Upcoming SlideShare
Loading in...5




IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit

IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit



Total Views
Views on SlideShare
Embed Views



0 Embeds 0

No embeds



Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
Post Comment
Edit your comment

Gs3112881292 Gs3112881292 Document Transcript

  • Surjeet Dalal, Gundeep Tanwar, Naveen Alhawat / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 Vol. 3, Issue 1, January -February 2013, pp.1288-1292 Designing CBR-BDI Agent for implementing Supply Chain system Surjeet Dalal*, Gundeep Tanwar**, Naveen Alhawat***, *(Research Scholar, Suresh Gyan Vihar University, Jaipur Rajasthan-302025) **(Associate Professor, BRCM CET Bahal, Bhiwani Haryana-127028) ***(Student, S.V.I.E.T, Banur Punjab-140601)ABSTRACT An agent is anything that can be viewed as The intelligent agent becomes perceiving its environment through sensors andsupplementary the budding technology of acting upon that environment through effectors. AArtificial Intelligence that is being used for human agent has eyes, ears, and other organs forscheming the enterprise applications. AI is sensors, and hands, legs, mouth, and other bodyconcerned in studying the mechanism of parts for effectors. A software agent has encoded bitintelligence (e.g., the ability to learn, plan) while strings as its percepts and actions. A generic agent isthe study of agents deals with integrating these diagrammed in Figure 1.same components. The reasoning capability &autonomous features of intelligent agents makesthe programmers to resolve the complexproblem. But the inability of learning from theenvironment is the main problem faced withintelligent agent. The case-based reasoning is themajor approach that enables to learn from theexisting environment. Hence the CBR-BDI agentare the variety of intelligent agents that utilize thefeatures of case-based reasoning to enable theintelligent agents to learn from the environment Figure 1: Intelligent Agent& make the decisions by using the existing Wooldridge and Jennings (1995) define ansolutions. intelligent agent as one that is capable of flexible In global market competitions, more autonomous action to meet its design objectives.companies know that without handling supply Flexible means:chain management they can survive for long time  Reactivity: intelligent agents perceive andperiod. Supply chain system consists of various respond in a timely fashion to changes thatentities, flows and relationships. In this paper, the occur in their environment in order to satisfyCBR-BDI agents are applied to imitate the their design objectives. The agent’s goals and/orentities of supply chain system. These agents assumptions that form the basis for a proceduremaintain the flows of material, products and that is currently executing may be affected by ainformation. This system helps managers to changed environment and a different set ofanalyze the business policies with respect to actions may be need to be performed.different situations arising in the supply chain.  Pro-activeness: reacting to an environment by mapping a stimulus into a set of responses is notKeywords - Intelligent Agent, Case-based enough. As we want intelligent agents to doReasoning, Supply chain, CBR-BDI agent things for us, goal directed behavior is needed. In a changed environment, intelligent agentsI. INTRODUCTION have to recognize opportunities and take the initiative if they are to produce meaningful Over the last two decades, various results. The challenge to the agent designer is to“intelligent technologies” analyses have significantly integrate effectively goal-directed and reactiveimpacted on the design and development of new behavior.decision support systems and expert systems in  Social ability: intelligent agents are capable ofdiverse disciplines such as engineering, science, interacting with other agents (and possiblymedicine, economics, social sciences and humans), through negotiation and/ormanagement. So far, however, barring a few cooperation, to satisfy their design objectives.noteworthy retailing applications reported in the Other properties sometimes mentioned in the contextacademic literature, the use of intelligent of intelligent agents include:technologies in retailing management practice is still  Mobility: the ability to move around anquite limited. electronic environment 1288 | P a g e
  • Surjeet Dalal, Gundeep Tanwar, Naveen Alhawat / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 Vol. 3, Issue 1, January -February 2013, pp.1288-1292  Veracity: an agent will not knowingly and intention are easy to understand. On the other communicate false information hand, its ma in drawback lies in finding a  Benevolence: agents do not have conflicting mechanism that permits its efficient implementation. goals and every agent will therefore always try Rao and Georgeff (1995) stated that the problem lies to do what is asked of it in the great distance between the powerful logic for  Rationality: an agent will act in order to BDI systems and practical systems. Another achieve its goals insofar as its beliefs permit problem is that this type of agent is not able to learn,  Learning/adaptation: the intelligent agents a necessary requirement for them since they have to improve performance over time [1]. be constantly adding, modifying or eliminating The architecture might be a plain computer, beliefs, desires and