The Emerald Research Register for this journal is available at www.emeraldinsight.

com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0144-3577.htm

IJOPM 24,5

Postponement: a review and an integrated framework

Biao Yang
Cardiff Business School, Cardiff University, Cardiff, UK, and

468

Neil D. Burns and Chris J. Backhouse

The Wolfson School of Mechanical and Manufacturing Engineering, Loughborough University, Leicester, UK
Keywords Mass customization, Electronic commerce, Supply chain management Abstract The concept of postponement has a long history of academic literature and practical applications. The recent literature has also well documented that the need for postponement is driven by todays market and business environment. However, its applications are still not as widespread as expected. Therefore, we are interested in what factors enable or hinder the successful implementation of postponement. After a review on postponement, we propose an integrated framework that will help to develop a set of general ideas on which further work can be based. The practical difculties associated with moving toward postponement are also included in this framework. The objective is also to understand better the challenges that exist in implementing postponement strategies.

International Journal of Operations & Production Management Vol. 24 No. 5, 2004 pp. 468-487 q Emerald Group Publishing Limited 0144-3577 DOI 10.1108/01443570410532542

Introduction Every company wishes to produce exactly what customers want at the proper place and time they wish to consume (Ballou, 1999). Ideally, they trigger the entire design-make-ship cycle only when a clear demand signal is available. This may be classied as the highest level of postponement, which enables companies to maximise their prots through fully understanding real customer requirements. In reality, it may be impossible to develop such a postponement strategy in many instances because of the tolerance time (that the customer is willing to wait). However, many companies are exploring ways toward this in response to constantly changing demands. For example, if customers demand delivery with a short lead time, having modules with standardised interfaces might be a desirable option. Combining standard modules into customer-specic nished products has been mentioned as one of the most efcient options of achieving the required amount of customisation. In other words, nal customisation is postponed until customers specify the exact mix and match. This allows a manufacturer to achieve the marketing benets of customisation while reaping the cost benets of standardised production in anticipation of future customer orders (Lampel and Mintzbert, 1996). Such strategy is called mass customisation, where the key is postponing the task of differentiating a product for a specic customer until the latest possible point (Feitzinger and Lee, 1997). Not surprisingly, in the literature of mass customisation research, many authors view postponement as a practical method to move toward mass customisation (Feitzinger and Lee, 1997; Kotha, 1995; Lampel and Mintzberg, 1996). The popularity of postponement is also reective of the increasing interest in agility and e-business. In todays highly competitive environments, companies have to achieve agility in order to better respond to the rapidly changing customer-driven markets (Iacocca Institute, 1991). With a high degree of conformance to the customers

ultimate requirements, postponement has been identied as an important approach for contributing to the attainment of agility, e.g. through its contribution to the: . customisation of products and services; . use of customer order information through the supply chain; and . cross functional efforts (Van Hoek, 2000). Christopher (2000) even states that postponement is a vital element in any agile strategy. Finally, the potential afforded by e-business is felt to be wide-reaching (e.g. in new cyber retailer Amazon.com and business-to-business online procurement network General Electric Global eXchange Services) and thus, there are opportunities to re-engineer the existing supply chains or design new supply chains using postponement strategies (Lee and Whang, 2001; Yang and Burns, 2003). E-business may offer a company the ability to more efciently capture and move information, and make it available more widely among supply chain partners. However, certain supply chain areas such as physical ow of goods are hardly beneting directly from e-business. For example, there is still a need for companies to keep surplus nished inventory to meet any unanticipated online demand. Agrawal and Pak (2001) also note that although business-to-business electronic marketplaces can help companies realise certain purchasing and transaction-processing benets in the short term, broader improvements (particularly reductions in inventory and faster time to market) are harder to achieve. In the end, real-time information ow further requires a heightened awareness of the importance of timely and cost-effective delivery of individual orders (Lee and Whang, 2001). Currently, it has been a big challenge for e-business companies to match material ow to information ow. In this regard, recent literature points out that e-business companies are big users of postponement (Lee and Whang, 2001; Richardson, 2000; Twede et al., 2000). For example, given the time lag between an online order placement and fulllment, companies may stock basic common components, based on forecast, and perform nal congurations (e.g. labelling, packaging or assembly) in response to individual online orders (Yang and Burns, 2002). However, this does not mean that postponement is a new concept or just a theoretical concept. In fact it was introduced in the literature in 1950 (Alderson, 1950) and its application can be traced back to the 1920s (CLM, 1995). The concept of postponement is about delaying activities (e.g. as to the form and/or place of goods) until the latest possible point in time (e.g. when exact attributes of demand can be identied). Faced with constant changes (e.g. volume change and variety change), it is a natural option for a company to seek opportunities for delaying some activities like the nal processing or manufacturing of product to as late a time as possible and hence reduce the risk. Postponement can enhance a companys exibility to effectively meet the requirements of the growing product variety and quick response (Lee et al., 1993). In recent times the amount of research addressing postponement has greatly increased. At the same time, this concept is being used by companies as diverse as Hewlett-Packard (HP) Company, Dell Computer, Motorola, National Bicycle Industry Company in Japan (NBIC), Toyota Motor Corporation, the Gillette Company, and clothing manufacturer Benneton. Despite the increasing attention to postponement, its applications are currently much less than expected (Battezzati and Magnani, 2000; Brown et al., 2000; Van Hoek,

Postponement: an integrated framework 469

IJOPM 24,5

470

1997). In the food industry, for example, Morehouse and Bowersox (1995) predict that by the year 2010, over 50 per cent of all inventory will be stored in a semi-nished state, and nal processing and shipment will be processed against a specic customer order. However, in a recent study by Van Hoek (1999), it is found that non-modularity of product and process offers a special challenge in applying postponement in the food industry. This is further compounded by the fact that the food industry is characterised by natural uctuation/variations in weather patterns (not only leading to changes in consumer preferences, but also changes in the yield of some food products supply). There are also many examples of failing to respond to the end customer requirements, where postponement does have a manufacturing operations basis. In a launch of a new cracker product by Nabisco in 1997, two variants of a new thin Triscuits cracker were introduced. One variant was regular-avoured. The other was onion-avoured, which was manufactured from the former regular one, but with additional avouring. Signicant inventory of each variant was produced for the expected holiday sales. The market enthusiastically received the regular-avoured cracker while the onion-avoured version was not well received. Unfortunately, Nabisco was unable to fully replenish the regular-avoured version, while high levels of onion-avoured stock remained on retail shelves until after the holiday season (Bowersox et al., 1999). Thus, our interest is what factors enable or hinder the successful implementation of postponement. In the following section, we review the relevant literature on postponement. Then, we propose an integrated framework to motivate academics and practitioners to further explore the subject. A review To clearly position the essential contributions of this paper, we review postponement research in logistics, manufacturing and supply chain areas. However, we do not purport to provide a comprehensive review on postponement. For a more detailed review, we refer the readers to Van Hoek (2001). The concept of postponement was rst introduced in the marketing literature by Alderson (1950), who notes that postponement can change the differentiation of goods (form, identity and inventory location) to as late a time as possible, and thus, it could be used to promote the efciency of a marketing system, e.g. by achieving savings both in costs related to uncertainty and in the physical movement of the goods. Aldersons concept of postponement was visionary for its time since long lead times in production and distribution made it difcult to rely on postponement. With the development of technology, especially information technology, transportation technology and manufacturing technology, postponement is increasingly appealing and available. Sometimes it is under the different labels like late customisation, delayed conguration and delayed product differentiation in the literature. Nevertheless, the logic remains the same. Logistics Logistics costs (including inventory, transportation, warehousing, and the like) can make up 10 per cent or more of putting a product in a customers hand (Davis and Sasser, 1995). Reducing logistics costs is often the central goal of companies implementing postponement. Postponement in logistics enables a company to keep its options open as to where inventory should be nally deployed, thus greatly reducing the risk of wrong time and place utility of products (Bowersox et al., 1993). The benets

of postponement also include reducing inventory levels across the supply chain while improving customer responsiveness. There is empirical evidence that postponed conguration and shipment, achieved largely through logistics, is the most prominent application of postponement at the present time (Van Hoek, 1998). In one of the early successful applications of postponement, Hewlett Packard realised over $3 million/month in logistics savings from their deskjet printers (Twede et al., 2000). Postponement was originally applied in the distribution channel, involving the delay of the forward movement of inventories (Bucklin, 1965). Bucklin (1965) views postponement as a means by which a supplier may shift risk to the buyer and he extends it to the postponement-speculation principle. He focuses on where and by which player in the distribution channel inventory should be positioned. Bowersox (1978) points out that knowing the exact level and location of the demand prior to product shipment offers the potential to reduce distribution cost. In Zinn and Bowersoxs (1988) category of time postponement, it is suggested to maintain anticipatory inventory at one or a few strategic locations. In recent literature, postponement further calls for stocking differentiated products at the strategically central location to achieve a balance between inventory cost and responsiveness (Bowersox and Closs, 1996). In the distribution channel, postponement is also involved in the repositioning of manufacturing activities, not limited to the relocation of inventories. This thinking is not surprising when it is considered that the success of postponement strategies, in part, depends on where in the process generic products should take on the form of specic end products. Postponement may relocate nal conguration from manufacturer plants closer to the end-use customer, allowing for rapid delivery of customised products and quick responsiveness to changes in display mixes. For example, packaging/labelling (as one of the variables in differentiating generic goods for different brands and geographic markets) may be delayed (until customer commitments are received) and be better carried out by local based distributors before distribution and delivery to the customer. Zinn and Bowersox (1988) provide insight on how four types of postponement (labelling, packaging, assembly and manufacturing) can be applied to improve the productivity of the physical distribution system. In addition, local based distributors have a real advantage in offering the exibility to t the changes in the local market, since they are already closer to the market. This is reective of Ferdowss (1997) view that a plant (or distribution centre) located near the important customers may be selected for customisation to contribute to local responsiveness. The nature of product ow and the production routeing may also have implications for the adoption of postponement (Yang and Burns, 2003). Postponement means that some phase, which usually precedes the point of differentiation, is prolonged. In terms of inventory costs, companies may tend to apply postponement upstream, because raw material inventories are usually cheaper than the end item inventories. Within other contexts, the best location for postponement might be at the convergence of the bill of materials in that it is the place in the supply chain where the amount of parts, components, sub-assemblies or assemblies is the lowest (Andries and Gelders, 1995). Taking this into consideration, it is viable to delay a products variety, value, volume and weight increases through the implementation of postponement to save on inventory carrying and holding, assorting, stock-out and obsolescence costs (Pagh and Cooper, 1998; Van Hoek, 1997; Zinn and Bowersox, 1988). Billington and Amaral (1999)

Postponement: an integrated framework 471

IJOPM 24,5

472

also note that postponement is valuable when capacity is unresponsive (i.e. supplier lead times are long, production scheduling is inexible, or manufacturing volumes are constrained). However, in many cases, postponement in logistics requires a fast and responsive transportation system and it can thus lead to a signicant increase in transport cost (Christopher, 1998). Its applications are also constrained by time windows specied by the market. Its maximum benets are obtainable when the distribution is not only physical transportation, but also contains a product customisation component (Battezzati and Magnani, 2000). Manufacturing The concept of postponement, as applied to manufacturing, retains the product in a neutral and non-committed status as long as possible in the manufacturing process. Because the inventory is generic, its exibility is greater, meaning that the same components, modules or platforms can be embodied in a variety of end products. Thus, postponement is one of the most benecial strategic mechanisms to manage the risks associated with product variety and uncertain sales (Aviv and Federgruen, 2001). In addition, since manufacturing postponement allows companies to operate with no nished inventory, companies may benet from maintaining the bulk of their inventories in the cheaper and/or pre-customised form by delaying expensive operations and point of product differentiation. In the international business, manufacturing postponement may be relevant also in those countries where the tariffs and taxes incurred on nished products are higher for semi-nished products. It is crucial for a company engaged in postponement strategies to consider product families and generations, and explore the commonality/modularity of products and processes as much as possible (Feitzinger and Lee, 1997; Zinn, 1990). Using common components and processes from initial stages implies that the products assume their unique attributes as late as possible (Lee, 1996). Additionally, component commonality results in inventory pooling effects (e.g. lower inventory levels and reduced risk of obsolete inventories). Leveraging commonality in parts may further enable purchasing postponement. Following this strategy, Dell does not buy monitors for its PCs until a customer order has been placed. In fact, monitors do not have to be shipped rst to Dell. Instead, United Parcel Service (UPS) gets an e-mail from Dell and pulls a monitor direct from supplier stocks and coordinates it to get to the customer along with the PC. It is estimated that this postponement strategy saves freight costs of $30 per display unit (McWilliams, 1997). However, introducing too much commonality (e.g. part standardisation which is easily perceived by the customer) can reduce product differentiation, leading to a cannibalisation effect (Swaminathan, 2001). In general, one wants to commonise/standardise features that have long lead times early and differentiate late those that have short lead times closer to the customer. Recent years have seen that short lead-time customisable features are postponed in the clothing industry, such as printing logos with T-shirts. Obviously, lead time requirement may inuence the scope of postponed activities (Bucklin, 1965; Van Hoek, 1997). The complexity of delayed manufacturing activities, like special capability, technological contents and restrictive economics of scale, are another relevant operational characteristic for the implementation of postponement (Bowersox and Closs, 1996; Van Hoek, 1998). The concept of commonality can be taken one step further in the

