so as to dampen the uncertainty coming from market-related variance. Among the various strategic levers to achieve the appropriate agility (i.e. internal processes redesign, workforce incentives, cooperation-oriented business models, etc.), the formation of virtual enterprises as a temporary alliance of companies, each providing a core competency, to take advantage of a market opportunity, is recognized as playing an important role. However, this makes materials management one of the most critical management areas, in that it plays a decisive role in continuous replenishment programs, vendor managed inventories programs, joint managed inventories programs, and collaborative planning, forecasting and replenishment business models (Johnson, 1999). For this reason, most companies, in recent years, have devoted a conspicuous amount of resources to implement and sometimes even re- implement enterprise resources planning and/or advanced planning systems, International Journal of Physical which allow them to effectively manage procurement activities to rapidly meet Distribution & Logistics customer needs (Stadtler and Kilger, 2000). The most popular available Management, Vol. 32 No. 7, 2002, pp. 556-576. software tools in this area often provide excellent algorithmic and technological # MCB UP Limited, 0960-0035 DOI 10.1108/09600030210442586 features to support management decisions, allowing customized planning

procedures and optimization algorithms. Notwithstanding, many Improving implementation case studies bear witnesses that advanced planning systems’ materials effectiveness is affected by a company’s databases consistency, which requires management appropriate parameters setting. The lack of consistency is relevant especially in the materials master files, e.g. mistaken definitions of lot-sizing, safety stock levels, and re-order points. This study aims to contribute by bridging the gap between theoretical 557 results and current industrial practice (Guide and Srivastava, 2000). It introduces a new method to ensure adequate safety stocks in materials requirements planning (MRP) environments, thus improving materials management effectiveness. The paper structure is as follows: section 2 presents the most popular dampening methods to reduce the effect of environmental uncertainty. Section 3 provides an analysis and classification of the safety stocks dampening method, while section 4 surveys the other dampening methods found in the literature. In section 5, the new model for safety stocks dimensioning, positioning and managing is presented. The effectiveness of the new model has been tested via simulation: section 6 introduces the experimental design, while results are discussed in section 7. Finally section 8 reports some concluding remarks and suggests future research paths. 2. Background In recent years many studies have been focused on uncertainty that plagues production systems and especially the ones managed via MRP systems (Yeung et al., 1998). A recommended direction for studying uncertainty is by focusing on dampening methods (Guide and Srivastava, 2000). Two different approaches have been proposed to reduce the effects of uncertainty (see Figure 1). One approach is based on slack resources, i.e. safety stocks, safety lead times and safety capacity. The popular method of safety stocks consists of defining a pre-determined inventory reserve, not considered in setting up master plans, and devoted to unexpected events. The safety lead time technique consists of using, for scheduling purposes, an average lead time incremented by a given amount. Safety capacity is defined as the capacity slack employed to prevent the production system from reaching complete saturation. Figure 1. Overall view of the dampening methods surveyed in literature

IJPDLM The second approach uses the technique of order management to create a 32,7 barrier against uncertainty, by resorting to master production schedule freezing or order over-planning. The master production schedule freezing technique consists of setting up a time horizon of the master production schedule (e.g. the next two weeks) within which no modification is allowed. The order over-planning technique consists of defining an additional quantity to be 558 added to each customer order when releasing the corresponding production order. Since this study is mainly focused on safety stocks, section 3 provides an analysis and classification of the main characteristics of the related studies, while section 4 presents a brief overview of the most relevant outcomes related to the other dampening methods. 3. Safety stocks dampening method: analysis and classification Table I provides a comprehensive overview of the most relevant and recent models from the literature, highlighting similarities and differences. The objective of this classification framework is twofold: it is aimed at assessing the most common options for each of the considered classification dimensions and it outlines the reasons for the prevalence of one option over the others. In Table I the models are listed on the columns and grouped according to the safety buffer treatment. The only exception relates to the model presented here, which has been listed as the last one because it is a special model from several points of view. In the rows of Table I, the classification dimensions of the models are listed, as well as all the options available for each dimension. It should be noted that the first dimension refers to the model focus, in terms of the safety buffer treatment type; the second group of dimensions relates to the environmental characteristics, while the last group relates to the type of approach. For each dimension a brief description is presented in the following subsections. 3.1 Safety buffer treatment When using safety stocks as a dampening method, three basic issues have to be handled, i.e. dimensioning, positioning and managing. The dimensioning issue deals with finding the appropriate value of safety stocks for each item. The positioning issue deals with finding the appropriate items in the bills of materials where safety stocks are to be placed. The managing issue deals with finding the appropriate time for safety stocks replenishments and with setting the appropriate delivery dates for replenishments. The majority of the surveyed studies carefully consider the dimensioning problem, while only a few contributions can be found about either positioning or managing. To this purpose, an interesting model is presented by Carlson and Yano (1986), who developed an algorithm that provides safety stocks guidelines for components whose schedules are re-planned each period and

Models Dimensions Options (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Models focus Safety buffer treatment Dimensioning X X X X X X X X X X X X X X Positioning X X X Managing X X Environmental characteristics Production management Push system X X X X X X X X X X Pull system X X X X X Production policy Make-to-stock X X X X X X X X X X X X X Make-to-order X X X Assemble-to-order X X X Production system Single product X X X X X X X X Multi-product X X X X X X X Product structure Single level BOM X X X X X X X X Multi-level BOM X X X X X X X Production capacity Unlimited X X X X X X X X X X Constrained X X X X X Sources of variance Demand X X X X X X X X X X X X X X Supply X Machines availability X X X Equipment performance X X Type of approach Performance measures Service level X X X X X X Overall cost X X X X X X X X X X X Type of model Analytic X X X X X X X X Simulation-based X X X X X X X ¨ Notes: (1) Brandolese and Cigolini (1999); (2) Dar-el and Malmborg (1991); (3) Wemmerlov and Whybark (1984); (4) Bourland and Yano (1994); (5) Hungh and Chang (1999): Carlson and Yano (1986); (7) Grubbstrom and Tang (1999); (8) Yano and Carlson (1987); (9) Molinder (1997); ¨ (10) Waker (1985); (11) Buzacott and Shanthikumar (1994); (12) Di Tillio (1998); (13) Vargas and Metters (1996); (14) Lagodimos and Anderson (1993); (15) Approach proposed here. BOM stands for ‘‘bill of materials’’ Classification found in literature uncertainty management models framework of the Table I. management Improving materials 559

IJPDLM whose successors have cyclic schedules. Buzacott and Shanthikumar (1994) 32,7 studied the influence of lead times and safety stocks on the performance of MRP systems and they concluded that safety stocks are more robust than safety lead times in coping with changes in customer requirements, when forecasts are unreliable. With reference to the positioning issue, Di Tillio (1998) introduced the notion 560 of the safety implicit availability as an instrument to face uncertainty: its value equals the difference between the stock availability at the beginning of the period and the cumulative requirements of LT successive periods, where LT represents the length of the production lead time. Lagodimos and Anderson (1993) studied the problem of positioning some budgeted safety stock in a multi-echelon network and they found that positioning all safety stock at the upper echelon is always optimal for serial networks. Finally, Vargas and Metters (1996) proposed a dual buffer approach to the safety stock problem: the former one is calculated for dimensioning purposes, the latter one for managing purposes. They found via simulation – supported also by data envelopment analysis – that the model outperforms the traditional ones in terms of service level, in the presence of a constrained capacity environment that allows backlogging. The approach presented in this paper aims at handling all the three issues of safety buffer treatment. 3.2 Production management The dimension related to production management represents the execution policy of the production system. In this area, the two most popular production management philosophies are pull and push systems. Under pull systems, production orders of each department are originated by consumption at the following department along the production chain. Under push systems, production orders are originated by the requirements (both in terms of customers orders and forecasts) of finished products processed through a MRP procedure. The literature review pointed out a number of relevant studies focusing on safety buffer dimensioning in a pull environment, probably due to the pioneering formula for safety stock dimensioning embedded in the basic EOQ model (Hax and Candea, 1984) which applies to pull systems. After Orlicky (1975) highlighted the main features related to safety stock dimensioning in push systems, a rich literature has been developed about this topic: in Table I some of the major contributions are noted. The approach proposed in this study refers to a push production system. 3.3 Production policy The make-to-stock production policy is employed when manufacturing flow time is much longer than the required delivery lead time, therefore production planning is driven by demand forecasts. Whenever the delivery lead time required by customers is longer than the manufacturing flow time, firms can

wait until customer orders are placed to start the production (make-to-order Improving policy). Finally, an assemble-to-order policy is a mixture of the previous two, in materials that the delivery lead time required by customers allows the firm to place management stocks on subassemblies. In this area, each literature contribution is focused on one specific policy, except for Yano and Carlson (1987), who analyzed the interactions between frequency of rescheduling and the role of safety stocks in terms of impact on 561 system’s performance. The results indicated that the fixed scheduling policy was more economical, even when the parameters would make flexibility advantageous. A further exception is Wacker (1985), who developed an analytical model for materials requirement planning systems under a multi- level bill of materials, and he proposed an empirical methodology to estimate the variance of components consumption. Notice that the approach proposed here is well suited for all the available production policies. 3.4 Production system Along the dimension devoted to the number of products, the models that consider either one product or different products without common subassemblies have been categorized as single-product, since each product can be treated separately from the others. Referring to Table I, notice that the literature of single-product models is not as extensive. The reason for this is twofold: on the one hand, single product models are always much simpler than the others; on the other hand, when dealing with safety buffers, products cannot be considered independent from each other. The proposed model is multi-product oriented. 3.5 Product structure As far as the product structure is concerned, when the bills of materials appear branched and extended, products exhibit a complex multi-level structure, while products are defined as single-level when they are manufactured starting from some basic components, all at the same level of the bill of materials. Among the most interesting contributions to this field, Grubbstrom (1998, ¨ 1999), Grubbstrom and Molinder (1996) and Grubbstrom and Tang (1999) ¨ ¨ applied a Laplace transform methodology to optimally dimension safety stocks, by striking a balance between consequences of expected stock-outs and expected inventory level, on the one hand, and set-up costs on the other one. While the single-level case is clearly the most studied, mainly in pull environments, in recent years interest has switched to the multistage case, which adequately fits realistic manufacturing systems without becoming intractable. This model pattern, at present, can be viewed as a best practice, since the safety buffer treatment at a given level in a bill of materials is tightly connected to the same problem at all the other levels. For this reason, the proposed model is a multi-level bill of materials one.

IJPDLM 3.6 Production capacity 32,7 All the production systems can produce – within a given planning period – a pre-defined maximum number of products, limited by the throughput rate of the bottleneck machine, and by the components availability. If market demand exceeds the system capacity, in at least one of the considered planning periods, the system is capacity constrained; otherwise the capacity is unlimited. 562 Although production capacity limits are an important driver of model likelihood, they also represent hurdles in model development, which is why most of the reviewed papers treat unconstrained production systems. Among the capacity constrained models, Bourland and Yano (1994) focused on the relative merits of capacity slacks (in terms of planned idle time in the schedule) and safety stocks, and they found that the former technique is not a cost- effective hedge against demand uncertainty. Brandolese and Cigolini (1999) presented an analytic approach to safety stocks dimensioning, based on the minimum inventory level needed to fulfill a sudden peak of finished products demand, where a two-stage capacity constrained production system is considered. 3.7 Sources of variance In the sources of uncertainty dimension, four types of variance have been considered in the literature, i.e. demand, supply, machine availability and equipment performance. Demand is recognized as the most difficult uncertainty source to control, as confirmed also by the great number of models addressing this problem. Few models even consider other sources of variance. Wacker (1985) considered both demand and supply uncertainty in quantity and over time. Hungh and Chang (1999) focused on machine availability and equipment performance and they presented a method for computing safety stocks level, motivated by complex manufacturing routes in semiconductor manufacturing. Finally, Molinder (1997) took into account uncertainty due to capacity loads, machines breakdowns, queuing effects, rework, etc. and applied the simulated annealing technique with the aim of optimizing safety stock level for each item. The proposed approach considers only demand as source of variance for the experimental campaign. 3.8 Performance measures The dimension linked to performance highlights two different measures. The first one relates to the service level, which is a customer-oriented criterion, in that it indicates customer satisfaction and it can be expressed in terms of either stock-out probability or stock-out quantity. The second measure refers to the total cost (e.g. production, overtime, sub-supply, stock-out, stock keeping, set- up cost). Given the well-known trade-off between cost and service, the majority of models consider the overall cost as the main performance measure, since stock- out costs are included. This alternative has been selected also for the proposed

approach. However, some models explicitly consider both measures (e.g. see Improving Yano and Carlson, 1987; Di Tillio, 1998). materials management 3.9 Type of model The type of model dimension focuses on whether the method of analysis of system behaviour is analytical or based on simulation. In the former case, the related calculations can be performed explicitly either in the time domain or in 563 the frequencies domain. In the latter case, such calculations are so complex that only computer simulation programs can perform them. For instance, Dar-el and Malmborg (1991) developed an analytical model where they compared the fixed safety stock approach to the practice of scheduling order releases earlier in the inventory cycle. This latter strategy resulted in lower expected carrying cost when combined with a fixed re-order quantity and when demand has a stationary probability distribution. On the other hand, Wemmerlov and Whybark (1984) studied 14 different single-stage ¨ production management procedures and suggested a simulation-based method to achieve a 100 percent service level. Given the inherent complexity of the considered environmental characteristics, the approach proposed in this paper is simulation-based. 4. Other dampening methods Although this study is mainly focused on the safety stock method, a brief overview of other approaches is provided for the sake of completeness. Within the area of order management, the main research paths that refer to the master production schedule freezing technique are directed towards the relationship between the frozen time interval, service level and stock holding costs, mainly depending on lot-sizing rules, forecasting models, cost structures and bill of materials structures (e.g. Sridharan and La Forge, 1989, 1994; Kadipasaoglu, 1995; Zhao et al., 1995; Zhao and Lam, 1997). A different technique in the area of order management is order over- planning (Verganti, 1997) which suggests orders to the production system of more than the average requests, to smooth down demand variance, with the objective of minimizing the overall cost and at the same time providing an adequate service level. Within the area of slack resources, a method for handling production uncertainty is the use of safety capacity, which is seldom recommended because of the extra-capacity carrying costs, even though Hungh and Chang (1999) pointed out that sandbagging by reserving capacity is very common and easy for manufacturing engineers. Another interesting technique is the use of safety lead time, which is often unconsciously practiced by many purchasing planners, just to get parts earlier and to prevent stock-outs (Ho et al., 1995). The majority of studies focused on comparing safety lead time and safety stocks, when demand quantity and supply time are uncertain. Various conclusions are reported in literature, depending on the considered manufacturing environments. Among the most

