Supply Chain Modelling, An Overview Of Practice And Methodology

Henry Burley La Trobe University School of Business Logistics & Operations Management Bundoora Victoria Australia

Supply Chain Modelling, An Overview Of Practice and Methodology

Henry Burley School of Business La Trobe University, Bundoora, Victoria

Abstract This paper views the essential elements of effective supply chain modelling as being the timely and accurate communication (about final demand) between the echelons, and optimisation considering the supply chain as a whole. The objective is to ensure that realworld supply chain is responsive and agile (but not over-responsive) to changes in final demand volume and in desired final product characteristics. A modellers view is taken of classifying and evaluating developments in supply chain methodology. Value is seen in drawing the attention of supply chain modellers to the requirements (of a core solution) in the analogous n-person (super-additive) game theoretic model. Supply chain echelons must, in fact, be able and be motivated to behave as the postulated supply chain views them to operate as actors in the virtual organisation. That is the overall performance objective implications for each link must be profitable, against other objectives, to motivate their delivery in practice. Additionally the culture of commercial and legal relationships (relating to fairness and equity between the links) is a very important determinant in the resolution of optimality in the global chain.

Keywords: supply-chain management, inventories, modelling

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 24

1. INTRODUCTION Supply chain analysis Until today, the focus of manufacturing production research has been on the various processes within the manufacturing supply chain individually. Recent approaches have focused on the performance, design and analysis of the supply chain as a whole. Beamon (1998), relates that the increased attention to the whole supply chain has arisen from: Rising costs of manufacturing Shrinking of resources of manufacturing bases Shortened product life cycles Leveling of manufacturing playing field Globalization of market economics

In this paper we emphasize that the purpose of supply chain modeling is not just to provide an information base (snapshot) of what each echelon (layer) is doing, or planning to do. The real purpose is to ensure that the whole, real-world supply chain is optimally responsive and agile (but not over-responsive) to changes in demand volume and to changes in product characteristics of final demand. These are the reasons for production. An operating supply chain model is to be judged on its performance in practice to guide the real supply chain to respond smoothly and inexpensively to these changes. Problems include the exaggerating in a bullwhip effect, similar to the effect of a railway engine shunting into a set of carriages without their brakes on. This involves a series of futile contractions and expansions of the train.

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 25

Hence communication and coordination play very important roles in effective supply chain modelling and in effective supply chains.

2. COMMUNICATION IN THE SUPPLY CHAIN Modes of communication All enterprises have a supply chain, in so far as the flows of goods (forward) and the flows of communications (backwards) between buyers and sellers are essential to maintain commerce in a contractual way. intranet or extranet communication. An Intranet is an internal network based on Internet and WWW technology. Whilst the Web is open to anyone, the intranet is private and is protected from public visits by firewalls - specialized software security systems to prevent outsiders from invading the private network. Companies are using intranets in many ways: customer profiles, policy manuals, product inventories. Intranets are a simple cost-effective communication tool. Like extranets intranets are immediately updateable from the central server. By utilizing Extranets some firms are allowing people and organizations from outside to have limited access to their internal intranets. Extranets are especially useful for linking organizations with customers or business partners, providing for example, product availability, pricing and shipment data. Centralized enterprise networks are more reliable and manageable, with many sites sharing data. Decentralized enterprise networking is sensitive to different versions of operating systems. Decision support system database and software tools A DSS software system contains software tools that are used for data analysis, for example Communications are often improved with vendor meetings with customers, formalized contact through solicitation of feedback, email,

