Dynamic supply chain modeling using a new fuzzy hybrid negotiation mechanism

ARTICLE IN PRESS Int. J. Production Economics 122 (2009) 319–328

Contents lists available at ScienceDirect

Int. J. Production Economics journal homepage: www.elsevier.com/locate/ijpe

Dynamic supply chain modeling using a new fuzzy hybrid negotiation mechanism Vipul Jain a,, S.G. Deshmukh b a b

Mechanical Engineering Department, Indian Institute of Technology Delhi, New Delhi 110 016, India ABV-Indian Institute of IT and Management, Gwalior, Morena Link Road, Gwalior 474010, India

a r t i c l e in fo

abstract

Available online 2 July 2009

The key part of dynamic supply chain management is negotiating with suppliers and with buyers. Coordination is essential for successful supply chain management. In order to model coordination among suppliers and buyers in a dynamic supply chain, this paper takes a step further and proposes a new fuzzy- logic-based hybrid negotiation mechanism. In most real-world negotiation situations, agents have a common interest to cooperate, but have conflicting interests over exactly how to cooperate. These situations involve restrictions and preferences that may be vaguely and partly defined. Therefore, this study takes the advantage of fuzzy logic and develops a hybrid negotiation-based mechanism, that combines both cooperative and competitive negotiations. Achieving effective coordination in a multi-agent system is non-trivial as no agent possesses the global view of the problem space. Moreover, the different strategies adopted by agents may produce conflicts. While agents coordinate with each other in the operations, they will negotiate about their strategies to reduce conflicts. The proposed fuzzy hybrid negotiation mechanism allows negotiation agents more flexibility and robustness in an automated negotiation system. The proposed mechanism not only helps sellers and buyers to explore various new choices and opportunities that the e-markets offer but also allows them to identify and analyze their resource constraints in a given schedule, and helps them to explore and exploit many alternatives for a better solution. The efficacy of the proposed approach is demonstrated through an illustrative example. & 2009 Elsevier B.V. All rights reserved.

Keywords: Coordination Negotiation Fuzzy logic Agents Supply chain management

1. Introduction and motivations Traditionally, marketing, distribution, planning, manufacturing, and purchasing of organizations along the supply chain operate independently. These organizations have their own objectives and they are often conflicting. Marketing objectives of high customer service and maximum sales dollars conflict with the manufacturing and distribution goals (Jain et al., 2007b). Many manufacturing

 Corresponding author. Fax: +9111 26582053.

E-mail addresses: [email protected] (V. Jain), [email protected] (S.G. Deshmukh). 0925-5273/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.ijpe.2009.06.034

operations are designed to maximize throughput and lower costs with little consideration for the impact on inventory levels and distribution capabilities. Purchasing contracts are often negotiated with very little information beyond historical buying patterns. The result of these factors is that there is no single, integrated plan for the organization. Clearly, there is a need for a mechanism through which these different functions can be integrated together (Jain, 2006a). Supply chain management is a strategy through which such integration can be achieved. Supply chain management is typically viewed to lie between fully vertically integrated firms, where the entire material flow is owned by a single firm, and where each channel member operates independently. Therefore,

ARTICLE IN PRESS 320

V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

coordination between the various players to the chain is the key in its effective management (Gan et al., 2004; Fung and Chen, 2005, Jain et al., 2007a). Effective supply chain management requires creating synergistic relationships between the supply and distribution partners with the objective of maximizing customer value and providing a profit for each supply chain member. However, often there is no effective control mechanism to coordinate the actions of the individual supply chain members such that their decisions are aligned with global supply chain objectives (Taylor, 2001; Chan and Chan, 2004). In this case, each supply chain member attempts to optimize a part of the system without giving full consideration to the impact of their myopic decisions on total system performance. This decentralized decisionmaking process is the traditional mode of operation in today’s business environment. Optimizing the portions of the system often yields sub-optimal performance, resulting in an inefficient allocation of scarce resources, higher system costs, compromised customer service, and a weakened strategic position (Jain et al., 2006b; Jennings et al., 2001). Supply chains are complex operations and their analysis requires a carefully defined approach. Moreover, with on increase in technological complexity, supply chains have become more dynamic and complex to solve. Consequently, it is easy to get lost in details and spend a large amount of effort for analyzing the supply chain. There is growing interest from industry and academic disciplines regarding coordination in supply chains, particularly addressing the potential coordination mechanisms available to eliminate sub-optimization within supply chains (Wang and Gerchak, 2001; Fung and Chen, 2005; Jain, 2006a). Coordination, the process by which an agent reasons about its local actions and the actions of others to try to ensure that the community acts in a coherent manner (Toledo Excelente and Jennings, 2002), is an important issue in multi-agent systems (MASs). There are three main reasons why it is necessary for agents to coordinate. First, there are dependencies between agents’ tasks or goals; second, there is a need to meet global constraints such as cost and time limits; and third, no individual agent has sufficient competence, resources, or information to solve the entire problem (Parunak et al., 1998; Toledo Excelente and Jennings, 2002). Member enterprises in the chain need to cooperate with their business partners in order to meet customers’ needs and to maximize their profit. Managing multi-party collaboration in a supply chain, however, is a very difficult task because there are so many parties involved in the supply chain operation, each with its own resources and objectives. There is no single authority over all the chain members. Cooperation is through negotiation rather than central management and control. The interdependence of multistage processes also requires real-time cooperation in operation and decision making across different tasks, functional areas, and organizational boundaries in order to deal with problems and uncertainties. The strategic shift of focus for mass customization, quick response, and highquality service cannot be achieved without more sophisticated cooperation and dynamic formation of supply chains

(Chan et al., 2004). One solution to this problem is to have intelligent interacting entities that can provide domainspecific information to validate the decision-making system. Therefore, MAS-based approaches for supply chain modeling are proposed (Swaminathan et al., 1998; Ertogral and Wu, 2000; Julka et al., 2002; Jain et al., 2007a; etc.). Agent-based modeling can be assumed to be a reasonable methodology for the examination of supply chains because in a supply chain a number of individual companies interact with each other using specific internal decision structures (Choi et al., 2001). Each of the players in the supply chain is modeled as an agent who negotiates with its immediate neighbor in pushing/pulling the part or product through the chain. The agents operate in MASs and situations often arise in which their plans conflict with the plans of other agents. For achieving effective multi-agent coordination, conflict resolution is crucial. Negotiation is a predominant tool for resolving conflict of interests (Jain et al., 2007a). However, with recent technological advances, the mechanisms available to carry out such activities have become increasingly sophisticated, and the environment in which these activities take place has become highly dynamic. A higher-level coordination mechanism with respect to distributed modeling of supply chains is generally not specified in the reported literature (Chan and Chan, 2004). A variety of research work exists on negotiation strategies in the areas of social science, game theory, negotiation support systems, agent technologies, and machine learning. Unfortunately, automated negotiation agents based on any of these techniques usually face two problems. First, agents are not as flexible and adaptive to different negotiation environments as desired. Negotiation environment is a set of pre-defined negotiation features that are not negotiable. This means an agent may work well under one set of negotiation features, but perform worse in others. Second, a fixed strategy or a static group of strategies may become known by competing agents as a result of negotiation processes, after which those agents can potentially exploit this knowledge in future negotiations (Jennings, 2000; Fung and Chen, 2005). The interacting network formation of a supply chain is inherently complex (Wilding, 1998), with the majority of firms operating simultaneously in multiple supply chains. The operational complexity of supply chains further complicates the structural complexity, which shows itself in supply chains as consistent and unpredictable materials and information flow. Knowledge in expert system is vague and modified frequently (Jain et al., 2005). Hence, there is a strong urge to design a dynamic knowledge inference system that is adaptable according to knowledge variation as human cognition and thinking (Jain et al., 2005). Therefore, unlike other researches, in this paper we take the advantage of fuzzy logic and develop a hybrid negotiation-based mechanism that combines both cooperative and competitive negotiations. The proposed fuzzy hybrid negotiation mechanism allows negotiation agents more flexibility and robustness in an automated negotiation system. The proposed mechanism not only helps sellers and buyers to explore various new choices and

ARTICLE IN PRESS V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

opportunities that the e-markets offer but also allows them to identify and analyze their resource constraints in a given schedule, and helps them to explore and exploit many alternatives for a better solution. Moreover, in the proposed hybrid fuzzy negotiation mechanism agents learn and make decisions on when to negotiate, with whom to negotiate, and how to negotiate based on the past negotiation experiences, current activities, and predictions. We choose the fuzzy negotiation framework for the following reasons:

 Consider the trade-off involved in agents deciding









whether they prefer to acquire exactly the preferred value of an attribute that is very important or several sets of less-good values for attributes that are less important to it. Such a trade-off can be modeled by fuzzy constraints. One of the primary things in negotiation is to represent trade-offs between the different possible values for parameters. A buyer’s preferences on trade-offs between different attributes of the desired product can be easily modeled by fuzzy constraints. For a single characteristic of the preferred product, a buyer might prefer certain values over others. Such a preference can be expressed as a fuzzy constraint over a single attribute, and the preference level at a certain value of the attribute is the constraint satisfaction degree for that value. Likewise, for multiple product attributes, a buyer might favor certain combinations of values over others. Such preferences can be expressed as fuzzy constraint over multiple attributes, and the preference level at a certain combination value of these attributes is the constraint satisfaction degree for the combination value. In several cases, buyers do not know the precise details of the products they want to buy, and so their requirements are often expressed by constraints over multiple issues. A buyer’s constraints are not for all time equally important. In order to deal with different levels of importance of different fuzzy constraints, researchers have introduced the concept of fuzzy in negotiation mechanisms. When buyers and sellers negotiate, it is hardly ever the case that an offer is totally acceptable or totally inconsistent with their respective constraints. Relatively, an offer usually satisfies the buyer’s constraints more or less. The proposed fuzzy negotiation framework is suited for capturing constraints of this kind because fuzzy constraints can be partially satisfied or violated.

The rest of the paper is arranged as follows: Section 2 deals with supply chains and fuzzy system modeling. Section 3 addresses an exhaustive literature review. Section 4 presents the proposed fuzzy hybrid negotiation mechanism. Section 5 discusses an illustrative numerical example. Finally, Section 6 concludes the paper with some perspectives.

321

2. Supply chain systems and fuzzy system modeling

A supply chain is also a network of facilities and distribution options that functions to procure materials, transform these materials into intermediate and finished products, and distribute these finished products to customers. Supply chains exist in both service and manufacturing organizations, although the complexity of the chain may vary greatly from industry to industry and firm to firm. Realistic supply chains have multiple end products with shared components, facilities, and capacities. The flow of materials is not always along an arborescent network; various modes of transportation may be considered, and the bill of materials for the end items may be both deep and large (Jain et al., 2004). Petrovic et al. (1999) highlighted the uncertainties in supply chain system as follows: ‘‘A real supply chain operates in an uncertain environment. Different sources and types of uncertainty exist along the supply chain. They are random events uncertainty in judgment, lack of evidence, lack of certainty in judgment, lack of evidence, lack of certainty of evidence that appear in customer demand, production, and supply. Each facility in the supply chain must deal with uncertainty demand imposed by succeeding facilities and uncertain delivery of the preceding facilities in the supply chain’’. Generally, supply chain networks include several subsystems with unlimited interfaces and relations. Every subsystem usually contains uncertainties. Obviously, uncertainties associated with each subsystem or components make the whole system vague. Also, the nature of interfaces in dynamic supply chains causes supply chains to function in completely imprecise and uncertain environment. These interfaces are rooted in the information flows, material flows, and supplier–buyer relations Goyal and Gopalakrishnan (1996). Moreover, relations among entities of dynamic supply chains critically depend on human activities. This fact forms the main reason why emergent dynamic supply chains necessitate fuzzy system modeling. Sugeno and Yasukawa (1993) state ‘‘Fuzzy algorithms are nothing but qualitative descriptions of human actions in decision making’’. Zadeh (1973) also states ‘‘As the complexity of a system increases, our ability to make precise and yet significant statements about its behavior diminishes until a threshold is reached beyond which precision and significance (or relevance) become almost mutually exclusive characteristics. It is in this sense that precise quantitative analyses of the behavior of humanistic systems are not likely to have much relevance to the real-world societal, political, economic, and other types of problems, which involve humans either as individuals or in groups’’. Sun (1999) develops a distribution constraint satisfaction problem formulation in the modeling of the supply chain as a total system using the fuzzy technology. Petrovic et al. (1999) examine uncertainties in supply chains by focusing on decentralized control of each inventory and partial coordination in the inventories. Turksen and Zarandi (1999) discussed many advantages of fuzzy system approach in

ARTICLE IN PRESS 322

V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

real-world applications that motivate the authors to apply fuzzy modeling in dynamic supply chains. 1. Fuzzy system models are flexible, and with any given system like dynamic supply chains, it is easy to handle it with fuzzy system models. 2. Nearly all nonlinear functions of arbitrary complexity can be captured by fuzzy system models. Also fuzzy models are conceptually simple to understand. 3. Superior communication between experts and managers is provided by Fuzzy system models. Moreover, these are based on natural languages and are tolerant of imprecise and vague data. 4. Fuzzy system models can be constructed on the top of the experience of experts and can be mingled with conventional control techniques. Dynamic supply chain network problems are characterized by their complexity and inherent decentralization. The application of fuzzy logic and MAS techniques to this problem seems appropriate. Therefore, in this paper, we have tapped the properties of fuzzy logic and propose fuzzy negotiation mechanism to capture dynamic negotiation between sellers and buyers. 3. Literature review Traditionally, supply chains have been formed and maintained over long periods of time by means of extensive human interactions. But the increasing demand for accelerated commercial decision making in the face of exploding growth and complexity of information is creating a need for more advanced support for automated supply chain formation (Reaidy et al., 2006). Companies ranging from auto makers to computer manufacturers are basing their business models on rapid development, build to order, and customized products to satisfy ever-changing consumer demand, and fluctuations in resource costs and availability mean that companies must respond rapidly to maintain production capabilities and profits. As these changes increasingly occur at speeds, scales, and complexity unmanageable by humans, the need for automated supply chain formation becomes acute (Durfee et al., 1989; Chopra and Meindl, 2007; Lo et al., 2008). The general agent-based modeling method is not new and it has already been applied in many different economic and social contexts. There has been much research work that deals with coordination in MASs. A framework is presented in (Toledo Excelente and Jennings, 2002) that enables agents to dynamically select the mechanism they employ in order to coordinate their inter-related activities. The agents also learn when and how to coordinate. In Xuan et al. (2001), various coordination aspects and a cooperative, multi-step negotiation mechanism are discussed. Collaboration and partnership are new leitmotivs in organizations and supply chain management, with emphasis on collaborative design and planning. Several major research projects emphasize the manufacturing and logistic aspects of the collaboration, addressing the

problem of enterprise and enterprise-wide modeling and integration. Examples are the CIMOSA project, the TOVE project (Fox et al., 2000), and the NetMan project (Sophie et al., 1999; Frayret et al., 2001). Yung and Yang (1999) proposed the integration of multi-agent technology and constraint network for solving supply chain management problems. Yan et al. (2000) developed a multi-agentbased negotiation support system for distributed electric power transmission cost allocation based on a network flow model and KQML. Gjerdrumm et al. (2001) showed how expert system techniques for distributed decision making can be combined with contemporary numerical techniques for supply chain optimization. Though the applications of MASs in networked manufacturing and supply chains are not brand new, the analytical models and algorithms for integrating the planning and coordinating the operations of these agents have not been fully addressed. An MAS consists of a group of agents that can take specific roles within an organizational structure. Different types of agents may represent different objects, with different authorities and capabilities, and perform different functions or tasks. MAS enhances performance along the dimensions of (1) computational efficiency because concurrency of computation is exploited (as long as communication is kept minimal, for example, by transmitting high-level information and results rather than low-level data); (2) reliability, that is, graceful recovery of component failures, because agents with redundant capabilities or appropriate inter-agent coordination are found dynamically (for example, taking up responsibilities of agents who fail); (3) extensibility because the number and the capabilities of agents working on a problem can be altered; (4) robustness, the system’s ability to tolerate uncertainty, because suitable information is exchanged among agents; (5) maintainability because it is easier to maintain a system composed of multiple-component agents because of its modularity; (6) responsiveness because modularity can handle anomalies locally, not propagate them to the whole system; (7) flexibility because agents with different abilities can adaptively organize to solve the current problem; and (8) reuse because functionally specific agents can be reused in different agent teams to solve different problems. With these properties, MAS then becomes a research area attracting a great amount of research efforts, especially, for a problem domain that is characterized by complexity, large scale, distribution and uncertainty (Jain, 2006a). Moreover, the agent system is an alternative technology for supply chain management because of the features such as distributed collaboration, autonomy, and intelligence (Nissen, 2001; Chan and Chan, 2004). The ability to meet the changing needs of customers requires changing the supply of product, including mix, volume, product variations, and new products. Meeting these needs in the supply chain requires flexibility in sourcing product from raw materials to outsourced finished product. Swaminathan et al. (1998) present a modeling and simulation framework for developing customized decision support tools for supply chain re-engineering. Agents may represent various supply chain entities, viz. customers,

ARTICLE IN PRESS V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

manufactures, and transportation. These agents use different protocols and help in simulation of material, information, and cash flows. The research findings of Tsay (2002) highlights how risk aversion affects both sides of the supplier-retailer relationships under various scenarios of relative strategic power, and how these dynamics are altered by the introduction of a return policy. Jennings et al. (1998) provide an overview of research and developments in the field of autonomous agents and multi-agent systems. Negotiation is a process by which a group of entities try and come to a mutually acceptable agreement on some matter (Pruitt, 1981). According to the cardinality and nature of the interaction, automated negotiation models can be classified into three main categories (Jennings et al., 2001). The first consists of many-to-one or many-tomany models in which multiple agents negotiate with either a single or many other agents. This category is predominantly handled using various auction-based models (Sandholm, 1999) and these models are widely used in the field of on-line retail, e.g., eBay (http:// www.ebay.com) and eMediator (Sandholm, 1999). The second category consists of one-to-one models in which a pair of agents negotiate directly with one another (Faratin et al., 2002). These models typically use a range of heuristic methods to cope with the uncertainties. The third category consists of argumentation/persuasion-based models (Kraus et al., 1998) in which agents use various types of argument, such as threats, rewards, and appeals, to persuade their opponent to accept a deal they would not previously have countenanced. For each of the three categories, the negotiation domain could be a single-issue one (e.g., price) or a multiple-issue one (e.g., price, quality, model, volume, delivery date, expiry date, after-sale service, warranty, and return policy). Agents in open and dynamic environments like dynamic supply chains would aim to produce the best possible result given their available processing, communication, and information resources to maximize system output (Garcı´a-Flores et al., 2000; Julka et al., 2002; Emerson and Piramuthu, 2004). While designing agent systems, it may be difficult to foresee all the potential situations an agent may encounter and specify an agent behavior optimally in advance. Agents therefore have to learn from, and adapt to, their environment, especially in multi-agent setting. However to build autonomous agents who improve their negotiation competence based on learning from their interaction with other agents is still an emerging area Mundhe and Sen (2000). The most intelligent agents will be able to learn, and will be able to adapt to their environment, in terms of user requests and the resources available to the agent (Papazoglou, 2001; Yoo and Kim, 2002). There are already a variety of information systems and networks working within and between chain members to facilitate the flows of materials, information, and funds. However, there is lack of coordination and integration between these systems. Pappas et al. (1996) presented conflict resolution architecture for multi-agent hybrid systems with emphasis on air traffic management systems, which has the ability to detect conflict dynamically by sensory informa-

323

tion available to aircraft. Wagner et al. (1999) presented different solutions to inter-agent, intra-agent, and metalevel conflicts: conversations and negotiations are used to resolve inter-agent conflicts that occur during the exchange of information, constraints, and formation of commitments among agents. Therefore, it is quite possible that no solution can be reached among a group of noncooperative agents in that the agent might be better off acting alone if no benefit is obtained from the negotiation. Therefore, the effectiveness of such systems in resolving conflicts among non-cooperative agents is questionable. Even in collaborative MASs, since no single agent has accurate and complete global knowledge, it is inevitable that agents enter into conflicts over actions, plans, or resources (that they select; Muller and Dieng, 2000). In addition, the uncertainties associated with supply chain networks result in an inefficient manufacturing enterprise. This is principally due to its imprecise interfaces and its real-world character, where uncertainties in various activities right from raw material procurement to the end user may make the supply chain imprecise. The true nature of the problem involves data that are often vague and imprecise. For example, in the production scenario, various elements like set-up time, processing time, mean time between failure, mean time to repair, etc. will be better expressed as fuzzy variables, as they are often expressed in imprecise, vague terms like ‘Processing time is high’ or ‘Set-up time is low’. Therefore, real-world production planning, inventory control, and scheduling problems are usually imprecise. However, managers are to interact in an intelligent way in this environment. Thus, they have to reach out for a new kind of reasoning based on imprecise knowledge. In real-world situations there is variability in order type, quantity, or frequency. There are effects due to incentives, lack of information, capacity constraints, demand forecasting errors, uncertainties in information flows, transportation scale economies, set-up and ordering costs. Henceforth, it is essential to be able to cope up dynamically with variations in the environment, i.e. inventory, demand, supply, spatial, temporal, monetary constraints, to overcome both costly interruption stock-out as well as over- stock problems. Thus, it can be seen that the dynamic supply chain network problems are characterized by their complexity and inherent decentralization. The application of fuzzy logic and MAS techniques to this problem seems appropriate.

4. Proposed fuzzy hybrid negotiation mechanism In this section, we discuss in detail the proposed fuzzy hybrid negotiation mechanism. First, we define the following notations for the fuzzy negotiation framework: VATITSEL value of attribute I at time T for the Seller agent. Its acceptable range is [VATIT(Low)SEL, VATIT(High)SEL] VATITBUY value of attribute I at time T for the Buyer agent. Its acceptable range is [VATIT(Low)BUY, VATIT(High)BUY] range of negotiation issues. It is given as RANI RANI ¼ VATI1SELVATI1BUY as shown in Fig. 1.

ARTICLE IN PRESS 324

V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

CRANITSEL appropriation range of attribute I at time T. It is given as CRANITBUY ¼ VATITSELVATI1SEL for the buyer and CRANITSEL ¼ VATITBUYVATI1BUY for the seller. weight associated with the attribute I, 0oWWTI TIo1 and sum of weights ¼ 1. The weight associated with seller agent and buyer agent is denoted as WTISEL and WTIBUY respectively. acceptable range for attribute I. For buyer, ARANI ARANIBUY ¼ VATI(High)BUYVATI(Low)BUY and for seller it is given as ARANISEL ¼ VATI(High)SELVATI(Low)SEL as shown in Fig. 2. SCRANISELscore gained from appropriation ratio of attributes I for the seller SPRANIBUY score gained from preference ratio of attributes I for the buyer The proposed negotiation mechanism uses two features, appropriation degree and preference degree, as a reference to find a new offer to the buyer. We tap the properties of fuzzy logic to develop two membership functions for these two features. Based on their fuzzy values, we allot fuzzy rules for the new offer to the buyers and sellers.

4.1. Calculation of appropriation degree During a negotiation process, if the appropriation of an attribute from the buyer is high, we would respond with a Buyer’s initial offer VATI 1BUY

Seller’s initial offer VATI1SEL

Buyer’s current offer VAT I 1BUY

better value for the issue to the buyer. The step-by-step procedure for calculating the appropriation degree is as follows: Step 1: Find the value of every attribute, from the present offer of the buyers and the sellers. Step 2: Calculate the range of appropriation CRANITSEL. Step 3: Calculate the appropriation ratio ACRANIT ¼ CRANSTSEL/RANS. From the appropriation ratio, find its corresponding score SCRANISEL as shown in Table 1 Step 4: Compute the total appropriation degree TCDTSEL for seller from the buyer’s offer using the formula P TCDTSEL ¼ IWTISELSCRANITSEL 4.2. Calculation of preference degree Each attribute has its acceptable range, from which we can divide the range into different proportions to show different preference degrees to opponent’s offer. The procedure for calculating the preference degree is as follows: Step 1: Every seller and buyer can find his/her acceptable range ARANI for each attributes like cost, quality, lead time, flexibility, etc. Step 2: From Step 1, seller and buyer can divide the acceptable range into different preference areas to show various preferences. Step 3: We can calculate the preference ratio PRANITSEL ¼ ((VATIT(High)SELVATITBUY)/ARANISEL) and PRANITBUY ¼ ((VATITSELVATIT(Low)BUY)/ARANIBUY) for seller and buyer, respectively. Step 4: From Step 3, we can find the corresponding score SPRANIBUY Step 5: Calculate the total preference degree TPDTSEL for seller from the buyer’s offer using the equation P TPDTSEL ¼ ¼ IWTISELSPRANITSEL 4.3. Calculations of counter offers

Range of negotiation RANI Fig. 1. Concept of appropriation (from seller’s viewpoint).

Seller’s lower limit VAT ISEL (Low)

Buyer’s current offer BUY VAT IT

Seller’s higher limit SEL VAT I (High)

Acceptable Range ARANI Fig. 2. Concept of Preference (from seller’s viewpoint).

For the above two parameters, appropriation degree and preference degree, we define two fuzzy membership functions. Each function has five values: very low (VL), low (L), medium (M), high (H), and very high (VH). Triangular membership functions are used to represent these two functions as shown in Figs. 3 and 4, respectively, and their universe of discourse for each fuzzy value is shown in Table 1. The step-by-step procedure for finding a counter-offer is as follows: Step 1: We can find the membership function values for each TCD and TPD, that is, m1(TCD), m2(TCD), m1(TPD), and

Table 1 Triangular fuzzy membership function.

Very low Low Medium High Very high

Very low

Low

Medium

High

Very high

Left (m ¼ 0)

Center (m ¼ 1)

Right (m41)

L,L,L L,M,H M,L,M M,H,L H,L,H

L,L,M L,H,L M,L,H M,H,M H,M,L

L,L,H L,H,M M,M,L M,H,H H,M,M

L,M,L L,H,H M,M,M H,L,L H,M,H

L,M,M M,L,L M,M,H H,L,M H,H,L

– 0.4 4.0 6.5 7.7

0.30 3.8 6.0 8.5 10.7

3.2 5.4 8.0 10.1 –

ARTICLE IN PRESS V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

µVL(TCD)

µL(TCD)

µM(TCD)

Low

Medium

µH(TCD)

µVH(TCD)

1 µ

Very Low

Very High

High

0 0 1

2

3

4

5

6

7

8

10

9

325

dynamically update their neighbor’s utility value using the formula VUa,b(t) ¼ (1b)VUa,b(t1)+b(Ca(t)/(Z+Ca(t))), where b is the learning rate (0rbr1), which determines the dependency degree of agent b on neighbor a’s current CPU allocation Ca(t) and Z is the balance parameter to get a percentage value for the computation of VUa,b(t). The flowchart of the proposed fuzzy negotiation mechanism is shown in Fig. 5.

Fig. 3. Fuzzy membership function for appropriation degree.

5. An illustrative numerical example µVL(TPD)

µL(TPD)

µM(TPD)

µH(TPD)

µVH(TPD)

µ

1 Very Low

Low

Medium

Very High

High

0 0

1

2

3

4

6

5

7

8

9

10

Fig. 4. Fuzzy membership function for preference degree.

m2(TPD). Each of these values can be either one of the five values: VL, L, M, H, and VH. Step 2: Calculate the fuzzy firing strength. The joint membership function between the two membership functions can be expressed as JMFV ¼ min(mm(TCD), mn(TPD)), where m, n ¼ 1 or 2, and n ¼ 1–4. Step 3: Each JMFv value corresponds to a fuzzy rule, as shown in Table 1, where each capital letters (L, M, and H) in each column stands for three different rules for three attributes. Step 4: A new value of attribute I for the counter-offer can be derived from the equation P VAT IT ¼ VAT It1  ARAN I

JMFv v P

RLv

v JMFv

where  is + for the buyer and  for the seller. Because of each agent’s incomplete view of the uncertain environment, our coordination strategy is not to obtain an optimal solution, but a ‘good enough, soon enough’ one. To increase the chance of success, we install learning mechanisms to our agents so that each agent can learn to coordinate better. Agents’ learning mechanisms consist of reinforcement learning (RL) and case-based learning (CBL). RL is helpful to decide with whom to coordinate while CBL is helpful to decide how to negotiate. To facilitate the two learning mechanisms, each agent dynamically profiles (1) each negotiation task—as a case in the case base, and (2) each neighbor—as a vector in the agent. In our research domain, agents also learn from the past negotiation results to improve coordination in the future since they work in a dynamic supply chain network. In such an environment, the optimal coordination is not guaranteed. To achieve the necessary degree of flexibility in coordination, an agent is required to dynamically make decisions on when to coordinate, with whom to coordinate, and how to coordinate. A case records the problem description of a task, its solution, its outcome, and its usage history. The agents

To exhibit the operation of our proposed negotiation mechanism and to show its practicality, we have generated a hypothetical example for evaluating suppliers. In this example, there are two suppliers S1 and S2 with attributes cost ($/item), lead time (days), quantity (items), reward, restrictions, and profit associated with them. Such a scenario is typical of semi-competitive environment. That is, both supplier agent and buyer agent attempt to acquire the best deal they can since they both are selfinterested. Towards this end, they should minimize the revelation of their private information since it could prevent them from getting good deals. However, as the seller desires to build or preserve his reputation (this should associate with more money in long term) and the buyer needs to settle down as soon as possible, it is also essential for them to cooperate to a certain extent in the negotiation. Table 2 shows the seller product model (i.e. the information about the available suppliers prepared by the seller agent). We assume the following ranges for attributes: Cost (in $): MIN ¼ OPT ¼ 12; MAX ¼ 67. Lead time (in days): OPT ¼ 2, 3, 4, 5, 6, 7, 8 MIN ¼ 1; MAX ¼ 20. Quantity (in items): MAX: OPT ¼ 127 MIN: 1 By varying the attributes and balancing with cost, we randomly generate 10 combinations as shown in Table 3. For different configuration of suppliers, the proposed fuzzy negotiation mechanism will represent trade-offs among the different probable values of the negotiation issues. The table with generated fuzzy rule is shown in Table 3. From the proposed negotiation procedure, we know that supplier configurations such as 1(S1), 2 (S1), 3 (S1), 4 (S1), 5(S1), 1(S2), 2(S2), 3(S3), and 4(S2) are not acceptable to the buyer agent, but configuration 5(S2) is acceptable. Therefore, from both the seller’s and buyer’s perspective 10 is the best solution. Our agents are autonomous in that they negotiate with each other on behalf of both sellers and buyers. Thus, they actually make the contract decision themselves. In this mechanism, the user’s requirements on attributes of a product/service that fuzzy constraints can easily capture are represented. The problem can be extended to m suppliers with n attributes and the proposed mechanism can be carried out over fuzzy constraints of multiple issues of products, which is more efficient than doing it over single solutions.

ARTICLE IN PRESS 326

V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

Start

Bid announcement with negotiable tangible and intangible attributes like cost, quality, lead-time etc

Evaluation of Bid based on both tangible and intangible attributes

Submit Bid

Bid submission

Start Negotiation

Calculate Preference degree

Calculate Appropriation degree

Calculate Membership value

Find negotiation strategy & fuzzy rule

Calculate New Counter offers Learned Pattern

Dynamic update

Yes

Learning

No

Training examples

No

Yes

Offers

Deal is materialized End End

Fig. 5. Flowchart for the proposed fuzzy hybrid negotiation mechanism.

Table 2 Product information held by supplier agent. Suppliers

Lead time

Quantity

Cost

Quality

Appropriation

S1

1–8 9–14 15–20

1–28 1.83 1–128

12–67 12–65 12–62

Low Medium High

2 units free No No

S2

1–6 7–12 13–20

1–13 1–63 1–128

12–67 12–64 12–63

Medium Low High

No 2 units free No

Since human negotiators are unwilling to disclose private information, decentralized methods for searching Paretooptimal solutions in negotiation problems are necessary. The proposed mechanism guarantees that the outcome of

the negotiation is Pareto optimal. Thus, it can be seen that the proposed mechanism not only helps sellers and buyers to explore various new choices and opportunities that the e-markets offer but also allows them to identify and analyze their resource constraints in a given schedule, and helps them to explore and exploit many alternatives for a better solution.

6. Conclusion and perspectives Designing efficient business processes throughout the supply chain, and controlling their speed, timing, and interaction with one another are decisive factors in a competitive and dynamic environment. Coordination, the process by which agents reason about their local actions and the actions of others to try to ensure that the

ARTICLE IN PRESS V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

327

Table 3 The negotiation round for several combinations of numerical example. Combination of suppliers

Lead time

Quantity

Cost

Quality

Preference ratio (PRANIT)

Score

Appropriation ratio (ACRANIT)

Score

Fuzzy rule

1(S1) 2(S1) 3(S1) 4(S1) 5(S1) 1(S2) 2(S2) 3(S2) 4(S2) 5(S2)

04 10 08 05 08 13 07 08 17 04

77 62 127 122 92 122 110 92 127 122

12 17 15 18 12 13 12 17 15 12

Low Medium Medium High Low Low Medium Medium Medium High

PRANITo1/3 1/3rPRANITo2/3 1/3rPRANITo2/3 1rPRANIT 1/3rPRANITo2/3 PRANITo1/3 2/3rPRANITo1/3 1/3rPRANITo2/3 2/3rPRANITo1/3 1rPRANIT

0.20 0.40 0.40 0.80 0.40 0.20 0.80 0.60 0.80 1.00

ACRANITo0.25 0.25rACRANITo0.40 0.40rACRANITo0.55 0.70rACRANITo0.85 ACRANITo 0.25 0.25rACRANITo0.40 0.55rACRANITo0.70 0.40rACRANITo0.55 0.55rACRANITo0.70 0.85rACRANITo1.00

0.10 0.30 0.30 0.90 0.10 0.30 0.50 0.30 0.50 1.00

VL M M H VL L L M H VH

community acts in a coherent manner is an important issue in MASs. There are three main reasons why it is necessary for agents to coordinate. First, there are dependencies between agents’ tasks or goals; second, there is a need to meet global constraints such as cost and time limits; and third, no individual agent has sufficient competence, resources, or information to solve the entire problem. Achieving effective coordination in an MAS is nontrivial as no agent possesses the global view of the problem space. Moreover, the different strategies adopted by agents may produce conflicts. In order to model coordination among suppliers and buyers in a dynamic supply chain, this paper takes a step further and proposes a new fuzzy-logic-based hybrid negotiation mechanism. In most real-world negotiation situations, agents have a common interest to cooperate, but have conflicting interests over exactly how to cooperate. These situations involve restrictions and preferences that may be vaguely and partly defined. Therefore, this study takes the advantage of fuzzy logic and develops a hybrid negotiation-based mechanism that combines both cooperative and competitive negotiations. While agents coordinate with each other in the operations, they will negotiate about their strategies to reduce conflicts. The proposed fuzzy hybrid negotiation mechanism allows negotiation agents more flexibility and robustness in an automated negotiation system. The proposed mechanism not only helps sellers and buyers to explore various new choices and opportunities that the e-markets offer but also allows them to identify and analyze their resource constraints in a given schedule, and helps them to explore and exploit many alternatives for a better solution. In our research domain, agents also learn from the past negotiation results to improve coordination in the future since they work in a dynamic and noisy environment. In such an environment, the optimal coordination is not guaranteed. To achieve the necessary degree of flexibility in coordination, an agent is required to dynamically make decisions on when to coordinate, with whom to coordinate, and how to coordinate. The proposed computational framework based on fuzzy constraints is suited for capturing the dynamics by modeling trade-offs between different attributes of a product, leading to a fair and equitable deal for both suppliers and buyers.

The proposed fuzzy negotiation mechanism is generic and can be used for wide range of domains, especially in negotiations pertaining to supply contracts for flexible production networks. The model ensures a high degree of flexibility; it avoids deadlocks and encourages the parties’ willingness to a compromise. It guarantees that the outcome of the negotiation is Pareto optimal, yet the participating agents reveal minimal information about their preferences and constraints. Efficacy and intricacy of the proposed model are demonstrated with the help of numerical examples. In future, the concept of game theory can be employed to deal with the insight of agent behavior for effective portrayal of the characteristics of the agents, especially in the emerging dynamic B2B eCommerce environment.

References Chan, F.T.S., Chan, H.K., 2004. A new model for manufacturing supply chain networks: a multiagent approach. Journal of Engineering Manufacture 218 (B), 443–454. Chan, F.T.S., Chung, S.H., Wadhwa, S., 2004. A heuristic methodology for order distribution in a demand driven collaborative supply chain. International Journal of Production Research 42 (1), 1–19. Choi, T.Y., Dooley, K.J., Rungtusanatham, M., 2001. Supply networks and complex adaptive systems: control versus emergence. Journal of Operations Management 19 (3), 351–366. Chopra, S., Meindl, P., 2007. Supply Chain Management Strategy, Planning, and Operations, third ed. Prentice-Hall, USA. Durfee, E.H., Lesser, V.R., Corkill, D.D., 1989. Trends in cooperative distributed problem solving. IEEE Transactions on Knowledge and Data Engineering 1 (1), 63–83. Emerson, D., Piramuthu, S., 2004. Agent based framework for dynamic supply chain configuration. In: Proceedings of the 37th Hawaii International Conference on System Sciences, HICSS-37, pp. 1–8. Ertogral, K., Wu, S.D., 2000. Auction-theoretic coordination of production planning in the supply chain. IIE Transactions 32 (10), 931–940. Faratin, P., Sierra, C., Jennings, N.R., 2002. Using similarity criteria to make issue tradeoffs in automated negotiations. Artificial Intelligence 142 (2), 205–237. Fox, M.S., Barbuceanu, M., Teigen, R., 2000. Agent-oriented supply-chain management. International Journal of Flexible Manufacturing System 12, 165–188. Frayret, J.M., Sophie, D., Benoit, M., Louis, C., 2001. A network approach to operate agile manufacturing systems. International Journal of Production Economics 74, 239–259. Fung, R.Y.K., Chen, T., 2005. A multiagent supply chain planning and coordination architecture. International Journal of Advanced Manufacturing Technology 25, 811–819. Gan, X., Sethi, S.P., Yan, H., 2004. Coordination of a supply chain with riskaverse agents. Production and Operations Management 13 (2), 135–149.

ARTICLE IN PRESS 328

V. Jain, S.G. Deshmukh / Int. J. Production Economics 122 (2009) 319–328

Garcı´a-Flores, R., Wang, X.Z., Goltz, G.E., 2000. Agent-based information flow for process industries supply chain modeling. Computers and Industrial Engineering 24 (2–7), 1135–1141. Gjerdrumm, J., Shan, N., Papageorgiou, L.G., 2001. A combined optimization and agent-based approach to supply chain modelling and performance assessment. Production Planning and Control 12 (1), 81–88. Goyal, S.K., Gopalakrishnan, M., 1996. Production lot-sizing model with insufficient production capacity. Production Planning and Control 7 (2), 222–224. Jain, V., 2006a. Hybrid approaches to model supplier related issues in a dynamic supply chain. Ph.D. Thesis, Mechanical Engineering Department, Indian Institute of Technology Delhi, India, unpublished. Jain, V., Tiwari, M.K., Chan, F.T.S., 2004. Evaluation of supplier performance using an evolutionary fuzzy based approach. Journal of Manufacturing Technology and Management 15 (8), 735–744. Jain, V., Wadhwa, S., Deshmukh, S.G., 2005. e-Commerce and supply chains: modelling of dynamics through fuzzy enhanced high level petri net. Sadhana 30 (2–3), 403–429. Jain, V., Wadhwa, S., Deshmukh, S.G., 2006b. Modeling and analysis of supply chain dynamics: a high intelligent time petri net based approach. International Journal of Industrial and Systems Engineering 1 (1/2), 59–86. Jain, V., Wadhwa, S., Deshmukh, S.G., 2007a. A negotiation-to-coordinate (N2C) mechanism for modeling buyer–supplier relationship in dynamic environment. International Journal of Enterprise Information Systems 3 (2), 1. Jain, V., Wadhwa, S., Deshmukh, S.G., 2007b. Supplier selection using fuzzy association rules mining. International Journal of Production Research 45 (6), 1323–1353. Jennings, N.R., Sycara, K., Wooldridge, M., 1998. A roadmap of agent research and development. Autonomous Agents and Multi-Agents Systems 1, 7–38. Jennings, N.R., 2000. On agent-based software engineering. Artificial Intelligence 117 (2), 277–296. Jennings, N.R., Faratin, P., Lomuscio, A.R., Parsons, S., Sierra, C., Wooldridge, M., 2001. Automated negotiation: prospects, methods and challenges. International Journal of Group Decision and Negotiation 10 (2), 199–215. Julka, N., Srinivasan, R., Karimi, I., 2002. Agent-based supply chain management a framework. Computers and Industrial Engineering 26 (12), 1755–1769. Kraus, S., Sycara, A.K., Evenchik, A., 1998. Reaching agreements through argumentation: a logical model and implementation. Artificial Intelligence 104 (1–2), 1–69. Lo, W-S., Hong, T-P., Jeng, R., 2008. A framework of E-SCM multi-agent systems in the fashion industry. International Journal of Production Economics 114, 594–614. Mu¨ller, H., Dieng, R., 2000. Computational Conflicts. Springer, New York. Mundhe, M., Sen, S., 2000. Evaluation concurrent reinforcement learners. In: Proceedings of the Fourth International Conference on Multiagent Systems, pp. 421–422. Nissen, M.E., 2001. Agent-based supply chain integration. Information Technology and Management 2, 289–312. Papazoglou, M.P., 2001. Agent-oriented technology in support of ebusiness. Communications of the ACM 44 (4), 71–78. Pappas, G.J., Tomlin, C., Sastry, S.S., 1996. Conflict resolution for multiagent hybrid systems. Decision and Control 2, 1184–1189. Parunak, H., Dyke, V., Savit, R., Riolo, R.L., 1998. Agent based modeling vs. equation-based modeling: a case study and user guide. In: Proceed-

ings of the Multi-Agent Systems and Agent-Based Simulation Conference, vol. 9, pp. 10–25. Petrovic, D., Roy, R., Petrovic, R., 1999. Supply chain modeling using fuzzy sets. International Journal of Production Economics 59 (3), 443–453. Pruitt, D., 1981. Negotiation Behavior. Academic Press, New York. Reaidy, J., Massotte, P., Diep, D., 2006. Comparison of negotiation protocols in dynamic agent-based manufacturing systems. International Journal of Production Economics 99, 117–130. Sandholm, T., 1999. Automated negotiation. Communications of the ACM 42 (3), 84–85. Sophie, D., Benoit, M., Pierre, L., Francois, S., 1999. Networked manufacturing: the impact of information sharing. International Journal of Production Economics 58, 63–79. Sugeno, M., Yasukawa, T., 1993. A fuzzy logic based approach to qualitative modeling. IEEE Transactions on Fuzzy Systems 1 (1), 7–31. Sun, R., 1999. Sensitivity analysis of constraint-based factory scheduling. Ph.D. Thesis, The University of North Carolina at Charlotte, Charlotte, NC. Swaminathan, M.J., Smith, S.F., Sadeh, N.M., 1998. Modeling supply chain dynamics: a multiagent approach. Decision Sciences 29 (3), 607–632. Taylor, T., 2001. Channel coordination under price protection, midlife returns and end-of-life returns in dynamic markets. Management Science 47 (9), 1220–1234. Toledo Excelente, C.B., Jennings, N.R., 2002. Learning to select a coordination mechanism. In: Proceedings of AAMAS’02, Bologna, Italy, July 15–19, pp. 1106–1113. Tsay, A., 2002. Risk sensitivity in distribution channel partnerships: implications for manufacturer return policies. Journal of Retailing 78 (2), 147–160. Turksen, I.B., Zarandi, M.H., 1999. Production Planning and Scheduling: Fuzzy and Crisp Approaches. Kluwer Academic publishers, Boston, pp. 479–529. Wagner, T., Shapiro, J., Xuan, P., Lesser, V., 1999. Multi-Agent Systems, UMass Computer Science Technical Report. Wang, Y., Gerchak, Y., 2001. Supply chain coordination when demand is shelf-space dependent. Manufacturing and Service Operations Management 3 (1), 82–87. Wilding, R., 1998. The supply chain complexity triangle: uncertainty generation in the supply chain. International Journal of Physical Distribution and Logistics Management 28 (8), 599–616. Xuan, P., Lesser, V., Zilberstein, S., 2001. Communication decisions in multi-agent cooperation: model and experiments. In: Proceedings of the Fifth International Conference on Autonomous Agents (Agents2001), January, 2001, Montreal, Canada, pp. 616–623. Yan, Y., Yen, J., Bui, T., 2000. A multi-agent based negotiation support system for distributed transmission cost allocation. In: Proceedings of the 33rd Hawaii International Conference on system Sciences, HICSS-33. Yoo, S.B., Kim, Y., 2002. Web-based knowledge management for sharing product data in virtual enterprises. International Journal of Production Economics 75, 173–183. Yung, S., Yang, C., 1999. A new approach to solve supply chain management problem by integrating multi-agent technology and constraint network. In: Proceedings of the 32nd Hawaii International Conference on system Sciences, HICSS-32. Zadeh, L.A., 1973. Outline of a new approach to the analysis of complex systems and decision processes. IEEE Transactions on Systems, Man and Cybernetics 3 (1), 28–44.