Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Supply chain management: notes on the capability and the limitations of a modern logistic paradigm

Supply chain management: notes on the capability and the limitations of a modern logistic paradigm Logist. Res. (2009) 1:71–82 DOI 10.1007/s12159-009-0013-x OR IGINAL PAPER Supply chain management: notes on the capability and the limitations of a modern logistic paradigm Wolf-Ru ¨ diger Bretzke Received: 20 November 2008 / Accepted: 21 May 2009 / Published online: 7 July 2009 Springer-Verlag 2009 Abstract The paper analyses the drawbacks and the ‘‘movement away from functional department suboptim- potential diseconomies of the attempt to optimize a supply ization to a holistic optimization of the entire supply chain in a holistic manner. The conclusion derived by the chain’’ [3, p. 21]. author is that this is neither a viable nor a beneficial idea. Obviously this definition leads to a dramatically exten- As a consequence he advocates a paradigm shift. The ded scope of the resources and activities that managers are concept of totally integrated supply chains should be expected not only to control but to ‘‘optimize’’: ‘‘supply replaced by the idea of loosely coupled processes and chain management spans the entire enterprise and beyond, planning systems run by companies who preserve their encompassing suppliers on one end and customers on the autonomy and use competition as an incentive system and a other’’ [4, p. 221]. In order to surpass the boundaries of source of energy and flexibility. These companies should individual rationality companies obviously have to estab- interconnect on a bilateral level in order to exchange lish something not seen or thought of before: a manage- information about updated demand forecasts and the ment beyond the limits of ownership. ‘‘This means treating availability of capacities. But they should not develop stages in the supply chain that a company does not own as strongly integrated networks who start competing with belonging to the company’’ [5, p. 41]. Before taking a other supply chains on the level of this new identity. critical look at the potential benefits and shortcomings of this concept we obviously have to ask ourselves how this can be arranged. 1 Introduction Under the headline ‘‘supply chain management’’ we are confronted with two different kinds of scientific work. At ‘‘Despite 20 years of ongoing research…there is no con- the bottom there is a kind of toolbox comprising models sensus on what SCM really is’’ [1, p. 70]. It is not an easy like vendor managed inventory (VMI), available/capable exercise to review and criticize a concept which is so iri- to promise (ATP) or collaborative planning, forecasting descent. In order to focus our line of thought we refer to a and replenishment (CPFR). These process models are definition from the early nineties provided by Cooper and based on an extension of the kind of information that Elram [2, p. 1], cited by Bechtel and Jayaram [3, p. 37]: companies should exchange while doing business with ‘‘supply chain management is an approach whereby the each other. Beyond transactional data like orders, delivery entire network from the supplier to the ultimate customer is advices or invoices companies are expected to inform analysed and managed in order to achieve the ‘best’ out- each other about updated demand forecasts, inventory come for the whole system’’. Within subsequent publica- days of reach, available production capacities and the tions the pretension was further increased by claiming a current status of all activities affecting the adherence to delivery dates. Based on this new visibility companies are provided with an early view of potential bottlenecks in the W.-R. Bretzke (&) supply chain, which in turn enables them to generate Jentgesallee 9, 47799 Krefeld, Germany robust plans. As a consequence the amount of surprise, e-mail: W.Bretzke@t-online.de emergency and costly ad hoc adjustments can be reduced. URL: http://www.bretzke-online.de 123 72 Logist. Res. (2009) 1:71–82 Processes get more synchronized (that is the famous The following comments result in the statement that this bullwhip effect will become less harmful) and more ori- holistic approach is neither viable nor desirable and ented to the end customer’s demand. It is a common therefore not likely to be turned into practice. In order to character of these models that they can be (and usually substantiate this converse proposition we will have to take are) implemented within a two-sided relationship. Due to a closer look at the implications and consequences of the the stronger integration within these partnerships as an idea of integrating companies across all stages of a value additional benefit one can expect a considerable reduction chain ‘‘from sheep to shop’’ in a rigid manner. The first and of transaction costs. What these models cannot provide is most basic issue highlights the question of the formation of any kind of a ‘‘holistic optimization’’ across several supply chains. This issue is less trivial than many authors in companies on different stages of a value chain, because this field suggest. without further construction work their implementation In a common work titled ‘‘defining supply chain man- does not lead to a system of a higher order which can be agement’’ seven renowned US-scientists start their analysis submitted to such an operation. with the finding, ‘‘that there remains considerable confu- On top of these process models many protagonists of sion as to its meaning’’ (see [8]. Nevertheless they feel sure supply chain management therefore propagate the idea that that supply chains ‘‘exist, whether they are managed or companies which are part of a supply chain should subject not’’. It is irritating to see someone claiming the existence their individual planning activities to a coordinating of something which is not clearly defined, because within authority which creates an additional value by taking into empirical sciences a clear distinction usually is regarded as account all kind of interdependencies between individual an indispensable prerequisite of any observation. Further- resources, restrictions, plans and activities that have for- more it remains unclear how an innovation like supply merly been neglected. This creates a larger solution space chain management can come into existence without being and opens the door to superior plans which were out of managed. reach as long as the single companies on their respective Probably the idea that supply chains already exist before stages were only following their narrow self-interest. Fur- any attempt to submit them to a ‘‘holistic optimization’’ thermore companies who are ready to share a common seems so self-evident because one always has the techno- destiny cannot only separate out more contingency of their logically predefinded linear sequence of value adding plans and orders but may develop a higher readiness to activities in mind that one has to run through in order to execute specific investments because the membership in bake bread, build mountain bikes or produce a yoghurt. In the supply chain seemingly lowers the financial risks. For this trivial sense the existence of supply chains of course the same reason one can argue that transaction costs can be cannot be doubted. What can be doubted is the (unman- aged) existence of networks which are ready to be managed further reduced. As a seemingly logical consequence the postulate arises or even optimized in a holistic manner. To call any network ‘‘that the whole supply chain should be managed as one of suppliers and customers surrounding a company a single entity’’ [6, p. 289]. Within the context of transac- ‘‘supply chain’’ would disentitle the notion of any inno- tion cost economics this implies that the market which vative meaning and render the word redundant. It would was used as coordination mechanism ex ante is entirely then be sufficient to use the word ‘‘logistics’’ and discuss replaced by a hierarchy. The situation strived for is called the issues of an intensified supplier and/or customer rela- a ‘‘win–win-situation’’ because it creates an additional tionship management. value which (if necessary after a redistribution between the participants) puts all members into a better position. From this fundament it is only a small step to postulate a 2 Network identity: where can ‘‘supply chains’’ fundamental change in the field of competition. Compe- be found? tition will/should shift from the level of single companies on their respective stage in the value chain to the level of In order to prevent misunderstandings we point out that we supply chains as whole entities and new players in the will not discuss about the following three issues: market: ‘‘the real competition is not company against 1. No business is an island. It is trivial to state that every company but rather supply chain against supply chain’’ [7, p. 18]. Some authors like Chopra and Meindl [5, p. 41] company is connected with numerous customers and suppliers and that the quality of these relations has an argue that this shift has already been carried out (It is a impact on the company’s ability to survive and create common weakness of many articles and books written value. about supply chain management that the authors do not 2. It is equally unquestioned that in a global economy clearly differentiate between observations, hypotheses, recommendations and predictions). companies face the challenge to manage complex 123 Logist. Res. (2009) 1:71–82 73 process chains involving an increasing number of If this formation is interpreted as an advice to all compa- players. nies regardless of their industry and their position in a 3. Everybody appreciates the opportunities offered by value chain, there would be as many networks (‘‘extended modern IT systems to improve the coordination of enterprises’’) as there are companies, which means there demand forecasts, production schedules and capacities would be none (Fig. 1). (including inventories) across the boundaries of single Already before any attempt to analyze this paradox in a companies. systematic manner, some doubts take possession of our mind. Should Airbus really take care about possible bot- All these issues have been and can be further discussed tlenecks in the capacities of brazilian companies occupied at length under the headline ‘‘logistics’’. Interconnected- with the extraction of the ore, because after several con- ness was not a foreign word in this context. But as already version steps transforming ore into steel in China the mentioned interconnectedness by itself does not lead to material is build into turbojet engines by Rolls and Royce institutional innovations or systems of a higher order which in Great Britain (who also supply other companies in the can be managed in a holistic manner—even if by means of aircraft industry)? Furthermore it is unclear who decides a single sourcing strategy the relations to suppliers get whether this is the supply chain of Airbus or the network of tighter and more companies seek a relief of complexity and Rolls and Royce. a reduction of transaction costs by accepting a longer Coming back to the question whether supply chains duration of supply contracts. exist out of themselves or have to be shaped before they When defining his concept of ‘‘strategical networks’’, can be further managed we base our arguments on the Sydow [9, p. 82] describes what is needed to constitute following perception of the current state of most german such a new organizational entity: ‘‘explicitly formulated industries: if we take a closer look at the mutual interde- goals,… a formal structure with the allocation of formal pendencies that connect companies operating on different roles… and an own identity’’. We have to add: without the stages of their respective value chain, we do not observe enabling limits of closed boundaries separating the orga- isolated chains, but instead we see overlapping open nization unambiguously from its environment a common polycentric networks. Many companies appear as cross- optimum is impossible for simple logical reasons. Open roads passed by different material and semi-finished goods opportunities (sourcing from or selling to third parties) which subsequently become part of totally different end would fragment the network and destroy the possibility to products. Glas yarn is needed to produce laminate, which derive a common optimum, because there would be no in turn is a material required by the manufacturers of cir- common agenda and no common, closed solution space. cuit boards that go into cars, mobile phones and industrial Even the more basic goal of surviving in an uncertain robots. Another output of the capacities used to produce environment could not be controlled because it remains glas yarn is fiber glass which, among other products, goes unclear who precisely should be kept living. Conversely into paperhangings. Consequently a car manufacturer or a this means: the price companies must pay for a holistic producer of mobile phones—without knowing from each optimum is a limitation of its freedom of action and a other—my be hit by a bottleneck in the capacity of the subordination. This characteristic cannot be graduated. supplier of glas yarn, and while depending on the same raw (Within every day speech the notion of ‘‘optimization’’ is material the automotive industry is connected with com- used in an inflationary manner denoting any kind of panies selling hangings. These overlapping, open and improvement. We use it in the strict sense in which it is polycentric networks cannot be subjected to a holistic used within the field of Operations Research, denoting optimization, because they are far too complex. They have solutions which, given a set of data and assumptions, cannot be further improved.) If companies are ready to abandon their autonomy and subsume their narrow self interest for the benefit of the supply chain as a whole they establish the prerequisites for relocating competition. They compete as an integrated supply chain against other supply chains. Christopher [7,p. 5] illustrates this ‘‘new competitive paradigm’’ with a simple drawing showing a company surrounded by its customers and its suppliers and their customers and sup- pliers. Together they form the ‘‘confederation of mutually complementary competencies and capabilities’’ [7, p. 286] that seems ready to be subjected to a holistic optimization. Fig. 1 The supply chain network 123 74 Logist. Res. (2009) 1:71–82 neither a distinct boundary nor (as a consequence) an own management the addressed goal conflict would not simply identity. This finding has three major implications: disappear. Building tightly coupled supply chains on the basis of a single sourcing strategy is not a logical impli- 1. The requirement that all supply chains should totally cation of systems thinking, but just one of several possible focus on the needs of the ultimate customer is ways to structure a system. The objective evidence that this misleading. On the pre-stages of production it is often arrangement creates more benefits than opportunity costs unclear, who the final customer will be and, if there are has to be provided in every single case. several, if their requirements concerning the perfor- Another problem is the generation of a strategy that can mance of ‘‘the’’ supply chain are identical. (For a be undersigned by all the members of the supply chain. manufacturer of hard disk drives like Western Digital Intrinsically this strategy is needed before the formation, the end customer can be a child using his paddle as because it is the basis for deriving the selection criteria and well as a large company using mainframe computers). defining the business case. On the other hand a legitimation 2. The idea of constantly feeding actual customer demand of the strategy ideally should be based on the approval of forward to all players in the supply chain can not be all members. The only way out of this circle is a company applied without a distinctive preliminary reduction of which has enough power to define the boundaries of a complexity (even than it will turn out to be extremely network which can than be subjected to a holistic optimi- difficult to find the way to the 5th or 7th tier suppliers zation. In this case the formation of the chain will be biased through a number of complex bills of material on all by the individual interests of the focal enterprise. An levels of the chain and to calculate the related lead ongoing conflict within the german consumer goods times per item, supplier and stage which must be industry exhibits the implications of this approach. known in order to allocate the adapted times of Within this sector it has been a long and unquestioned delivery that have to be achieved by all suppliers on all tradition that the supply of retail outlets is the genuine task stages). of the manufacturers. Accordingly they have established 3. Supply chains can not be found in reality as existing their distribution systems, which show the pattern of objects for managers or researchers (like fruit flies can uprooted trees and whose efficiency strongly depends on be found by a genetic researcher). Instead they must be the annual transportation volume a single supplier controls created by carving a limited number of companies on (economies of scale and density). Since the end of the last different stages of production out of an unmanageable century not only in Germany more and more retail chains network of networks. have discovered the opportunity to get control over their For several reasons the latter is not a trivial act. One inbound goods flows in order to exploit the benefits of reason points to an organizational problem. Until today the bundling for themselves. Because this change of system organizational pattern of many companies still follow the leadership requires a change in the delivery terms this idea of a functional specialization. This means: in order to model is also called ‘‘factory gate pricing’’ (Germans use become a member of a single supply chain, people from the the word ‘‘Selbstabholung’’ which means selfcollection). purchasing and the sales department have to be convinced Figure 2 illustrates the conflict behind these competing that their way to see the world is not as relevant as the claims to define ‘‘the’’ supply chain. The two models world outlook of logisticians. Purchasers e.g. tend to like cannot tolerate each other because both create economies competition in the buying market which would have to be of scale which can only be generated and exploited once. abandoned in order to make a holistic optimization of ‘‘their’’ supply chain possible. Multiple sourcing could Distribution system of create ‘‘economies of substitution’’, but it would fragment a supplier the chain in the same way in which multiple distribution M1 channels would do (in the latter case there is no predefined R1 reference object for the ‘‘optimal’’ allocation of scarce capacities). This problem cannot just be solved by broadening the concept of SCM, e.g. by defining SCM ‘‘as being more than just logistics’’ and extending it to ‘‘the management of all business processes’’ [10, p. 5]. Companies would have M m Rn to reorganize themselves via a fundamental shift of inter- Procurement system of a retailer faces and a related redistribution of power and influence. But even if logistics and purchasing are organized as subsections of a newly established superior supply chain Fig. 2 Competing supply chain definitions 123 Logist. Res. (2009) 1:71–82 75 In order to fully understand the picture, imagine that • Who is/should be in charge of deciding which compa- the supplier in the middle is Proctor and Gamble deliv- nies should be accepted or excluded as member of the ering pampers to the retailer in the middle named supply chain? METRO. The dark arrows represent the distribution sys- • If the coherence of the chain and the channelizing of behaviour is not only guaranteed by incentives and tem of Proctor and Gamble which is bundling orders from different retail organizations and perhaps some healthcare mutual trust (which seems to be favoured by the advocates of ‘‘collaboration’’) but by negotiated organizations on the way to the outlets, regional ware- houses or clinics where they are needed (the real archi- explicit reciprocal obligations written down in long term contracts (which might be regarded as necessary tecture of such a network of course is more complex, what matters here is the fact that the commodity flows are in order to foster and protect specific investments): who entirely controlled be the manufacturer). If METRO is can/should play the role of the contractual partner? successful in claiming that all their inbound flows from • If a member steps out of line by exploiting the different manufacturers (comprising among others pam- opportunity to source some material from outside pers, CD-players, mountain bikes, power saws and food of (because the products are cheaper or better) or by all kind) have to be regarded and treated as their supply allocating part of his capacities to a third party because they are ready to pay higher prices: who has the chain, the dark arrows are replaced by the bright arrows and a totally different, formerly unknown network with a authority to inhibit this or force the disloyal to compensate the victims? (Would a prevention of this different architecture arises. The boundaries of this alter- native ‘‘supply chain’’ are marked by the dotted line. (By opportunism be compatible with the idea of ‘‘optimi- the way: METRO was very successful in following this zation’’? And: are the related opportunity costs path). included in the calculation of the ‘‘win–win-situation’’? Although this is an interesting question, this is not the • Who takes on the costs for observing compliance with place to discuss which side has the stronger arguments or commitments or enforcing contracts? whether this is a zero sum game. What matters in the given • Who measures/controls the benefits or losses which collaboration initially (that is before any redistribution) context is the finding that generates within the four walls of each member? (This • The confining of a supply chain is a foregoing act presupposes the absence of opportunistic behaviour of which cannot be ‘‘optimized’’ in itself. all partners while exhibiting their data and it raises • The thesis that ‘‘collaboration’’ in supply chains always some delicate measurement problems: for instance the and by necessity unlocks win-win-situations cannot be question to what extent the number of unplanned ad hoc kept up. changes of short term production schedules formerly • Starting the reflection about supply chains based on the provoked by unforeseen bottlenecks in the capacities of assumptions that there is (or should be?) ‘‘a common suppliers can be reduced and what the exact bottom line agreed agenda driving the achievement of the supply effects of this stabilisation are). chain goals’’ [7, p. 293] and a ‘‘joint ownership of • Who decides about the reallocation of the benefits/ decisions’’ is audacious. losses if (which has to be expected) these outcomes initially are spread among the members in an asym- metric manner? 3 Governance structures: how can supply chains • If the results of the before mentioned activities be managed? (incentives, subsidies and/or guaranties) must be known in advance in order to facilitate the decision of a The idea of a management beyond the limits of ownership company to join a supply chain: who can claim the is not as trivial as may look like. Keeping in mind that the legitimation to do this work before the supply chain has concept of an ‘‘extended enterprise’’ is an institutional assumed a concrete shape and how could anyone get innovation and that institutions are defined by roles and access to the sensitive data base needed to simulate the rules, it is amazing to see that most articles on SCM focus outcomes for different constellations? • If, as Lambert et al. (2008) correctly point out, a on process models and on the potential outcomes and benefits of supply chain management while losing sight of process-oriented organization overcoming the internal the problems connected with the organization of leader- functional silos (procurement, production, marketing/ ship. Before discussing this issue we list up some of the sales) within the individual firm is a prerequisite for any problems that have to be addressed and solved in order to supply chain-wide process integration: who can stim- make a supply chain work: ulate the necessary transformation in order to create 123 76 Logist. Res. (2009) 1:71–82 supply chain readiness within companies which are still based on the mutual trust between equal partners. Like organized around business functions? power trust is a means to reduce complexity. It averts the • If a coordinated production planning across the whole risk of opportunistic behaviour, reduces uncertainty, allows supply chain requires modern IT-tools like so-called for decreased controlling efforts and thus can reduce Advanced Planning Systems installed in all companies transaction costs. But if trust is a prerequisite of collabo- plus the associated adjustment of master data (like bills ration it must be in place before the supply chain can be of material): can all suppliers be forced to accept the constructed. The problem whether a leap of faith pays off considerable expenditures actuated by these measures? has been discussed at length within game theory (see for instance the insights delivered by Axelrod [12] regarding Basically there are two solutions offered to shape the human behaviour in a situation called ‘‘prisoner’s governance structure of a supply chain, namely (a) the dilemma’’). But unlike people playing games real compa- subordination to a company endowed with enough power nies have a long history in their role as customers and/or to orchestrate the coordination across all other participants, suppliers which strongly affect their expectations con- and (b) a heterarchical coordination based on a mutual cerning the behaviour of their ‘‘partners’’, and an equal partnership among companies with equal rights called distribution of power is uncommon in most markets. This ‘‘collaboration’’ (in the latter case the organizational leads Christopher [7, p. 286] to the insight that ‘‘perhaps arrangement would be some kind of a steering committee). the one of the biggest challenges to the successful estab- The question whether the idea of self-organizing, poly lishment of marketing networks is the need to break free centric actors forming a supply chain presupposes a sym- from the often adversarial nature of buyer/supplier rela- metric distribution of power (a situation which can hardly tionships that existed in the past’’. be found in reality) or can be combined with a distinct If one takes a closer look at the way in which original power-based leadership remains unanalyzed because it is equipment manufacturers in the automotive industry or barely raised. This is not a big drawback because both lager retail chains interact with their suppliers one soon approaches do not solve the problem. realizes that via severe negotiations on prices these com- If the orchestration is done by a strong leading company panies still fight for their part of the value creation in their it remains unclear respective supply chain, using all the power they have. Do (a) why a powerful company should give up its individual they need a re-education because they have not yet optimum for the benefit of a superordinate ‘‘supply understood that eliminating the bullwhip-effect and chain profitability’’ [5, p. 6], and reducing transactions costs by means of collaboration can (b) what could make the powerless members of the supply yield higher benefits than letting suppliers fight for orders chain believe that after long years of a fierce compe- and exploiting economies of substitution? tition on the shares of the value added the supply chain Of course one has to change things if one wants to leader while orchestrating the chain will always improve them. The current performance of managers respect their interests and not capture the surplus? (If therefore may be the wrong benchmark for the evaluation in the face of market power suspicion established by of an innovation. But the advocates of collaboration have previous experience cannot be replaced by trust the to ask themselves some fundamental questions: can an functioning of information sharing cannot be expected. organization that abstains from any hierarchy be strong Instead opportunism will spread and generate the kind enough to create and run something like an ‘‘extended of problems which have been elaborated in detail enterprise’’ reaching out from the production of raw within the agency theory [see e.g. 11], in particular material to the end customer consuming the end product? there will be hidden information and hidden action). Can an arrangement led by some kind of a steering com- mittee (with the inclination to endless debates about the Power comes along with legitimation problems, reduces right strategy, the distribution of benefits, etc.) really loyalty, provokes resistance and the striving for indepen- compete with more rigid organizations which rely on dence (and/or the building of countervailing power), and in empowered managers taking over personal responsibility? most cases it looses strength when it is extended If one hesitates to impute ignorance, inertia or irrational ‘‘upstream’’ to the supplier’s supplier. Probably these are behaviour to managers who do not ‘‘break free from the main reasons why most exponents of supply chain adversarial relationships’’ and accept a shared destiny with management prefer the second option for the design of a their partners the conclusion from what we can observe is governance structure which ideally presupposes the different. Obviously the concept of collaboration is based absence of power: Collaboration. on unrealistic assumptions and at the same time too con- Unfortunately while solving one problem the absence of fuse in order to be used as (or instead of) a governance power creates some others. Collaboration is/should be structure which is strong enough to create and manage 123 Logist. Res. (2009) 1:71–82 77 whole supply chains as if they were legal entities. This Probably his partner and former customer would not even assessment corresponds with a number of recent empirical notice the opportunity for a price cut because in order to findings. reduce transaction costs (which is always highlighted as Kampstra et al. (2006, p. 315) diagnose that ‘‘paradox- one of the major benefits of integration) he has dismantled ically, SCC is immensely popular both in business and his procurement department. Within the so called ‘‘new academia and at the same time most collaborative initia- institutional economics’’ [17] the harmfulness of dissy- tives end up in failure’’. A joint study of Capgemini, the metrically distributed information has been elaborated in Georgia (Southern) University and the University of detail under headlines like ‘‘adverse selection’’ and ‘‘moral Tennessee states: ‘‘supply chain collaboration is needed, hazard’’. With their captive internal markets supply chains but not happening’’ [13], and a study delivered by Forrester are the ideal nutrient medium to generate such dysfunc- comes to the conclusion that the ‘‘much-hyped concept in tional effects. The external competitiveness of supply the late 1980s and 1990s’’… ‘‘did not live up to the chains will diminish because inside the network competi- industries expectations’’ [14]. Vereecke and Muylle [15,p. tion has been suspended as a controlling mechanism and as 2] do not only cast doubt on the dispersion of collaboration, a source of energy. but also on the capability of this model to deliver better For the same reason the advocates of rigidly integrated results: ‘‘Empirical support for the relationship between supply chains should be careful with their assertion that supply chain collaboration and performance improvement this institutional arrangement would lead to a stronger is scarce’’, they state and sum up their own investigation orientation towards the needs of the ‘‘ultimate customer’’. with the conclusion ‘‘performance improvement is only As long as purchasers and suppliers compete for their part weakly related to the extent of collaboration with cus- of the value added the market mechanism makes sure that tomers or suppliers’’. at the end of the day all increases in efficiency will be handed on to the end customer. In contrast to this while reading about the win-win-situations that supply chain 4 Incentive systems: more competitiveness based management creates and about the necessity to share these on less competition? effects among the partners in a fair manner one gets the impression that supply chains should take possession of the Within integrated supply chains market prices have to be additional gains as ‘‘relational rents’’ in order to stabilize substituted by internal transfer prices. This operation the arrangement. This is definitely not in the interest of comes along with a huge loss of valuable information. customers and it can turn out to be a competitive disad- Looked at from outside a market price looks like a rather vantage in relation to those companies who refrain from joining any collective. poor information: it consists of only one number. What is making it rich and valuable as an incentive and as the A side impact of internal transfer prices is the effect that linchpin of a steering mechanism is it’s mode of formation. in a dynamic environment the associated partners after a Market prices reflect the currently prevailing conditions of short time will no longer know whether the allocation rules the production and distribution of a good, including for fixed with these prices comply with the requirements of instance updated oilprices, increased road charges, the fairness and justness. This can generate as much dissatis- pricing policies of suppliers and competitors, and the faction as the impression that one could buy or sell prod- shortness of material and capacities. ucts outside the supply chain at better conditions. Market prices work as signals leading managers to adapt Astonishingly neither the question whether such an even to facts that are out of their sight. By producing and arrangement fits in the shareholder value concept nor the spreading these signals markets permanently create situa- question whether it is compatible with anti-trust-law has tions in which all decision makers can use much more ever been raised und discussed. information than any single one of them has at his disposal. Disengaging oneself from the market mechanism as The market mechanism uses a knowledge which does not means of coordination obviously generates considerable exist as an entity, and it therefore is a much better instru- opportunity costs. The presumably most negative effect of ment for the usage of scattered information than any central the idea of vertically integrated supply chains in this regard planning institution could ever be (for a fundamental has not yet been mentioned. Competition stimulates and analysis see [16]. encourages innovation. Eliminating competition and The replacement of market prices by an internal pricing entrepreneurship therefore means eliminating an important system inevitably leads to delayed and biased adjustments. source of competitive advantages. Suppliers whose sales A supplier facing overcapacities in his original market seem secured through the membership in a supply chain do would not feel the need to adjust the internally defined no longer have to prove their right to exist by positioning transfer prices he was granted under previous conditions. themselves at the top of technological progress. They 123 78 Logist. Res. (2009) 1:71–82 would also find it relatively hard to get there because they relations between processes on different levels which are lack the permanent confrontation with the demand and sequentially interdependent one also has to look at inter- expectations of different customers. Perhaps they are also dependencies between resources which can have the effect discouraged by the expectation that in a tight cooperation it that short capacities of one company affects the capacity is hard to prevent an unattended leakage of knowledge or utilization of another. With regard to planning processes even that the potential benefit of an innovation would be these two interdependencies are interconnected. If all socialized within the network they are part of. In the end possible constraints across all members of a supply chain they are no longer willing to do what they were capable of. were known to one central planning authority this instance In any case it seems most unlikely that integrated supply could replace local suboptima (which formerly resulted chains could originate suppliers with the calibre of Intel or from taking the output of other local decisions as given Bosch. They are also unable to integrate them ex post restrictions) by a total optimum which makes the best use because such companies derive their strength from their out of the joint capacities of the whole chain and in addi- autonomy. In the end this is for the benefit of the whole tion (by taking out uncertainty) could make plans more economy because within open networks innovative prod- stable. This expectation seems to bare an undisputable ucts are made accessible to more companies and to more logic. But there are dysfunctions and a severe spillover that customers. Facing such proud suppliers supply chains will are often ignored. make the distressing experience that they either have to do The attempt to map and represent all interdependencies without the market leaders or to give up their holistic in one decision model leads to the necessity to cover and management approach. represent a huge number of variables, restrictions and Last but not least we have to point out that a single parameters that have formerly been hidden behind the sourcing strategy practiced within a supply chain in order interfaces which had separated the companies into smaller to enable an end to end optimization of the whole network legal entities (i.e. they were absorbed by the market). destroys economies of scale and scope that can be Ideally this is a way to gain control over something that exploited by companies which preserve the option of pur- formerly was an annoying source of surprise and emer- chasing from different sources and accept vendors sup- gency. But unfortunately many of the parameters now plying their competitors (which is common practice within included in the model are not just there (ready to be col- the German automotive industry). Risk reducing pooling lected as fixed data) but have to be forecasted (the relevant effects on the level of an inventory which is carried by a environment of a whole supply chain comprises much supplier for several customers can be regarded as part of more potential sources of a disturbance than the environ- these economies. For simple logical reasons a holistic ment of one single company). The implication is that with the extension of the model a lot of additional uncertainty is optimization requires dedicated resources (if not specific investments that cannot be redeployed for alternative use). imported (an uncertainty that formerly other people cared This confinement inevitably increases the risks of major about). This makes the outputs of the model more and more investments, raises capital costs and scales down the unstable and creates the necessity of more and more shareholder value. To advocate this in a world in which the amendments of plans. coincidence of globalization and the internet reduces The result looks paradoxical: the attempt to represent transaction costs and creates sourcing opportunities com- the complexity of a whole network in a central decision panies never had before seems to be a peculiar idea. model does not only produce the need to collect and pro- cess a huge amount of data (as multi-level bills of mate- rial), but it also foils the intention to gain control over the 5 Flexibility: are supply chains better prepared whole system: instead of increasing the stability of plans to cope with uncertainty? the amount of uncertainty rises. In the end the organization will not gain, but loose control and has to handle one ‘‘Responsive supply chains are by definition highly inte- exception after the other. Instead of reliability and resil- grated’’, says Christopher [7, p. 280]. Besides the fact that ience the result is vulnerability. There are no more local this is not a matter of definition we argue in the following problems which can be solved on a local level. Instead each chapter that the opposite is true. We start our argumenta- local problem raises the necessity of readjusting the overall tion with a little mental exercise putting ourselves into the plan (otherwise one would neglect interdependencies and position of a mathematician trying to build an integrated give up ‘‘systems thinking’’). Because of its inability to planning tool which covers all the interdependencies incapsulate and absorb local disturbances and resolve them between the activities and capacities of the companies ad hoc, the system creates one domino effect after another. forming a supply chain. Basically there are two kind of (Experts in macroeconomics will see remarkable analogies interdependencies that have to be covered: besides to the way socialistic economies were run. They were also 123 Logist. Res. (2009) 1:71–82 79 based on a linear, mechanical and unworkable notion of Figure 3 illustrates the functioning of an ‘‘available-to- control). promise’’-model (for further details see [18]. The model shown in Fig. 3 replaces the standard lead times which traditionally structure the coordination within 6 A paradigm change: loosely coupled systems order fulfilment processes. Standard lead times, usually cross-linked by modern IT systems defined by the number of days required to deliver a prod- uct, do not consider the availability of resources (inventory In the beginning of our investigation we have separated the or production capacity) at the time orders are accepted. As ‘‘tool box’’ which has been developed under the headline of no supplier can afford to cover infrequent demand peaks ‘‘supply chain management’’ (comprising concepts like with additional capacity it usually is an unexpressed part of VMI and ATP) from the philosophy propagated on top of the agreement that on time delivery rates do not exceed a these models. This analysis has led to the conclusion that level of 95 ? x% (depending on the particular industry). the idea of optimizing networks in a holistic manner The remaining risk comes not by surprise but is intended (embracing all companies of a value chain from the sup- and expected (and usually defined by dimensioning safety pliers of raw material to the ultimate customer) is all uto- stocks), thus using the customer as a final buffer. pian. For several reasons it cannot be put into practice, and The ATP-model shown in Fig. 3 can be described as an even if it would be viable it would be unwise to pursue it. extended feedback loop, which replaces the common A further consequence of this analysis is the prediction that sequence ‘‘plan–act–check–replan’’ by the sequence the often propagated shift of competition to the level of ‘‘plan–check (availability)–replan–act’’. The original entire supply chains in many important industries is unli- equipment manufacturer (OEM = tier n) propagates his kely to evolve. initial production schedule (‘‘unconstrained forecast’’) We therefore have to replace the paradigm of optimizing upstream to his first tier suppliers and calls on them to ‘‘extended enterprises’’ by another paradigm which com- check possible constraints either on the level of inventory prises the feasible idea of an improved mutual visibility or on the level of production capacities (in the latter case concerning updated demand forecasts and the availability one can use the notion ‘‘capable-to-promise’’). As a feed- of resources without generating the detrimental side effects back he receives a capacity update (‘‘best-we-can-do’’- of a rigid vertical integration. We advocate the idea of message). Ideally the suppliers have modern software tools loosely coupled companies maintaining their autonomy called ‘‘Advanced Planning Systems’’ in place which within polycentric overlapping networks because in a enable them to derive stressable commitments out of a complex and dynamic environment loose coupling is not a constraint based planning (for further details see [19]. In case of disclosed bottlenecks the OEM either adjusts his defect but a precondition of stability and resilience. Within these open networks which do not strive for a new superior plans and propagates them as viable planning basis identity as ‘‘supply chains’’ efficiency can be considerably upstream again (‘‘constrained forecast’’) or exploits econ- increased by implementing the above mentioned models in omies of substitution and sources part of his demand from numerous bilateral arrangements. In some cases this may another supplier. Ideally this results in a situation where the lead to close relations between purchasers and suppliers, to involved suppliers only produce what is really needed long term contracts and to the development of mutual trust. whereas the OEM is disburdened from the risk of ad Fig. 3 ‘‘Available-to-promise’’ as an extended feedback loop 2. Capacity-Update Check capacities, („Best we can do“) Customer Commitments and Supply Allocations, Apply Priority Rules tier n tier n-1 Typical Response Options:  Schedule order to next period (month or week) 1. Initial Forecast 3. Forecast-Update  Re-balance supply from („unconstrained“) („constrained“) another customer  Offer a product substitute 123 80 Logist. Res. (2009) 1:71–82 hoc-adjustments and/or the belated completions of unfin- never be completely transferred from the location where ished work. Within the cooperation the activation of pro- problems first appear to the top of a central planning cesses changes from a push- to a pull-mode. authority. Hierarchies are built to communicate directives The model delivers an ostensive description of a top down but for several reasons they usually show a poor mechanism capable of exploiting the benefits of an performance if a transfer of data in the opposite direction is increased visibility between partners in a value chain. It required. The consequence is that loosely coupled networks represents that part of knowledge delivered under the are faster in adapting to changing conditions and that they headline of ‘‘supply chain management’’ that we regard as can develop a superior learning aptitude. Endowed with useful and viable (although for several reasons the com- modern planning tools they can build feedback loops plexity of their implementation by far exceeds the com- comprising customers and/or suppliers (in general not plexity of their logic). These models do not lead into a beyond the first level) which enable them to take con- perfect world (e.g. the uncertainty reduced by an ATP- straints into account which were formerly hidden behind message will not disappear completely but show up again the four walls of their partners. But the companies which before the next query, suppliers may hesitate to submit a are embedded in such open networks do not have to commitment if the risk of a cancellation is not compen- anticipate the huge number of changes in the environment sated, and the chain is still exposed to an unknown risk as which a whole supply chain would have to take into long as some potentially critical suppliers are excluded account and which would put their planning system into a from the game). Nevertheless one can proceed in this state of a permanent great nervousness. (To point this out direction and we therefore base the paradigm advocated more clearly: The number of contacts to the outside world here on the idea of companies supplying each other with per company is smaller but an integrated supply chain as a more information needed for a stable allocation of whole has much less contacts than the open, overlapping, resources. polycentric networks we are talking about. Living in a But keep in mind: within these models the companies do richer context means receiving more impulses to learn and not completely give up the flexibility of exchanging part- improve. Conversely supply chains are endowed with less ners and using the market as a means to provide economies sensitive sensing mechanisms). of substitution. They understand that in a dynamic and The institutional arrangement that is recommended here uncertain environment indeterminateness is a strength. can derive several competitive advantages from using the Maintaining the opportunities the market offers is part of market as a coordination mechanism. Some of these have their risk management. In contrast supply chains would to do with the incentives which have an impact on the make the experience that rigid coupling and an overdone behaviour of managers. While the determined membership in a closed network can be interpreted as an invitation to an connectivity foster structural inertia and narrow the number of options if major changes in the environment require opportunistic behaviour competition drives suppliers into a major internal changes in order to survive. Supply chains continuous improvement process. Innovation is stimulated lack what the cybernetician Ross Ashby [20] once has by the aspiration to be rewarded by relational rents. But called ‘‘requisite variety’’. (Further analogies cannot only this motivation only works if these rents are not granted be found in the fields of education (see [21] and artificial and remain uncertain, that is to say if competition is not intelligence (see the current discussion of ‘‘swarm intelli- abandoned. The permanent threat to loose a customer may gence’’ (Kennedy et al. 2001) and the description of multi- be regarded as a rather negative kind of motivation because agent systems and the ‘‘internet of things’’ (e.g. with [22], it is associated with fear. But this is the price that suppliers but also within evolution biology: ‘‘it is not the ability to have to pay for preserving their autonomy and uniqueness adapt, but the ability to decouple which explains the which many of them regard as their most important source enormous stability and resilience of life and of all systems of motivation. build upon it’’ says the German sociologist Luhman [23,p. The ability to solve local problems locally does not only 556] referring to the work of leading biologists like the preserve more diversity in responding to changes in the author of the ‘‘Principles of Biological Autonomy’’ [24]). environment and prevent trouble from spreading (‘‘domino Using the market mechanism will equip companies effects’’). It also reduces the intensity of communication working together in loosely coupled networks with a higher and thus helps to reduce transactions costs. Nevertheless intelligence which is not embodied in a single location, the total effect on transaction costs can be regarded as organization or IT system. From an SCM-perspective ambiguous. On the one hand there is no need to establish a polycentric networks are the incarnation of redundancy. secondary organization on top of the supply chain which But for the companies embedded in these networks this is can accomplish the holistic optimization. Furthermore the basis of their adaptiveness. They can use a wealth of there will be fewer goal conflicts to be resolved and there information (among them meaningful prices) which can are less frictions arising from different corporate cultures 123 Logist. Res. (2009) 1:71–82 81 and less inconsistencies between actions to be discussed. chains), and a community of scientist who confirm each And there is no controlling effort needed in order to pre- other in their way of thinking. The classification of this part vent members from exploiting the lock-in-situation created of the literature on supply chain management as a paradigm by supply chains through opportunistic actions. But of is further confirmed by the observation that many authors course on the other hand the permanent need to adjust who propagate the related ideas do not realize that their prices forces companies embedded in open networks to concept might produce opportunity costs (compared to establish larger sales and purchasing departments. Thus the alternatives not seen on their radar screen) or some balance of different impacts on transaction costs may be unforeseen spillover. classified as unclear. Fortunately this is not the main cri- The analysis of the drawbacks and potential disecono- terion needed to resolve the cost-benefit-equation we are mies of the attempt to optimize supply chains in a holistic discussing. manner as if they were legal entities has led to the con- In his famous article on ‘‘The Architecture of Com- clusion that this is neither a viable nor a beneficial idea (see plexity’’ Simon [25] has pointed out that modularity can be also [28]). The author therefore advocates the shift to a a strong means to avoid the drawbacks of interlaced pro- paradigm which favours loosely coupled processes and cess architectures. This idea leads us to complete the pic- planning systems and the usage of competition as an ture by highlighting the role which standardized interfaces incentive system and a source of flexibility. Within this can play in the field of collaboration. Standardized inter- paradigm the above mentioned process models have their faces facilitate integration and replaceability concurrently place because they can support decentralized planning via (for a more detailed discussion see 26). Integration must no feeding the decision making processes with updated longer lead to lock-in-situations, and one could combine demand data and information about the availability of the benefits of interconnectedness with the benefits of using inventory or production capacities sourced from suppliers. markets and competition. Transaction cost could then be In order to exploit the potentiality of an increased visibility cut to an absolute minimum. But on the other hand mod- the building of supply chains is a sufficient, but not a ularity and standardization are not compatible with the necessary precondition. individual strive of companies for competitive advantages As a logical consequence the prediction/recommenda- based on a unique process design. Probably it is not least tion that competition will be/should be relocated to the this ambivalence that has made the implementation of level of entire supply chains is disputed. There are some concepts like a demand and capacity management inte- valid arguments favouring IT-based cross-links between grated across the interfaces between purchasers and sup- companies within open polycentric networks. What we pliers so delicate in industries like the German automotive neither observe nor expect is the emergence of strongly integrated networks which develop their own superior sector. However modularity can be regarded as an intelli- gent doorway to the flexibility which organizations need in identity by defining clear boundaries to their environment order to survive in a dynamic environment. Unfortunately and establishing an exclusive ‘‘collaboration’’ inside these modularity fragments supply chains and are a threat for borders. Whoever advocates such systems of a higher those companies who seek competitive advantages or order should keep in mind that there is a price to pay relational rents out of closed networks, specific investments for the development of the institutional arrangements and rigidly coupled processes. recommended under the headline of ‘‘supply chain management’’. 7 Summary In this article a separation was made between two kinds of References outcome of the past research on the subject-matter of supply chain management. On top of a number of sub- 1. Kotzab H, Friis A (2006) Implementing supply chain manage- ment in organizations. In: Jacquemin M, Pibernik R, Sucky E stantiated process models a widely shared philosophy has (eds) Quantitative Methoden der Logistik und des supply chain been developed that shows all the elements which Kuhn management. Hamburg, p 65 [27] has described as attributes of a ‘‘paradigm’’: a shared 2. Cooper MC, Ellram LM (1993) Characteristics of supply chain world view (which is rather a perspective predefining our management and the implications for purchasing and logistics strategy. Int J Logist Manag 4(2):13 perception of reality than a theory about it), a number of 3. Bechtel Ch, Jayaram J (1997) Supply chain management: a standard problems (like in this case the ‘‘bullwhip-effect’’), strategic perspective. Int J Logist Manag 8(1):15 a number of solution statements (like the calling for 4. Simchi-Levi D, Kaminski P, Simchi-Levi E (2000) Designing ‘‘systems thinking’’, the mantra of total integration or the and managing the supply chain—concepts, strategies and case advice to shift competition to the level of whole supply studies. McGraw-Hill, Boston 123 82 Logist. Res. (2009) 1:71–82 5. Chopra S, Meindl P (2007) Supply chain management: strategy, empirical findings, Paper provided by Ghent University, Faculty planning and operation, 3rd edn. Pearson Prentice Hall, Upper of Economics and Business Administration (February 2005), Saddle River Nr. 05/291 6. Sandberg E (2007) Logistics collaboration in supply chains: 16. Von Hayek FA (1996) Die Anmaßung von Wissen. Tubingen practice vs. theory. Int J Logist Manag 18(2):274 17. Brousseau E, Glachant J-M (2008) New institutional economics. 7. Christopher ML (2005) Logistics and supply chain management, A guidebook, Cambridge University Press, Cambridge creating value-adding networks, 3rd edn. Prentice Hall, Harlow 18. Bretzke W-R (2007) Available to promise: Der schwierige Weg 8. Mentzer JT, DeWitt W, Keebler JS, Min S, Nix NW, Smith CD, zu einem berechenbaren Lieferservice. In: Logistik Management, Zacharia ZG (2001) Defining supply chain management. J Bus vol 2, p 8 Logist 22(2):1 19. Stadtler H, Kilger C (eds) (2005) Supply chain management and 9. Sydow J (1992) Strategische Netzwerke, Wiesbaden advanced planning, 3rd edn. Springer, Berlin 10. Cooper MC, Lambert D, Pagh JD (1997) Supply chain manage- 20. Ashby WR (1952) Design for a brain. Chapman and Hall, London ment: more than a new name for logistics. J Bus Logist 18(1):1 21. Weick KE (1976) Educational organizations as loosely coupled 11. Jensen MC, Meckling WH (1976) Theory of the firm: managerial systems. Administrative Science Quarterly 21 (March 1976), p 1 behaviour, agency costs and ownership structure. J Financ Econ 22. Bullinger H-J, ten Hompel M (2007) Internet der Dinge. Springer, 3:305 Berlin 12. Axelrod R (2000) Die Evolution der Kooperation, 5. Aufl. 23. Luhmann N (1991) Die Wissenschaft der Gesellschaft. 2. Aufl, Mu ¨ nchen 2000 (engl.: The Evolution of Cooperation) Frankfurt a.M 13. Staff LT (2005) Supply chain collaboration is needed, but not 24. Varela F (1979) Principles of biological autonomy. Elsevier, happening, article based on a joint study by Capgemini, Georgia North Holland Southern University and the University of Tennessee. Published 25. Simon HA (1962) The architecture of complexity. Proc Am in http://www.logisticstoday.com Philos Soc 106:467 14. Tohamy N (2005) Supply chain collaboration checkup, Forrester 26. Garud R, Kumaraswamy A, Langlois RN (2003) Managing in the report, published by Forrester Research inc. 2005. http://www. modular age. Blackwell, Malden forrester.com/research (executive summary) 27. Kuhn T (1962) The structure of scientific revolutions. University 15. Vereecke A, Muylle S (2005) Performance improvement through of Chicago Press, Chicago supply chain collaboration: conventional wisdom versus 28. Bretzke W-R (2008) Logistische Netzwerke. Springer, Berlin http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Logistics Research Springer Journals

Supply chain management: notes on the capability and the limitations of a modern logistic paradigm

Logistics Research , Volume 1 (2) – Jul 7, 2009

Loading next page...
 
/lp/springer-journals/supply-chain-management-notes-on-the-capability-and-the-limitations-of-0ymrtu6d9I
Publisher
Springer Journals
Copyright
Copyright © 2009 by Springer-Verlag
Subject
Engineering; Engineering Economics, Organization, Logistics, Marketing; Logistics; Industrial and Production Engineering; Simulation and Modeling; Operation Research/Decision Theory
ISSN
1865-035X
eISSN
1865-0368
DOI
10.1007/s12159-009-0013-x
Publisher site
See Article on Publisher Site

Abstract

Logist. Res. (2009) 1:71–82 DOI 10.1007/s12159-009-0013-x OR IGINAL PAPER Supply chain management: notes on the capability and the limitations of a modern logistic paradigm Wolf-Ru ¨ diger Bretzke Received: 20 November 2008 / Accepted: 21 May 2009 / Published online: 7 July 2009 Springer-Verlag 2009 Abstract The paper analyses the drawbacks and the ‘‘movement away from functional department suboptim- potential diseconomies of the attempt to optimize a supply ization to a holistic optimization of the entire supply chain in a holistic manner. The conclusion derived by the chain’’ [3, p. 21]. author is that this is neither a viable nor a beneficial idea. Obviously this definition leads to a dramatically exten- As a consequence he advocates a paradigm shift. The ded scope of the resources and activities that managers are concept of totally integrated supply chains should be expected not only to control but to ‘‘optimize’’: ‘‘supply replaced by the idea of loosely coupled processes and chain management spans the entire enterprise and beyond, planning systems run by companies who preserve their encompassing suppliers on one end and customers on the autonomy and use competition as an incentive system and a other’’ [4, p. 221]. In order to surpass the boundaries of source of energy and flexibility. These companies should individual rationality companies obviously have to estab- interconnect on a bilateral level in order to exchange lish something not seen or thought of before: a manage- information about updated demand forecasts and the ment beyond the limits of ownership. ‘‘This means treating availability of capacities. But they should not develop stages in the supply chain that a company does not own as strongly integrated networks who start competing with belonging to the company’’ [5, p. 41]. Before taking a other supply chains on the level of this new identity. critical look at the potential benefits and shortcomings of this concept we obviously have to ask ourselves how this can be arranged. 1 Introduction Under the headline ‘‘supply chain management’’ we are confronted with two different kinds of scientific work. At ‘‘Despite 20 years of ongoing research…there is no con- the bottom there is a kind of toolbox comprising models sensus on what SCM really is’’ [1, p. 70]. It is not an easy like vendor managed inventory (VMI), available/capable exercise to review and criticize a concept which is so iri- to promise (ATP) or collaborative planning, forecasting descent. In order to focus our line of thought we refer to a and replenishment (CPFR). These process models are definition from the early nineties provided by Cooper and based on an extension of the kind of information that Elram [2, p. 1], cited by Bechtel and Jayaram [3, p. 37]: companies should exchange while doing business with ‘‘supply chain management is an approach whereby the each other. Beyond transactional data like orders, delivery entire network from the supplier to the ultimate customer is advices or invoices companies are expected to inform analysed and managed in order to achieve the ‘best’ out- each other about updated demand forecasts, inventory come for the whole system’’. Within subsequent publica- days of reach, available production capacities and the tions the pretension was further increased by claiming a current status of all activities affecting the adherence to delivery dates. Based on this new visibility companies are provided with an early view of potential bottlenecks in the W.-R. Bretzke (&) supply chain, which in turn enables them to generate Jentgesallee 9, 47799 Krefeld, Germany robust plans. As a consequence the amount of surprise, e-mail: W.Bretzke@t-online.de emergency and costly ad hoc adjustments can be reduced. URL: http://www.bretzke-online.de 123 72 Logist. Res. (2009) 1:71–82 Processes get more synchronized (that is the famous The following comments result in the statement that this bullwhip effect will become less harmful) and more ori- holistic approach is neither viable nor desirable and ented to the end customer’s demand. It is a common therefore not likely to be turned into practice. In order to character of these models that they can be (and usually substantiate this converse proposition we will have to take are) implemented within a two-sided relationship. Due to a closer look at the implications and consequences of the the stronger integration within these partnerships as an idea of integrating companies across all stages of a value additional benefit one can expect a considerable reduction chain ‘‘from sheep to shop’’ in a rigid manner. The first and of transaction costs. What these models cannot provide is most basic issue highlights the question of the formation of any kind of a ‘‘holistic optimization’’ across several supply chains. This issue is less trivial than many authors in companies on different stages of a value chain, because this field suggest. without further construction work their implementation In a common work titled ‘‘defining supply chain man- does not lead to a system of a higher order which can be agement’’ seven renowned US-scientists start their analysis submitted to such an operation. with the finding, ‘‘that there remains considerable confu- On top of these process models many protagonists of sion as to its meaning’’ (see [8]. Nevertheless they feel sure supply chain management therefore propagate the idea that that supply chains ‘‘exist, whether they are managed or companies which are part of a supply chain should subject not’’. It is irritating to see someone claiming the existence their individual planning activities to a coordinating of something which is not clearly defined, because within authority which creates an additional value by taking into empirical sciences a clear distinction usually is regarded as account all kind of interdependencies between individual an indispensable prerequisite of any observation. Further- resources, restrictions, plans and activities that have for- more it remains unclear how an innovation like supply merly been neglected. This creates a larger solution space chain management can come into existence without being and opens the door to superior plans which were out of managed. reach as long as the single companies on their respective Probably the idea that supply chains already exist before stages were only following their narrow self-interest. Fur- any attempt to submit them to a ‘‘holistic optimization’’ thermore companies who are ready to share a common seems so self-evident because one always has the techno- destiny cannot only separate out more contingency of their logically predefinded linear sequence of value adding plans and orders but may develop a higher readiness to activities in mind that one has to run through in order to execute specific investments because the membership in bake bread, build mountain bikes or produce a yoghurt. In the supply chain seemingly lowers the financial risks. For this trivial sense the existence of supply chains of course the same reason one can argue that transaction costs can be cannot be doubted. What can be doubted is the (unman- aged) existence of networks which are ready to be managed further reduced. As a seemingly logical consequence the postulate arises or even optimized in a holistic manner. To call any network ‘‘that the whole supply chain should be managed as one of suppliers and customers surrounding a company a single entity’’ [6, p. 289]. Within the context of transac- ‘‘supply chain’’ would disentitle the notion of any inno- tion cost economics this implies that the market which vative meaning and render the word redundant. It would was used as coordination mechanism ex ante is entirely then be sufficient to use the word ‘‘logistics’’ and discuss replaced by a hierarchy. The situation strived for is called the issues of an intensified supplier and/or customer rela- a ‘‘win–win-situation’’ because it creates an additional tionship management. value which (if necessary after a redistribution between the participants) puts all members into a better position. From this fundament it is only a small step to postulate a 2 Network identity: where can ‘‘supply chains’’ fundamental change in the field of competition. Compe- be found? tition will/should shift from the level of single companies on their respective stage in the value chain to the level of In order to prevent misunderstandings we point out that we supply chains as whole entities and new players in the will not discuss about the following three issues: market: ‘‘the real competition is not company against 1. No business is an island. It is trivial to state that every company but rather supply chain against supply chain’’ [7, p. 18]. Some authors like Chopra and Meindl [5, p. 41] company is connected with numerous customers and suppliers and that the quality of these relations has an argue that this shift has already been carried out (It is a impact on the company’s ability to survive and create common weakness of many articles and books written value. about supply chain management that the authors do not 2. It is equally unquestioned that in a global economy clearly differentiate between observations, hypotheses, recommendations and predictions). companies face the challenge to manage complex 123 Logist. Res. (2009) 1:71–82 73 process chains involving an increasing number of If this formation is interpreted as an advice to all compa- players. nies regardless of their industry and their position in a 3. Everybody appreciates the opportunities offered by value chain, there would be as many networks (‘‘extended modern IT systems to improve the coordination of enterprises’’) as there are companies, which means there demand forecasts, production schedules and capacities would be none (Fig. 1). (including inventories) across the boundaries of single Already before any attempt to analyze this paradox in a companies. systematic manner, some doubts take possession of our mind. Should Airbus really take care about possible bot- All these issues have been and can be further discussed tlenecks in the capacities of brazilian companies occupied at length under the headline ‘‘logistics’’. Interconnected- with the extraction of the ore, because after several con- ness was not a foreign word in this context. But as already version steps transforming ore into steel in China the mentioned interconnectedness by itself does not lead to material is build into turbojet engines by Rolls and Royce institutional innovations or systems of a higher order which in Great Britain (who also supply other companies in the can be managed in a holistic manner—even if by means of aircraft industry)? Furthermore it is unclear who decides a single sourcing strategy the relations to suppliers get whether this is the supply chain of Airbus or the network of tighter and more companies seek a relief of complexity and Rolls and Royce. a reduction of transaction costs by accepting a longer Coming back to the question whether supply chains duration of supply contracts. exist out of themselves or have to be shaped before they When defining his concept of ‘‘strategical networks’’, can be further managed we base our arguments on the Sydow [9, p. 82] describes what is needed to constitute following perception of the current state of most german such a new organizational entity: ‘‘explicitly formulated industries: if we take a closer look at the mutual interde- goals,… a formal structure with the allocation of formal pendencies that connect companies operating on different roles… and an own identity’’. We have to add: without the stages of their respective value chain, we do not observe enabling limits of closed boundaries separating the orga- isolated chains, but instead we see overlapping open nization unambiguously from its environment a common polycentric networks. Many companies appear as cross- optimum is impossible for simple logical reasons. Open roads passed by different material and semi-finished goods opportunities (sourcing from or selling to third parties) which subsequently become part of totally different end would fragment the network and destroy the possibility to products. Glas yarn is needed to produce laminate, which derive a common optimum, because there would be no in turn is a material required by the manufacturers of cir- common agenda and no common, closed solution space. cuit boards that go into cars, mobile phones and industrial Even the more basic goal of surviving in an uncertain robots. Another output of the capacities used to produce environment could not be controlled because it remains glas yarn is fiber glass which, among other products, goes unclear who precisely should be kept living. Conversely into paperhangings. Consequently a car manufacturer or a this means: the price companies must pay for a holistic producer of mobile phones—without knowing from each optimum is a limitation of its freedom of action and a other—my be hit by a bottleneck in the capacity of the subordination. This characteristic cannot be graduated. supplier of glas yarn, and while depending on the same raw (Within every day speech the notion of ‘‘optimization’’ is material the automotive industry is connected with com- used in an inflationary manner denoting any kind of panies selling hangings. These overlapping, open and improvement. We use it in the strict sense in which it is polycentric networks cannot be subjected to a holistic used within the field of Operations Research, denoting optimization, because they are far too complex. They have solutions which, given a set of data and assumptions, cannot be further improved.) If companies are ready to abandon their autonomy and subsume their narrow self interest for the benefit of the supply chain as a whole they establish the prerequisites for relocating competition. They compete as an integrated supply chain against other supply chains. Christopher [7,p. 5] illustrates this ‘‘new competitive paradigm’’ with a simple drawing showing a company surrounded by its customers and its suppliers and their customers and sup- pliers. Together they form the ‘‘confederation of mutually complementary competencies and capabilities’’ [7, p. 286] that seems ready to be subjected to a holistic optimization. Fig. 1 The supply chain network 123 74 Logist. Res. (2009) 1:71–82 neither a distinct boundary nor (as a consequence) an own management the addressed goal conflict would not simply identity. This finding has three major implications: disappear. Building tightly coupled supply chains on the basis of a single sourcing strategy is not a logical impli- 1. The requirement that all supply chains should totally cation of systems thinking, but just one of several possible focus on the needs of the ultimate customer is ways to structure a system. The objective evidence that this misleading. On the pre-stages of production it is often arrangement creates more benefits than opportunity costs unclear, who the final customer will be and, if there are has to be provided in every single case. several, if their requirements concerning the perfor- Another problem is the generation of a strategy that can mance of ‘‘the’’ supply chain are identical. (For a be undersigned by all the members of the supply chain. manufacturer of hard disk drives like Western Digital Intrinsically this strategy is needed before the formation, the end customer can be a child using his paddle as because it is the basis for deriving the selection criteria and well as a large company using mainframe computers). defining the business case. On the other hand a legitimation 2. The idea of constantly feeding actual customer demand of the strategy ideally should be based on the approval of forward to all players in the supply chain can not be all members. The only way out of this circle is a company applied without a distinctive preliminary reduction of which has enough power to define the boundaries of a complexity (even than it will turn out to be extremely network which can than be subjected to a holistic optimi- difficult to find the way to the 5th or 7th tier suppliers zation. In this case the formation of the chain will be biased through a number of complex bills of material on all by the individual interests of the focal enterprise. An levels of the chain and to calculate the related lead ongoing conflict within the german consumer goods times per item, supplier and stage which must be industry exhibits the implications of this approach. known in order to allocate the adapted times of Within this sector it has been a long and unquestioned delivery that have to be achieved by all suppliers on all tradition that the supply of retail outlets is the genuine task stages). of the manufacturers. Accordingly they have established 3. Supply chains can not be found in reality as existing their distribution systems, which show the pattern of objects for managers or researchers (like fruit flies can uprooted trees and whose efficiency strongly depends on be found by a genetic researcher). Instead they must be the annual transportation volume a single supplier controls created by carving a limited number of companies on (economies of scale and density). Since the end of the last different stages of production out of an unmanageable century not only in Germany more and more retail chains network of networks. have discovered the opportunity to get control over their For several reasons the latter is not a trivial act. One inbound goods flows in order to exploit the benefits of reason points to an organizational problem. Until today the bundling for themselves. Because this change of system organizational pattern of many companies still follow the leadership requires a change in the delivery terms this idea of a functional specialization. This means: in order to model is also called ‘‘factory gate pricing’’ (Germans use become a member of a single supply chain, people from the the word ‘‘Selbstabholung’’ which means selfcollection). purchasing and the sales department have to be convinced Figure 2 illustrates the conflict behind these competing that their way to see the world is not as relevant as the claims to define ‘‘the’’ supply chain. The two models world outlook of logisticians. Purchasers e.g. tend to like cannot tolerate each other because both create economies competition in the buying market which would have to be of scale which can only be generated and exploited once. abandoned in order to make a holistic optimization of ‘‘their’’ supply chain possible. Multiple sourcing could Distribution system of create ‘‘economies of substitution’’, but it would fragment a supplier the chain in the same way in which multiple distribution M1 channels would do (in the latter case there is no predefined R1 reference object for the ‘‘optimal’’ allocation of scarce capacities). This problem cannot just be solved by broadening the concept of SCM, e.g. by defining SCM ‘‘as being more than just logistics’’ and extending it to ‘‘the management of all business processes’’ [10, p. 5]. Companies would have M m Rn to reorganize themselves via a fundamental shift of inter- Procurement system of a retailer faces and a related redistribution of power and influence. But even if logistics and purchasing are organized as subsections of a newly established superior supply chain Fig. 2 Competing supply chain definitions 123 Logist. Res. (2009) 1:71–82 75 In order to fully understand the picture, imagine that • Who is/should be in charge of deciding which compa- the supplier in the middle is Proctor and Gamble deliv- nies should be accepted or excluded as member of the ering pampers to the retailer in the middle named supply chain? METRO. The dark arrows represent the distribution sys- • If the coherence of the chain and the channelizing of behaviour is not only guaranteed by incentives and tem of Proctor and Gamble which is bundling orders from different retail organizations and perhaps some healthcare mutual trust (which seems to be favoured by the advocates of ‘‘collaboration’’) but by negotiated organizations on the way to the outlets, regional ware- houses or clinics where they are needed (the real archi- explicit reciprocal obligations written down in long term contracts (which might be regarded as necessary tecture of such a network of course is more complex, what matters here is the fact that the commodity flows are in order to foster and protect specific investments): who entirely controlled be the manufacturer). If METRO is can/should play the role of the contractual partner? successful in claiming that all their inbound flows from • If a member steps out of line by exploiting the different manufacturers (comprising among others pam- opportunity to source some material from outside pers, CD-players, mountain bikes, power saws and food of (because the products are cheaper or better) or by all kind) have to be regarded and treated as their supply allocating part of his capacities to a third party because they are ready to pay higher prices: who has the chain, the dark arrows are replaced by the bright arrows and a totally different, formerly unknown network with a authority to inhibit this or force the disloyal to compensate the victims? (Would a prevention of this different architecture arises. The boundaries of this alter- native ‘‘supply chain’’ are marked by the dotted line. (By opportunism be compatible with the idea of ‘‘optimi- the way: METRO was very successful in following this zation’’? And: are the related opportunity costs path). included in the calculation of the ‘‘win–win-situation’’? Although this is an interesting question, this is not the • Who takes on the costs for observing compliance with place to discuss which side has the stronger arguments or commitments or enforcing contracts? whether this is a zero sum game. What matters in the given • Who measures/controls the benefits or losses which collaboration initially (that is before any redistribution) context is the finding that generates within the four walls of each member? (This • The confining of a supply chain is a foregoing act presupposes the absence of opportunistic behaviour of which cannot be ‘‘optimized’’ in itself. all partners while exhibiting their data and it raises • The thesis that ‘‘collaboration’’ in supply chains always some delicate measurement problems: for instance the and by necessity unlocks win-win-situations cannot be question to what extent the number of unplanned ad hoc kept up. changes of short term production schedules formerly • Starting the reflection about supply chains based on the provoked by unforeseen bottlenecks in the capacities of assumptions that there is (or should be?) ‘‘a common suppliers can be reduced and what the exact bottom line agreed agenda driving the achievement of the supply effects of this stabilisation are). chain goals’’ [7, p. 293] and a ‘‘joint ownership of • Who decides about the reallocation of the benefits/ decisions’’ is audacious. losses if (which has to be expected) these outcomes initially are spread among the members in an asym- metric manner? 3 Governance structures: how can supply chains • If the results of the before mentioned activities be managed? (incentives, subsidies and/or guaranties) must be known in advance in order to facilitate the decision of a The idea of a management beyond the limits of ownership company to join a supply chain: who can claim the is not as trivial as may look like. Keeping in mind that the legitimation to do this work before the supply chain has concept of an ‘‘extended enterprise’’ is an institutional assumed a concrete shape and how could anyone get innovation and that institutions are defined by roles and access to the sensitive data base needed to simulate the rules, it is amazing to see that most articles on SCM focus outcomes for different constellations? • If, as Lambert et al. (2008) correctly point out, a on process models and on the potential outcomes and benefits of supply chain management while losing sight of process-oriented organization overcoming the internal the problems connected with the organization of leader- functional silos (procurement, production, marketing/ ship. Before discussing this issue we list up some of the sales) within the individual firm is a prerequisite for any problems that have to be addressed and solved in order to supply chain-wide process integration: who can stim- make a supply chain work: ulate the necessary transformation in order to create 123 76 Logist. Res. (2009) 1:71–82 supply chain readiness within companies which are still based on the mutual trust between equal partners. Like organized around business functions? power trust is a means to reduce complexity. It averts the • If a coordinated production planning across the whole risk of opportunistic behaviour, reduces uncertainty, allows supply chain requires modern IT-tools like so-called for decreased controlling efforts and thus can reduce Advanced Planning Systems installed in all companies transaction costs. But if trust is a prerequisite of collabo- plus the associated adjustment of master data (like bills ration it must be in place before the supply chain can be of material): can all suppliers be forced to accept the constructed. The problem whether a leap of faith pays off considerable expenditures actuated by these measures? has been discussed at length within game theory (see for instance the insights delivered by Axelrod [12] regarding Basically there are two solutions offered to shape the human behaviour in a situation called ‘‘prisoner’s governance structure of a supply chain, namely (a) the dilemma’’). But unlike people playing games real compa- subordination to a company endowed with enough power nies have a long history in their role as customers and/or to orchestrate the coordination across all other participants, suppliers which strongly affect their expectations con- and (b) a heterarchical coordination based on a mutual cerning the behaviour of their ‘‘partners’’, and an equal partnership among companies with equal rights called distribution of power is uncommon in most markets. This ‘‘collaboration’’ (in the latter case the organizational leads Christopher [7, p. 286] to the insight that ‘‘perhaps arrangement would be some kind of a steering committee). the one of the biggest challenges to the successful estab- The question whether the idea of self-organizing, poly lishment of marketing networks is the need to break free centric actors forming a supply chain presupposes a sym- from the often adversarial nature of buyer/supplier rela- metric distribution of power (a situation which can hardly tionships that existed in the past’’. be found in reality) or can be combined with a distinct If one takes a closer look at the way in which original power-based leadership remains unanalyzed because it is equipment manufacturers in the automotive industry or barely raised. This is not a big drawback because both lager retail chains interact with their suppliers one soon approaches do not solve the problem. realizes that via severe negotiations on prices these com- If the orchestration is done by a strong leading company panies still fight for their part of the value creation in their it remains unclear respective supply chain, using all the power they have. Do (a) why a powerful company should give up its individual they need a re-education because they have not yet optimum for the benefit of a superordinate ‘‘supply understood that eliminating the bullwhip-effect and chain profitability’’ [5, p. 6], and reducing transactions costs by means of collaboration can (b) what could make the powerless members of the supply yield higher benefits than letting suppliers fight for orders chain believe that after long years of a fierce compe- and exploiting economies of substitution? tition on the shares of the value added the supply chain Of course one has to change things if one wants to leader while orchestrating the chain will always improve them. The current performance of managers respect their interests and not capture the surplus? (If therefore may be the wrong benchmark for the evaluation in the face of market power suspicion established by of an innovation. But the advocates of collaboration have previous experience cannot be replaced by trust the to ask themselves some fundamental questions: can an functioning of information sharing cannot be expected. organization that abstains from any hierarchy be strong Instead opportunism will spread and generate the kind enough to create and run something like an ‘‘extended of problems which have been elaborated in detail enterprise’’ reaching out from the production of raw within the agency theory [see e.g. 11], in particular material to the end customer consuming the end product? there will be hidden information and hidden action). Can an arrangement led by some kind of a steering com- mittee (with the inclination to endless debates about the Power comes along with legitimation problems, reduces right strategy, the distribution of benefits, etc.) really loyalty, provokes resistance and the striving for indepen- compete with more rigid organizations which rely on dence (and/or the building of countervailing power), and in empowered managers taking over personal responsibility? most cases it looses strength when it is extended If one hesitates to impute ignorance, inertia or irrational ‘‘upstream’’ to the supplier’s supplier. Probably these are behaviour to managers who do not ‘‘break free from the main reasons why most exponents of supply chain adversarial relationships’’ and accept a shared destiny with management prefer the second option for the design of a their partners the conclusion from what we can observe is governance structure which ideally presupposes the different. Obviously the concept of collaboration is based absence of power: Collaboration. on unrealistic assumptions and at the same time too con- Unfortunately while solving one problem the absence of fuse in order to be used as (or instead of) a governance power creates some others. Collaboration is/should be structure which is strong enough to create and manage 123 Logist. Res. (2009) 1:71–82 77 whole supply chains as if they were legal entities. This Probably his partner and former customer would not even assessment corresponds with a number of recent empirical notice the opportunity for a price cut because in order to findings. reduce transaction costs (which is always highlighted as Kampstra et al. (2006, p. 315) diagnose that ‘‘paradox- one of the major benefits of integration) he has dismantled ically, SCC is immensely popular both in business and his procurement department. Within the so called ‘‘new academia and at the same time most collaborative initia- institutional economics’’ [17] the harmfulness of dissy- tives end up in failure’’. A joint study of Capgemini, the metrically distributed information has been elaborated in Georgia (Southern) University and the University of detail under headlines like ‘‘adverse selection’’ and ‘‘moral Tennessee states: ‘‘supply chain collaboration is needed, hazard’’. With their captive internal markets supply chains but not happening’’ [13], and a study delivered by Forrester are the ideal nutrient medium to generate such dysfunc- comes to the conclusion that the ‘‘much-hyped concept in tional effects. The external competitiveness of supply the late 1980s and 1990s’’… ‘‘did not live up to the chains will diminish because inside the network competi- industries expectations’’ [14]. Vereecke and Muylle [15,p. tion has been suspended as a controlling mechanism and as 2] do not only cast doubt on the dispersion of collaboration, a source of energy. but also on the capability of this model to deliver better For the same reason the advocates of rigidly integrated results: ‘‘Empirical support for the relationship between supply chains should be careful with their assertion that supply chain collaboration and performance improvement this institutional arrangement would lead to a stronger is scarce’’, they state and sum up their own investigation orientation towards the needs of the ‘‘ultimate customer’’. with the conclusion ‘‘performance improvement is only As long as purchasers and suppliers compete for their part weakly related to the extent of collaboration with cus- of the value added the market mechanism makes sure that tomers or suppliers’’. at the end of the day all increases in efficiency will be handed on to the end customer. In contrast to this while reading about the win-win-situations that supply chain 4 Incentive systems: more competitiveness based management creates and about the necessity to share these on less competition? effects among the partners in a fair manner one gets the impression that supply chains should take possession of the Within integrated supply chains market prices have to be additional gains as ‘‘relational rents’’ in order to stabilize substituted by internal transfer prices. This operation the arrangement. This is definitely not in the interest of comes along with a huge loss of valuable information. customers and it can turn out to be a competitive disad- Looked at from outside a market price looks like a rather vantage in relation to those companies who refrain from joining any collective. poor information: it consists of only one number. What is making it rich and valuable as an incentive and as the A side impact of internal transfer prices is the effect that linchpin of a steering mechanism is it’s mode of formation. in a dynamic environment the associated partners after a Market prices reflect the currently prevailing conditions of short time will no longer know whether the allocation rules the production and distribution of a good, including for fixed with these prices comply with the requirements of instance updated oilprices, increased road charges, the fairness and justness. This can generate as much dissatis- pricing policies of suppliers and competitors, and the faction as the impression that one could buy or sell prod- shortness of material and capacities. ucts outside the supply chain at better conditions. Market prices work as signals leading managers to adapt Astonishingly neither the question whether such an even to facts that are out of their sight. By producing and arrangement fits in the shareholder value concept nor the spreading these signals markets permanently create situa- question whether it is compatible with anti-trust-law has tions in which all decision makers can use much more ever been raised und discussed. information than any single one of them has at his disposal. Disengaging oneself from the market mechanism as The market mechanism uses a knowledge which does not means of coordination obviously generates considerable exist as an entity, and it therefore is a much better instru- opportunity costs. The presumably most negative effect of ment for the usage of scattered information than any central the idea of vertically integrated supply chains in this regard planning institution could ever be (for a fundamental has not yet been mentioned. Competition stimulates and analysis see [16]. encourages innovation. Eliminating competition and The replacement of market prices by an internal pricing entrepreneurship therefore means eliminating an important system inevitably leads to delayed and biased adjustments. source of competitive advantages. Suppliers whose sales A supplier facing overcapacities in his original market seem secured through the membership in a supply chain do would not feel the need to adjust the internally defined no longer have to prove their right to exist by positioning transfer prices he was granted under previous conditions. themselves at the top of technological progress. They 123 78 Logist. Res. (2009) 1:71–82 would also find it relatively hard to get there because they relations between processes on different levels which are lack the permanent confrontation with the demand and sequentially interdependent one also has to look at inter- expectations of different customers. Perhaps they are also dependencies between resources which can have the effect discouraged by the expectation that in a tight cooperation it that short capacities of one company affects the capacity is hard to prevent an unattended leakage of knowledge or utilization of another. With regard to planning processes even that the potential benefit of an innovation would be these two interdependencies are interconnected. If all socialized within the network they are part of. In the end possible constraints across all members of a supply chain they are no longer willing to do what they were capable of. were known to one central planning authority this instance In any case it seems most unlikely that integrated supply could replace local suboptima (which formerly resulted chains could originate suppliers with the calibre of Intel or from taking the output of other local decisions as given Bosch. They are also unable to integrate them ex post restrictions) by a total optimum which makes the best use because such companies derive their strength from their out of the joint capacities of the whole chain and in addi- autonomy. In the end this is for the benefit of the whole tion (by taking out uncertainty) could make plans more economy because within open networks innovative prod- stable. This expectation seems to bare an undisputable ucts are made accessible to more companies and to more logic. But there are dysfunctions and a severe spillover that customers. Facing such proud suppliers supply chains will are often ignored. make the distressing experience that they either have to do The attempt to map and represent all interdependencies without the market leaders or to give up their holistic in one decision model leads to the necessity to cover and management approach. represent a huge number of variables, restrictions and Last but not least we have to point out that a single parameters that have formerly been hidden behind the sourcing strategy practiced within a supply chain in order interfaces which had separated the companies into smaller to enable an end to end optimization of the whole network legal entities (i.e. they were absorbed by the market). destroys economies of scale and scope that can be Ideally this is a way to gain control over something that exploited by companies which preserve the option of pur- formerly was an annoying source of surprise and emer- chasing from different sources and accept vendors sup- gency. But unfortunately many of the parameters now plying their competitors (which is common practice within included in the model are not just there (ready to be col- the German automotive industry). Risk reducing pooling lected as fixed data) but have to be forecasted (the relevant effects on the level of an inventory which is carried by a environment of a whole supply chain comprises much supplier for several customers can be regarded as part of more potential sources of a disturbance than the environ- these economies. For simple logical reasons a holistic ment of one single company). The implication is that with the extension of the model a lot of additional uncertainty is optimization requires dedicated resources (if not specific investments that cannot be redeployed for alternative use). imported (an uncertainty that formerly other people cared This confinement inevitably increases the risks of major about). This makes the outputs of the model more and more investments, raises capital costs and scales down the unstable and creates the necessity of more and more shareholder value. To advocate this in a world in which the amendments of plans. coincidence of globalization and the internet reduces The result looks paradoxical: the attempt to represent transaction costs and creates sourcing opportunities com- the complexity of a whole network in a central decision panies never had before seems to be a peculiar idea. model does not only produce the need to collect and pro- cess a huge amount of data (as multi-level bills of mate- rial), but it also foils the intention to gain control over the 5 Flexibility: are supply chains better prepared whole system: instead of increasing the stability of plans to cope with uncertainty? the amount of uncertainty rises. In the end the organization will not gain, but loose control and has to handle one ‘‘Responsive supply chains are by definition highly inte- exception after the other. Instead of reliability and resil- grated’’, says Christopher [7, p. 280]. Besides the fact that ience the result is vulnerability. There are no more local this is not a matter of definition we argue in the following problems which can be solved on a local level. Instead each chapter that the opposite is true. We start our argumenta- local problem raises the necessity of readjusting the overall tion with a little mental exercise putting ourselves into the plan (otherwise one would neglect interdependencies and position of a mathematician trying to build an integrated give up ‘‘systems thinking’’). Because of its inability to planning tool which covers all the interdependencies incapsulate and absorb local disturbances and resolve them between the activities and capacities of the companies ad hoc, the system creates one domino effect after another. forming a supply chain. Basically there are two kind of (Experts in macroeconomics will see remarkable analogies interdependencies that have to be covered: besides to the way socialistic economies were run. They were also 123 Logist. Res. (2009) 1:71–82 79 based on a linear, mechanical and unworkable notion of Figure 3 illustrates the functioning of an ‘‘available-to- control). promise’’-model (for further details see [18]. The model shown in Fig. 3 replaces the standard lead times which traditionally structure the coordination within 6 A paradigm change: loosely coupled systems order fulfilment processes. Standard lead times, usually cross-linked by modern IT systems defined by the number of days required to deliver a prod- uct, do not consider the availability of resources (inventory In the beginning of our investigation we have separated the or production capacity) at the time orders are accepted. As ‘‘tool box’’ which has been developed under the headline of no supplier can afford to cover infrequent demand peaks ‘‘supply chain management’’ (comprising concepts like with additional capacity it usually is an unexpressed part of VMI and ATP) from the philosophy propagated on top of the agreement that on time delivery rates do not exceed a these models. This analysis has led to the conclusion that level of 95 ? x% (depending on the particular industry). the idea of optimizing networks in a holistic manner The remaining risk comes not by surprise but is intended (embracing all companies of a value chain from the sup- and expected (and usually defined by dimensioning safety pliers of raw material to the ultimate customer) is all uto- stocks), thus using the customer as a final buffer. pian. For several reasons it cannot be put into practice, and The ATP-model shown in Fig. 3 can be described as an even if it would be viable it would be unwise to pursue it. extended feedback loop, which replaces the common A further consequence of this analysis is the prediction that sequence ‘‘plan–act–check–replan’’ by the sequence the often propagated shift of competition to the level of ‘‘plan–check (availability)–replan–act’’. The original entire supply chains in many important industries is unli- equipment manufacturer (OEM = tier n) propagates his kely to evolve. initial production schedule (‘‘unconstrained forecast’’) We therefore have to replace the paradigm of optimizing upstream to his first tier suppliers and calls on them to ‘‘extended enterprises’’ by another paradigm which com- check possible constraints either on the level of inventory prises the feasible idea of an improved mutual visibility or on the level of production capacities (in the latter case concerning updated demand forecasts and the availability one can use the notion ‘‘capable-to-promise’’). As a feed- of resources without generating the detrimental side effects back he receives a capacity update (‘‘best-we-can-do’’- of a rigid vertical integration. We advocate the idea of message). Ideally the suppliers have modern software tools loosely coupled companies maintaining their autonomy called ‘‘Advanced Planning Systems’’ in place which within polycentric overlapping networks because in a enable them to derive stressable commitments out of a complex and dynamic environment loose coupling is not a constraint based planning (for further details see [19]. In case of disclosed bottlenecks the OEM either adjusts his defect but a precondition of stability and resilience. Within these open networks which do not strive for a new superior plans and propagates them as viable planning basis identity as ‘‘supply chains’’ efficiency can be considerably upstream again (‘‘constrained forecast’’) or exploits econ- increased by implementing the above mentioned models in omies of substitution and sources part of his demand from numerous bilateral arrangements. In some cases this may another supplier. Ideally this results in a situation where the lead to close relations between purchasers and suppliers, to involved suppliers only produce what is really needed long term contracts and to the development of mutual trust. whereas the OEM is disburdened from the risk of ad Fig. 3 ‘‘Available-to-promise’’ as an extended feedback loop 2. Capacity-Update Check capacities, („Best we can do“) Customer Commitments and Supply Allocations, Apply Priority Rules tier n tier n-1 Typical Response Options:  Schedule order to next period (month or week) 1. Initial Forecast 3. Forecast-Update  Re-balance supply from („unconstrained“) („constrained“) another customer  Offer a product substitute 123 80 Logist. Res. (2009) 1:71–82 hoc-adjustments and/or the belated completions of unfin- never be completely transferred from the location where ished work. Within the cooperation the activation of pro- problems first appear to the top of a central planning cesses changes from a push- to a pull-mode. authority. Hierarchies are built to communicate directives The model delivers an ostensive description of a top down but for several reasons they usually show a poor mechanism capable of exploiting the benefits of an performance if a transfer of data in the opposite direction is increased visibility between partners in a value chain. It required. The consequence is that loosely coupled networks represents that part of knowledge delivered under the are faster in adapting to changing conditions and that they headline of ‘‘supply chain management’’ that we regard as can develop a superior learning aptitude. Endowed with useful and viable (although for several reasons the com- modern planning tools they can build feedback loops plexity of their implementation by far exceeds the com- comprising customers and/or suppliers (in general not plexity of their logic). These models do not lead into a beyond the first level) which enable them to take con- perfect world (e.g. the uncertainty reduced by an ATP- straints into account which were formerly hidden behind message will not disappear completely but show up again the four walls of their partners. But the companies which before the next query, suppliers may hesitate to submit a are embedded in such open networks do not have to commitment if the risk of a cancellation is not compen- anticipate the huge number of changes in the environment sated, and the chain is still exposed to an unknown risk as which a whole supply chain would have to take into long as some potentially critical suppliers are excluded account and which would put their planning system into a from the game). Nevertheless one can proceed in this state of a permanent great nervousness. (To point this out direction and we therefore base the paradigm advocated more clearly: The number of contacts to the outside world here on the idea of companies supplying each other with per company is smaller but an integrated supply chain as a more information needed for a stable allocation of whole has much less contacts than the open, overlapping, resources. polycentric networks we are talking about. Living in a But keep in mind: within these models the companies do richer context means receiving more impulses to learn and not completely give up the flexibility of exchanging part- improve. Conversely supply chains are endowed with less ners and using the market as a means to provide economies sensitive sensing mechanisms). of substitution. They understand that in a dynamic and The institutional arrangement that is recommended here uncertain environment indeterminateness is a strength. can derive several competitive advantages from using the Maintaining the opportunities the market offers is part of market as a coordination mechanism. Some of these have their risk management. In contrast supply chains would to do with the incentives which have an impact on the make the experience that rigid coupling and an overdone behaviour of managers. While the determined membership in a closed network can be interpreted as an invitation to an connectivity foster structural inertia and narrow the number of options if major changes in the environment require opportunistic behaviour competition drives suppliers into a major internal changes in order to survive. Supply chains continuous improvement process. Innovation is stimulated lack what the cybernetician Ross Ashby [20] once has by the aspiration to be rewarded by relational rents. But called ‘‘requisite variety’’. (Further analogies cannot only this motivation only works if these rents are not granted be found in the fields of education (see [21] and artificial and remain uncertain, that is to say if competition is not intelligence (see the current discussion of ‘‘swarm intelli- abandoned. The permanent threat to loose a customer may gence’’ (Kennedy et al. 2001) and the description of multi- be regarded as a rather negative kind of motivation because agent systems and the ‘‘internet of things’’ (e.g. with [22], it is associated with fear. But this is the price that suppliers but also within evolution biology: ‘‘it is not the ability to have to pay for preserving their autonomy and uniqueness adapt, but the ability to decouple which explains the which many of them regard as their most important source enormous stability and resilience of life and of all systems of motivation. build upon it’’ says the German sociologist Luhman [23,p. The ability to solve local problems locally does not only 556] referring to the work of leading biologists like the preserve more diversity in responding to changes in the author of the ‘‘Principles of Biological Autonomy’’ [24]). environment and prevent trouble from spreading (‘‘domino Using the market mechanism will equip companies effects’’). It also reduces the intensity of communication working together in loosely coupled networks with a higher and thus helps to reduce transactions costs. Nevertheless intelligence which is not embodied in a single location, the total effect on transaction costs can be regarded as organization or IT system. From an SCM-perspective ambiguous. On the one hand there is no need to establish a polycentric networks are the incarnation of redundancy. secondary organization on top of the supply chain which But for the companies embedded in these networks this is can accomplish the holistic optimization. Furthermore the basis of their adaptiveness. They can use a wealth of there will be fewer goal conflicts to be resolved and there information (among them meaningful prices) which can are less frictions arising from different corporate cultures 123 Logist. Res. (2009) 1:71–82 81 and less inconsistencies between actions to be discussed. chains), and a community of scientist who confirm each And there is no controlling effort needed in order to pre- other in their way of thinking. The classification of this part vent members from exploiting the lock-in-situation created of the literature on supply chain management as a paradigm by supply chains through opportunistic actions. But of is further confirmed by the observation that many authors course on the other hand the permanent need to adjust who propagate the related ideas do not realize that their prices forces companies embedded in open networks to concept might produce opportunity costs (compared to establish larger sales and purchasing departments. Thus the alternatives not seen on their radar screen) or some balance of different impacts on transaction costs may be unforeseen spillover. classified as unclear. Fortunately this is not the main cri- The analysis of the drawbacks and potential disecono- terion needed to resolve the cost-benefit-equation we are mies of the attempt to optimize supply chains in a holistic discussing. manner as if they were legal entities has led to the con- In his famous article on ‘‘The Architecture of Com- clusion that this is neither a viable nor a beneficial idea (see plexity’’ Simon [25] has pointed out that modularity can be also [28]). The author therefore advocates the shift to a a strong means to avoid the drawbacks of interlaced pro- paradigm which favours loosely coupled processes and cess architectures. This idea leads us to complete the pic- planning systems and the usage of competition as an ture by highlighting the role which standardized interfaces incentive system and a source of flexibility. Within this can play in the field of collaboration. Standardized inter- paradigm the above mentioned process models have their faces facilitate integration and replaceability concurrently place because they can support decentralized planning via (for a more detailed discussion see 26). Integration must no feeding the decision making processes with updated longer lead to lock-in-situations, and one could combine demand data and information about the availability of the benefits of interconnectedness with the benefits of using inventory or production capacities sourced from suppliers. markets and competition. Transaction cost could then be In order to exploit the potentiality of an increased visibility cut to an absolute minimum. But on the other hand mod- the building of supply chains is a sufficient, but not a ularity and standardization are not compatible with the necessary precondition. individual strive of companies for competitive advantages As a logical consequence the prediction/recommenda- based on a unique process design. Probably it is not least tion that competition will be/should be relocated to the this ambivalence that has made the implementation of level of entire supply chains is disputed. There are some concepts like a demand and capacity management inte- valid arguments favouring IT-based cross-links between grated across the interfaces between purchasers and sup- companies within open polycentric networks. What we pliers so delicate in industries like the German automotive neither observe nor expect is the emergence of strongly integrated networks which develop their own superior sector. However modularity can be regarded as an intelli- gent doorway to the flexibility which organizations need in identity by defining clear boundaries to their environment order to survive in a dynamic environment. Unfortunately and establishing an exclusive ‘‘collaboration’’ inside these modularity fragments supply chains and are a threat for borders. Whoever advocates such systems of a higher those companies who seek competitive advantages or order should keep in mind that there is a price to pay relational rents out of closed networks, specific investments for the development of the institutional arrangements and rigidly coupled processes. recommended under the headline of ‘‘supply chain management’’. 7 Summary In this article a separation was made between two kinds of References outcome of the past research on the subject-matter of supply chain management. On top of a number of sub- 1. Kotzab H, Friis A (2006) Implementing supply chain manage- ment in organizations. In: Jacquemin M, Pibernik R, Sucky E stantiated process models a widely shared philosophy has (eds) Quantitative Methoden der Logistik und des supply chain been developed that shows all the elements which Kuhn management. Hamburg, p 65 [27] has described as attributes of a ‘‘paradigm’’: a shared 2. Cooper MC, Ellram LM (1993) Characteristics of supply chain world view (which is rather a perspective predefining our management and the implications for purchasing and logistics strategy. Int J Logist Manag 4(2):13 perception of reality than a theory about it), a number of 3. Bechtel Ch, Jayaram J (1997) Supply chain management: a standard problems (like in this case the ‘‘bullwhip-effect’’), strategic perspective. Int J Logist Manag 8(1):15 a number of solution statements (like the calling for 4. Simchi-Levi D, Kaminski P, Simchi-Levi E (2000) Designing ‘‘systems thinking’’, the mantra of total integration or the and managing the supply chain—concepts, strategies and case advice to shift competition to the level of whole supply studies. McGraw-Hill, Boston 123 82 Logist. Res. (2009) 1:71–82 5. Chopra S, Meindl P (2007) Supply chain management: strategy, empirical findings, Paper provided by Ghent University, Faculty planning and operation, 3rd edn. Pearson Prentice Hall, Upper of Economics and Business Administration (February 2005), Saddle River Nr. 05/291 6. Sandberg E (2007) Logistics collaboration in supply chains: 16. Von Hayek FA (1996) Die Anmaßung von Wissen. Tubingen practice vs. theory. Int J Logist Manag 18(2):274 17. Brousseau E, Glachant J-M (2008) New institutional economics. 7. Christopher ML (2005) Logistics and supply chain management, A guidebook, Cambridge University Press, Cambridge creating value-adding networks, 3rd edn. Prentice Hall, Harlow 18. Bretzke W-R (2007) Available to promise: Der schwierige Weg 8. Mentzer JT, DeWitt W, Keebler JS, Min S, Nix NW, Smith CD, zu einem berechenbaren Lieferservice. In: Logistik Management, Zacharia ZG (2001) Defining supply chain management. J Bus vol 2, p 8 Logist 22(2):1 19. Stadtler H, Kilger C (eds) (2005) Supply chain management and 9. Sydow J (1992) Strategische Netzwerke, Wiesbaden advanced planning, 3rd edn. Springer, Berlin 10. Cooper MC, Lambert D, Pagh JD (1997) Supply chain manage- 20. Ashby WR (1952) Design for a brain. Chapman and Hall, London ment: more than a new name for logistics. J Bus Logist 18(1):1 21. Weick KE (1976) Educational organizations as loosely coupled 11. Jensen MC, Meckling WH (1976) Theory of the firm: managerial systems. Administrative Science Quarterly 21 (March 1976), p 1 behaviour, agency costs and ownership structure. J Financ Econ 22. Bullinger H-J, ten Hompel M (2007) Internet der Dinge. Springer, 3:305 Berlin 12. Axelrod R (2000) Die Evolution der Kooperation, 5. Aufl. 23. Luhmann N (1991) Die Wissenschaft der Gesellschaft. 2. Aufl, Mu ¨ nchen 2000 (engl.: The Evolution of Cooperation) Frankfurt a.M 13. Staff LT (2005) Supply chain collaboration is needed, but not 24. Varela F (1979) Principles of biological autonomy. Elsevier, happening, article based on a joint study by Capgemini, Georgia North Holland Southern University and the University of Tennessee. Published 25. Simon HA (1962) The architecture of complexity. Proc Am in http://www.logisticstoday.com Philos Soc 106:467 14. Tohamy N (2005) Supply chain collaboration checkup, Forrester 26. Garud R, Kumaraswamy A, Langlois RN (2003) Managing in the report, published by Forrester Research inc. 2005. http://www. modular age. Blackwell, Malden forrester.com/research (executive summary) 27. Kuhn T (1962) The structure of scientific revolutions. University 15. Vereecke A, Muylle S (2005) Performance improvement through of Chicago Press, Chicago supply chain collaboration: conventional wisdom versus 28. Bretzke W-R (2008) Logistische Netzwerke. Springer, Berlin

Journal

Logistics ResearchSpringer Journals

Published: Jul 7, 2009

References