modularity toward postponement applications. Modularity can take different forms as follows: (1) Modularity in design refers to dening the design boundaries of a product and of its components so that design features and tasks avoid creating strong interdependencies among specic components (modules) design (Baldwin and Clark, 1997). A fully modular architecture means that a change made to one component does not require a change to other components (Ulrich, 1995). (2) Modularity in production refers to designing the production process in order to make complicated products by designing and developing modules at different sites and then bringing them together to create a complete system (Baldwin and Clark, 1997). This modularity breaks down the whole production process into sub-processes that can be performed concurrently or in a different sequential order (Lee, 1998). When certain process components are performed at a different time or location (e.g. decoupled from the other process components with geographical distance, ownership separation and different cultural features), then process modularity increases (Fine, 1998). Apparently, the production process must allow for a ow of discrete steps in order for postponement to take place. In this regard, a prerequisite to manufacturing postponement is modularity in production (decoupling the process in at least two sub-processes, separated in time and/or place) (Feitzinger and Lee, 1997; Van Hoek, 1998). The choice of different postponement strategies depends on the extent to which it is possible for a company to modularise its products and processes (Chiou et al., 2002; Lee, 1998; Ulrich, 1995; Van Hoek, 1997; Yang et al., 2003). For example, Ulrich (1995) shows that a high degree of product modularity coupled with component-process exibility could render postponement possible and effective. More recently, Yang et al. (2003) also provide a conceptual framework for making postponement choice decisions based on an uncertainty-modularity matrix. The literature has well documented what effects manufacturing postponement has on distributors (or retailers). For example, distributors/retailers are increasingly performing some nal postponed manufacturing (Lowson, 2001). In this context, issues such as outsourcing, power and control, are raised around manufacturing postponement (Bucklin, 1965; Yang and Burns, 2003; Zinn and Levy, 1988). At the same time, manufacturing postponement may also inuence supplier relationships. Since manufacturing postponement means that customer orders will be fullled through production rather than through stockholding nished products, a reliable supplier network that can supply parts and services plays a key role in its application (Feitzinger and Lee, 1997). In the case of SmartCar, the extensive use of manufacturing postponement leads the suppliers to take over part of the assembly factory, where they supply components directly into the nal assembly line from their workshop adjacent to the factory (Van Hoek, 1998). Here postponement gives rise to a need for more strategic use of geographical proximity and outsourcing (e.g. providing additional services such as technical and project management support), although they are often justied on the basis of cost reductions/effectiveness (e.g. with suppliers often paying lower wages, achieving greater economies of scale and lowering transportation costs). This is also partly echoed by Sako (2002), who comments that the geographical proximity, either on a supplier part or as part of a modular consortium where the

Postponement: an integrated framework 473

IJOPM 24,5

474

module supplier has the responsibility of assembling the module directly on the automotive manufacturers assembly line (Collins and Bechler, 1997; Doran, 2003), is combined with tight synchronisation of operations between the assembler and suppliers, putting pressure on the exploitation of such cost reduction as lower wages of suppliers (since the proximity might make wage differentiations close). Close proximity can also be important for new and/or complex products that require a high level of interaction between the manufacturer and its suppliers. The application of postponement in manufacturing is a logical extension of implementing logistics postponement (Battezzati and Magnani, 2000). Using survey data from US manufacturing rms, Rabinovich and Evers (2003) also suggest that the implementation of logistics postponement supports the adoption of manufacturing postponement. In the meantime, competition in the current business environment has forced companies to recognise that efciencies in logistics have currently become as important to strategic planning as are improvements in manufacturing and marketing (Ernst and Kamrad, 2000). Accordingly, postponement strategies are going towards integrating both logistics and manufacturing activities (Pagh and Cooper, 1998). Van Hoek (1997) also notes that the integration of manufacturing postponement into logistics postponement has been coming to the forefront as an innovative supply chain arrangement. Supply chain Over time, the scope of postponement consideration has expanded from marketing to the whole supply chain. For example, in new product development, a design company was asked to develop a single integrated circuit. During the design process, the micro-processor, which is crucial to the design, was improved by the manufactures so that it ran at a 50 per cent higher clock rate. This change resulting from a technology shift led to a complete redesign which eventually added 15 per cent to the total development time (Thomke and Reinertsen, 1998). In this case postponement should be considered not directly with regard to customer needs, but technology evolution during the design process. Rao (1997) also describes how IBM applied postponement in designing asynchronous transfer mode (ATM) networking switches in response to the uncertainty about standards and protocols, as the industry had not yet fully developed standards and protocols at that moment. Thus, companies should rst consider every possible postponement opportunity along the supply chain and then balance the trade-off not from an individual player, but the whole supply chain. Accordingly, a perspective spanning the length of the supply chain should be applied in studying postponement (Van Hoek, 1999). To extend the insight into where and when in the supply chain this concept is applied, a variety of types of postponement have been proposed from different perspectives, for example, time postponement, place postponement and manufacturing/form postponement (Bowersox and Closs, 1996), and pull postponement, logistics postponement and form postponement (Lee, 1998). However, postponement applications so far mainly involve downstream activities in the supply chain (Van Hoek, 1998). Postponement enables the customer requirements to affect product design and production process, and it may thus be described as a pull approach (Shapiro, 1984). As supply chains have evolved from traditional forecast-driven push to demand-driven pull systems over time, postponement is playing an increasingly important role in a

supply chain. Postponement fosters a new way of thinking about the supply chain (Yang and Burns, 2003) and it has been identied as an important characteristic of modern and competitive supply chains (Van Hoek, 1999). For example, considering the potential of postponement in saving inventory holding and carrying costs, the longer and wider the supply chain, the greater the potential benets of postponement (Richardson, 2000; Twede et al., 2000). Cooper (1993) identied a set of operational characteristics as criteria for assessing the viability of different types of postponement in products with a global brand. Accordingly, postponement is increasingly explored as a glocalisation strategy (achieving global efciency and local responsiveness simultaneously) in the practice of international business (Taylor, 1996; Twede et al., 2000), particularly where local forecasts will be more accurate than a forecast for the worldwide volume. Figure 1 shows different supply chain strategies which are closely related to the concept of postponement. For a more detailed review of this, see Yang and Burns (2003). The dotted line in Figure 1 reects how postponement is associated with the customer decoupling point, where in the supply chain the customer order penetrates and that distinguishes forecast and order-driven activities. Obviously, leagility, mass customisation and globalisation all hinge heavily on the identication of the decoupling point, because the decoupling must be strategic to establish the required lead times to meet the customer demands. This highlights the role of postponement in the supply chain. In the leagile supply chain, for example, postponement is used to move the decoupling point closer to the end-user and increase the efciency and the effectiveness of the supply chain (Naylor et al., 1999). However, it always remains a challenge to nd an optimal balance in which upstream activities in the supply chain are performed using lean/standardisation/centralisation and downstream activities are postponed until customer orders are received. More recently, Yang and Burns (2003) investigated how postponement could help locate the decoupling point from both an information ow perspective and a material ow perspective. However, this thinking has yet to be substantiated empirically. This is further compounded by the fact that, although the literature has identied the factors inuencing the location of the DP, little attention has been directed to how these factors can be balanced. Framework It is clear from the literature that features and benets of postponement have existed for many years, but postponement has recently received more attention. This development has become mandatory to many companies in view of the current levels of market globalisation, increasing demand for product variety and customisation, rapid technological innovation, shortening product life cycles and intense competition. Apparently, the research on postponement is at present at the stage of investigating and understanding how the postponement concept can be operationalised across sectors (in order to motivate the practitioners), rather than investigating whether companies should or should not adopt strategies. In this sense, further development in postponement research tends to point towards more applied issues. In the literature, much is written to generate insights into the relevance of such factors for the implementation of postponement as high sales uctuation and short product life cycles (marketing characteristics), high degree of customisation, commonality or modularity (product characteristics) and decouplability of product process (production

Postponement: an integrated framework 475

IJOPM 24,5

476

Figure 1. Postponement and different supply chain strategies

characteristics). However, the current postponement literature lacks any study in depth on how to (re-) congure the existing organisational structures and cultures into postponement. Further, little research has been focused on the formulation of methodologies that realign company strategies, capabilities, process and resources. Here we propose an integrated framework in Figure 2, where we are mainly concerned with the development of a set of general ideas upon which future work can be based. The practical difculties (and some potential solutions) associated with moving towards postponement are also included in this framework. Precondition Before embarking upon postponement, companies need to identify and fully understand the marketplace requirements. There are certain conditions and situations where the implementation of postponement is more likely to succeed. Its benets vary with the competitive environment. Although highly competitive markets and short product life cycles rarely make sense to use stocks to buffer supply/market uncertainty, some manufacturers are forced to do so (rather than looking to postponement) in order to remain competitive with other manufacturers who are able to offer shorter lead times and/or variety (Newman et al., 1993). The degree of uncertainty is signicant for selecting an appropriate postponement strategy. The strength of postponement lies in its capability of coping with those uncertainties inherent in dynamic and changing markets, which companies may have to accommodate in their business strategies (Yang et al., 2003). In easily predictable environments, companies would obviously gain little from postponement. In new product development, postponement is favourable due to uncertainty about customer requirements and technical evolution during the development process. For innovative products, the supply chain should respond quickly to the unpredictable demand to minimise stock-outs, forced markdowns, and obsolete inventory (Fisher, 1997). In this situation, it will be appropriate to postpone the nal manufacturing and logistics operations (Pagh and Cooper, 1998). For postponement to work, companies need to rst gure out what can be well forecasted and what cannot be forecasted. Postponement considerations are only given

Postponement: an integrated framework 477

Figure 2. Integrated framework for postponement application

IJOPM 24,5

478

to those decisions about unpredictable items. That is, postponement is a strategy to intentionally delay the execution of a task instead of starting it with incomplete or unreliable information input. At this stage, it is essential to anticipate the extent of the variability of those unpredictable items, leading to an exhaustive set of solutions in the sense that any future change of solutions (responding to real demands) fall within its boundaries. The logic behind this is to defer commitment of the most uncertain items to a later time when these could be dened more efciently in response to such possible major changes in customer needs, technologies and competitors action. In this way, postponement maps various decisions to different time horizons and alleviates the need to make some decisions prior to the availability of adequate information. In this context, postponement is relevant to the principle that, in general, short-term forecasts for an item are more accurate than long-term ones. For global products tailored to local markets, it is increasingly difcult to forecast demand for every product variation, which results from technical requirements and customer preferences, in every sales location. However, once companies choose to target certain markets, the extent of product variability is clear and forecasting is easier at the generic level than at the level of the nished item (Christopher, 1998). As often cited in the literature as the practitioners of postponement, HP has to forecast total demand of printers and Benneton must forecast aggregate demand for garments of a certain size. Further information (e.g. information on local demands) must be available in the delay period. Postponement is valuable only if the information about the customers need can be captured quickly and accurately (Lee, 1998). Postponement does not necessarily rely on advanced IT (Van Hoek et al., 1999) and companies can combine postponement with simple information sharing to achieve low supply chain costs and high customer service level (Billington and Amaral, 1999). However, latest information technology may extend the application of postponement or even enhance the value of postponement. The information for postponement, in many cases, does not only refer to the customers order information which is transmitted in a timely manner without distortion from the players closer to the customers. Companies sometimes need some information about the ongoing demand. The precision is a major concern for this kind of information. The Internet provides a way for richer understanding of customers requirements. An electronic broker can increase the precision of demand information and thus, enhance the value of postponement (Elofson and Robinson, 1998; Robinson and Elofson, 2001). Approaches Considering the application of postponement, there should be postponable points. At these points, companies have to examine product/production characteristics and analyse the costs and benets of postponement. Analytical models are often powerful as a means to help motivate or justify postponement (Lee, 1998). Once postponable points have been chosen, companies can begin to create those points. Modular/standard design in product and production is conducive to creating postponable points (Lee, 1998; Lee and Tang, 1997). However, modularity in product and production is necessary but not in itself sufcient for postponement applications. For postponement to work, product design may need to further isolate the most variable portion of the functionality from other functions in order to be added last. This becomes obvious when one poses the question:

what is the motivation behind modularity? For modularity in design, product designers are interested in reducing lead times and costs for design and development (e.g. in parallel development of modules by independent design teams). For modularity in production, managers try to increase operational efciency. Modular production/design may enable development/production of the modules to be carried out in parallel processes, thus reducing production cycle time (Lee, 1998). The more recent initiative to introduce production modules is also associated with the manufacturers wish to cope with in-line complexity due to ever-increasing product variety. Furthermore, a strong linkage exists between the design of the product, the design of the production processes and the accompanying supply chain of participating organisations. Fine (1998) claims that integral products tend to be developed and built by integral processes and supply chains, whereas modular products tend to be designed and built by modular processes and supply chains. However, pre-existing organisation structures and capabilities also inuence product architecture/design (Gulati and Eppinger, 1996). For example, some highly specialised companies are closely linked to within-modular innovation (Chesbrough and Kusunoki, 2001). They are unlikely to promote modular product architecture that provides competitors with advantages within modules and renders their integrative skills less valuable. Also, under uncertainty, it is better to err on the side of integrality (e.g. integrality in organisation architecture and system integration capability) than on modularity in product architecture to promote innovation (Ethiraj and Levinthal, 2003), as it requires addressing unpredictable inter-module or inter-system interdependencies. For this reason, companies may well choose to retain their systems integration capabilities and remain a product architect (with a product-wide system knowledge). In the postponement literature, Feitzinger and Lee (1997) also address this concurrency where product architecture and manufacturing processes should be designed so that they consist of independent modules that can be moved or rearranged easily to support different distribution-network designs. On this basis, it is tempting to characterise modularity as an organisational and managerial broad issue, linked to the implementation of postponement. Modularity in organisation relates to those organisational processes, governance structures and contracting procedures that are adopted or utilised to accommodate modular production at both the intra- and inter-rm context (Schilling, 2000). Adding this element to the postponement implementation might offer insights into further progressing the level of practice. Since successful postponement applications rst lie in organising product architecture, it is necessary to incorporate postponement thinking all the way up to product development. However, in the early phases of a product development project, detailed information on the product attributes (e.g. due to uncertainty about customer preferences and new technology) may not be available. Thus, product development may have to look to postponement in itself (Yang et al., 2003), avoiding major changes later in the development process (which are inherently expensive). Further, with the eeting windows of marketing opportunities, companies may not be able to simply wait until all aspects of exact information are available. For example, facing technological uncertainties, companies may delay their commitment to a single technology by pursuing simultaneous commercialisation of two or more competing technologies, as opposed to simply waiting to enter later until the dominant design emerges (Hateld et al., 2001). That is, during the delay, alternative options (e.g. about

Postponement: an integrated framework 479

IJOPM 24,5

480

technology and thus product architectures) should be seriously considered, and companies thus need to have more knowledge than needed for what they make. In doing so, companies may gain industry- and product-level knowledge and experience in applying postponement before the dominant technology emerges. However, there is a big challenge when the emphasis shifts from the capability for focused and rapid project execution to the capability to gather and incorporate new technical and market information during the course of the project itself. Not surprisingly, some projects that attempted to investigate a broad set of concept alternatives were forced to focus early and freeze the product concept (Iansiti, 1995). In the postponement example at Toyota, Toyota relied more on its suppliers in their product development: its suppliers (designers) deeply and simultaneously explored broad regions of the design space, and provided a set of possibilities, while Toyota continuously compiled information on the market and technological developments, various constraint and trade-offs between the alternatives (Ward et al., 1995). However, this co-design heavily hinges on a long-term experience together (Dyer and Ouchi, 1993), where Toyotas suppliers even invest in developing in ideas and plans for the next model in advance. Many companies will not be capable of such inter-company coordination necessary to implement such a postponement strategy. In fact, some authors (e.g. Hsuan, 1998) argue that the above case at Toyota may only be existing in Japanese practices where the operations and planning are highly integrated between manufacturers and suppliers. In addition, postponement implementation may have to consider the nature of design information. Baldwin and Clark (1997) suggest that designers achieve modularity by portioning information into visible rules (e.g. an architecture and interfaces) and hidden design parameters. Companies may choose to control the visible information (e.g. by designing proprietary architectures) or design modules (where they conform to the architecture, interfaces and test protocols of others) (Baldwin and Clark, 1997). Generally visible rules affect subsequent design decisions. For example, an architecture species what modules will be part of the system and what their functions will be. Interfaces describe in detail how the modules will interact. Therefore, the visible design rules should be established early in a design process. Conversely, hidden elements can be chosen late and changed often and do not have to be communicated to anyone beyond the local module design team. That is, companies may look for postponement opportunities in hidden design rather than visible design. At the same time, companies have to understand how to tackle obstacles in the way to an effective adoption of postponement. Companies generally nd it difcult to change routines and knowledge that is deeply embedded in the organisation. Exploring postponement is no exception. An organisations administrative heritage and the lack of an overall supply chain vision can be a major bottleneck (Van Hoek et al., 1999). Consequently, traditional culture companies will deliberately distort order information to mask their intent not only to competitors, but even to their own suppliers and customers (Towill, 1996). Thus, even though now it is possible for companies to capture, validate, and forward information in real time, the players upstream in the supply chain may receive distorted information. The latest techniques like the Internet, electronic data interchange (EDI) and point-of-sale system may improve the information ow by reducing the data collection errors and moving data quickly. However, they do not eliminate distortions or identify good information in the information ow. In this regard, it may not be enough simply to rely on the technology

to cope with distortion in the information ow. Even when market information is shared by all supply chain players in an undistorted way, the performance of the whole supply chain might still be poor (if all players do not share a common target) (Hong-Minh et al., 2000). To further enhance information ow towards postponement strategies, companies need to recognise the need for open communication, and manage the information whilst being willing to act as partners in the supply chain (Mason-Jones and Towill, 1999). Some other managerial challenges are also posed for postponement implementation, because it may have adverse implications for the existing managerial practices. For example, there is a fundamental belief that projects will get done faster and better if resource utilisation is maximised (e.g. everyone should be busy) over all parts of the project (Elliott and Hughes, 2000). With the earliest possible start date in mind, this also forces companies to specify items with incomplete information (Thomke and Reinertsen, 1998), exposing them to making changes when accurate information becomes available. At Toyota, product development postponement seems completely inconsistent with the speed, economy, and effectiveness of Toyotas development process. Therefore, a critical part of the managers job is to prevent people from making decisions too quickly (Ward et al., 1995). Based on the above approaches, postponement can be achieved by delaying operations or by re-sequencing processes, each of which is associated with a separate attribute (Lee and Tang, 1997), e.g. by re-sequencing knitting and dyeing at Benetton. It is important to note that differences in the internal organisation and the external demands for products require different postponement strategies (Van Hoek et al., 1999). Companies at the different starting points require different postponement strategies. Also companies implementing postponement may need to modify their postponement strategies over time, according to changes in the competitive environments, stages of product life cycle, technological advances and so on. To select an appropriate postponement strategy, for instance, companies may need to consider different phases of the product life cycle. Bowersox et al. (1999) investigated the role of postponement in the new product launch (i.e. at the introduction stage), and concluded that postponement enables better management of inventory levels, and improves replenishment times and in-stock availability of products. Postponement can also reduce launch inventory exposure to cut losses in product launch failure. Twede et al. (2000) notice that currently HP (as one of early successful adopters of postponement) mainly applies postponement strategies for new products during the growth stage of the product life cycle. However, considering the downstream part of the supply chain (from factory level to end customer), Pagh and Cooper (1998) propose that starting from the full speculation strategy will be more appropriate in the introduction and growth stages, while starting from the full postponement strategy will be more appropriate in the maturation and decline stages. We argue intuitively that postponement might not always be relevant during the decline stage. For example, companies may make a downward substitution offer (providing the customer with one of the available models that have a superset of features required by the customer) (Swaminathan, 2001). In this way, companies may reduce the negative impact of customisation related to product proliferation both in terms of reducing costs for managing existing products (e.g. inventory reductions) and the number of product design changes over the life cycle.

Postponement: an integrated framework 481

IJOPM 24,5

Therefore, it may be worth further investigating the issues around postponement in different product life cycles. Fit other concepts into postponement While postponement applications are going through the whole supply chain, negative effects are associated with it. Typically the implementation of postponement might lead to reducing economies of scale and increasing cycle time. There is a need for integrating other related concepts/strategies into postponement, supplementing rather than replacing each other. This is consistent with the increasing calls for a supply chain wide perspective in postponement research (Van Hoek, 2001; Zinn and Levy, 1988). Following this, further research needs to consider the postponement concept in holistic terms. To accomplish postponement with time and processing cost constraints, some attempts have recently been made in relation to just in time (Van Hoek, 2001; Waller et al., 2000), consolidation in transportation (Ballou, 1999; Bhatnagar and Viswanathan, 2000; Bowersox and Closs, 1996), merge-in-transit (Bowersox et al., 1999), cross-docking (Yang et al., 2003), lead time reduction and quick response (Aviv and Federgruen, 2001), transshipment (Herer et al., 2002) and exibility (Yang et al., 2003). Apparently, some of the above concepts have similar characteristics as postponement. For example, transshipment can also lead to cost reductions and improved services by enabling the sharing of stock among different locations, thereby reducing the inventory in the supply chain. Further, transshipment can be viewed as logistics postponement, because it delays the point of differentiation which transforms a generic item (an item at any location) into a specic item (an item at a specic location) (Herer et al., 2002). In a slightly different way, Yang et al. (2003) investigate the role of postponement for the management of uncertainty, suggesting that some managerial practices (e.g. towards the reduction of internal uncertainties) have a positive relationship with the adoption of postponement. Therefore, one idea to pursue postponement strategies might be based on the earlier methods such as quality function deployment (QFD), which has considerable customer involvement. Nevertheless, most of the above research is still highly conceptual, and thus empirical validation is a direction for further research. Furthermore, in contrast with the relatively rich literature on postponement enabling practices and technologies, it is less easy to nd information about what factors hinder the adoption and implementation of postponement. This may deserve more investigation in terms of removing barriers to postponement. Finally, although a lot of research has pointed out the link between postponement and such business strategies as supply chain management, mass customisation, agility and e-business, there are still insufcient studies to fully investigate this link. Conklin (1994) asserts that no single enterprise in todays global marketplace is able to realise market opportunities in a timely and cost-effective way, mainly due to the lack of solid skills and experience bases. This assertion is consistent with the view that competition no longer takes place between individual businesses, but between supply chains. When increasingly adopting strategies to improve overall supply chain management, companies may look to wider implementation of postponement (Bowersox and Morash, 1989; Davis and Sasser, 1995). However, this does not mean that players in a supply chain may have natural tendencies to develop cooperative (rather than competitive) relationships. In manufacturing postponement, manufacturers may

482

require distributors to take on additional responsibilities (due to the relocation of nal manufacturing downstream to distributors/retailers). Manufacturers may also need to foster and develop manufacturing capabilities (e.g. skills to assemble the products) within a particular distributor/retailer, since a companys success is partly tied to the strength of its weakest supply chain partner. However, in many industries like the automotive, aerospace and personal computer industries, many original equipment manufacturers share common distributors/retailers. The existence of common or overlapping distributors/retailers in different supply chains may limit the manufacturers ability to invest in the related facilities and human training programs of distributors/retailers. The reason for this is that any investment in this will provide a free benet for competitors (manufacturers in other supply chains) (Rice and Hoppe, 2001). Obviously, this added competitive aspect of the relationship sometimes complicates the effective employment of postponement strategies. Additionally, as demonstrated in the introductory section, important consideration is being given to mass customisation, agile and e-business strategies in many companies at present. These strategies should result in more interest in postponement. However, there has been absence of empirical research supporting this implication. It is still unclear whether companies adopting a higher degree of postponement are likely to be more successful than their competitors adopting a lower degree of postponement. In other words, should companies adopt postponement strategies because this may lead to better performance? Conclusion After a literature review, this paper proposes a framework covering a variety of important and unexplored facets of postponement that should motivate academics and practitioners to further explore the subject. Particularly, we concentrate our analysis on how to better understand the challenges that exist in implementing postponement strategies. For example, we have extended the work on modular design for postponement by taking it a step further and suggesting that companies may need to incorporate the appropriate organisational changes with the development of postponement strategies. This may corroborate the idea that postponement is not every companys best strategy, since existing organisation structure and culture of some companies may not t the business opportunities inherent in postponement. This paper has attempted to address the question of why postponement is much less applied than expected. Meanwhile, we have also raised many more questions. We hope that this could bring added interest into postponement research. Finally, it should be noted that while further development in postponement research tends to point towards more applied issues, there are pending issues regarding the theoretical aspects of postponement. For example, there is a need to further develop the postponement concept to match the extending scope of its applications (e.g. in product development, purchasing and supply chain paradigm).
References Agrawal, M.K. and Pak, M.H. (2001), Getting smart about supply chain management, The McKinsey Quarterly, Vol. 2, pp. 22-5. Alderson, W. (1950), Marketing efciency and the principle of postponement, Cost and Prot Outlook, Vol. 3, pp. 15-18.

Postponement: an integrated framework 483

IJOPM 24,5

484

Andries, B. and Gelders, L. (1995), Time-based manufacturing logistics, Logistics Information Management, Vol. 8 No. 3, pp. 30-6. Aviv, Y. and Federgruen, A. (2001), Design for postponement: a comprehensive characterization of its benets under unknown demand distributors, Operations Research, Vol. 49 No. 4, pp. 578-98. Baldwin, C.Y. and Clark, K.B. (1997), Managing in an age of modularity, Harvard Business Review, Vol. 75 No. 5, pp. 84-93. Ballou, R.H. (1999), Business Logistics Management: Planning, Organising, and Controlling the Supply Chain, 4th ed., Prentice-Hall International, London. Battezzati, L. and Magnani, R. (2000), Supply chains for FMCG and industrial products in Italy: practices and the advantages of postponement, International Journal of Physical Distribution & Logistics Management, Vol. 30 No. 5, pp. 413-24. Bhatnagar, R. and Viswanathan, S. (2000), Re-engineering global supply chain, International Journal of Physical Distribution and Logistics Management, Vol. 30 No. 1, pp. 13-34. Billington, C. and Amaral, J. (1999), Investing in product design to maximize protability through postponement, Achieving Supply Chain Excellence through Technology, Montgomery Research, San Francisco, CA, pp. 64-71. Bowersox, D.J. (1978), Logistical Management, Macmillan, New York, NY. Bowersox, D.J. and Closs, D.J. (1996), Logistical Management: The Integrated Supply Chains Process, McGraw-Hill, New York, NY. Bowersox, D.J. and Morash, E.A. (1989), The integration of marketing ows in channels of distribution, European Journal of Marketing, Vol. 23 No. 2, pp. 58-67. Bowersox, D.J., Carter, P.L. and Monczka, R.M. (1993), Materials logistics management, International Journal of Physical Distribution & Logistics Management, Vol. 23 No. 5, pp. 46-51. Bowersox, D.J., Stank, T.P. and Daugherty, P.J. (1999), Lean launch: managing product introduction risk through response-based logistics, Journal of Production Innovative Management, Vol. 16, pp. 557-68. Brown, A.O., Lee, H.L. and Petrakian, R. (2000), Xilinx improves its semiconductor supply chain using product and process postponement, Interfaces, Vol. 30 No. 4, pp. 65-80. Bucklin, L.P. (1965), Postponement, speculation and the structure of distribution channels, Journal of Marketing Research, February, pp. 26-31. Chesbrough, H.W. and Kusunoki, K. (2001), The modularity trap: innovation, technology phase shifts and the resulting limits of virtual organizations, in Nonaka, I. and Teece, D.J. (Eds), Managing Industrial Knowledge: Creation, Transfer and Utilization, Sage, Thousands Oaks, CA, pp. 77-91. Chiou, J.S., Wu, L.Y. and Hsu, J.C. (2002), The adoption of form manufacturing strategy in a global system: the case of Taiwanese information technology industry, Journal of Business Logistics, Vol. 23 No. 3, pp. 107-24. Christopher, M. (1998), Logistics and Supply Chain Management, 2nd ed., Financial Times/Pitman, London. Christopher, M. (2000), The agile supply chain: competing in volatile markets, Industrial Marketing Management, Vol. 29 No. 1, pp. 37-44. Collins, R. and Bechler, K. (1997), Outsourcing in the automotive industry: from JIT to modular consortia, European Management Journal, Vol. 15 No. 5, pp. 498-508. Conklin, J.M. (1994), Extending capabilities through contract manufacturing, Proceedings of Electro International Conference, pp. 145-53. Cooper, J.C. (1993), Logistics strategy for global business, International Journal of Physical Distribution & Logistics Management, Vol. 24, pp. 12-23.

Council of Logistics Management (CLM) (1995), World Class Logistics: The Challenge of Managing Continuous Change, CLM, Oak Brook, IL. Davis, T. and Sasser, M. (1995), Postponing product differentiation, Mechanical Engineering, November, pp. 105-7. Doran, D. (2003), Supply chain implications of modularisation, International Journal of Operations and Production Management, Vol. 23 No. 3, pp. 316-26. Dyer, J.H. and Ouchi, W.G. (1993), Japanese style partnerships: giving companies a competitive edge, Sloan Management Review, Vol. 35 No. 1, pp. 51-63. Elliott, S.S. and Hughes, B. (2000), High-velocity new product development, Proceedings of the 9th International Conference on Management of Technology, Miami, FL, 20-25 February. Elofson, G. and Robinson, W.N. (1998), A custom mass-production on the Internet, Communications of the ACM, Vol. 41 No. 3, pp. 56-62. Ernst, R. and Kamrad, B. (2000), Evaluation of supply chain structures through modularization and postponement, European Journal of Operational Research, Vol. 124 No. 3, pp. 495-510. Ethiraj, S.K. and Levinthal, D. (2003), Modularity and innovation in complex systems, working paper, available at: http://lem.sssup.it/WPLem/les/2003-15.pdf Feitzinger, E. and Lee, H.L. (1997), Mass customization at Hewlett-Packard: the power of postponement, Harvard Business Review, Vol. 75 No. 1, pp. 116-21. Ferdows, K. (1997), Making the most of foreign factories, Harvard Business Review, Vol. 75 No. 2, pp. 73-86. Fine, C.H. (1998), Clockspeed Winning Industry Control in the Age of Temporary Advantage, Perseus Books, Reading, MA. Fisher, M.L. (1997), What is the right supply chain for your product, Harvard Business Review, Vol. 75 No. 2, pp. 105-16. Gulati, R. and Eppinger, S.D. (1996), The coupling of product architecture and organizational structure decisions, Working Paper No. 3906, Sloan School of Management, MIT, Boston, MA. Hateld, D.E., Tegarden, L.F. and Echols, A.E. (2001), Facing the uncertain environment from technological discontinuities: hedging as a technology strategy, Journal of High Technology Management Research, Vol. 12 No. 1, pp. 63-76. Herer, Y.T., Tzur, M. and Yucesan, E. (2002), Trans-shipments: an emerging inventory recourse to achieve supply chain leagility, International Journal of Production Economics, Vol. 80 No. 3, pp. 201-12. Hong-Minh, S.M., Disney, S.M. and Naim, M.M. (2000), The dynamics of emergency trans-shipment supply chains, International Journal of Physical Distribution and Logistics Management, Vol. 30 No. 9, pp. 788-815. Hsuan, J. (1998), Modularization in black-box design: implications for supplier-buyer partnerships, DRUID Winter Conference, Holte, 7-9 January. Iacocca Insitute (1991), 21st Century Manufacturing Enterprise Strategies, Lehigh University, Bethlehem, PA. Iansiti, M. (1995), Shooting the rapids: managing product development in turbulent environments, California Management Review, Vol. 38 No. 1, pp. 37-58. Kotha, S. (1995), Mass customisation: implementing the emerging paradigm for competitive advantage, Strategic Management Journal, Vol. 16, pp. 21-42. Lampel, J. and Mintzberg, H. (1996), Customizing customization, Sloan Management Review, Vol. 38 No. 1, pp. 21-30. Lee, H.L. (1996), Effective inventory and service management through product and process redesign, Operations Research, Vol. 44 No. 1, pp. 151-9.

Postponement: an integrated framework 485

IJOPM 24,5

Lee, H.L. (1998), Postponement for mass customization: satisfying customer demands for tailor-made products, in Gattorna, J. (Ed.), Strategic Supply Chain Alignment, Gower, Brookeld, VT, pp. 77-91. Lee, H.L. and Tang, C.S. (1997), Modelling the costs and benets of delayed product differentiation, Management Science, Vol. 43 No. 1, pp. 40-53. Lee, H.L. and Whang, S. (2001), Winning the last mile of e-commerce, MIT Sloan Management Review, Vol. 42 No. 4, pp. 54-62. Lee, H.L., Billington, C. and Carter, B. (1993), Hewlett-Packard gains control of inventory and service through design for localisation, Interfaces, Vol. 23 No. 4, pp. 1-11. Lowson, R.H. (2001), Customized operational strategies for retailers in fast-moving consumer industries, The International Review of Retail Distribution and Consumer Research, Vol. 11 No. 2, pp. 201-24. McWilliams, G. (1997), Whirlwind on the Web, Business Week, pp. 51-3. Mason-Jones, R. and Towill, D.R. (1999), Using the information decoupling point to improve supply chain performance, The International Journal of Logistics Management, Vol. 10 No. 2, pp. 13-26. Morehouse, J.E. and Bowersox, D.J. (1995), Supply Chain Management: Logistics for the Future, Food Marketing Institute, Washington, DC. Naylor, J.B., Naim, M.M. and Berry, D. (1999), Leagility: integrating the lean and agile manufacturing paradigms in the total supply chain, International Journal of Production Economics, Vol. 62 No. 1/2, pp. 107-18. Newman, W.R., Hanna, M. and Maffei, M.J. (1993), Dealing with the uncertainties of manufacturing: exibility, buffers and integration, International Journal of Operations & Production Management, Vol. 13 No. 1, pp. 19-34. Pagh, J.D. and Cooper, M.C. (1998), Supply chain postponement and speculation strategies: how to choose the right strategy, Journal of Business Logistics, Vol. 19 No. 2, pp. 13-33. Rabinovich, E. and Evers, P.T. (2003), Postponement effects on inventory performance and the impact of information systems, International Journal of Logistics Management, Vol. 14 No. 1. Rao, S.S. (1997), Chips that change their spots, Forbes, Vol. 160 No. 1, pp. 294-6. Rice, J.B. and Hoppe, R.M. (2001), Supply chain versus supply chain: the hype and the reality, Supply Chain Management Review, September/October, pp. 46-54. Richardson, H.L. (2000), Packaging for todays demanding fullment, Transportation and Distribution, Vol. 41 No. 7, pp. 72-4. Robinson, W.N. and Elofson, G. (2001), Electronic broker impacts on the value of postponement in a global supply chain, Journal of Global Information Management, Vol. 9 No. 4, pp. 29-43. Sako, M. (2002), Modularity and outsourcing: the nature of co-evolution of product architecture and organisation architecture in the global automotive industry, working paper, Said Business School, Oxford University, available at: www.sbs.ox.ac.uk/downloads/SakoModularity.pdf Schilling, M. (2000), Toward a general modular systems theory and its application to interrm product modularity, Academy of Management Review, Vol. 25 No. 2, pp. 312-34. Shapiro, R.D. (1984), Get leverage from logistics, Harvard Business Review, Vol. 62 No. 3, pp. 119-26. Swaminathan, J.M. (2001), Enabling customization using standardized operations, California Management Review, Vol. 43 No. 3, pp. 125-35.

486

Taylor, J.C. (1996), International marketing and the role of logistics: logistics strategies and the implications for management and government, Advances in International Marketing, Vol. 7, pp. 83-98. Thomke, S. and Reinertsen, D. (1998), Agile product development: managing development exibility in uncertain environments, California Management Review, Vol. 41 No. 1, pp. 8-30. Towill, D.R. (1996), Time compression and supply chain management a guided tour, Supply Chain Management: An International Journal, Vol. 1 No. 1, pp. 15-27. Twede, D., Clarke, R.H. and Tait, J.A. (2000), Packaging postponement: a global packaging strategy, Packaging Technology Science, Vol. 13 No. 3, pp. 105-15. Ulrich, K. (1995), The role of product architecture in the manufacturing rm, Research Policy, Vol. 24 No. 3, pp. 419-40. Van Hoek, R.I. (1997), Postponed manufacturing: a case study in the food supply chain, Supply Chain Management, Vol. 2 No. 2, pp. 63-75. Van Hoek, R.I. (1998), Logistics and virtual integration: postponement, outsourcing and the ow of information, International Journal of Physical Distribution & Logistics Management, Vol. 28 No. 7, pp. 508-23. Van Hoek, R.I. (1999), Postponement and the reconguration challenge for food supply chains, Supply Chain Management, Vol. 4 No. 1, pp. 18-34. Van Hoek, R.I. (2000), The thesis of leagility revisited, International Journal of Agile Management Systems, Vol. 2 No. 3, pp. 196-201. Van Hoek, R.I. (2001), The discovery of postponement: a literature review and directions for research, Journal of Operations Management, Vol. 19 No. 2, pp. 161-84. Van Hoek, R.I., Vos, B. and Commandeur, H.R. (1999), Restructuring European supply chains by implementing postponement strategies, Long Range Planning, Vol. 32 No. 5, pp. 505-18. Waller, M.A., Dabholkar, P.A. and Gentry, J.J. (2000), Postponement, product customization, and market-oriented supply chain management, Journal of Business Logistics, Vol. 21 No. 2, pp. 133-60. Ward, A., Liker, J.K., Cristiano, J.J. and Sobek, D. II (1995), The second Toyota paradox: how delaying decisions can make better cars faster, Sloan Management Review, Vol. 36 No. 3, pp. 43-71. Yang, B. and Burns, N.D. (2002), Postponement in new product introduction, Proceedings of the 10th International Manufacturing Conference in China, Xiamen. Yang, B. and Burns, N.D. (2003), Implications of postponement for the supply chain, International Journal of Production Research, Vol. 41 No. 9, pp. 2075-90. Yang, B., Burns, N.D. and Backhouse, C.J. (2003), The management of uncertainty through postponement, International Journal of Production Research, Vol. 42 No. 6, pp. 1049-64. Zinn, W. (1990), Developing heuristics to estimate the impact of postponement on safety stock, International Journal of Logistics Management, Vol. 1 No. 2, pp. 11-16. Zinn, W. and Bowersox, D.J. (1988), Planning physical distribution with the principle of postponement, Journal of Business Logistics, Vol. 9 No. 2, pp. 117-36. Zinn, W. and Levy, M. (1988), Speculative inventory management: a total channel perspective, International Journal of Physical Distribution and Materials Management, Vol. 18 No. 5, pp. 34-9.

Postponement: an integrated framework 487