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 26

various OLAP (online analytical processing tools). Optimization models (such a linear programming) are employed to minimize such objectives as cost and time, or a manufacturing advance planning system or an integrated supply chain model. All of the above communication systems are only as good as the data accuracy and relevance to up-to-date final demand levels. After taking into account their cost, each system of communication must be assessed in terms of savings to the organization. This is not a simple question. Under some very general conditions, speedier communications give rise to greater volatility. Communication systems for successful virtual chain organization All management information systems (MIS) should be designed to enable a decision support system (DSS). In this case the DSS is that of the virtual organization to coordinate the chain itself. Additional to the communication conditions lies the question as to who to communicate with, what message/s to convey, and how to ensure the desired production deliveries from each link further down the chain. objective problem. Hilton (2000), in a review of modern commercial production and procurement practices, observes that Western style procurement agencies retain the trust-destroying, costly, but politically defensible, system of compulsory competitive tendering (CCT). He adds CCT ensures arms length relationship between the ultimate user and the supplier. In CCT the suppliers are dealt with through the legalistic, unfriendly and costly processes, which focus on fairness and equity between suppliers. Hilton further observes that this focus is not on efficiency and effectiveness. By contrast Fruin (1992) describe the culture of a Japanese vertical Keiretsu network where firms through cross investment of physical and intellectual resources build partnerships based on mutual commitment. Fruin describes Japanese Primes as posting staff at all levels to work with suppliers and in reciprocation suppliers post staff to them. Even supplier selection (the particular links to be in the chain) by purchasing teams in a management environment is a multi-

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 27

He sees the idea to create an operational ethos based on genuine partnerships rooted in a deep knowledge of each others culture. This in turn is rooted in shared experiences of overcoming the unforeseen and unexpected in the tough minded but mutually advantageous manner indicative of a good marriage. Fruin concludes that the Japanese type of collaborative relationship can not exist with CCT and Western style outsourcing reliant on confrontational litigation. Convergence of vertical Keiretsu and competitive tendering Morita & Nakahara (2001), and others have recently asserted that rapid advances in information technology, in particular three dimensional computer aided design (3D CAD) which dramatically increases the efficiency of manufacturer-suppliers communication in designing customised parts, will strengthen vertical Keiretsu. The Japanese cooperative relationship with a few suppliers could become stronger due to the information-technology revolution. Arguments that Western manufacturing should establish Japanese-type cooperative relationships with their suppliers may be influenced by this IT induced trend. Sharing software enables closer cooperation. It appears there will be some converging of the two different systems.

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 28

3. OPTIMIZATION OF A SUPPLY CHAIN Optimization levels and criteria Li and OBrien (1999) remind us that supply chains should be optimised at two levels: At the chain level objectives are set for each stage so that the SC performance can meet the customer service target. At the operations level manufacturing and logistic activities are optimised under given targets. In managing a supply chain model Li and OBrien (1999) also point out that supply chain efficiency is not just about minimizing costs: For example: Four important criteria need to be addressed: Profit Lead-time performance Delivery promptness and Waste elimination

A reasonable solution concept The suppliers, buyers and processors (or holders) of intermediate products in a supply chain are generally independent players in the market and have their own objectives and constraints. One cannot presume, having made a model of a supply chain, that the links in the chain, will necessarily behave in the way the modeler would like them to. Optimization across echelons, and across firms within echelons, raises the question of determining standards of rational, actual and optimal behaviour for a group in which the outcomes depend upon the actions of independent players. An important development in economic theory was the formulation of the Theory of Games and Economic Behaviour by von Neumann and Morgenstern (1953). The purpose of their work was to find mathematically complete principles, which define rational behaviour for the participants in a social economy, and to derive from them the general characteristics of that behaviour. The immediate concept of a solution is plausibly a set of rules for each participant, which tell him/her how to behave in every situation which may conceivably arise.

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 29

The problem that we encounter is that even if we have a concept of individual rationality i.e. profit maximization, can that concept be extended to situations with two or more individuals? As noted in Burley (1978), this is an exceptional, even if extremely complex problem. Even in the two-person, non constant sum game, the simple concept of individual strategy domination fails to achieve a sensible group result in the simple case of the Prisoners Dilemma (Winston, p850). An equilibrium solution to a cooperative game should guarantee each player (firm) as much as it can guarantee itself (alone) and also reward each coalition as much as it could guarantee itself. Otherwise the equilibrium would be unstable, that is, outside the core.

Supply Chain Operations Reference Model The Supply Chain Council (SCOR), www.supply-chain.org, is an unofficial international standard for supply chain management. It does not prescribe solutions, or a specific model, but does give a framework for defining the supply chain and thus the all-important relationships and information flows. It provides a series of metrices that become the building blocks for benchmarking in the SC management process. I feel that the Models documentation is a bit weak on implementation, but that is where consulting service organisations come in. One Australian firm in this strategic innovation, Lexian Solutions, www.lexian.com.au, is convinced from their experience that relationship factors are the biggest single determinant of supply chain management success. Dominant players who adopt a cooperative supply chain management approach to their trading partners, even without a supply chain modelling technology, will achieve more than a technically excellent but relationally weak SC operator.

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 30

4. DECISION VARIABLES AND MODELLING APPROACHES Decision variables Inventories exist throughout the supply chain. Carrying of inventories can cost from twenty per cent to forty per cent of their (turnover) value per year. Managing inventories in a scientific manner to maintain minimum levels of total inventory policy cost makes economic sense. Ganeshan (1999) demonstrates a simple supply chain model as a synthesis of three components: Inventory analysis at the retailers The demand process at the warehouse The inventory analysis at the warehouse

Decision variables in supply chain modelling Beamon (1998), in a comprehensive analysis of supply chains, demonstrates the complexity of a supply chain in that it contains: Production/distribution scheduling Inventory levels Number of echelons Distribution centre to customer assignment Plant-product assignment Buyersupplier relationship Product differentiation step specification Number of product types held in inventory

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 31

Modelling approaches Cohen and Lee (1998), in a strategic analysis of integrated production distribution systems, categorize four different modelling approaches in supply chain modelling: Table 1: SC Modelling approaches and emphasis of interest

Supply chain Modelling approach Deterministic analytical models Stochastic analytical models Economic models Cost-based simulation of: material control Cost-based simulation of: production control Cost-based simulation of: finished goods stockpile Cost-based simulation of: distribution

Paradigm or parameters of concern Variables are known and specified At least one variable is unknown and follows a probability distribution Game theoretic framework for modelling buyer-supplier relationship Order quantities, response times, cost data

Production lot sizes, lead times, material response times EOQ, demand data, production lead times

Ordering policies, transportation time requirements, demand data, cost data, fill rate objectives

Cohen and Lee (1998), in a strategic analysis of integrated production distribution systems, categorize four different modelling approaches in supply chain modelling:

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 32

5. ISSUES IN SUPPLY CHAIN MODELLING Uncertainty in supply chain modelling Petrovic, et al. (1999) considers there are three major sources of uncertainty in supply chains. They can be associated with: Customer demand Supply deliveries along the SC and External or market supply

Uncertainties can be represented by fuzzy sets, which interpret vague and imprecise phrases. Relationship between SC echelons and Innovation in UK clothing industry An empirical study, by Bruce and Mogor (1999), into innovation supporting and blocking factors examines the perceptions of stakeholders within the supply chain of the UK clothing industry. It is interesting that the paper title is Dangerous Liaisons as the experience of being a link in a chain can be both supportive or limiting. The data revealed strategic intentions and assumptions that support the emergence of two distinct trajectories of change in the industry based on co-partnership and ad-hoc relationships respectively. Conclusions of UK clothing industry study The co-partnership trend offered some scope for incremental innovation. The ad-hoc relationship offered less scope for incremental innovation. The absence of R&D capabilities may be an important factor in the industrys difficulties in innovating. A study by Cox (1997) into business success contends that the control of relatively scarce material resources is the major factor in ensuring sustainable business success, and that:

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 33

Owning, controlling and leveraging critical supply chain assets - those resources that cannot easily be imitated - is the key to success. Recent supply chain modelling issues Beamon (1998) has drawn attention to a number of associated issues receiving increasing attention. These are: 1. Product postponement 2. Global vs single nation supply chain modelling 3. Demand distortion and variance amplification. Product postponement is the practice of delaying one or more operations to a later point in the supply chain. For example, indenting brand name of product only when the country of retail destination is determined, or making appliances that can be installed with either a left or right-hand opening door. Product postponement in a reduction in amount and value of held inventory. Global vs. single nation supply chains raises additional structural considerations in formulation. These include: export regulations, exchange rates, and duty rates Demand distortion and variance amplification Demand distortion is the phenomenon in which orders to the supplier have a larger variance than sales to the buyer. As demand propagates upstream the variance amplification continues, (hence need for information feedback through multiple echelons). This is known as the bullwhip effect and is common in real world supply chains. This hog-cycle effect is well known in US agricultural markets, and is also evident in mineral markets for copper and aluminium.