intentions. It would beor it might include special-purpose hardware for convenient to have a reasoning mechanism thatcertain tasks, such as processing camera images or would enable the agent to learn and adapt in realfiltering audio input. It might also include software time, while the computer program is executing,that provides a degree of insulation between the raw avoiding the need to recompile such an agentcomputer and the agent program, so that we can whenever the environment changes.program at a higher level. In general, the architecture In order to overcome these issues, wemakes the percepts from the sensors available to the propose the use of a case-based reasoning (CBR)program, runs the program, and feeds the program’s system for the development of deliberative agents.action choices to the effectors as they are generated. The proposed method facilitates the automation ofThe relationship among agents, architectures, and their construction. Implementing agents in the formprograms can be summed up as follows: of CBR systems also facilitates learning and agent = architecture + program adaptation, and therefore a greater degree of Before we design an agent program, we autonomy than with a pure BDI architecture. If themust have a pretty good idea of the possible percepts proper correspondence between the three mentaland actions, what goals or performance measure the attitudes of BDI agents and the informationagent is supposed to achieve, and what sort of manipulated by a CBR system is established, anenvironment it will operate in. There exists a variety agent with beliefs, desires, intentions and a learningof basic agent program designs, depending on the capacity will be obtained. Our approach to establishkind of information made explicit and used in the the relationship between agents and CBR systemsdecision process. The designs vary in efficiency, differs from other proposals, as we propose a directcompactness, and flexibility. The appropriate design mapping between the agent conceptualization and itsof the agent program depends on the percepts, implementation, in the form of a CBR system [3].actions, goals, and environment [2] The intentions are plans of actions that the agent has to carry out in order to achieve its objectives, so anII. CBR-BDI AGENTS intention is an ordered set of actions; each change Agents must be able to reply to events, from state to state is made after carrying out anwhich occur in their environment, take the initiative action (the agent remembers the action carried out inaccording to their goals, interact with other agents the past, when it was in a specified state, and the(even human), and to use past experiences to achieve subsequent result). A desire will be any of the finalcurrent goals. Several architectures have been states reached in the past (if the agent has to dealproposed for building deliberative agents, most of with a situation, which is similar to a past one, it willthem being based on the BDI model. In this model, try to achieve a similar result to the previouslyagents have mental attitudes of Beliefs, Desires and obtained one).Intentions. In addition, they have the capacity to Case: <Problem, Solution, Result>decide what to do and how to get it according to Problem: initial_statetheir attitudes. In BDI architecture, agent behavior is Solution: sequence of <action, [intermediate_state]>composed of beliefs, desires, and intentions. The Result: final_statebeliefs represent its information state, what the agent BDI agentknows about itself and its environment. The desires Belief: stateare its motivation state, what the agent is trying to Desire: set of <final_state>achieve. And the intentions represent the agent’s Intention: sequence of <action>deliberative states. Intentions are sequences of Based on this relationship, agentsactions; they can be identified as plans. (conceptual level) can be implemented using CBR These mental attitudes determine the systems (implementation level). This means, aagent’s behaviour and are critical in attaining proper mapping of agents into CBR systems. The advantageperformance when the information about the of this approach is that a problem can be easilyproblem is scarce. A BDI architecture has the conceptualized in terms of agents and thenadvantage that it is intuitive and relatively simple to implemented in the form of a CBR system. So onceidentify the process of decision-making and how to the beliefs, desires and intentions of an agent areperform it. Furthermore, the notions of belief, desire identified, they can be mapped into a CBR system. 1289 | P a g e
  • Surjeet Dalal, Gundeep Tanwar, Naveen Alhawat / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 Vol. 3, Issue 1, January -February 2013, pp.1288-1292III. RELATED WORK proposal. The Beliefs-Desires-Intentions (BDI) approach to design deliberative agents can be Cindy Olivia et al. (1999) described a improved with the learning capabilities of Case Baseframework at integrates case-based reasoning Reasoning (CBR) techniques [9].capabilities in a BDI agent architecture as well as its Javier Bajo et al. (2006) presented aapplication to the design of Web information deliberative agent that incorporates a hybrid systemretrieval agents. Their search proposed in this paper as a reasoning motor in the frame of CBR systems.generates two key insights. First, it shows that the To solve this problem the agent incorporates neuralintegration of case-based reasoning in BDI agent networks to implement the stages of the CBRarchitecture is a non-trivial exercise that suggests system. Our agent can acquire knowledge and adaptinteresting ways of building BDI agents with itself to environmental changes. The hybrid systemlearning capabilities. Second it demonstrates the has been applied for evaluating the interactionefficacy of the resulting framework by presenting between the atmosphere and the ocean. The systemthe design of intelligent Web information retrieval has been tested successfully, and the results obtainedagents that are effective in well-demarcated domain are presented in this paper [10]. Mart´ Navarro et al.[4]. Cindy Olivia et al. (2003) introduces a robust (2009) stated that In real-time Multi-Agent Systems,mathematical formalism for the definition of Real-Time Agents merge intelligent deliberativedeliberative agents implemented using a case-based techniques with real-time reactive actions in areasoning system. The concept behind deliberative distributed environment. CBR has been successfullyagents is introduced and the case-based reasoning applied in Multi-Agent Systems as deliberativemodel is described using this analytical formalism. mechanism for agents. However, in the case of Real-Variational calculus is introduced in this chapter to Time Multi-Agent Systems the temporal restrictionsfacilitate to the agents the planning and preplanning of their Real-Time Agents make their deliberationof their intentions in execution time, so they can process to be temporally bounded. Therefore, thisreact to environmental changes in real time [5]. J. M. paper presents a guide to temporally bound the CBRCorchado et al. (2003) shows how autonomous to adapt it to be used as deliberative mechanism foragents may be constructed with the help of case- Real-Time Agents [11].based reasoning systems. The advantages and Costin Bădică et al. (2011) provided andisadvantages of deliberative agents are discussed, overview of the rapidly developing area of softwareand it is shown how to solve some of their agents serving as a reference point to a large body ofinconvenient, especially those related to their literature and to present the key concepts of softwareimplementation and adaptation. It shows how agent technology, especially agent languages, toolsautonomous agents may be constructed with the help and platforms. Special attention is paid to significantof case-based reasoning systems. The advantages languages designed and developed in order toand disadvantages of deliberative agents are support implementation of agent-based systems anddiscussed, and it is shown how to solve some of their their applications in different domains. Afterwards, ainconvenient, especially those related to their number of useful and practically used tools andimplementation and adaptation [6]. Rosalia Laza et platforms available are presented, as well as supportal. (2003) showed how deliberative agents can be activities or phases of the process of agent-orientedbuilt by means of a case-based reasoning (CBR) software development [12].system. This system is used for providing advice tothe user and its functioning is guided by a human IV. ARCHITECTURE OF SUPPLY CHAINexpert. The most important features described are SYSTEMrelated with the implementation of retrieve andretain phases of the CBR cycle [7]. The object of SCM obviously is the supply Juan M. Corchado-Rodríguez et al. (2004) chain which represents a “network of organizationspresented a practical application of an agent-based that are involved, through upstream and downstreamarchitecture which has been developed using the linkages, in the different processes and activities thatmethodological framework defined by case-based produce value in the form of products and servicesreasoning systems. The deliberative agents in the hands of the ultimate consumer”. In a broaddeveloped within the framework of this research sense a supply chain consists of two or more legallyhave been used to construct a multi-agent separated organizations, being linked by material,architecture used in an industrial application. The information and financial flows. These organizationsdeveloped architecture is presented together with the may be firms producing parts, components and endresults obtained [8]. Juan M. Corchado-Rodríguez et products, logistic service providers and even theal. (2005) presented a model of an agent that (ultimate) customer himself. So, the above definitioncombines both BDI and CBR techniques. We of a supply chain also incorporates the target group –discuss the development of this kind of agent and the ultimate customer.present a case study. We use a real application of a In a narrow sense the term supply chain iswireless tourist guide system to illustrate the also applied to a large company with several sites often located in different countries. Coordinating 1290 | P a g e
  • Surjeet Dalal, Gundeep Tanwar, Naveen Alhawat / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 Vol. 3, Issue 1, January -February 2013, pp.1288-1292material, information and financial flows for such a  Increased cost pressures based on globalmultinational company in an efficient manner is still competition and shareholder demands to reducea formidable task. Decision-making, however, working capital;should be easier, since these sites are part of one  A new level of complexity brought on by morelarge organization with a single top management complicated distribution models, increasedlevel. A supply chain in the broad sense is also outsourcing, and “new technologies thatcalled an inter-organizational supply chain, while promise efficiency but can increasethe term intra-organizational relates to a supply complexity.”chain in the narrow sense. Irrespective of this While supply chains are getting more difficult todistinction, a close cooperation between the different manage, the competitive environment means thatfunctional units like marketing, production, most companies need to further reduce costs.procurement, logistics and finance is mandatory – aprerequisite being no matter of course in today’s V. IMPLEMENTING SUPPLY CHAINfirms. THROUGH CBR-BDI AGENT We propose a model of a learning agent whose interaction with the environment which allows the agent to project itself into future situations before it takes real action. A supply chain is a network of autonomous entities, or agents, engaged in procurement of raw materials, manufacturing and converting raw materials into finished Products and distribution of finished products. Distribution, manufacturing and purchasing organizations along the supply chain often operates independently and have their own objectives, which may be in conflict. The supply chain management (SCM) should ensure the objectives to deliver the right product, at the appropriate time, at the competitive cost, and with customer satisfaction in order to keep the competitive advantages. A perfect coordination Figure 2: Supply chain system among various functions always ensures success of The coordination of flows along the supply SCM to achieve its main objective. We identify thechain can be executed efficiently by utilizing the elements in a supply chain, their features, and thelatest developments in information andcommunication technology. These allow processes challenges associated with SCM. We classify the elements in a supply chain as entities and flows.formerly executed manually to be automated. Aboveall, activities at the interface of two entities can be Entities include all manufacturers, logisticsscrutinized, while duplicate activities (like keying in providers, electronic exchanges and all their internal departments that participate in the business process.the data of a consignment) can be reduced to a single These entities are essentially the operators in theactivity. Process orientation thus often incorporates supply chain. Flows are of three types: material,a redesign followed by a standardization of the new information and finance, and these are the operandsprocess. in the supply chain. These entities have three For executing customer orders, the common features:availability of materials, personnel, machinery andtools has to be planned. Although production and 1. Dynamic: The supply chains are more flexibledistribution planning as well as purchasing have now. In today’s business environment, there arebeen in use for several decades, these mostly have no obligations for companies to be part of abeen isolated and limited in scope. Coordinating supply chain for a certain time period and theyplans over several sites and several legally separated may join or leave based on their own interest.organizations represents a new challenge that is This changes the structure and flows in thetaken up by Advanced Planning (Systems). There supply chain. Information in the supply chainare three major factors that have dramatically e.g. prices, demands, technologies, etc. is alsoincreased the stress on supply chains: changing continuously.  Fragmenting customer needs, resulting in a 2. Distributed: The elements are distributed broader selection of SKUs (stock keeping units) across various geographical locations. The aimed at specific consumer segments, different planning and operating systems used by an price points, shorter product life-cycles, and less entity may also be geographically distributed predictable demand patterns; e.g. there may be a dedicated inventory database residing at each warehouse of a manufacturer. 1291 | P a g e
  • Surjeet Dalal, Gundeep Tanwar, Naveen Alhawat / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 Vol. 3, Issue 1, January -February 2013, pp.1288-1292 The SCM related information might even reside REFERENCES as rules-of-thumb with the people responsible for performing the various tasks in the business [1] Stuart Russell and Peter Norvig, Artificial process. Intelligence: A Modern Approach. 3. Disparate: The entities in a supply chain use Englewood Cliffs, NJ: Prentice Hall, 1995. different systems built on different platforms for [2] Martin L. Griss, Software Agents as Next planning and management of their business. Generation Software Components, Chapter Information pertaining to the various elements 36, Component-based software is also disparate in form. engineering, Addison-Wesley Longman Publishing Co., Inc. pp. 641-657. [3] J. M. Corchado, Agent-based Web Engineering, Proc. ICWE03 Proceedings of the 2003 international conference on Web engineering, 2003, 17-25 [4] Cindy Olivia, Chee-Fon Chang, Case- Based BDI Agents: An Effective Approach For Intelligent Search On the World Wide Web Proc. AAAI - National Conference on Artificial Intelligence, 1999, 20-27. [5] M. G. Bedia and J. M. Corchado, Constructing autonomous distributed systems using CBR-BDI agents, Innovations in Knowledge Engineering, Chapter 11, (Advanced Knowledge International, Magill, S. