Journal of Materials Processing Technology 107 (2000) 398±403
A blackboard-based negotiation for collaborative supply chain system Teruaki Ito*, Mohd Rizal Salleh Department of Mechanical Engineering, University of Tokushima, 2-1 Minami-Josanjima, Tokushima, 770 8506 Japan
Abstract Supply chain management system (SCM) is an integrated network of suppliers, factories, warehouses, distribution centers and retailers, through which the whole chain of logistic processes is managed so that it should aim at a faster and more ¯exible co-ordination between a company and its customers and suppliers within the chain. Collaboration among the members of SCM plays a critical role to implement an effective SCM, but its implementation would not be easy only with the conventional information sharing mechanism. This study employs the idea of open tender concept with blackboard-based negotiation to develop a collaborative supply chain system. The open tender concept is designed to provide an open and free opportunity to all participants, which make a sound competition to obtain the tender. Blackboardbased negotiation is derived from the mechanism of negotiation among intelligent agents using a blackboard as a media where each agent exchange information. The study applies the idea to a parts/material supply system, which is designed to function as a part of SCM. For replenishment of parts/materials, a company issues a tender and publishes it to a public blackboard for information exchange via the Internet using electronic tender submission form to ®nd appropriate suppliers. Candidate suppliers to the tender are invited to issue their quotations, which compete one another under an open environment and the most appropriate candidate is selected as a result of open competition. # 2000 Elsevier Science B.V. All rights reserved. Keywords: Supply chain system; Intelligent agents; Negotiation techniques; Collaborative supply chain system; Blackboard; Internet; Open tender
1. Introduction Rapid development of new products today increases the dif®culty of production for manufacturing companies. Dynamic markets may sometimes require the companies to restructure their function and position in industry. Under these circumstances, the companies need to take appropriate actions so that they could remain competitive in the global market situation [1]. Concurrent engineering (CE) is one of the approaches to be competitive in the global market. CE can be realized by simultaneous work of experts from various functions in an enterprise, concerned with the production of speci®c products, in order to achieve high quality, functionality and manufacturability of products in the shortest time with the minimum cost. The idea of supply chain management system (SCM) also attracts their attention as an effective approach to comply with these dif®cult situations. SCM is an integrated network of suppliers, factories, warehouses, distribution centers and retailers, through which the whole chain of logistic processes is managed so that it should aim at a faster and more ¯exible co-ordination between a company and its customers and *
Corresponding author. E-mail addresses:
[email protected],
[email protected] (T. Ito).
suppliers within the chain. Collaboration among the members of SCM plays a critical role to implement an effective SCM, but its implementation would not be easy only with the conventional information sharing mechanism. This study employs the idea of open tender concept with blackboard-based negotiation to develop a collaborative supply chain system (CSCS). Fig. 1 shows an illustrative view of open tender concept with blackboard-based negotiation. The open tender concept is designed to provide an open and free opportunity to all participants, which make a sound competition to obtain a tender. Blackboard-based negotiation is based on the mechanism of negotiation among intelligent agents (IAs) using a blackboard as a media where each agent exchange information for collaboration. CSCS is designed to enhance the supply speed and certainty through all relevant key members of SCS. The key members communicate and interact each other in order to generate the solution for ef®cient material supply process. Although the paper mainly focuses on a parts/materials supply system in SCM, ultimate objective of the study is to integrate all of these techniques and systems including agents, negotiation techniques and CSCS so that realization of CE should be achieved (Fig. 2). This paper describes the roles of IAs in CSCS, and their interaction which works for the success of communication and negotiation process between the manufacturers and the
0924-0136/00/$ ± see front matter # 2000 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 4 - 0 1 3 6 ( 0 0 ) 0 0 7 3 0 - 5
T. Ito, M.R. Salleh / Journal of Materials Processing Technology 107 (2000) 398±403
Fig. 1. Open tender process between manufacturers and suppliers.
suppliers. Then, the paper addresses IAs negotiation life cycle, describes the collaboration and negotiation process in CSCS and presents an example of open trading for parts/ materials replenishment through prototype system. 2. Roles of agents in CSCS Agent technologies are increasingly used in a wide range of industrial and commercial domains. This is because of the reason that IAs have a high degree of self-determination capabilities, and they can decide for themselves when, where and under what condition their action should be performed. Interactions among IAs may even make them work towards a particular course of action, modify a planned course of action and come to an agreement on a common course of action. Furthermore, collaboration among agents is used for streamlining and integrating the entire process of CSCS. Several kinds of IAs play an important role in CSCS for smooth communication and collaboration within the key
Fig. 2. Contributors to realize CE.
399
members of SCS. IA identi®es the critical information from incoming resources, monitors the information and trigger actions based on the contents of information. According to the contents, IA executes collective tasks on behalf of users as an autonomous process. The application of IA in CSCS strongly promotes co-operation among the members of supply chain. Interactions among IAs, however, may sometimes lead to con¯icts and coalitions among them. Negotiation techniques are used to overcome these con¯icts and coalitions, and to come to an agreement among agents, instead of persuading them to accept a ready solution. In the negotiation techniques, operation for negotiation is viewed as actual interactions among participants that lead to mutual commitment [2]. IAs bring their goals to a bargaining table, share the information, and seek for the goals' alternatives. IAs assist manufacturers to speed up their product ¯ows along the supply route, and to ensure that the products should be available in the markets at the required time. The function of IAs in CSCS is to simulate members of supply team and to perform the chores according to users' requirements. As shown in Fig. 3, each agent interacts and shares information through a blackboard (BB). BB regulates and provides the information to each agent, which works for the key members of CSCS. For inventory control application, an inventory control agent works to control the inventory levels and negotiates with the other agents for the smooth materials ¯ow through information exchange. For internal information exchange, the Intranet can be used to provide internal communication services to achieve better results than the conventional means of data access and transfer. Its application also allows the users to simply access for day to day information, and the company can easily screen the critical information from external users. In the mean time, the Internet allows the users to reach each other on a global basis for the information exchanges. The Internet is becoming used in the various industrial ®elds for improving manufacturing processes [4]. For example,
Fig. 3. Agents interaction in CSCS.
400
T. Ito, M.R. Salleh / Journal of Materials Processing Technology 107 (2000) 398±403
the Internet can be used to offer such an opportunity to compare a number of suppliers, to choose a suitable supplier and to ensure the selected supplier should satisfy requirements. The roles of IAs are essential for the success of communication and negotiation processes between a company and its suppliers so as to gain the information required for their desired goals. IAs support the open tender procedure for the group problem solving through the negotiation and collaboration techniques, increase the network communication for the users' satisfactions and enable both the manufactures and the suppliers to communicate effectively. 3. IAs negotiation life cycle Negotiation is essential to reach an agreement about the provision of service by one agent to another, and this approach has been used for managing the con¯ict which arises among the IAs. IAs start negotiation when they begin communicating their goals, and ®nish when all of them agree to a speci®ed decision. IAs employ a range of strategies and tactics to generate initial offers, evaluate proposals and offer counter proposals. Negotiation uses the speci®c set of techniques for the problem solving to derive agreements among IAs, instead of persuading them to accept a ready solution. Negotiation techniques assist group problem solving in various domains [5]. As shown in Fig. 4, the negotiation life cycle participants are composed of (1) analyst and owner: to describe and formalize the goals, (2) designer: plan for achieving the goals through interactions among the agents, and (3) implementor: to implement the designed strategy using the appropriate protocols and tools for the goals [3]. In the ®rst step of negotiation life cycle, negotiation analysis needs to procure and model individual preferences for
Fig. 4. The negotiation life cycle phase and negotiation participants.
the negotiation protocols and strategies. In the second step, negotiation design seeks to de®ne interaction protocols and strategies for the goals' requirements. In the ®nal step, negotiation implementation seeks to reach group commitment through communication exchanges. IAs work on behalf of participants, and they are equipped with a special skill to manipulate the other members according to their task's function. IAs use the plan and particular negotiation protocol by which they intend to interact with the other agents. When IAs seek for negotiation, they disclose their goals to the other members, identify the con¯icts and search the con¯ict resolution, and select the ideal resolution for their goals. Negotiation techniques assist CSCS for the realization of effective materials ¯ow. 4. Collaboration and negotiation in CSCS Companies are required to comply with customer orders even if they may be hard to do so. Companies have to respond to the orders quickly and ef®ciently in the limited time available to ful®ll the customer's requirements. Unexpected rush orders, however, may come sometimes, which causes the delay of delivery and decreases ef®ciency in all of the supporting members. The collaboration of CSCS members and the introduction of negotiation techniques provide a solution to those problems [6]. Collaboration and negotiation works to ®nd an appropriate supplying company in open tender trading. For parts/ materials replenishment, a company publishes a tender to BB, which can be reviewed by the entire aviation community of suppliers. This is the start of open tender competition. Supplier agents (SAs) for each supplying company scour the Internet to ®nd any appropriate tenders. When an agent ®nd one, it will urge its host company to submit a quotation. BB publishes quotations submitted by candidate supplying companies, which may urge revision of quotations, or negotiation among these companies. BB closes the tender when the most appropriate quotation is selected based on selection criteria. IAs work for communication and negotiation processes between a company and its supplier to achieve their mutual goal. Collaboration and negotiation in CSCS also works to ®nd a solution for supplier replacement processes. For example, when a delivery problem happens in one of the suppliers, IAs work as a catalyst of collaboration and negotiation for effective supply chain, and ®nd an alternative suppliers to replace the supplier in problem. IAs select the alternative supplier according to the selection criteria prepared by the manufacturer. Fig. 5 shows how this kind of activities take place when the selected supplier fails to deliver the materials at the right time. In this activity, the supplier in problem, the manufacturer and the alternative supplier are controlled by supplier agent 1 (SA1), manufacturer agent (MA) and supplier agent 2 (SA2), respectively. When the problem
T. Ito, M.R. Salleh / Journal of Materials Processing Technology 107 (2000) 398±403
401
Fig. 5. Supplier replacement process by collaboration and negotiation. Fig. 6. IAs interaction in open tender process.
happens, SA1 sends a delay message to MA. After receiving the delay message, MA refers to BB to ®nd an alternative supplier for the replacement. If the supplier 2 is regarded as the alternative one, MA makes an inquiry to SA2 and asks for the delivery situation. Then SA2 quotes the cost of materials and time for the delivery and sends them back to MA. If SA2 does not accept the offer, MA takes the same procedure to ®nd another alternative supplier. The collaboration and negotiation activities among these agents assist the manufacturer and the suppliers to solve the problems during the delivery process. 5. An example of open trading for parts replanishment in CSCS This section presents an example of open trading for parts replenishment in CSCS. Toyo Manufacturing Industries is a car stereo company, and is looking for a supplier of screw for car stereo top plate. The company has some business relations with several supplying companies because of business transactions in the past. The company, however, wants to ®nd the most appropriate supplier in each occasion in a dynamic manner using collaboration of agents, which include inventory stock control agent (ISCA), MA, SA, supplier stock control agent (SSCA), and blackboard control agent (BBCA), as shown in Fig. 6. The company sets up the minimum inventory level so that the agent can control the inventory. When the inventory stock reaches to the level, an ISCA sends a signal to Purchasing Department so that they can take an appropriate action to place an order to suppliers. When it happens, an MA urges Purchasing Department to ®ll out an order request form (ORF) including several key items, such as part name, part number, selection criteria, delivery point, order record, etc. as shown in Fig. 7(1). MA veri®es ORF and submits it to BB as a tender. When a new tender is published, supplying companies are noti®ed by each SA working for each supplier so that they can consider their supply positions. Details of the tender are available on BB, so that not only the issuer of
tender but also supplying companies can refer to the details if necessary. If a supplying company wants to obtain the tender, Sales Department of the company checks availability of the parts through SSCA. If it is available, then an SA urges the department to ®ll out a quotation form (QF) as shown in Fig. 7(3). SA veri®es the QF and submits it to BB, which means that the company shows an intention to join a competition for the tender. In this way, list of supplying companies to the tender and details of each quotation are open through BB so that every participant can understand what is going on regarding the tender until the ®nal stage of open competition and, of course, which company obtains the tender in the end. Updated information is fed back to each company either by MA or SA. BB in Fig. 7(2) shows that Kurosawa Manufacturing, Bersekutu Light Industries, International Materials Supply, etc. are among the supplier list. At the end of open competition, BBCA ranks up these quotations based on the selection criteria, and recommends Kurosawa Manufacturing Industries as the most appropriate supplier to the tender. Even though the other companies may have lost the competition this time, they know which company is the winner and what kind of quotation the winner has submitted. They also know why they have lost the competition, so that they may obtain some hints for an effective counter measure to be more competitive in the next opportunity. 6. Concluding remarks The paper presented the idea of open tender concept with BB-based negotiation to implement a CSCS in order to achieve an ef®cient material ¯ow and to shorten the production lead-time. IAs carry the tasks through the support of collaboration and negotiation techniques for co-ordinating interactions, generating resolutions, and deriving agreements to ful®ll the users' requirements. Through the support of IAs and negotiation, CSCS enables a company and its
402
T. Ito, M.R. Salleh / Journal of Materials Processing Technology 107 (2000) 398±403
Fig. 7. Sample of electronic forms for manufacturer and supplier usage in the prototype system.
supplier to utilize the full framework of potential communications. There are various kinds of data that may be exchanged to improve the integration of trading partners' operations and to reduce supply uncertainty, inventory and cost [7]. BB plays an important role to maximize the function of all the partners in the joint management of CSCS by providing the information for open competition. IAs can exchange messages and co-ordinate their activities such as con¯icts resolution, effective planning, execution control, negotiation strategies, etc. The bene®ts of CSCS can be summarized as follows: (1) to increase the chance to ®nd the most appropriate supplier by open competition among the suppliers, (2) to increase the ef®ciency of transaction in the
supply chain, and (3) to increase the productivity and customer satisfaction. The future work includes the development of knowledge based subsystem that works on decision making process to ®nd the appropriate supplier based on the selection criteria. For the realization of this subsystem, application of decision analysis tools such as MAUT or ISMAUT is under consideration.
References [1] T. McGuffog, Managing the supply chain with speed and certainty, 1996. http://www.ana.org.uk/ana/.
T. Ito, M.R. Salleh / Journal of Materials Processing Technology 107 (2000) 398±403 [2] C. Beam, A. Segev, Automated negotiations: a survey of the state of the art, CMIT Working Paper 97-WP-1022, 1997. http://haas.berkeley.edu/edu/citm/wp-1022-summary.html. [3] W.N. Robinson, et al., Supporting the negotiation life cycle, Commun. ACM 41 (5) (1998) 95±102. [4] H. Robert Frost, et al., Design for manufacturability via agent interaction, ASME Design for Manufacturing Conference, 96-DETC/ DFM-1302, Irvine, CA, 1996.
403
[5] S. Franklin, A. Graesser, Is it an agent, or just a program? in: Proceedings of the Third International Workshop on Agent Theories, Architectures, and Languages, Springer, Berlin, 1996. http:// www.msci.memphis.edu/franklin/AgentProg.html. [6] E. Van Heck, P. Vervest, How should CIOs deal with web-based auctions? Commun. ACM 41 (7) (1998) 99±100. [7] J.M. Tenenbaum, WISs and electronic commerce, Commun. ACM 41 (7) (1998) 89±90.