IJPDLM relevant contributions are those of Grasso and Taylor (1984), Graves (1988), 32,7 Wijngaard (1989), Buzacott and Shanthikumar (1994) and Molinder (1997). 5. The model The new model proposed here aims at providing answers to the three most popular issues among production managers whenever safety stocks are 564 considered (Vollman et al., 1992): (1) sizing safety stocks (dimensioning problem); (2) selecting the items to be endowed with safety stocks (positioning problem); (3) setting a releasing time for safety stocks replenishment orders (managing problem). The model proposes to split safety stocks into two components, i.e. ‘‘traditional’’ safety stocks and strategic stocks. Strategic stocks can help in managing the risk of stock-out whenever companies face a lumpy demand pattern, i.e. a steady demand where – at unpredictable time intervals – a significant peak takes place. For this purpose, according to Orlicky (1975), safety stocks (in MRP environments) should be sized on the basis of forecast error distribution. However, if error distribution is calculated, even considering the peaks, its standard deviation result will be considerably higher than the value representing the majority of the distribution (Bradford and Sugrue, 1997). This leads to higher safety stocks, which are costly and, unfortunately, unable to face the peaks of error whenever they occur. As a consequence, the proposed methodology suggests keeping two different kinds of buffer for two different objectives: (1) safety stocks for normal demand forecast error; and (2) strategic stocks for high forecast errors in important orders. 5.1 Definitions and assumptions The model assumes that the inventory policy has been stated, i.e. it is known whether each item is managed according to a push or pull policy. The MRP procedure renews demand forecasts in each period and it handles both push and pull managed items, either according to the specific lot-sizing policy, or according to a lot-for-lot (or similar) policy, with the aim of keeping stocks as low as possible. The practice of speculative stock keeping is treated as an exception. The requirements chain time (RCT) of pull managed items is defined as the length, over time, of the deepest bill of materials branch made up of push managed items: e.g. assume that the finished product manufacturing lead time equals two periods, and the raw materials procurement lead time equals three periods, if raw materials are push managed items, while the finished product is pull managed, the requirements chain time equals five periods. As a

consequence, under a given inventory policy, each pull managed item is Improving provided with its own requirements chain time. materials Moreover – once the inventory policy has been stated – for a given finished management product, the highest level pull managed items can be determined. These items are defined here as LIMP ones (i.e. LInked to the Market in a Pull fashion), since their consumption is connected in a straightforward manner – i.e. it is pegged – to a specific customer order: e.g. under a make-to-stock production policy, only 565 finished products are LIMP ones, while under a make-to-order production policy, there are no LIMP items. For the sake of clarity, the following notation is introduced: t = time period index; t = LIMP item index; RCTj = requirements chain time of item j; P = number of finished products; Bp = number of branches that make up the bill of materials of finished product p, p = 1, . . ., P; LTh = lead time of item h; Ib(j,p) = set of items that are between j and p along branch b; forp(x,y) = demand forecast of p for period x formulated in period y; reqj(t) = requirement of item j in period t. The aggregate utilization factor of item j in product p along branch b, AUFjpb, is defined as the multiplication of quantity per parent item (UF) for all the items between j and p along branch b: Y AUFjpb ¼ UFj Á UFh Á ð1Þ h2Ib ðjÀpÞ For example, assume that a finished product is made up of two units of a subassembly, that in turn is made up of three units of a component, which is a LIMP item: the AUF equals 2 6 3 = 6 units. The branch lead time of the LIMP item j in product p along branch b, LTb(j, p) is defined as the sum of the lead times of the items that are between j and p along branch b: X LTb ðj; pÞ ¼ LTh þ LTp Á ð2Þ h2Ib ðjÀpÞ The branch lead time expresses how many periods before the delivery of p, that item j is consumed. If finished product p has to be delivered in period t, item j’s consumption takes place in as many periods as the distinct values of LTb(j, p) in the bill of materials. Since j is a LIMP item, the LTb(j, p) value is an upper bound on the delivery lead time of p, at least for some b. The replenishment process of item j lasts

IJPDLM RCTj time periods and the decision of releasing a replenishment order of j in 32,7 period t is based upon the forecast formulated in period t and related to period j’s consumption from period t + 1 to period t + RCTj. In turn, consumption of j from t + 1 to t + RCTj depends upon the finished product consumption from t + LTb(j, p) + 1 to t + LTb(j, p) + RCTj for each branch b. 566 Item j’s safety stocks should shelter production system whenever j’s effective consumption during RCTj is greater than the one previously forecast. The general formula for the calculus of the difference between the effective consumption and the forecast requirement of a LIMP item, during its entire requirements chain time, is therefore: RCTj X X X P Bp Áj ðtÞ ¼ AUFjpb Á ½forp ðt þ i þ LTb ðj; pÞ; t þ iÞÀ i¼1 p¼1 b¼1 ð3Þ forp ðt þ i þ LTb ðj; pÞ; tÞ: Formula (3) expresses the cumulative effect of forecast errors in RCTj periods, which are uncovered by a replenishment order of j released in t. The two forecast values in (3) concern the same delivery period. The former one determines an effective consumption of j in period t + 1, since it is processed LTb(j, p) periods before the delivery of p. The latter one is considered when a decision to release a replenishment order of j in period t is to be taken. By multiplying the difference of the finished product forecasts by AUFjpb, the gap between the actual and the forecast consumption of j, is obtained. This gap is due to the branch b of the bill of materials of p. To dimension safety stocks, we need to extend this gap to all the positions of j in the bills of materials of all the finished products. This is the reason for summing over all the branches b and over all the finished products p. The last step consists of summing the gaps, previously calculated, over the periods that fall within the RCTj, starting from period t. The cumulative gap within RCTj measures the global unforeseen consumption, against which the company can protect itself only by safety stocks, because a production order placed in period t will be delivered only after a complete requirements chain time. 5.2 Dimensioning problem Given the previous definitions and notations, now the new safety stock dimensioning algorithm is introduced, considering some LIMP item j. (1) Calculate Áj ðtÞ for each period t of the time series. (2) Using the Áj ðtÞ values as a sample space, build the cumulative " probability curve of Áj and calculate its mean value Áj over time.

(3) Calculate the value of coefficient j , according to (4): Improving " Áj materials j ¼ ð4Þ management XXP Bp AUFjpb Áfor p ðt þ LTb ðj; pÞ; tÞ p¼1 b¼1 567

IJPDLM where for p ðt þ LTb ðj; pÞ; tÞ represents the mean value over time of 32,7 product p’s demand forecasts related to LTb(j, p) periods in the future. Notice that, when forecasts are biased, j absolute value is high. (4) If jj j > Ã, where Ã is a convenient threshold set by experiments, a systematic error in the forecast requirement of j is reflected in the 568 withdrawal orders of its PLMs on items. A calibration of the forecast instrument is needed: Áj should be subtracted to Áj ðtÞ for each t, thus moving forecast average to zero. In the meanwhile, j’s requirement in period t is corrected by adding the ratio Áj =RCTj . The adjusted requirement of j, req Ãj ðtÞ is calculated as: XX P Bp Áj req Ãj ðtÞ ¼ AUFjpb Á forp ðt þ LTb ðj; pÞ þ RCTj ; tÞ þ : p¼1 b¼1 RCTj ð5Þ If jj j < Ã, forecasts are unbiased and no requirements’ adjustment is needed. (5) Given the cumulative probability curve of Áj (built at step 2 and eventually corrected at step 4), find out the value ÁÃj whose cumulative probability is equal to the service level required by the company. Safety stocks of item j should be set equal to ÁÃj . In order to dimension the strategic stocks for some LIMP item j, the following algorithm can be introduced: . The absolute forecast error must be calculated, i.e. the difference between the expected and the actual consumption of j in each time period. . Then the probability density function of the absolute error must be drawn, together with its mean value. A peak of demand is a period whose forecast error exceeds a so-called upper bound, where is the double of the mean absolute forecast error, diminished by the minimum absolute forecast error. The reason for using the absolute error, rather than the one with sign, lies in that the absolute error provides a more severe evaluation of forecast error, since the mean absolute error is higher than the mean error with sign. Notice that, when a peak of demand occurs, the demand is remarkably higher than the forecast, and the peak is a possible source of stock-out, which makes it dangerous. If period t is a peak of demand, we equal its forecast error to . In this way, we make forecast error distribution quasi-symmetric and we also force its maximum value to equal . . For each peak period t, the exposition due to the peak can be calculated as the difference between the error at period t and the mean error. So, we can calculate the exposition of each peak.

. Then the cumulative probability curve of the exposition values is built. Improving On the vertical axis, the value of the required service level is found out materials and the corresponding value on the abscissa is sought. Strategic stocks management should amount to this value. In this way the dimensioning of both safety and strategic stocks can be performed for all the LIMP items considered. 569 5.3 Positioning problem When handling the safety (and strategic) stocks positioning problem in MRP environments, the first issue to be considered lies in that they should be positioned on pull managed items. This method seems to be the most effective, since usually push managed items are customized (e.g. standard gears marked with customers’ trademarks) or they are hard to keep as stock (e.g. large-size or fragile products, perishable goods, etc.). Moreover, safety stocks are to be placed on LIMP items. This allows the manufacturing system to react on time whenever a forecast error occurs. If safety stocks are positioned on lower levels of bills of materials, they are useless when a forecast error occurs, since the lead time required to manufacture finished products starting from components (kept as safety stocks) is longer than the delivery time required by the market. For example in make-to-stock environments, protections against uncertainty have to be placed at the finished product level, whereas, under assemble-to-order production policies, safety stocks have to be placed on subassemblies. Lower-level pull managed items do not require to be provided with their own safety stocks. This is true because the amount of safety stock placed on LIMP items is high to help cope with forecast errors during the cumulative requirements chain time of pull managed items. Alternatively, safety stocks can be split among all the pull managed items provided that each safety stock covers the forecast error during each pull managed item requirements chain time. However the optimal course of action in this field is system-specific and it should be carefully evaluated through simulation. Finally, recalling that the consumption of LIMP items is directly pegged to finished products demand, strategic stocks should be positioned, at least, on the LIMP items, so giving the manufacturing system enough time to assemble finished products starting from stocks, whenever a forecast error takes place. 5.4 Managing problem Another important issue is the definition of the most convenient delay in releasing replenishments. When a replenishment of safety stocks is required (e.g. demand has been repetitively under-estimated), the related order can miss the needed components, since it generates an unexpected withdrawal of lower level items, which is normally uncovered by appropriate inventory levels, thus introducing nervousness in the MRP procedure (Ho and Carter, 1996; Ho and Ireland, 1993; Ho et al., 1995; Portioli, 1997). A way of avoiding MRP

32,7 570 Table II. IJPDLM strategic stocks both safety and Overall view of the guidelines proposed for Production policy Issue Make-to-stock Assemble-to-order Make-to-order Dimensioning Safety stocks Proportional to: actual demand error Proportional to: actual requirements Proportional to: actual requirements distribution and (for lower level, if error distribution; requirements chain error distribution; requirements chain any) actual requirements error time down to pipeline length time down to pipeline length distribution;requirements chain time (depending on simulation results) (depending on simulation results) down to pipeline length (depending on simulation results) Strategic stocks Proportional to the actual distribution Proportional to the actual distribution Proportional to the actual distribution of demand error peaks of requirements error peaks of requirements error peaks Positioning Safety stocks Finished products: always. Lower Subassemblies: always. Lower levels Components and/or raw materials levels down to components/raw down to components/raw materials: materials: to be evaluated via to be evaluated via simulation simulation Strategic stocks Finished products Subassemblies Components and/or raw materials Managing Safety stocks From as soon as possible to never, From as soon as possible to never, From as soon as possible to never, depending on simulation results depending on simulation results depending on simulation results Strategic stocks As soon as possible As soon as possible As soon as possible

nervousness lies in setting replenishment orders due dates as far in the future Improving as the cumulative requirements chain time. materials Even in this case trade-offs occur. On the one hand, replenishment orders management can be released as soon as safety stocks are withdrawn to guarantee the appropriate service level. Yet this practice introduces nervousness into the production plan. On the other hand, safety stocks can never be replenished to smooth down MRP nervousness, but this policy runs the risk of stock-out. 571 Although this latter case does not fit very well in practical manufacturing environments, the alternative is clearly stated and the optimum compromise highly depends on the specific production system considered, again requiring some form of simulation evaluation. Finally, as far as the strategic stocks are concerned, replenishment orders should be released as soon as possible, regardless the production policy considered. 5.5 Guidelines summary Table II provides the reader with an overall view of the strategies proposed for both safety and strategic stocks with reference to all the three issues, i.e. dimensioning, positioning and managing. With reference to safety stocks treatment, the appropriate set of recommended guidelines follows: . Whatever the manufacturing environment, LIMP items should be endowed with safety stocks. Even the lower level items can be endowed with safety stocks: this decision impacts on the safety stocks dimensioning and it is system-specific, so it should be treated via simulation, which also helps in defining the replenishment policy (i.e. orders due date and release date). . Only pull managed items should be considered as candidates for safety stocks positioning. . Safety stocks sizes are proportional to RCT, to avoid stock-out on the lower-level push managed items. . The proposed algebraic formula for LIMP items’ safety stock dimensioning recalls the one embedded in the basic EOQ model (Hax and Candea, 1984), with the main difference that in materials requirements planning environments safety stock calculus is based on the forecast error Áj instead of on demand pattern, as early suggested by Orlicky (1975). Moreover, the actual forecast error distribution is to be considered, instead of the Normal distribution assumed by the traditional safety stock dimensioning formula (Wild, 1997), but seldom encountered in real-life manufacturing environments. . Finally, the safety stock dimensioning procedure should consider the renewal of forecasting quantities per item and per period causes disturbances during MRP execution, which is fairly handled by the proposed algebraic formula.

IJPDLM With reference to strategic stocks treatment, the appropriate set of recommended 32,7 guidelines follows: . In a similar way to what was assessed about safety stocks, and for the same reasons, strategic stocks are also to be placed on LIMP items: in a make-to-stock environment, finished product should be endowed with strategic stocks; in an assembly-to-order environment subassemblies 572 should do, etc. . Strategic stocks are to be sized on the basis of the statistical distribution of the peaks of forecast error, to reach a given service level. 6. Experimental design The safety stocks dimensioning, positioning and managing model is tested to prove its effectiveness via simulation. The experimental environment is an assemble-to-order system with four finished products types. Only LIMP items are endowed with safety and strategic stocks, and the replenishment policy is ‘‘as soon as possible’’. Service level is set at 95 percent and the mean demand is set at 1,000 units per period, with a probability of demand peak occurrence set at 10 percent per period. The test horizon is made up of 52 periods, i.e. a year-long horizon is considered, broken in weekly periods. A 34 factorial design with replicates (Montgomery, 1991) has been set-up to study the effects of four factors on mean safety stock amount and total stock- Traditional model New model Mean stock Mean stock-out Mean stock Mean stock-out i1 16338 285 12939 1291 i2 11144 0 5895 148 Table III. i3 5087 36 3705 112 Experimental results i4 7860 585 9915 233 Figure 2. How trade-off between stock level and stock-out quantity (in terms of actual service level) changes passing from traditional model application to the new proposed model application (arrow’s direction)

out quantity, assumed as performance indexes. The factors can be divided into Improving two groups, i.e. decision variables and environmental variables. Three levels materials are selected for each factor. The environmental variables are: management (1) the forecast error coefficient of variation (levels: 10 percent, 50 percent, 100 percent); (2) the ratio between basic demand standard deviation and mean demand 573 (levels: 1 percent, 10 percent, 20 percent); and (3) the percentage exposition due to demand peak out of demand mean (levels: 0 percent, 150 percent, 300 percent). The levels of the environmental variables have been selected from a larger number of tested values. The only decision variable is the stocks dimensioning model and its levels are: zero safety stocks, traditional model (i.e. safety stock dimensioned under the hypothesis of normal distribution of forecast error; absence of strategic stock) and the new proposed model (i.e. both safety and strategic stocks sized). In each experiment, safety stock amount and total stock-out quantity of four item types have been tallied. Type i1 is representative of first level raw materials and it is present in 75 percent of bills of materials. Type i2 appears as raw material in 75 percent of bills of materials, at third level in 50 percent and at fourth level in 25 percent of bills of materials. Type i3 appears as raw material in 25 percent of bills of materials, at second level. Type i4 appears as second level subassembly in 25 percent of bills of materials. We suppose that i1, i2, i3, i4 are the sole LIMP item types of the system and that all the other item types are push managed items. 7. Results Table III shows the experimental results corresponding to the traditional model and the new proposed one. Figure 2 shows the trade-off between the mean stock level and the stock-out amount, as resulted from the experimental analysis, in case of traditional model and new model application. The new model performs more for item types i2 and i3. By comparison to the traditional model, that over-sizes safety stocks obtaining a service level that is more than 95 percent, the new model reduces stock (–47 percent for i2 and –27 percent for i3), so bringing service level closer to the target. This is a consequence of the presence of error peaks which determine, in the traditional model, an increase in forecast error standard deviation and, as a consequence, in safety stock. As far as type i4 is concerned, the traditional model has relatively low performance. This is due to the fact that type i4 is representative of subassemblies, which are parents of push managed items. As a consequence, sizing i4’s safety stocks on the basis of the sole i4’s lead time is not enough to balance forecast error, whereas in the new model, i4’s safety stocks are sized on

IJPDLM the basis of the requirements chain time. This determines an increase in safety 32,7 stocks value (+26 percent), so that the target service level is reached. Finally, type i1 case is slightly more complex. As for i2 and i3, safety stocks are over-sized and the actual service level is close to 99 percent in the traditional model. Applying the new methodology, safety stocks level is reduced (–21 percent) and stock-out increases. Although the new model sets 95 percent 574 service level, demand generation determines an effective service level (94.6 percent) which is slightly lower than the target. 8. Concluding remarks and future research paths The research reported in this paper considers an appropriate MRP parameters setting as a prerequisite for achieving agility in the manufacturing management area. To this purpose, both lean production management techniques and agility-oriented decision support systems usually provide plant managers with a remarkably wide set of planning strategies, each of them assuming a conspicuous data consistency. For this reason, this research has focused on developing a new methodology for buffering against uncertainty coming from demand variance. In developing the model, the so-called LIMP items (i.e. the items whose consumption is connected in a straightforward manner to a specific customer order) are initially introduced, in that they prove to be the most critical ones to be sandbagged by resorting to safety stocks, since they are directly exposed to forecast errors. Furthermore, from the managing perspective, additional safety stocks may be required also on other pull managed items below LIMP items in the bill of materials, to avoid system nervousness in case a replenishment order is placed to refill safety stocks. While safety stocks are dimensioned to cope with random variation of the steady component of demand, strategic stocks should be kept to face unforeseen peaks of demand. Also strategic stocks are recommended to be placed on LIMP items. From the dimensioning perspective, the new model is based on two steps. The first one consists in analyzing the forecast errors time series, to detect demand peaks. By studying the probability distribution of peaks, under a given service level, the strategic stock to protect against unforeseen demand peaks is calculated. Therefore demand peaks do not need to be taken into account any more. The second step of the procedure consists in determining the safety stocks level to face the variability of steady demand. The actual statistical distribution of the forecasting errors is to be calculated. Then, once the service level for steady demand is established (which may be different from that one for peaks) the proposed algorithm allows to calculate the appropriate safety stocks level for LIMP items. The difference between the traditional safety stocks model and the new one presented here is almost threefold. First, the new model separately calculates the peak and the steady component of demand, thus resulting in two different types of buffering approaches. Strategic and safety stocks are computed

differently since their purposes for sandbagging are different, i.e. protecting Improving against unpredictable peaks in the former case, and dampening the forecast materials error steady demand in the latter one. In this situation the traditional model management considers only one source of variance. Second, the new model is based on the actual distribution of forecast error, on which the service level is measured. The traditional approach hypothesizes a normal distribution, widely encountered in literature, but far from being true 575 in real-life environments; this results in an actual service level either lower or higher than the required value. Third, in safety stocks dimensioning, the new model takes into account the renewal of the forecast quantities per item and per period. Indeed the forecasts’ renewal is popular among industrial practitioners, and yet introduces relevant disturbances, during MRP execution, in the replenishment plans of items that are produced and consumed on the basis of the forecasts. The literature review highlights that the majority of researchers have seldom considered this practice. To test the effectiveness of the new proposed model in improving materials management in push production systems, a full factorial design has been employed. Simulation, performed after a series of pilot tests to select the appropriate values for the variables, highlighted that the new model overcomes the traditional one. In particular, for about 50 percent of the considered sample items, it significantly reduces stock amount, without increasing the stock-out quantity. Despite the strengths of the model when compared to the currently available literature, the way the model itself can be improved is significant. First of all, the methodology should be extended to pull managed items that are not LIMP ones. Key factors suggesting which pull managed items are to be endowed with safety stocks should be provided, eventually by carrying out an economical analysis of experimental results. Second, the proposed model should be enhanced, by introducing lead time uncertainty and by considering the eventual interactions between demand and lead time variance. Third, simulation should be extended to different scenarios and parameter settings, mainly to make-to-stock environments, where safety stocks are placed on the finished products, and to different product structures, so to allow more exhaustive conclusions. Fourth, the implementation of the model in a real-life manufacturing environment could lead to a definitive judgment about its effectiveness. Organizational problems originating from safety and strategic stocks separate management could be observed and solved in this real context and a thorough comparison between separate and joint management of the two types of stocks could be made.

IJPDLM References 32,7 Bourland, K.E. and Yano, C.A. (1994), ‘‘The strategic use of capacity slack in the economic lot scheduling problem with random demand’’, Management Science, Vol. 40 No. 12, pp. 1690-704 Bradford, J.W. and Sugrue, P.K. (1997), ‘‘Estimating the demand pattern for C category items’’, Journal of the Operation Research Society, Vol. 48, pp. 530-2. 576 Brandolese, A. and Cigolini, R. (1999), ‘‘A new model for strategic management of inventories subject to peaks in market demand’’, International Journal of Production Research, Vol. 37 No. 8, pp. 1859-80. Buzacott, J.A. and Shanthikumar, J.G. (1994), ‘‘Safety stock versus safety time in MRP controlled production systems’’, Management Science, Vol. 40 No. 12, pp. 1678-89. Carlson, R.C. and Yano, C.A. (1986), ‘‘Safety stocks in MRP-systems with emergency set-ups for components’’, Management Science, Vol. 32 No. 4, pp. 403-12. Dar-El, E.M. and Malmborg, C.J. (1991), ‘‘Improved strategy for service level based safety stock determination’’, Production Planning and Control, Vol. 2 No. 2, pp. 116-21. Di Tillio, S. (1998), ‘‘Safety stocks and lot-sizing rule in MRP environments under uncertainty. A proposal and evaluation of new resolution approaches’’, Master’s thesis (in Italian), Politecnico di Milano. Grasso, E.T. and Taylor, B.W. (1984), ‘‘A simulation based experimental investigation of supply/ timing uncertainty in MRP systems’’, International Journal of Production Research, Vol. 22 No. 3, pp. 485-97. Graves, S.C. (1988), ‘‘Safety stocks in manufacturing systems’’, Journal of Manufacturing and Operations Management, Vol. 1, pp. 67-101. Grubbstrom, R.W. (1998), ‘‘A net present value approach to safety stocks in planned production’’, ¨ International Journal of Production Economics, Vol. 56-57, pp. 213-29. Grubbstrom, R.W. (1999), ‘‘A net present value approach to safety stocks in a multi-level MRP ¨ system’’, International Journal of Production Economics, Vol. 59, pp. 361-75. Grubbstrom, R.W. and Molinder, A. (1996), ‘‘Safety production plans in MRP systems using ¨ transform methodology’’, International Journal of Production Economics, Vol. 46-47, pp. 297-309. Grubbstrom, R.W. and Tang, O. (1999), ‘‘Further developments on safety stocks in an MRP ¨ system applying Laplace transforms and input-output analysis’’, International Journal of Production Economics, Vol. 60-61, pp. 381-7. Guide, V.D.R. and Srivastava, R. (2000), ‘‘A review of techniques for buffering against uncertainty with MRP systems’’, Production Planning and Control, Vol. 11 No. 3, pp. 223-33. Hax, A. and Candea, D. (1984), Production and Inventory Management, Prentice-Hall, New York, NY. Ho, C. and Carter, P.L. (1996), ‘‘An investigation of alternative dampening procedures to cope with MRP systems nervousness’’, International Journal of Production Research, Vol. 34, pp. 137-56. Ho, C. and Ireland, T.C. (1993), ‘‘A diagnostic analysis of the impact of forecast errors on production planning via MRP system nervousness’’, Production Planning and Control, Vol. 4, pp. 311-22. Ho, C., Law, W.K. and Rampal, R. (1995), ‘‘Uncertainty-dampening methods for reducing MRP system nervousness’’, International Journal of Production Research, Vol. 33 No. 2, pp. 483-96.

The research register for this journal is available at The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/researchregisters http://www.emeraldinsight.com/0960-0035.htm Measuring supply chain agility Measuring supply chain in the virtual organization agility Mary Margaret Weber Anderson Schools of Management, University of New Mexico, 577 Albuquerque, New Mexico, USA Received June 2001 Keywords Agile production, Flexibility, Market share, Direct marketing Revised April 2002 Abstract The need for increased flexibility in responding to market demand is driving a heightened interest in virtual, or agile organizations. However, agile response in the supply chain may not always be necessary and may not always be a better alternative than more traditional organizational structures. The model proposed in this paper provides a means of measuring both the need for agility and how agile an organization actually is. This is accomplished through the use of a hierarchical model that details with increasing specificity sources and levels of variance in the supply chain. As the ability to control specified variances increase, the need for agility decreases. Increasingly, business organizations are finding it necessary to be extremely flexible in responding to changes in the market environment. This need for flexibility has heightened, and continues to heighten, interest in the virtual organization. However, it is also apparent the use of virtual organizations may not always be a reasonable, or better, alternative than more traditional organizations. Therefore, a means of determining if the development of a virtual organization is necessary, or even practical, is needed. The purpose of this paper is to propose a means of measuring an organization’s need for and ability to develop an agile business strategy within the context of a virtual organization. Virtual organizations Virtual organizations, also referred to as agile, outsourced, or seamless organizations, have been variously defined in the literature. Greis and Kasarda (1997, p. 58) define them as ‘‘legally separate but operationally interdependent companies focused on responding to a market opportunity’’. Similarly, Fitzpatrick and Burke (2000, p. 13), state that ‘‘virtual organizations attempt to create a network or coalition of suppliers, manufacturers, and administrative services to accomplish specific objectives’’. According to Hunt (2000, p. 18), a virtual organization describes ‘‘almost any association of people who are linked, not by face-to-face relationships but by sharing information through electronic networks’’. Kavan et al. (1999, p. 73) describe the outsourced organization as ‘‘intrusting elements of a firms’ value chain to an external corporate entity’’. What all these definitions have in common is the idea that a virtual International Journal of Physical organization is a loosely related group of companies formed to enable Distribution & Logistics Management, collaboration toward mutually agreed on goals (Greis and Kasarda, 1997). Vol. 32 No. 7, 2002, pp. 577-590. # MCB UP Limited, 0960-0035 Virtual organizations promote adaptability, flexibility, and the ability to react DOI 10.1108/09600030210442595

IJPDLM quickly to changes in the market (Grabowski and Roberts, 1999). They may 32,7 also provide substantial cost reductions, higher productivity, and greater satisfaction of both employees and customers (Kavan et al., 1999) because they provide the means to create greater focus and integration (Magretta, 1998). Agility in virtual organizations can be contrasted with the move toward leanness. Leanness is about accomplishing more with fewer resources. 578 However, lean strategies work best in environments with high volume and high predictability (Christensen, 2000). Many organizations that have pursued leanness have also become very fragile (Greis and Kasarda, 1997), leading to an inability to respond quickly and flexibly to changes in the market. Thus, the move toward agility is, in part, a response to the recognition that an organization can be too lean. It is also an attempt to manage the changes that are occurring as companies continue the move toward e-business and seek to drive down inventory and logistics costs (Schwartz, 2001). However, the ability to build agile relationships has not developed as rapidly as anticipated, perhaps because as noted by both Nellore and Motwani (1999) and Biggs (2001), the development of technology to manage the agile relationships is still in progress. Further, virtual organizations may not work on the ‘‘bleeding edge’’ because of the requirement for management coordination among separate businesses (Christensen, 2000). Measuring variances in the supply chain The push toward greater collaboration and coordination of channel relationships is well documented (for example, Ellram, 1993; La Londe et al., 1988; Lambert et al., 1998). Lambert and Cooper (2000, p. 65) note that ‘‘One of the most significant paradigm shifts of modern business management is that individual businesses no longer compete as solely autonomous entities, but rather as supply chains’’. Given that developing agile, or virtual, relationships in the supply chain are a significant departure from more traditional business strategies, it seems reasonable that organizations should determine the need for and ability to develop agility in supply chain relationships before attempting changes in business strategy. The purpose of this paper is to propose a model that can be used to measure the need for agility in an organization as well as isolate the sources of unpredictability that drive the organization toward agile responsiveness. The model described below is based on a hierarchical approach to analyzing sources of variance in the supply chain. Hierarchical approaches to analyzing variances have been suggested for use in both management accounting (Shank and Churchill, 1977; Shank and Govindarajan, 1989) and marketing management (Hulbert and Toy, 1977; Bentz and Lusch, 1980; Lusch and Bentz, 1984; Weber, 1996). They depend on an analysis of activity that adds increasing levels of complexity as sources of variance are explored in order that large, off-setting variances might be discovered. The work in the field of marketing management is significant because it made use of a revised plan; a technique recognized as based on the generally accepted notion of flexible

budgeting (Ijiri et al., 1968; Hewitt, 1994; Ratnatunga et al., 1988). By using this Measuring technique, the analysis is no longer limited to comparisons between planned supply chain and actual performance. Instead, two types of comparisons can be made. The agility first is a comparison between actual performance and what should have been done given the circumstances encountered during the period. The second is a comparison between the original plan and what should have been planned, thus revealing forecasting errors. The idea of a revised plan to analyze system 579 variance was then applied to inventory-related variances in the buyer-supplier relationship to determine the opportunity cost of poor planning in the supply chain as well as the cost of failure to perform (Weber, 2000). This work combines the inventory-related variance analysis with the marketing performance analysis to provide a means of measuring agility, how well the organization was able to respond to actual market conditions; and predictability, how well the organization can plan for future circumstances. An increase in predictability will reduce the need for an organization to achieve high levels of agility; instead it can focus on leanness in the supply chain. Likewise, as unpredictability increases, so to will the need for agile response. Actual vs required vs planned performance Traditionally, variances in the supply chain have been measured by comparing actual to planned performance. The weakness of this approach is that it fails to address the problem of planning inaccuracies. Small differences between planned and actual performance do not guarantee that actual demand in the market was satisfied. Likewise, while large differences between planned and actual performance may be the result of failure to perform, they could also be due to superior performance in meeting unexpected and unplanned for demand. The first two levels of the model are shown in Figure 1. It makes use of three types of measurements: actual, required, and planned (see Table I for variable and subscript definitions). Variables with the subscript ‘‘a’’ define actual performance during the period. For example Xa defines the percentage of market demand actually satisfied through production or delivery of inventory. Variables with the subscript ‘‘r’’ define the required performance needed to satisfy the market during the period. Note that required performance may not equal actual performance. Required performance measures what should have been done – even if it was not planned for or actually accomplished. For example, Xr represents the percentage of market demand that occurred and that should have been met – even if the organization could not do so. Finally, variables with the subscript ‘‘p’’ represent planned performance and are Figure 1. Performance versus planning variances

IJPDLM Variable definitions Subscript definitions 32,7 P Price per unit a Actual performance during the period V Variable cost per unit r Required performance based on the market C P-V; per unit contribution margin demand that occurred S Percentage of units produced in usable p Planned performance as determined at the condition beginning of the period 580 M Market size measured in units Q M*S; potential sales in units N Q*C; total contribution value of Table I. inventory available when needed Variable and subscript T Percentage of time inventory available definitions when needed determined a priori as part of a strategic or tactical plan. For example, Xp specifies the percentage of market demand that the organization planned on satisfying at the beginning of the period. Making the distinction among actual, required, and planned performance provides a means of measuring both agility and predictability. Performance variance measures agility – the difference between actual performance and required demand in the market. Small performance variances indicate that an organization is very agile in responding to unexpected events. In other words, the organization was able to correctly and adequately respond to market conditions – regardless of whether they had been forecast. Planning variance measures predictability – the difference between performance originally planned at the beginning of the period and required demand. Small planning variances indicate that the organization was able to accurately forecast actual events in the market. Consistently small planning variances over time demonstrate the ability of an organization to move toward lean operations rather than developing a need to be agile. In order to isolate sources of variances, both performance and planning variances can be broken down into more and more detailed variances in subsequent levels of the model (Figures 2 and 3) as described below. Variance descriptions Total variance At the first level of the model, total variance measures the difference between planned and actual needs. To understand where the sources of total system variance lie, total variance needs to be broken into its components, performance variance and planning variance, via the use of the required measurements as shown at the second level of the model. Performance variance At level II, performance variance measures system agility. A negative variance indicates an overall system shortfall in the right value of total inventory available at the right time. A positive variance represents a system

Measuring supply chain agility 581 Figure 2. Performance variances

IJPDLM 32,7 582 Figure 3. Planning variances

over-response, or the creation of excess inventory value. As shown below, this Measuring variance can be broken down into more detailed variances. The closer these supply chain individual performance variances are to zero the more agile the logistics system in responding to unexpected changes in demand. However, as these variances agility are aggregated at higher levels of the model, care must be taken to note any large, offsetting variances at lower levels of the model. 583 Planning variance Also at level II, planning variance measures the difference between planned events and what should have been planned if all unforeseen circumstances had been taken into account. Measuring the original plan against requirements allows an organization to determine where planning has been inadequate. Improvements in planning can then reduce variability throughout the system. A positive planning variance means requirements exceeded planned performance. A negative variance shows that actual market requirements were less than what had been planned. It can be expected that as planning variance increases, so will the need for system agility. As shown in Figures 2 and 3, both performance and planning variance can be broken down into three more detailed variances: time, contribution, and quantity variance (level III). Contribution and quantity variances can be further detailed at level IV of the model. Time variance The first variance detailed at level III of the model is time variance, measuring the percentage of time inventory is available when needed. A positive performance time variance indicates the percentage of time in the period that overages existed. A negative variance shows the percentage of time inventory was not available – stock-outs. A positive planning time variance indicates that needed availability exceeded planned availability; a negative variance that planned availability was over-estimated given the circumstances experienced. Contribution variance The second type of variances detailed at level III, performance and planning contribution variances, represent the difference between actual or planned contribution margins, respectively, and requirements given the environment experienced during the period in question. A positive performance contribution margin variance means contribution margin was higher than it should have been given the conditions experienced in the market. A negative performance contribution margin variance shows that the organization did not reach the achievable contribution margin. A positive planning performance variance means that the required contribution necessary to meet goals exceeded the planned contribution; a negative variance, the reverse. Price and cost variances At the fourth level of the model, contribution variances can be further separated into price and cost variances. This level of detail allows examination

IJPDLM of whether variances in contribution margins are due to changes in price or 32,7 cost, or a combination of both. A positive performance price variance indicates that actual unit price received exceeded the market requirements – the price was too high. A negative performance price variance shows that actual price was less than what the market would accept. A positive planning price variance denotes that the price the market would accept exceeded the planned 584 price; a negative variance the reverse. Cost variances can be similarly defined. A positive performance cost variance demonstrates that the organization was able to achieve greater cost efficiency than what should have been realized under the conditions experienced. A negative cost performance variance demonstrates cost inefficiency. A positive planning cost variance indicates that planned costs exceeded required; a negative variance than planned costs were not sufficient given actual market conditions. Quantity variance The last variance detailed at level III is quantity variance. Performance quantity variance measures the difference between the quantity of inventory actually available and the inventory quantity needed during the period. A positive variance indicates excess inventory; a negative variance, shortages. Planning quantity variance shows the difference between planned and required inventory availability and indicates a failure to adequately predict inventory needs during the period. A positive planning quantity variance shows that inventory needed fell short of planned; a negative variance demonstrates an under-prediction of inventory needs. Condition variance Performance condition variance represents a difference in the percentage of usable inventory available and the percentage of usable inventory needed to satisfy demand during the period. A positive performance condition variance indicates that necessary production standards were exceeded; a negative performance condition variance represents a failure to meet production standards. Similarly, positive and negative planning condition variances represent a greater or lesser need to meet production standards, respectively, than that initially planned. Amount variance The amount variance measures the difference between actual or planning production/delivery levels, measured in number of units, and what was actually needed. A positive performance amount variance shows that actual production/delivery exceeded needs; a negative variance means the number of units produced/delivered was insufficient. A positive planning amount variance shows that planned production/delivery amounts were insufficient to meet needs; a negative variance, that plans exceeded needs.

Example Measuring ABC Company forecast annual potential sales for its product to be 50,000 units. supply chain Based on current production capacity, ABC believed that it could produce agility enough units in usable condition to satisfy 98 percent of that demand. It also believed that, based on past experience, it could deliver on time with 99 percent accuracy. It was estimated that the market would accept a price of $10/unit and variable costs of $4 per unit would be achievable. 585 During the year, one of ABC’s main competitors experienced labor problems and was not able to produce as expected. Consequently, the demand for ABC’s product increased to 65,000 units and the market price increased to $11.50. Initially, ABC believed that expanded production would increase unit variable costs to $5.50 and that it would be able to meet 97 percent of unit demand. ABC Company was not able to increase production while maintaining quality of output and delivery. Consequently its on-time delivery fell to 97 percent and the percentage of products produced in usable condition fell to 94 percent. However, through an unexpected and innovative restructuring of its production process ABC was able to hold variable costs to $5 per unit. Because of penalties for late deliveries, the per unit price for which ABC was able to sell its products was $10.50. Summary data for the analysis is shown in Table II. Figures 4 and 5 show the detailed performance and planning variances. Analysis Total variance The total variance is the result of a large positive planning variance being partially offset by a negative performance variance. The positive planning variance indicates that market requirements exceeded planned amounts because of ABC Company’s inability to foresee demand correctly. Thus, it is clear that a high level of organizational agility is needed to compensate. The negative performance variance shows that the organization was only partly able to respond with sufficient agility and experienced shortfalls in inventory needed to meet market demand. Planned Actual Required T (%) 99.00 97.00 99.00 P ($) 10.00 10.50 11.50 V ($) 4.00 5.00 5.50 S (%) 99.00 94.00 98.00 M 50,000 65,000 65,000 C ($) 6.00 5.50 6.00 N ($) 297,000.00 336,050.00 378,300.00 Table II. Q 49,500.00 61,100.00 63,050.00 Variance example

IJPDLM 32,7 586 Figure 4. Performance variance example

Measuring supply chain agility 587 Figure 5. Planning variances example

IJPDLM Performance variances 32,7 The performance variance of –$48,548.50 is a combination of time, contribution, and quantity variances. The time variance of –$6,721.00 shows a small level of failure in on-time delivery given market requirements. More concern should be generated by the contribution and quantity variances of –$30,244.50 and –$11,583.00, respectively. 588 By breaking down the contribution variance into its component parts, it becomes apparent that ABC lost $60,489.00 in sales because of its failure to charge the market price. The positive cost variance of $30,244.56 may be interpreted as good news. ABC Company was able to quickly expand production capacity without the expected increase in variable cost per unit. This indicates agile performance in production. The quantity variance of –$11,583.00 was entirely due to condition variance. ABC Company was agile in its ability to increase production levels to meet the new level of demand. However, it was unable to maintain quality standards at that level. Planning variances ABC Company experienced a large planning variance indicating an inability to foresee competitive changes in the market. At level III of the model it appears that the entire variance is due to a failure to predict unit demand. In particular, ABC was inaccurate in predicting the number of units demanded (amount variance) but relatively accurate in planning for the correct condition of inventory produced. However, by examining a more detailed breakdown of contribution variance at the fourth level of the model, it can be seen that large but offsetting price and cost variances existed. Had ABC been able to plan for the higher demand level it might have been able to take advantage of the increase in market price without also increasing costs. Conclusion The model proposed in this paper provides a means of measuring both the extent to which agile performance is required, and the agility of an organization’s response to the market environment. This is done by separating the standard comparison of actual versus planned performance into two types of variances. Planning variance measures the need for agility by comparing planned performance with market requirements. Larger planning variances indicate fluctuations in the market demand that either were not or could not be foreseen. Thus, larger planning variances indicate a need for agile response. As planning variances decrease, less agile response in needed and the organization may be able to become leaner. Performance variance measures how agile an organization was given the environment experienced. Low performance variances coupled with high planning variances indicate agile response. An organization that experiences both low performance and planning variances should have a greater likelihood

of lean performance. High performance and low planning variances Measuring demonstrate that the organization is unable to accomplish its planned goals, supply chain even though it is operating in an environment where leanness should be agility attainable. Finally, high performance variances as well as high planning variances show that the organization needs to be agile but has not yet reached required levels of agility. This model should be useful for organizations wishing to assess the need for 589 and the ability to react with agility to changes in the marketplace. By comparing the costs associated with developing agile performance to the costs of failure to be agile, organizations should be able to make a determination of whether or not agility is a useful strategy. References Bentz, W.F. and Lusch, R.F. (1980), ‘‘Now you can control your product’s market performance,’’ Management Accounting, January, pp. 17-25. Biggs, M. (2001), ‘‘The technologies that 2000 forgot,’’ InfoWorld, 29 January, p. 74. Christensen, C.M. (2000), ‘‘Limits of the new corporation’’, Business Week, 21 August, p. 180. Ellram, L. (1993), ‘‘Total cost of ownership: elements and implementation’’, International Journal of Purchasing and Materials Management, Vol. 29 No. 4, pp. 3-9. Fitzpatrick, W.M. and Burke, D.R. (2000), ‘‘Form, functions, and financial performance realities for the virtual organization’’, SAM Advanced Management Journal, Vol. 65 No. 3, pp. 13-22. Grabowski, M. and Roberts, K.H. (1999), ‘‘Risk mitigation in virtual organizations’’, Organization Science, Vol. 10, pp. 704-21. Greis, N.P. and Kasarda, J.D. (1997), ‘‘Enterprise logistics in the information era’’, California Management Review, Vol. 39 No. 3, pp. 55-78. Hewitt, F. (1994), ‘‘Supply chain redesign’’, International Journal of Logistics Management, Vol. 5 No. 2, pp. 1-9. Hulbert, J.M. and Toy, N.E. (1977), ‘‘A strategic framework for marketing control’’, Journal of Marketing, April, pp. 12-20. Hunt, J.W. (2000), ‘‘The virtual challenge’’, The Financial Times (London), 24 November, p. 18. Ijiri, Y., Kinard, J.C. and Putney, F.B. (1968), ‘‘An integrated evaluation system for budget forecasting and operating performance with a classified budgeting bibliography’’, Journal of Accounting Research, Vol. 6 No. 1, pp. 1-28. Kavan, C.B. (1999), ‘‘Virtual@Virtual.org’’, in Stoak Saunders, C. and Nelson, R.E. (Eds), Business Horizons, Vol. 42 No. 5, pp. 73-83. La Londe, B.J., Cooper, M.C. and Noordeweier, T.G. (1988), Customer Service: A Management Perspective, Council of Logistics Management, Oak Brook, IL. Lambert, D.M. and Cooper, M.C. (2000), ‘‘Issues in supply chain management’’, Industrial Marketing Management, Vol. 29 No. 1, pp. 65-84. Lambert, D.M., Cooper, M.C. and Pagh, J.D. (1998), ‘‘Supply chain management: implementation issues and research opportunities’’, The International Journal of Logistics Management, Vol. 9 No. 2, pp. 1-19. Lusch, R.F. and Bentz, W.F. (1984), ‘‘A market-accounting framework for controlling product profitability’’, in Shapiro, S.J. and Kirpalani, V.H. (Eds), Marketing Effectiveness: Insights from Accounting and Finance, Allyn and Bacon, Boston, MA, pp. 481-508.

IJPDLM Magretta, J. (1998), ‘‘Fast, global, and entrepreneurial: supply chain management, Hong Kong style’’, Harvard Business Review, Vol. 76 No. 5, pp. 102-14. 32,7 Nellore, R. and Motwani, J. (1999), ‘‘Procurement commodity structures: issues, lessons and contributions’’, European Journal of Purchasing and Supply Management, Vol. 5 No. 3-4. Ratnatunga, J., Pike, R. and Hooley, G.J. (1988), ‘‘The application of management accounting techniques to marketing’’, Accounting and Business Research, Vol. 18 No. 72, pp. 363-70. 590 Schwartz, E. (2001), ‘‘Skipping steps’’, InfoWorld, Vol. 23 No. 6, 5 February, pp. 1, 29. Shank, J.K. and Churchill, N.C. (1977), ‘‘Variance analysis: a management-oriented approach’’, The Accounting Review, Vol. 52 No. 4, pp. 950-7. Shank, J.D. and Govindarajan, V. (1989), Strategic Costs Analysis: The Evolution from Managerial to Strategic Accounting, Irwin, Homewood, IL. Weber, M.M. (1996), ‘‘A framework for analyzing sources of variance in the supplier-buyer relationship: determining the contribution of buyer planning and supplier performance to total variance’’, Journal of Marketing Theory and Practice, Vol. 4 No. 2, pp. 61-70. Weber, M.M. (2000), ‘‘Calculating the cost of variances in the supply chain’’, Industrial Marketing Management, Vol. 29 No. 1, pp. 57-64.

The research register for this journal is available at The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/researchregisters http://www.emeraldinsight.com/0960-0035.htm A multi-dimensional empirical A multi- dimensional exploration of technology exploration investment, coordination and 591 firm performance Received August 2001 Anthony Ross Revised March 2002 Department of Marketing and Supply Chain Management, Eli Broad Accepted May 2002 School of Management, Michigan State University, East Lansing, Michigan, USA Keywords Co-ordination, Performance, Technology Abstract As economic activities span the supply chain boundary, the effective use of technology as the medium for coordination (or integration) among and within organizations has received much attention. In the US manufacturing sector, IT usage is increasingly becoming a source of sustained competitiveness and an opportunity for improvement. And there is a growing demand to achieve conflicting performance objectives (revenue versus profitability versus efficiency, for example). This article explores the relationships between information technology investment, performance, and productivity. While management should continue to evaluate IT investments by any practical means that satisfies company needs, the development of IT competencies and investment policies so as to optimize the firm’s performance seems to be a worthwhile goal. Our empirical findings clearly suggest that IT investment has a positive impact on market performance as a result of better coordination in the value chain, but that larger investments do not seem to lead to higher financial performance. Additionally, coordination productivity seems to benefit from increased investment by reducing, say, working capital requirements. Given the diversity of firms represented, we conclude that the way in which these firms compete may also have a direct influence on the extent of IT investment and competencies. Introduction IT usage is increasingly becoming a source of sustained competitiveness and an opportunity for improvement. In fact, key drivers for IT adoption have included growing interactivity in supply chains, financial opportunity, efficiency and cost savings, enhanced customer and market penetration, and increased competition (Sood et al., 1999). The effective use of technology as the medium for coordination (or integration) among and within organizations has received much attention with many unclear or mixed conclusions (Narasimhan and Kim, 2002). Examples from industry include: . the launch of RetailLink by Wal-Mart for Internet-based transactions; . the recent alliance between the big three car manufacturers (Ford, General Motors, and DaimlerChrysler) to establish Coviscint (an electronic market for automotive parts suppliers); and International Journal of Physical Distribution & Logistics Management, Vol. 32 No. 7, 2002, pp. 591-609. The author wishes to gratefully acknowledge the assistance of three anonymous reviewers and # MCB UP Limited, 0960-0035 to the special issue editors for their comments and suggestions on strengthening the paper. DOI 10.1108/09600030210442603

IJPDLM . Procter & Gamble’s failure to merge the physical and cyber 32,7 marketplaces which has jeopardized corporate value and direction (Wall Street Journal, 2000, p. 1). Indeed, they are evidence of technology’s role as an agent of change in the approach to economic activity. Alliances such as these suggest that the industrial future does not belong to 592 large bureaucratic organizations, but to networks of affiliated companies transacting/exchanging with one another using advanced forms of information technologies. These informal networks are significantly less capital intensive than traditional enterprise structures. They consume less of everything (input resources) than do traditional enterprises. Davidow and Malone (1992) coined these networks ‘‘virtual corporations’’. Successful virtual corporations, by necessity, collect and integrate massive information flows through their organizational components and intelligently act upon that information. They suggest that information technology (IT), the core of virtual corporations, reduces coordination costs between buyers and sellers, and leads supply chain members to organize economically and efficiently. In fact, Byrnjolfsson et al. (1994) found that investments in technology were associated with declining firm size, and decentralized decision making, and others (Zenger and Hesterly, 1997; Lawler, 1988) contend that all firms use information technology to reduce transaction costs and/or disaggregate activities by substituting technology for aggregate activities. Clearly, virtual organizations, too, appear to have a growing impact on value-adding processes as goods speed into customers’ hands. This evidence of information technology’s role in shaping practice in today’s organizations (virtual or otherwise) makes it clear that technology is no longer an impediment in our information-rich society. Instead, information technology may be the important link to business success (Argyres, 1999). Therefore, organizations must emphasize information technology’s relationship to coordination, control, and business performance. This study explores the relationships between information technology investments, market performance and financial performance, and coordination productivity for a set of leading manufacturers. However, a comparison between Internet virtual companies and traditional manufacturers, represents a logical extension. The motivation for the current study stems from the rapidly intensified interaction among buyers and sellers and the continuing debate surrounding the exact relationship of firm performance to information technology investment (Byrne, 1993; Byrnjolfsson, 1993; Kivijarvi and Saarinen, 1995; Rai et al., 1997; Strassman, 1998). Therefore, three specific questions are addressed by this study: (1) Do IT investment and IT competence influence financial and market performance? (2) What is the relationship of IT investment to coordination productivity? (3) Does IT investment lead to higher efficiency?

This paper is organized as follows. First, the relevant literature on coordination A multi- and IT investment is presented. Next, key dimensions of performance are dimensional derived from this literature. The methodology of this study is then discussed. exploration Finally, analyses are presented along with managerial implications. Technology, coordination and performance The supply chain management literature (Buzzell, 1985; Evans et al., 1993, 593 1995; Ettlie, 1995; Levary, 2000) concludes that the use of IT to coordinate is blurring the flow of information and products among trading partners, and that successful adoption of IT can lead to success in the marketplace. Indeed, several authors (Tallon et al., 1997; Shin, 1999) conclude that any positive impact results from the ability to coordinate value-adding activities. For example, the diffusion of IT in the workplace has taken on many forms to either increase the flow of relevant information between trading partners and coordinate boundary-spanning processes, or coordinate value-chain processes using such technologies as Internet-based customer transaction processing systems, electronic marketplaces, barcode/scanning technologies, in-house systems for resource planning, data interchange, extranets, and Web-storefronts (Malone et al., 1989; Croom, 1999). Key elements of these technologies include: . electronic data interchange; . integrated information systems; and . production control systems. Therefore, since IT has emerged as the backbone of economic activity, investment in it has strategic and transactional impact on coordination. Many studies have empirically evaluated IT’s relationship to overall business performance with no relationship (Markus and Soh, 1993; Loveman, 1994; Hitt and Brynjolfsson, 1996), a slight relationship (Weil, 1992; Barua et al., 1995; Rai et al., 1997) or a positive relationship (Bender, 1986; Barua and Lee, 1997). These results have furthered the debate over the precise contributions made by information technology. Shafer and Byrd (2000) explore the relationship between ‘‘soft’’ IT resource inputs (IT budget, processor value, and IT training) and compounded growth in market performance (revenues and profits) for the period 1990-1994 and attempt to benchmark efficient performance for the firms studied. As is shown in the succeeding sections, we Figure 1. An IT processing view of manufacturing

IJPDLM go further in a number of ways related to the type of resource inputs and the 32,7 variety the output forms considered in the context of our IT processing view in Figure 1. Following Rai et al. (1997), our view is that IT investments are quite different from traditional capital expenditures in heavy manufacturing. IT capital typically has shorter life cycles (of say 6-12 months), than do traditional capital expenditures that can extend over several years. Therefore, lag effect 594 assumptions do not exist for these IT investments. The precise relationship between business success, such as working capital requirements and other forms of productivity, remain unclear and will be addressed in the paper. Dimensions of coordination Many direct internal and external benefits result from linking value-adding processes to deliver products and services. They can include sharing of detailed supplier process data, cost accounting data, material release coordination, increased/early visibility to product requirements, and so on. To operationalize the coordinating nature of a firm, this paper uses two intermediate constructs, labor productivity and administrative productivity. Labor productivity, LABPROD, is postulated to increase with technology use. This occurs as technology frees labor staff to focus on core activities. As more labor is freed up, sales generation potential is assumed to increase. Therefore, LABPROD is computed as sales generated per employee. Administrative productivity, ADMPROD, is the ratio of sales to total administrative costs of the firm. Many econometric studies have found no significant relationship between IT investment and administrative productivity (Strassman, 1988; Rai et al., 1997). In fact, Strassman (1988) notes, ‘‘Computers will not make a [poorly] managed business better. The expenses for computerization . . . are likely to accelerate the decline of incompetent management.’’ Despite this finding, we revisit this relationship with a view of resource efficiency (discussed in detail later). Business performance dimensions This study does not address the traditional measures of manufacturing performance such as cost, time, product reliability and flexibility that have been well studied (Vickery et al., 1997; Vokurka et al., 1998; Ward et al., 1995; Youndt et al., 1996). Rather, our purpose here is to investigate those dimensions of coordination, investment, and performance supported by the relevant literature and most relevant to a broad set of firms representing the manufacturing sector. Financial performance (profit-related) and market performance (revenue-related) have been widely used as measures of firm performance (Bender, 1986; Boyer et al., 1997; Boyer, 1999). As elaborated in Table I, this exploratory study uses firm revenues, net income and operating income before depreciation as measures of market performance, an external orientation. Return on assets and working capital are internal financial performance measures.

Variable/ A multi- acronym Relationship to study Definition dimensional exploration IT-STAFF The number of IT-specific employees in Number of IT employees the firm; is a resource input variable in the DEA model TOT-STAFF The total number of employees in the firm; Number of total employees 595 a resource input variable in the DEA model IT investment That portion of total budget allocated to IT Percent of IT budget to salaries, hardware, software and other IT salaries support; resource input variable in the DEA model TCOMP A self-reported index of the extent of IT IT competence index sophistication or development within the firm; resource input variable in the DEA model WKGCAP A financial ratio of the available financial Sales/(accounts receivable + resources; as coordination improves, this inventories + accounts ratio should increase. Therefore it is an payable) output variable in the DEA model LABPROD A financial ratio of total sales (less returns Sales/total staff and allowances) per employee. An output variable in the DEA model ADMPROD A financial measure of administrative Sales/(selling and general productivity in generating revenues; an administrative expenses – output variable in the DEA model IT budget) Revenue Total sales (less returns and allowances); Gross sales an output variable in the DEA model OIBDP The level of operating income before Operating income before applying depreciation expenses; an output depreciation variable in the DEA model ROA Return on assets; a financial ratio of sales Return on assets to fixed assets in the firm Table I. Definition of Net income Net income earned by the company Return on assets variables used Methodology Since the initial work of Farrell (1957), a body of literature has analyzed the efficiency of operating units in the pubic and private sectors. DEA is a generally accepted methodology for firm-level and supply chain-level analyses (Kleinsorge et al., 1992; Weber and Desai, 1996; Narasimhan et al., 2001; Ross and Droge, 2002). This study does not extend the DEA methodology, but rather we use it to investigate the issues raised earlier in this paper. Namely: whether there are significant differences in market performance, financial performance, and coordination productivity among specific groups of firms; and whether

IJPDLM there exists a precise relationship between IT investment and the level of 32,7 efficiency that ensues for a firm. This non-parametric, multi-factor approach enhances our ability to capture the multi-dimensionalities of performance, and proceeds as follows: (1) Compute the DEA scores using the Banker et al. (1984) (BCC) formulation, known as BCC, and the reduced Charnes et al. (1978) 596 (RCCR). (2) Rank the firms by IT investment levels (descending) and partition into groups (high, medium, low). (3) Perform parametric tests of significance between the groups to evaluate market, financial, and coordination productivity performance of the firms. (4) Determine whether significant differences exist. Consistent with our IT processing view of the firm in Figure 1, notice that the analysis considers the output factors/measures of firm performance and several IT-related input factors presented and defined in Table I. A summary of the model is presented in the Appendix. For other background details on DEA theory and formulations, the interested reader may consult such references as Charnes et al. (1994), or the literature review of Seiford (1996). The study A total of 51 firms in the US manufacturing sector were studied. The 1999 Information Week 500TM rankings of the most innovative and IT-enabled firms in the manufacturing sector were used to identify the firms involved in this study. This firm-level data was then matched with other data from CompustatTM financial databases. The data reported in these sources are comparable to that reported by the Bureau of Economic Analysis, a division of Variable Mean Std deviation Low High Revenue ($00,000s) 13,424.12 31,571.79 4.53 158,514 Net Income ($00,000s) 1,713.98 3,320.19 0 22,825.8 OIBDP ($0,000s) 4,484.86 5,863.67 2,242.19 34,533 Total staff 55,818.31 103,045.98 2,900 608,000 IT staff 640.18 963.74 12 5,000 IT Investment ($mil) 56.144 9.138 14.77 77.55 ADMPROD 7.73 4.73 2.05 26.88 LABPROD 1.227 0.112 1.001 1.556 Table II. WKGCAP 2.56 0.88 0.54 5.16 Overall descriptive statistics ROA 21.57 5.87 0 30.98

the US Commerce Department. The firm sizes range from roughly $453,000 to A multi- $15 billion in revenues, with workforce from 2,900 to 608,000 employees. Other dimensional summary statistics are reported in Table II. Primary SIC codes were not used exploration because the sample size was too small to permit meaningful analysis. The data Although the four input variables and six output variables are categorized into 597 three groups, they are integrated into a single DEA evaluation model. The first group, LABPROD and ADMPROD, captures coordination productivity of internal processes or procedures in the organization that may have been manual procedures, but may now be automated using various IT applications. These applications might include order generation/management, barcoding, scanning, electronic data interchange, and enterprise-wide systems among others. They support product coordination activities related to inbound procurement, production support, and outbound delivery through information sharing. In the second group, REVENUES, WKGCAP, OIBDP, and ROA capture the many aspects of business profitability that may result from better demand/ order generation and order-fulfillment (e.g. inbound/outbound warehousing, inventory management, shipping, customer service). Third, to develop insights to a firm’s technological competence, we also collected self-reported data on the firms’ state of development in four broadly defined IT categories using a three-point scale (1 = highly advanced, 2 = moderately advanced, 3 = basic implementation). These self-reported data were also a part of the Information Week 500TM survey where IT managers were asked to evaluate their firm’s efforts in several categories. The categories are: . the extent to which the firm is sharing data with its trading partners (e.g. using EDI, virtual marketplaces); . the stage of ERP use/implementation within the firm (initial implementation, intermediate implementation, advanced implementation); . the extent of the firm’s reliance upon intranet and Internet-based systems for business transactions from supply chain members (e.g. private exchanges); and . the scale of capabilities for the development of internal applications systems by its IT staff. An index of technological competence for each company, TCOMP, is now defined as the sum of each firm’s score in these four categories. A low index score suggests high overall competence, while a high index score indicates basic overall competence. The methodology discussed in the Appendix requires that data for the DEA model be classified as either inputs or outputs (see Figure 1). Table III presents the measured intercorrelations (p < 0.05) of the variables used in the study. As one might expect, technology competence (TCOMP) would be positively

32,7 598 Table III. IJPDLM study variables Intercorrelations of Revenue NetInc. OIBDP ADMPR LABPR WKGCAP ROA TCOMP TOTSTF ITSTAF ITBUDG Revenue 1 NetInc 0.353 1 OIBDP 0.756 0.525 1 ADMPR 0.155 0.066 –0.012 1 LABPR 0.273 0.176 0.221 0.149 1 WKGCAP –0.576 –0.307 –0.543 –0.081 –0.343 1 ROA –0.155 0.577 –0.014 0.072 0.084 0.149 1 TCOMP 0.398 0.403 0.46 –0.198 0.09 –0.125 0.148 TOTSTF 0.814 0.372 0.806 0.041 –0.18 –0.439 –0.159 0.447 1 ITSTAF 0.6523 0.3501 0.5804 0.034 0.071 –0.451 –0.078 0.517 0.7285 1 ITBUG 1 0.3532 0.7562 0.1549 0.2727 –0.5756 –0.155 0.3988 0.8135 0.6523 1 Note: p < 0.05

correlated with IT-related resources (TOT-STAFF, IT-STAFF, and IT-INV). A multi- Surprisingly, market performance (measured as revenues, net income, and dimensional operating income before depreciation) appears to be very positively correlated exploration with IT-related resource levels. The financial performance measures are negatively correlated with IT. Finally, coordination productivity appears slightly positively correlated. To our knowledge, these findings have yet to be so clearly supported empirically. The paper explores these relationships further 599 in the sections that follow. Results This section reports the empirical findings on the relationships among IT investment efficiency, productivity, performance, and IT competence based on the initial DEA results in Table IV. As organizations adopt one or more of the various forms of technology alluded to earlier, learning and readjustment occur. For the analysis, we define investment intensity as the monetary level of IT investment. Technology competence and investment The 51 firms were ranked in descending order of investment, split evenly into thirds, and finally categorized as low, medium or high investment intensity. We assumed the firms in each group to differ significantly in administrative and labor productivity, return on assets and technological competence measures. Figure 2 illustrates general relationships among some of the constructs. The level of IT investment appears to be positively associated with both IT competence and labor productivity. The former suggests that organizational efficiency improves due to individual learning (competence) and improved utilization of labor resources (labor productivity). The remainder of the paper examines such phenomena. Nonparametric tests were used since the data for these variables were not symmetric. Technological competence and investment intensity were found to have significant differences in certain scenarios. For example, we found significant differences for low intensity firms compared to the medium and high groups. The separation of both high (p = 0.001) and medium (p = 0.002) competence from the low group is significant (Table V). Similarly, the relationship holds for investment intensity where we detected significant differences between the low-medium and low-high (slight) comparisons in Table VI. This suggests that some baseline investment intensity might be required to develop IT competencies, but beyond some level diminishing returns will result. Investment intensity and coordination productivity The second phase of analysis evaluated IT investment against technology competence and intermediate performance (ADMPROD, LABPROD). Labor productivity was unaffected by investment level. Revenue generated per capita remains relatively constant despite increases in IT investment. However,

IJPDLM DMU BCC RCCR CCR 32,7 Unit 1 1.000 0.9965 0.9965 Unit 2 1.000 1.005 1.000 Unit 3 1.000 1.0034 1.000 Unit 4 0.9999 0.9887 0.9887 Unit 5 1.000 1.0008 1.000 600 Unit 6 1.000 0.9921 0.9921 Unit 7 1.000 0.9988 0.9988 Unit 8 1.000 0.9901 0.9901 Unit 9 1.000 0.9949 0.9949 Unit 10 1.000 0.9966 0.9969 Unit 11 1.000 1.4973 1.000 Unit 12 1.000 0.9981 0.9981 Unit 13 1.000 0.9985 0.9985 Unit 14 1.000 0.9945 0.9945 Unit 15 1.000 0.9955 0.9955 Unit 16 0.9999 1.0008 1.000 Unit 17 1.000 0.9957 0.9957 Unit 18 1.000 3.3119 1.000 Unit 19 1.000 1.011 1.000 Unit 20 1.000 0.9929 0.9929 Unit 21 1.000 1.0305 1.000 Unit 22 1.000 1.0058 1.000 Unit 23 1.000 0.9976 0.9976 Unit 24 1.000 0.9954 0.9954 Unit 25 1.000 1.3194 1.000 Unit 26 1.000 0.9989 0.9989 Unit 27 1.000 0.9989 0.9989 Unit 28 1.000 0.9972 0.9972 Unit 29 0.9999 0.9865 0.9865 Unit 30 1.000 1.3848 1.000 Unit 31 1.000 0.9982 0.9982 Unit 32 1.000 0.9992 0.9992 Unit 33 0.9999 0.9932 0.9932 Unit 34 1.000 1.0057 1.000 Unit 35 1.000 0.9902 0.9902 Unit 36 1.000 0.9931 0.9931 Unit 37 1.000 0.9923 0.9923 Unit 38 1.000 0.9931 0.9931 Unit 39 1.000 1.4084 1.000 Unit 40 1.000 0.9846 0.9846 Unit 41 1.000 0.9856 0.9856 Unit 42 1.000 0.9961 0.9961 Unit 43 0.9999 0.9975 0.9975 Unit 44 0.9999 0.9855 0.9855 Unit 45 1.000 0.9949 0.9949 Unit 46 1.000 0.9972 0.9972 Unit 47 1.000 0.9867 0.9867 Unit 48 1.000 1.0448 1.000 Table IV. Unit 49 0.9999 1.0594 1.000 Manufacturers’ Unit 50 1.000 0.9862 0.9862 efficiency scores Unit 51 1.000 0.9929 0.9929

administrative productivity, revenues divided by the expense of generating A multi- orders (S&GA) minus IT investment improves as investment increases. dimensional Though not reported here, our comparison tests found no significant difference exploration (p = 0.05) between the low and medium groups but significance for the high investment group on these productivity measures. In fact, Figure 2 suggests a positive relationship between investment, IT competence and coordination. 601 When combined with Table II’s finding of a strong negative relationship between working capital requirements (WKGCAP) and IT investment, the results suggest that higher IT investments reduce the need for working capital through efficient coordination and control of procurement, production, and distribution to final customer. This seems to suggest that IT improves the capacity to further squeeze wasted time out of the value-adding process. Figure 2. IT investment vs coordination productivity Competence group t-statistic p-value Low-medium 3.295 0.002 Low-high 4.018 0.001 Medium-high 0.037 0.971* Table V. Multiple comparisons: Notes * Not significant IT competence Investment group t-statistic p-value Low-medium –1.923 0.063* Low-high –2.38 0.026* Medium-high –0.600 0.553** Table VI. Notes * Levene’s test showed unequal group variances (F = 14.466; p = 0.001); Multiple comparisons: ** Not significant IT investment intensity

IJPDLM Firm profitability and investment intensity 32,7 The third phase evaluates financial performance and IT investment. Two financial ratios were considered (see Figure 3). The descriptive statistics in Table I summarize the differences in size/operating scale of the firms considered. Because the dataset includes both large and small manufacturers, 602 net income and revenue were normalized by each firm’s total workforce size to give NIPERCAP (net income per capita) and REVPERCAP (revenue per capita), respectively. We found that NIPERCAP seems to have a parabolic relationship with investment level while REVPERCAP increased linearly. And this suggests that IT facilitates sales generation as a result of better coordination in the value chain; but with this comes the added expense, at least in the short term, of acquisition, development and management of technology (i.e. hardware, software, applications development). We then examined return on assets (ROA), the broadest measure of profitability and management effectiveness. Independent of financing strategy, which we do not consider, ROA is frequently used to compare managers’ relative performance. Figure 4 shows that return on assets appears to decrease as investment level increases. Firms with higher ROA are much better at selecting IT investments, all other things equal. Figure 3. IT investment vs financial performance Figure 4. ROA by investment level

Investment intensity and DEA efficiency A multi- Table IV reports the DEA computations for the three contexts discussed in the dimensional Appendix. Of the 51 leading manufacturers, 15 considered in the analysis were exploration CCR efficient – with a score of 1.00 – while all others were inefficient (see Table IV). In the context of our competence/productivity model, this suggests that the inefficient firms must do one of the following: . reduce the size of their inputs (IT investment or staff, total workforce); or 603 . increase output levels for the same input level. However, since IT competence (an input) is akin to knowledge or learning, reducing its level cannot be accomplished without significant change in IT strategy. But to further differentiate investment intensity for the efficient firms, we ranked the firms in descending order by their RCCR score (Table IV). RCCR allows for scores higher than 1.00 (or super efficiency) and the opportunity to further discriminate among the 15 efficient firms (units 2, 3, 5, 11, 16, 18, 19, 21, 22, 25, 30, 34, 39, 48 and 49). Table VII reports descriptive statistics for the ‘‘super’’ efficient firms. The mean NIPERCAP and REVPERCAP for these firms were $7,007 and $8,449, respectively. By comparison to the overall data in Table I, the super efficient firms on average invest less in IT, are more productive and achieve higher returns on assets (ROA). But to be sure, several additional t-tests (for differences in mean values) were performed to compare the super efficient firms to the remaining sample of firms on the dimensions of IT competence, IT investment, ROA, and ADMPR. It was determined that the null of hypothesis of no difference could only be rejected for the ROA variable Variance Minimum Maximum Mean Std. dev. ROA (%) 14.65 27.58 22.48 4.23 LABPROD 1.08 1.56 1.23 0.13 Table VII. Admin. prody. 3.26 15.47 7.27 3.10 Summary of super IT investment ($ millions) 49.21 76.99 57.97 7.07 efficient firms Figure 5. IT investment efficiency

IJPDLM ( = 0.01). Clearly, high IT investment does not necessarily lead to higher 32,7 financial performance. But such investment decisions are reflected in management’s ROA effectiveness and seem to be captured by the RCCR ranks. There was a significant difference in ROA with super efficient firms realizing higher returns. Furthermore, the ROA and IT investment sample variances among these groups were statistically different ( = 0.01) with less variance 604 among super efficient firms. Finally, we return to the BCC formulation, which assumes that financial performance, productivity and efficiency increase disproportionately with IT investment. Figure 5 examines the mean efficiency for the investment groups. Based on these results, we conclude a positive relationship between IT investment and efficiency. Concluding remarks This study has analyzed performance of a set of leading manufacturers using financial data matched with survey data of their IT practices. While we cannot rule out all alternative explanations, our results are consistent with some previous empirical arguments regarding IT investment and coordination, and contradictory with others regarding financial performance and IT investment. The cross-section of manufacturers represented here may provide some explanation for this contradiction. But these empirical results do show that market performance and coordination productivity are positively affected by IT investments. Furthermore, despite IT’s slightly negative correlation to financial performance, investment intensity (Figures 3-4) suggests a complex relationship exists. This might imply that any two firms making the same IT investment, whether competitors or partners, can derive quite different performance benefits. Therefore, it may not be possible to assume that all firms are equally effective. But rather, there exist a variety of factors beyond the scope of this study that influences firms’ IT strategy formulation. We call for a longitudinal examination of the relationships explored here. One such study currently under way is to explore the temporal changes in performance related to changing investments. This study concludes by returning to the three questions raised early in the paper. First, efficiency does not require that firms lead the way in IT investment. Such large IT investments did not seem to lead to higher efficiency and profitability in several cases. Whereas several colleagues have studied specific sectors of manufacturers, this study included manufacturers from aircraft, and automobile sectors, to those in rubber and paper. These sectors are characterized by different rates of evolution and diffusion in product technology, process technology, and organizational transformation. As a result, efficiency and profitability may have more to do with precise resource allocation and economic cycles, respectively, than purely with IT investment itself. For example, certain industries (like food or automotive) have a long history with electronic data interchange use and development, while in other industries (such as electronics or semiconductors) there is a prevalence to adopt open systems like enterprise resource planning. The competencies acquired

with closed systems are not necessarily the same competencies required or A multi- developed using open systems. However, companies must be very effective dimensional with the products, processes and organizational coordination structures used to exploration create competitive advantage. As Charles Fine (1998) notes, ‘‘. . . all advantage is temporary’’. IT investment and the ensuing competencies may mediate not only internal productivity, but also revenue generation. Because IT enhances product and 605 information coordination in the supply chain, one immediate outcome is market penetration as a result, for example, of viral marketing and other forms of Internet-based economic activity. Second, the degree and types of IT investment required to maintain such systems vary. What ensues, however, is a set of capabilities in product, process, or organization technologies that defines each firm’s competitive advantage – at least temporarily. In sum, the impact of investment and competence on performance is on coordination – measured as labor and administrative productivity, and working capital. This study found a positive relationship between investment, the ensuing competence, and their impact on administrative productivity in supply chains. Although the impact on profitability remains mixed, our results suggest a positive relationship. Extensions of this study are investigating the effect of IT on growth in performance and productivity. However, we attempt to summarize the range of possibilities using Table VII. This range suggests a set of typologies of the units on the empirical frontier. Thus, it may not be necessary to lead in all input categories such as IT investment, or workforce size. These firms may compete differently within their own sector, and the requisite IT capabilities (electronic brokerage, communication) will differ. There is much debate over using an aggregate measure of IT investment versus specific allocations to its components (hardware, applications software, systems software, environment, etc.). Our position, however, is that performance benefits should also be measured based on a composite of IT spending. This paper has suggested a methodology through which managers can objectively evaluate its overall IT effort relative to other companies, rather than specific IT-related projects. Third, we have also shown that there is a positive relationship between efficiency and IT investment, and that coordination productivity improves with increased investment and use of IT. These results help us step towards a better understanding of the empirical relationships between business profitability, coordination, and efficiency. There are several extensions of this work. Though not the focus of this study, we suggest a few here. Is there a precise functional form to the relationship between, say, IT investment and market or financial performance? This study has shown that higher investments seem to correspond to higher market performance, but does this curve shift up and to the right (graphically speaking), extending the investment returns? Few studies have shown whether a functional form actually exists between IT spending and various dimensions of performance. Second, the notion of value chain coordination has been raised

IJPDLM since administrative productivity is affected by information sharing. As a 32,7 result, an investigation of supply chain coordination costs for electronic brokerage (procurement and physical distribution) and the e-supply chain might be worthwhile extensions. What are the antecedents for successful technology investments in business-to-business environments? Finally, an expanded study comparing IT uses and productivity among several industries 606 together (e.g. manufacturing, pharmaceuticals, utilities, banking) also has merit. Such a study would use the DEA methodology to measure efficiency differences among the firms and across the industries to detect comparative IT advantage and disadvantage between industries. Furthermore, it might also investigate performance differences between internet companies that sell physical products and those that sell digital products (music or services). References Anderson, P. and Peterson, N. (1993), ‘‘A procedure for ranking efficient units in data envelopment analysis’’, Management Science, Vol. 39 No. 10, pp. 1261-4. Argyres, N. (1999), ‘‘The impact of information technology on coordination: evidence from the B-2 ‘Stealth’ bomber’’, Organization Science, Vol. 10 No. 2, pp. 162-80. Banker, R., Charnes, A. and Cooper, W. (1984), ‘‘Some models for estimating technical and scale efficiencies in data envelopment analysis’’, Management Science, Vol. 39 No. 10, pp. 1265-74. Barua, A. and Lee, B. (1997), ‘‘The IT productivity paradox revisited: a theoretical and empirical investigation in the manufacturing sector’’, The International Journal of Flexible Manufacturing Systems, Vol. 9 No. 2, pp. 145-66. Barua, A., Kriebel, C. and Mukhopadhyay, T. (1995), ‘‘Information technologies and business value: an analytic and empirical investigation’’, Information Systems Research, Vol. 6 No. 1, pp. 3-23. Bender, D. (1986), ‘‘Financial impact of information processing’’, Journal of Management Information Systems, Vol. 3 No. 2, pp. 22-32. Boyer, K.K. (1999), ‘‘Evolutionary patterns of flexible automation and performance: a longitudinal study’’, Management Science, Vol. 45 No. 6, pp. 824-42. Boyer, K.K., Leong, G., Ward, P. and Krajewski, L. (1997), ‘‘Unlocking the potential of advanced manufacturing technologies’’, Journal of Operations Management, Vol. 15 No. 4, pp. 331-47. Buzzell, R. (1985), Marketing in an Electronic Age, Harvard Business School Press, Boston, MA. Byrne, J. (1993), ‘‘The virtual corporation’’, Business Week, 8 Februrary, pp. 98-104. Byrnjolfsson, E. (1993), ‘‘The productivity pardox of information technology’’, Communications of the ACM, Vol. 35, pp. 66-77. Byrnjolfsson, E., Malone, T., Gurbaxani, V. and Kambil, A. (1994), ‘‘Does information technology lead to smaller firms?’’, Management Science, Vol. 40 No. 12, pp. 1628-44. Charnes, A., Cooper, W. and Rhodes, E. (1978), ‘‘Measuring the efficiency of decision making units’’, European Journal of Operational Research, Vol. 43 No. 2, pp. 429-44. Charnes, A., Cooper, W., Lewin, A. and Seiford, L. (Eds) (1994), Data Envelopment Analysis: Theory, Methodology and Applications, Kluwer Academic Publishers, Boston, MA. Croom, S. (1999), ‘‘The implication of electronic procurement for major account management’’, Journal of Selling and Major Account Management, Vol. 1 No. 4, pp. 47-63. Davidow, W. and Malone, M. (1992), The Virtual Corporation, HarperCollins, New York, NY.

Evans, G., Naim, M. and Towill, D. (1993), ‘‘Dynamic supply chain performance: assessing the A multi- impact of information systems’’, Logistics Information Management, Vol. 6 No. 4, pp. 15-25. dimensional Evans, G., Naim, M. and Towill, D. (1995), ‘‘Business process re-engineering the supply chain’’, exploration Production Planning and Control, Vol. 6 No. 3, pp. 227-37. Ettlie, J.E. (1995), ‘‘Product-process development integration in manufacturing’’, Management Science, Vol. 41 No. 7, pp. 1224-37. 607 Farrell, M. (1957), ‘‘The measurement of productive efficiency’’, Journal of Royal Statistical Society, Series A, pp. 253-90. Fine, C. (1998), Clockspeed: Winning Industry Control in the Age of Temporary Advantage, Perseus Books, Reading, MA. Gardener, C. (2000), The Valuation of Information Technology: A Guide for Strategy Development, Valuation, and Financial Planning, John Wiley & Sons, New York, NY. Hitt, L. and Brynjolfsson, E. (1996), ‘‘Productivity, business profitability, and consumer surplus: three different measures of information technology value’’, MIS Quarterly, Vol. 20 No. 2, pp. 121-37. Kivijarvi, H. and Saarinen, T. (1995), ‘‘Investment in information systems and the financial performance of the firm’’, Information and Management, Vol. 28 No. 2, pp. 143-64. Kleinsorge, I., Schary, P. and Tanner, R. (1992), ‘‘Data envelopment analysis for monitoring customer supplier relationships’’, Journal of Accounting and Public Policy, Vol. 11, pp. 357-72. Lawler, E. (1988), ‘‘Substitutes for hierarchy’’, Organization Dynamics, Vol. 17, pp. 477-91. Levary, R. (2000), ‘‘Better supply chains through information technology’’, Industrial Management, May-June, pp. 24-30. Loveman, G. (1994), ‘‘An assessment of the productivity impact of information technologies’’, in Allen, T.J. and Morton, M.S. (Eds), Information Technology and the Corporation of the 1990s: Research Studies, MIT Press, Cambridge, MA. Malone, T., Yates, J. and Benjamin, R. (1989), ‘‘The logic of electronic markets’’, Harvard Business Review, May-June, pp. 166-70. Markus, M. and Soh, C. (1993), ‘‘Banking on information technology: converting IT spending into firm performance’’, in Banker, R., Kauffman, R. and Mahmood, R. (Eds), Strategic Information Technology Management: Perspectives on Organizational Growth and Competitive Advantage, Idea Group, Harrisburg, MA. Narasimhan, R. and Kim, S. (2002), ‘‘Effect of supply chain integration on the relationship between diversification and performance: evidence from Japanese and Korean firms’’, Journal of Operations Management, forthcoming. Narasimhan, R., Talluri, S. and Mendez, D. (2001), ‘‘Supplier evaluation and rationalization via data envelopment analysis’’, Journal of Supply Chain Management, Vol. 37 No. 3, pp. 28-37. Rai, A., Patnayakuni, R. and Patnayakuni, N. (1997), ‘‘Technology investment and business performance’’, Communications of the ACM, Vol. 40 No. 7, pp. 89-97. Roach, S. (1988), ‘‘Technology and the service sector: the hidden competitive challenge’’, Technological Forecasting and Social Change, Vol. 34 No. 4, pp. 388-403. Ross, A. and Droge, C. (2002), ‘‘An integrated benchmarking approach to distribution center perfomance using DEA modeling’’, Journal of Operations Management, Vol. 20 No. 1, pp. 19-32. Seiford, L. (1996), ‘‘Data envelopment analysis: the evolution of the state of the art (1978-1995)’’, The Journal of Productivity Analysis, Vol. 7, pp. 99-137.

IJPDLM Shafer, S. and Byrd, T. (2000), ‘‘A framework for measuring the efficiency of organizational investments in information technology using data envelopment analysis’’, Omega, Vol. 28 32,7 No. 2, pp. 125-41. Shin, N. (1999), ‘‘Does information technology improve coordination? An empirical analysis’’, Logistics Information Management, Vol. 12 No. 1, pp. 138-44. Sood, R., Friedman, J. and Parekh, M. (1999), ‘‘BUSINESS-TO-BUSINESS: 2B or Not 2B?: BUSINESS-TO-BUSINESS e-markets’’, Version 1.1, in Goldman Sachs & Co., Investment 608 Research, New York, NY. Strassman, P.A. (1988), ‘‘Management productivity as an IT measure’’, in Berger, P., Kobielus, J. and Sutherland, D. (Eds), Measuring Business Value Information Technologies, ICIT Press, Washington, DC, pp. 17-55. Tallon, P., Kraemer, K. and Gurbaxani, V. (1997), ‘‘A multidimensional assessment of the contribution of information technology to firm performance’’, Proceedings Of the 5th European Conference on Information Systems, Cork, June. Vickery, S., Droge, C. and Markland, R. (1997), ‘‘Dimensions of manufacturing strength in the furniture industry’’, Journal of Operations Management, Vol. 15 No. 4, pp. 317-30. Vokurka, R., O’Leary-Kelly, S. and Flores, B. (1998), ‘‘Approaches to manufacturing improvement: use and performance implications’’, Production & Inventory Management Journal, Vol. 39 No. 2, pp. 42-8. Wall Street Journal (2000), 8 March, p. 1. Ward, P.T., Leong, G. and Boyer, K. (1995), ‘‘Manufacturing proactiveness and performance’’, Decision Sciences, Vol. 25 No. 3, pp. 337-55. Weber, C.A. and Desai, A. (1996), ‘‘Determination of paths to vendor market efficiency using parallel coordinates representation: a negotiation tool for buyers’’, European Journal of Operational Research, Vol. 90, pp. 142-55. Weil, P. (1992), ‘‘The relationship between investment in information technology and firm performance: a study of the valve-manufacturing sector’’, Information Systems Research, Vol. 3, pp. 307-33. Youndt, M., Snell, S., Dean, J. and Lepak, D. (1996), ‘‘Human resource management, manufacturing strategy, and firm performance’’, Academy of Management Journal, Vol. 39 No. 4, pp. 836-51. Zenger, T. and Hesterly, M. (1997), ‘‘The disaggregation of corporations: selective intervention, high-powered incentives and modular units’’, Organization Science, Vol. 8, pp. 209-22. Further reading Byrnjolfsson, E. and Hitt, L. (1996), ‘‘Paradox lost? Firm-level evidence on the return to information systems spending’’, Management Science, Vol. 42 No. 4, pp. 541-58. Mukhopadhyay, T., Kekre, S. and Kalathur, S. (1995), ‘‘Business value of information technology: a study of electronic data interchange’’, MIS Quarterly, pp. 137-56. Appendix. Data envelopment analysis DEA is used to evaluate a grouping/population of firms and determine which members are efficient relative to the remaining firms. Maximum efficiency for a firm is generally constrained to a value of 1. This means that no other firm is more efficient than the unit being evaluated, for the given weights. Together, the set of maximally efficient firms form an efficient frontier. For example, in two-dimensional space (one input and one output variable) an imaginary line segment connects the efficient firm, or decision-making unit (DMU). This line represents the efficient frontier and all firm’s scores positioned below this line represent inefficiencies. Although the consideration of several inputs and outputs positions the problem in multi-dimensional

space, the same frontier analogy still holds. In this paper, our attention focuses on identifying the A multi- extent to which resource inputs may be reduced for a given level of output. Thus the orientation is on the resource inputs. Model A1 below, developed by Banker et al. (1984) and referred to as dimensional the BCC model, assumes that outputs do not increase proportionately with inputs (variable exploration returns to scale). Let there be m input factors, p output factors and n firms. It is summarized mathematically by: Model A1: X m X p 609 Min EFFO ¼ Â À ei À Sr ð1Þ i¼1 r¼1 s.t. X n Xj j þ ei ¼ Âxi0 ; i ¼ 1; 2; . . . ; m; ð2Þ j¼1 X n Yj j À sr ¼ yr0 ; r ¼ 1; 2; . . . ; p; ð3Þ j¼1 X n j ¼ 1 ð4Þ j¼1 Â; ei ; sr ; j ! 0; 8i; r; j ð5Þ where Â = allocative efficiency ratio for COMPANYo, the unit under consideration. yr0 = observed value of output r (e.g. wkgcap,labprod,admprod,etc.) at COMPANYo. xi0 = observed value of input i (e.g. it-staff,tot-staff,it-inv,etc.) at COMPANYo. xj = ðx1j ; X2j ; . . . ; xmj Þ, vector of resource inputs for COMPANYj. Yj = ðy1j ; y2j ; . . . ; ypj Þ, vector of actual output values for COMPANYj. ei = amount of excess input i for DMUo. sr = amount of slack in output r for DMUo. j = the weights assigned to the input and output variables. Model A1 is an extension of the original DEA model first developed by Charnes et al. (1978) and to which the name CCR has been accorded after its developers. Because of their close relationship, the CCR model is easily derived from Model A1, the BCC model, by excluding constraint (4) above to allow for the assumption of constant returns to scale. The reader can consult Banker et al. (1984) or Charnes et al. (1978) for further details. Continuing with Model A1, the objective function computes the efficiency score, EFFO. Constraint set (2) ensures that the input level for input i is a linear combination of the inputs from the analysis set plus the excess input of i. Constraint (3) specifies that the optimal output of r at DMUo should also be a linear combination of the outputs from the analysis set minus its slacks. The weights, j , are generated by solving the n DEA models simultaneously. We also modify model A1 a second time by dropping DMUo from constraint sets (2)-(3) as proposed in Anderson and Peterson (1993) to get the ‘‘reduced’’ CCR formulation (RCCR) – allowing efficiency scores greater than 1.00 primarily for purposes of ranking the units (firms) by their corresponding RCCR score. The rankings allow for discrimination among the technically efficient firms in our study.

The research register for this journal is available at The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/researchregisters http://www.emeraldinsight.com/0960-0035.htm IJPDLM 32,7 Improving electronics manufacturing supply chain agility through outsourcing 610 Scott J. Mason and Michael H. Cole Received June 2001 Department of Industrial Engineering, University of Arkansas, Revised January 2002 Fayetteville, Arkansas, USA Brian T. Ulrey Wal-Mart Logistics, Arkansas, USA, and Li Yan Advanced Technology Engineering, Avaya Inc., Denver, Colorado, USA Keywords Supply chain, Agile production, Outsourcing, Electronics Abstract The highly competitive electronics manufacturing marketplace demands that suppliers provide low-cost, high-quality products to their customers in a timely fashion. Shortened product life cycles and increasingly global competition have caused traditional manufacturers to focus on their company core competencies, such as product design and development, choosing to outsource the actual manufacturing of their products to contract manufacturers. Although the decision to outsource can have both positive and adverse effects on key areas of the manufacturing supply chain, one positive effect is that the manufacturer’s supply chain agility is increased. Outsourcing has caused an increase in the amount of information that is shared between supply chain partners. As a result, a greater reliance on suppliers and alliance partners has become essential for company survival. We examine the ways in which contract manufacturing has increased the agility of the electronics manufacturing supply chain. 1. Introduction Electronics manufacturing has accounted for at least 30 percent of the USA’s Gross National Product since the Second World War (Landers et al., 1994). Since that time, the importance of electronics manufacturing to the world’s economy has continued to grow. For the year 2000, the worldwide semiconductor industry enjoyed sales of US$204 billion (Semiseek News, 2001). Dataquest, Inc., a leading semiconductor industry research group, projects this market to reach US$375 billion by 2005 (Semiconductor Business News, 2001). In the past, original equipment manufacturers (OEMs) designed, built, tested, and serviced all of their own products. According to Hassig (1995), OEMs bought equipment that became obsolete quickly, cyclically hired people during economy upswings, and fired these same people during production slumps. The pace of new technology has quickened, making electronics manufacturers reluctant to invest in expensive manufacturing equipment and to hire/fire skilled workers at the first sign of a shift in market direction. As International Journal of Physical electronics manufacturers realized that they were unable to keep up with every Distribution & Logistics aspect of industrial change, contract manufacturing started to grow. Management, Vol. 32 No. 7, 2002, pp. 610-620. # MCB UP Limited, 0960-0035 This research effort was supported by Avaya Incorporated (www.avaya.com) through the DOI 10.1108/09600030210442612 Logistics Institute (www.tli.uark.edu) at the University of Arkansas.

Conklin (1994) asserts that no single enterprise in today’s global marketplace Electronics is able to realize market opportunities in a timely and cost-effective way, manufacturing mainly due to the lack of solid skills and experience bases. When electronics supply chain and contract manufacturers work together to harness the combined knowledge of both parties, the result can be powerful. Calaf (1995) cites an increasing trend of companies choosing to outsource many of their products to other companies that specialize in manufacturing products. Carter and Narasimhan (1996) 611 suggest that outsourcing will be one of the eight most important factors in the future for supply chain management. Lakhal et al. (2001) define a superior supply chain as one that maximizes the value of internal activities while developing strong partnerships that lead to high value external activities. Martin (1999) estimates that the outsourcing of electronics manufacturing through contract manufacturing will continue to grow approximately 25 percent per year for the foreseeable future. Today, a typical integrated device manufacturer outsources approximately 20 percent of its chip production (Smith, 2001). The main factors that support the continuation of this trend include shortened product life cycles, increasingly global competition, cost reductions achieved through large volume procurement of consumables and components, and better utilization of high cost capital infrastructure. Veeramani and Joshi (1997) define the concept of agility as the ability to respond quickly and effectively to satisfy customers. Manufacturing agility has become a defining characteristic of competition in industry today. Companies must quickly identify, design, manufacture, and deliver products that meet customer desires, while maintaining stringent cost and quality standards. More specifically, greater importance is being placed on agility in terms of producing a broad range of low-cost, high-quality products with short lead times in varying lot sizes, built to individual customer specifications (Narasimhan and Das, 1999). To be competitive from an agility standpoint, companies must adapt their supply chains efficiently and build strong relationships with customers and suppliers more quickly (Tolone, 2000). A company cannot become agile unless its relationships with the supply chain are also agile. Supply chain agility is a key to inventory reduction, adapting to market variations more efficiently, enabling enterprises to respond to consumer demand more quickly, and integrating with suppliers more effectively. The agile supply chain is market sensitive – it is capable of reading and responding to real demand. Most organizations are forecast-driven rather than demand-driven. In other words, because they are not market sensitive in terms of actual customer requirements, they tend to make forecasts based upon historical sales data and use those forecasts to determine their inventory requirements (Christopher, 1999). The sharing of information between supply chain partners has become increasingly important as companies focus on their core competencies, choosing to outsource all other activities. In this environment, a greater reliance on suppliers and partners becomes inevitable, and new types of relationships are needed.

IJPDLM There is a growing recognition that individual businesses no longer compete 32,7 as stand-alone entities but rather as supply chains. Individual businesses that previously competed as stand-alone entities are now aligning themselves in network relationships and competing as supply chains. Organizations that can effectively coordinate and manage relationships with their partners in a network committed to closer and more agile relationships with their final 612 customers should quickly gain the advantage. Emphasis will be placed on leveraging the strengths of network partners to achieve a greater response to marketplace demands (Christopher, 1999). This paper examines the impact of contract manufacturing in terms of improving electronics manufacturing supply chain agility. The remaining sections of this paper are organized as follows. Section 2 provides a brief introduction to electronics manufacturing and the electronics manufacturing supply chain. Section 3 highlights several ways that contract manufacturing improves electronics manufacturing supply chain agility. We propose a conceptual model for assessing the impacts of contract manufacturing on the agility of the electronics manufacturing supply chain in Section 4. Finally, some conclusions and areas for future research are given in Section 5. 2. Electronics manufacturing Electronics manufacturing comprises the process of design, development, fabrication, assembly, and testing of electronics parts, tools, technology, components, and systems (Landers et al., 1994). The history of electronics manufacturing can be divided into three eras: vacuum tube era (1920-1950); transistor era (1950-mid-1960s); and integrated circuit era (mid-1960s-present). The evolution of electronics manufacturing occurred with the discovery of better techniques to produce smaller, more reliable electronics components at lower costs. However, the manufacturing process has become more complicated and costly to develop. For example, Kumar (1999) states that an initial investment of US$3.5 million is required to establish a world-class surface mount technology line for circuit board population. Current generation wafer fabrication facilities can cost upwards of US$2 billion. Further, areas of facility support such as process engineering and quality control also come with their own significant costs. As a result, this high price of admission has caused many start-up electronics manufacturers to seek the assistance of contract manufacturers. Though contract manufacturing began primarily in printed circuit board (PCB) assembly, the significant cost reduction and efficiencies offered by contract manufacturers have led to expanded contract offerings in other areas beyond assembly. Today’s contract manufacturers are capable of taking care of almost every aspect of electronics manufacturing. In fact, some contract manufacturers, such as Taiwan Semiconductor Manufacturing Corporation (TSMC), have mastered the latest processing technologies that rival all but the largest electronics manufacturers. Some of the larger players in the contract manufacturing marketplace today include Solectron, Celestica, Flextronics, TSMC, and United Microelectronics Corporation.

Ballou (1999) asserts that a company’s products must be in the possession of Electronics the customer at the proper place and time that they wish to consume them. manufacturing Effective management of the electronics manufacturing supply chain is supply chain important due to the existence of short product life cycles and the resulting cyclical demand. In addition, the heightened expectations of customers for electronics products have made managing the electronics products supply chain a very challenging task. The increasing use of contract manufacturers 613 has totally altered the electronics manufacturing supply chain in both positive and adverse ways. Although the outsourcing of manufacturing to contract manufacturers typically results in a significant reduction of cost and production time and an increase in supply chain agility, contract manufacturing can also complicate the electronics manufacturing supply chain. 3. Contract manufacturing’s impact on the electronics manufacturing supply chain Contracting or partnering with an external manufacturer involves a significant commitment from a company. This decision is not one that is taken lightly by OEM firms. The firm must know both the benefits and risks of outsourcing in order to outsource intelligently (Vining and Globerman, 1999). Outsourcing production to contract manufacturers can cause changes to the product development process, manufacturing strategy, and labor needs, as well as the loss of market visibility, the ability to manufacture, control of the manufacturing process, the ability to repair, and the ability to monitor the inventory levels of the OEM’s products. According to Kumar (1999), OEMs have no direct control or ability to benchmark their own contract manufacturer’s product quality, flexibility, and cost management. More and more in today’s high-tech industry, contract manufacturers are being used not only to produce components, but also to design and build entire systems. Contract manufacturers have proven over time that they are able to produce at equivalent or even higher quality levels when compared to OEMs (Carbone, 1999). The reasons for this shift in philosophy are the reduced manufacturing costs associated with contract manufacturing, a new focus on core competencies, and the need for electronics manufacturers to build and integrate agile supply chains into their operations. In today’s environment, many OEMs do not consider manufacturing as a core competency (Carbone, 1999). By effectively managing their supply chain and contract manufacturing partnerships, OEMs can secure manufacturing capacity without expensive capital investments. A typical distribution system in a supply chain consists of a supplier, a manufacturer, and a warehouse (see Figure 1). However, the distribution and assembly system network becomes very complicated if the company decides to outsource some of its system’s parts and manufacture the rest in-house. Some examples of different distribution system scenarios under a contract manufacturing partnership are as shown in Figure 1.

IJPDLM The assembly system contained within the supply chain of an electronics 32,7 manufacturing process will also be altered. The manufacture-versus-outsource decision may be based upon factory capacity constraints during high seasonal demand, unforeseen changes in demand, and/or the proprietary nature of the product(s) to be manufactured. The decision could also be due to technological ability of the company. A company may look to contract manufacturing to 614 increase its supply chain agility without investing capital. A manufacturer must often reformulate its logistics strategy due to the changes caused by contract manufacturing. The following sections focus on the effects of contract manufacturer location, and various transportation, distribution, and warehousing strategies during outsourcing on the electronics manufacturer’s supply chain. 3.1 Location The geographical location of facilities, warehouses, and supplier are the base elements of a logistics network. For an electronics manufacturer, the location of the contract manufacturer is key. Many contract manufacturers set up their organizations in low-cost manufacturing areas. However, OEMs not only look to cut manufacturing costs, but also look to streamline their supply chain. This leads to selecting contract manufacturers not only in low-cost manufacturing locations, but also in locations with strong transportation facilities. Also, contract manufacturers that are located in close proximity to the company are often selected in order to reduce transportation cost and lead time. Companies can receive significant cost benefits as well as dramatically increase their own supply chain’s agility by utilizing a geographically close contract manufacturer (Kumar, 1999). Close proximity can also be important for new and/or complex products that require a high level of interaction between the OEM and the contract manufacturer. 3.2 Transportation, distribution, and warehousing Transportation is one of the most important elements in the execution of the supply chain. Transportation decisions include modal selection (e.g. rail, truck, air, or water), shipment size, vehicle routing, and scheduling, all of which are directly related to the location of warehouses, customers, and plants. The strategy for distributing outsourced products is also a major concern for companies who employ contract manufacturers. The introduction of contract Figure 1. Potential distribution system scenarios with contract manufacturing

manufacturing into the logistics network can alleviate the need for electronics Electronics manufacturers’ warehouses, traditionally used for storing both raw material manufacturing and finished product inventory. There are three main distribution and supply chain warehousing options for the OEM who chooses to outsource. First, the OEM could choose to allow the contract manufacturer to ship products directly to end customers. If the OEM uses a direct shipment strategy, no warehouses are needed, as the finished products will be stored in the 615 contract manufacturer’s warehouse. This decision can potentially increase the agility of the OEM’s supply chain by reducing both on-hand inventory and lead-time. However, OEMs often receive customer orders that, if shipped directly from the responsible contract manufacturer, would not be large enough to fill an entire trailer. These less-than-full truckload (LTL) shipments sent directly from the contract manufacturer are very costly. Also, manufacturer transportation costs would increase, when compared to simply shipping from one large warehouse, because the manufacturer would have to send more small trucks to more locations. Economies of scale favor large shipment sizes, which lower the transportation cost on a per-unit basis. A second option for OEMs is to ship finished products from the contract manufacturer to one of their own warehouses or distribution centers. Although the outsourcing of the manufacturing process alleviates the need for raw material warehouses and potentially the need for finished goods storage, some companies prefer contract manufacturers to ship the finished products to their warehouses before they are distributed to customers. This enables the company to monitor the inventory level of finished products. Most companies that deal with low volume orders use lower-rate, full-truckload shipments to ship finished product from the contract manufacturers to intermediate distribution warehouses located near their customers. These warehouses improve supply chain agility by being able to respond more quickly to customer demand. Finally, electronics manufacturers may choose to employ cross-docking at their warehouses. This strategy is especially useful for companies who have outsourced many different components to numerous contract manufacturers. By setting up a cross-dock facility, an OEM can consolidate components of an entire system without the components ever coming to rest as static inventory in the warehouse. This value-added technique impacts the supply chain in a positive way by minimizing inventory levels and implementing a very flexible ‘‘build to order’’ aspect in the warehouse. However, cross docking requires a fast, responsive transportation system, as all pick-ups and deliveries need to be made within specified time windows. Failure to meet these stringent delivery windows can dramatically reduce a cross-docking warehouse’s effectiveness. 3.3 Order processing Order processing is an important information flow along the logistics network. Information from customers’ orders determines the demand for products. Order processing consists of customer order transmittal, order processing and assembly (warehouse), and order delivery. If the warehouse is out of stock, the

IJPDLM process requires transmittal of backorder items, factory time to manufacture 32,7 and/or supply the product, and express order delivery. When an OEM partners with a contract manufacturer, the customer’s order is first transmitted to the OEM, then on to the contract manufacturer. One problem that can occur during order processing is the mismatching of system component quantities. For example, suppose Acme Computer Company decides to outsource the 616 manufacturing of a monitor to a contract manufacturer. Further, assume the company receives an order of 1,000 computer systems. Even though Acme manufactures 1,000 central processing units (CPUs) and keyboards, the contract manufacturer only ships 500 monitors to Acme’s warehouse. Now, Acme can only satisfy 500 customer orders for this situation. Although order batching, a process of collecting orders into batch processes in order to reduce processing cost, could cause this problem, the fact that there is a lead-time associated with getting the product from the contract manufacturer to Acme necessitates more advanced order processing techniques. To avoid such potential problems, electronics manufacturers often share information more frequently with contract manufacturers. By increasing the frequency of information sharing with contract manufacturers, supply chain agility is increased as companies can maintain and/or improve their ability to quickly respond to customer orders and requests. In fact, the explosion of the Internet in business-to-business transactions has led to significant gains in this arena. By implementing electronic order transmission, the electronics manufacturer can sharply reduce production lead-times, add flexibility in order modification, and speed the supply chain. 3.4 Purchasing The growing acceptance of contract manufacturing has also altered the purchasing policies of many electronics manufacturers. Traditionally, electronics manufacturers purchased raw material and manufactured them into finished products. Outsourcing often eliminates the OEM’s need to purchase raw materials. Today, electronics manufacturers must decide which processes should be outsourced, as well as how much of each product to outsource. As a result, electronics manufacturers have somewhat lost their relationship with suppliers. In the realm of forging strong supply chain relationships, the contract manufacturer selection process has replaced the vendor-selection process. Start-ups and partnership changes with contract manufacturers are expensive and time consuming. Electronics manufacturers need to select the appropriate contract manufacturer and the right quantity of product to be outsourced for every product in their portfolio. The right purchase quantity at the right time will hopefully lead to proper fulfillment of customer requirements. Mistakes made on the selection of a contract manufacturer can be catastrophic, resulting in the loss of customers, market share and, potentially, company reputation. Building the right relationship with the contract manufacturer, one based on shared information and leveraging each other’s strengths, can lead to a strong,

agile supply chain for the electronics manufacturer and improved customer Electronics satisfaction. manufacturing supply chain 4. Modeling the impacts of contract manufacturing on supply chain agility Cole et al. (2001) present a detailed mathematical model for minimizing the total cost of a logistics network that includes contract manufacturers. Their model is 617 an extension of Wu and Golbasi (1999), which does not explicitly consider contract manufacturing. For extensive surveys on similar models in the literature, see Vidal and Goetschalckx (1997). The model of Cole et al. (2001) is a multi-period, multi-commodity network model with complicating side constraints. It can be used to model the agility implications of dynamic capacity management (hiring/firing of contract manufacturers). Further, the model forms an integral part of an approach to consider agility implications of uncertainty. In words, the model can be described as follows: Minimize: Production cost (fixed, ‘‘minimum revenue’’, hire/fire, variable) + warehousing cost (fixed, variable) + transportation cost (variable) + inventory cost (variable) subject to: production capacity warehousing capacity service requirements (e.g. demand satisfaction). Although the model is comprehensive, we are most interested in the production features. The cost elements for production comprise both fixed and variable components: . Variable cost per unit produced. . Fixed cost: independent of production level. . ‘‘Minimum revenue’’ fixed cost in which a factory is guaranteed a certain dollar amount of revenue. Variable costs are charged against this fixed cost. Thus, the company pays the higher of the ‘‘minimum revenue’’ or total variable cost. This fee structure is popular among large contract manufacturers because it enables them to better plan their capacity usage. . Hire/fire cost to open or close a production line. As for production capacity, contract manufacturers are often modeled as having infinite capacity. The contract manufacturer often is required to produce a predetermined dollar volume of products by the OEM (i.e. ‘‘minimum revenue’’) without the need for the OEM to worry about the contract manufacturer’s capacity limitations/issues.

IJPDLM The fixed cost structure and capacity features of the model have an interesting 32,7 effect. Since both the original manufacturer and the contract manufacturer must allocate fixed costs over the units actually produced, the cost model tends to find extreme solutions in which all products are either produced in-house or outsourced. This is sometimes called the ‘‘internal manufacturing death spiral.’’ See Cole et al. (2001) for numerical examples. One mitigating factor is that the 618 original manufacturer often must maintain at least some capability for manufacturing prototypes of new designs, or to handle warranty service. In experiments based on real-world data, inventory costs (especially for work-in-process inventory held over between planning periods) do not play a big role in the final objective value. Due to very high obsolescence costs, it rarely makes sense to keep finished goods or work-in-process from one planning period to the next. This aids modeling because precise estimation of inventory cost parameters is difficult. Similarly, warehousing is not a dominating factor since product is not stored for long periods of time in the fast-changing electronics industry. Since electronics products tend to have high value to size ratios, transportation and warehousing costs are usually a small portion of total logistics costs. Furthermore, transportation and warehousing are relatively easy to obtain for such products; thus transport and warehouse capacities do not drive the model solution. Service is often the most difficult feature to model in a logistics network. For the agile logistics networks considered in this research, service is defined by fraction of demand satisfied within a certain time. As mentioned before, electronics products become obsolete rapidly. Thus, the time period for demand satisfaction and the production-planning period are set to be approximately equal. One of the key difficulties in finding the ‘‘best’’ logistics system design using mathematical modeling is that model parameters (demands, costs, capacities) are often subject to uncertainties. Given the ability to design a minimum cost logistics network for a specific scenario, a suggested approach to design an agile logistics network design is as follows: (1) Develop a number of scenarios with features related to agility (e.g. low demand variability over time, high demand variability over time). (2) Optimize each scenario separately (i.e. determine minimum cost solutions). (3) Evaluate each individual scenario optimum against every other scenario. (4) Evaluate results to determine which designs are ‘‘agile’’. Depending on the company’s definition of agile, this can include designs which meet any one of the following (incomplete) list of criteria: . acceptable performance over a wide range of scenarios; . results in the minimum expected total costs (based on the likelihood of each scenario); . produces the minimum probability of a poor result; . results in the maximum probability of a great result.

Note that no single design will likely meet all criteria. In fact, each criterion Electronics might best be satisfied by a different design. In the end, the human decision manufacturing maker retains a key role in actually deciding which design is ‘‘best’’ for the supply chain company. Conclusions and future work 619 The introduction and rapid acceptance of outsourcing production to contract manufacturers in electronics manufacturing has altered the electronics manufacturing supply chain. We have discussed the impacts of outsourcing on supply chain agility, such as facility location, customer service, and product distribution. Outsourcing has impacted both capital investment plans and the management cost of the manufacturing process. Future advances in the development of information technology-intensive decision support systems will play an important part in helping to improve the agility of the electronics manufacturing supply chain. Some areas for future work on the outsourcing in electronics manufacturing problem include developing a better understanding of the ‘‘business’’ side of the contractual agreements with contract manufacturers. For example, how much time is required before the company’s products are qualified for production in the contract manufacturer’s facility? What is the usual length of a contract with a contract manufacturer? Is there any minimum amount of time in which the company may not terminate their agreement with the contract manufacturer? Finally, who is responsible for insuring the proper subcomponents are on hand at the contract manufacturer’s facility so that the specified product mix can be manufactured, the company or the contract manufacturer? Answers to these and other open questions will allow for future model refinements that will foster a better understanding of the importance of contract manufacturing on the electronics manufacturing supply chain. References Ballou, R.H. (1999), Business Logistics Management, 4th ed., Prentice-Hall, Upper Saddle River, NJ. Calaf, J.E. (1995), ‘‘Value-added network in contract manufacturing’’, Annual International Conference Proceedings, American Production and Inventory Control Society, Falls Church, VA, pp. 521-4. Carbone, J. (1999), ‘‘High-tech buyers see tidal wave of opportunity’’, Purchasing, Vol. 126 No. 10, pp. 36-9. Carter, J.R. and Narasimhan, R. (1996), ‘‘Purchasing and supply management: future directions and trends’’, International Journal of Purchasing and Materials Management, Vol. 32 No. 4, pp. 2-12. Christopher, M. (1999), ‘‘Creating the agile supply chain’’, The ASCET Project, 1 April, available at: www.ascet.com/documents.asp?d_ID=200 Cole, M., Mason, S.J. and Yan, L. (2001), ‘‘Decision models for contract manufacturing’’, submitted to Computers and Industrial Engineering.

IJPDLM Conklin, J.M. (1994), ‘‘Extending capabilities through contract manufacturing’’, Electro International Conference Proceedings, pp. 145-53. 32,7 Hassig, R. (1995), ‘‘The case for contract manufacturing’’, IEEE International Test Conference, p. 296. ILOG, Incorporated (2000), CPLEX 7.0 User’s Manual, ILOG, Incorporated, Mountain View, CA. Kumar, K. (1999), ‘‘Contract manufacturing of electronic hardware’’, Electronic Information and 620 Planning, Vol. 26 No. 8-9, pp. 359-407. Lakhal, S., Martel, A., Kettani, O. and Oral, M. (2001), ‘‘On the optimization of supply chain networking decisions’’, European Journal of Operational Research, Vol. 129, pp. 259-70. Landers, T.L., Brown, W.D., Fant, E.W., Malstrom, E.M. and Schmitt, N.M. (1994), Electronic Manufacturing Processes, Prentice-Hall, Englewood Cliffs, NJ. Martin, S. (1999), ‘‘The ten most dreaded questions for the contract manufacturer: a view from the inside looking out’’, Proceedings of the Technical Program, NEPCON East ’99, pp. 87-94. Narasimhan, R. and Das, A. (1999), ‘‘Manufacturing agility and supply chain management practices’’, Production and Inventory Management Journal, Vol. 40 No. 1, pp. 4-10. Semiconductor Business News (2001), ‘‘Worldwide chip sales to fall 17% in 2001, says Dataquest,’’ Semiconductor Business News, 8 May, available at: www.siliconstrategies. com/story/OEG20010508S0059 Semiseek News (2001), ‘‘Semiconductor Industry Association reports global semiconductor market tops $200 billion mark for first time,’’ Semiseek News, 5 February, available at: www.semiseeknews.com/press_release2499.htm Smith, T.W. (2001), ‘‘Semiconductor equipment’’, Standard and Poor’s Industry Surveys, 1 February. Tolone, W.J. (2000), ‘‘Virtual situation rooms: connecting people across enterprises for supply- chain agility’’, Computer Aided Design, Vol. 32 No. 2, pp. 109-17. Veeramani, D. and Joshi, P. (1997), ‘‘Methodologies for rapid and effective response to requests for quotation (RFQs)’’, IIE Transactions, Vol. 29 No. 10, pp. 825-38. Vidal, C.J. and Goetschalckx, M. (1997), ‘‘Strategic production-distribution models: a critical review with emphasis on global supply chain models’’, European Journal of Operational Research, Vol. 98, pp. 1-18. Vining, A.R. and Globerman, S. (1999), ‘‘A conceptual framework for understanding the outsourcing decision’’, European Management Journal, Vol. 17 No. 5, pp. 644-54. Wu, S.D. and Golbasi, H. (1999), ‘‘Manufacturing planning over alternative facilities: modeling, analysis, and algorithms’’, Technical Report 99T-10, Department of IMSE, Lehigh University, Bethlehem, PA. Further reading Fourer, R., Gay, D.M. and Kernighan, B.W. (1993), AMPL: A Modeling Language for Mathematical Programming, Scientific Press, San Francisco, CA. Hau, T., Mason, S.J., Yan, L. and Cole, M.H. (2001), ‘‘Contract manufacturing’s impact on the electronics manufacturing supply chain’’, 10th Annual Industrial Engineering Research Conference, Dallas, TX. Pendrous, R. (2000), ‘‘Virtual future’’, Manufacturing Management, pp. 33-4. Schmidt, G. and Wilhelm, W.E. (2000), ‘‘Strategic, tactical, and operational decisions in multinational logistics networks: a review and discussion of modeling issues’’, International Journal of Production Research, Vol. 38 No. 7, pp. 1501-23. Simchi-Levi, D., Kaminsky, P. and Simchi-Levi, E. (1997), Designing and Managing Supply Chains, Irwin McGraw-Hill, Boston, MA.

Physical Distribution & Logistics Management

0 comments

Notes on slide 1

Favorites, Groups & Events