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 34

6. CONCLUSIONS The overall aim of supply chain modelling is to achieve better coordination of the supply chain. Recent investigations by Xu, Dong and Evers (2001) highlight the negative impact that independent actions taken by members of a conventional supply chain have on expanding the volatility of forecast errors and order release. Increasing volatility may also be associated with increased speed of communications. The communication system and optimisation criteria utilized in a supply chain must reflect actual behaviour in response to the communicated signals. The culture of commercial and legal relationships (relating to fairness and equity between the links) is a very important determinant in the resolution of optimality in the global chain. Many decision variables and modelling approaches can be considered. Emphases should be placed on those particular mechanisms which are important in determining the size and volatility of the variables of interest. Online communication of precise product details through three dimensional computer aided drafting between buyer and supplier forms using the same 3D CAD software systems will not only facilitate communications, but may also widen the potential supplier market and thus change the nature of the game between buyers and suppliers.

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 35

References Beamon, B. (1998). Supply Chain Design and Analysis, Models and Methods, International Journal of Production Economics, Vol 55, No 3, pp.281-294. Bruce, M. and Mogor, S. (1999). Dangerous Liaisons: An Application of Supply Chain Modelling for Studying Innovation within the UK Clothing Industry, Technology Analysis and Strategic Management, Vol 11, No 1, pp.113-125. Burley, H. T. (1978). Game Theory and Gaming Applications in Economics. Kozesnik (ed), 8th Prague Conference on Information Theory, Statistical Decision Functions and Random Processes , Academia, Prague, Vol C, pp.143-157. Cohen, M. A. & Lee H. L. (1988).Strategic Analysis of Integrated Production Distribution Systems, Operations Research Vol 36, No 2, pp.216-228.

Cox, A. (1997). Business Success: A Way of Thinking about Strategy, Critical Supply Chain Assets and Operational Best Practice, Earlsgate Press, Stratford-on Avon. Fruin, M. (1992). The Japanese Enterprise System: Competitive Strategies and Cooperative structures. Clarendon Press, Oxford. Ganeshan, R. (1999). Managing Supply Chain Inventories: A Multiple Retailer, One Warehouse, Multiple Supplier Model, International Journal of Production Economics, Vol 59, No 3, pp.341-354. Hilton, B. J. (2000). The Impact of Modern Commercial Production and Procurement Practices on Cost Estimating and Forecasting, Journal of Cost Analysis and Management. Winter, pp.31-57 Li, D. and OBrien, C. (1999). Integrated decision modelling of supply chain efficiency, International Journal of Production Economics, Nol 59 No 1-3, pp.147-157. Morita, H. & Nakahara, H. (2001). Impacts of Information-technology Revolution on Manufacturer-suppliers Relationships: Theory and Evidence from Japan. Proceedings of Seminar at University of New South Wales, March, 2001. Petrovic, et al. (1999). Managing Supply Chain Inventories, International Journal of Production Economics, Vol 59, pp.443-453. von Neumann, J & Morgenstern, O. (1953). Theory of Games and Economic Behaviour, Princeton U. P., Princeton, NJ. Winston, W. L. (1994). Operations Research Applications and Algorithms, Duxbury, Belmont, California.

Supply Chain Modelling, An Overview Of Practice and Methodology Henry Burley

Page 36

Xu, K., Dong, Y., and Evers, P. T. (2001) Towards Better Coordination of the Supply Chain, Transportation Research Vol E, No 37, 35-54.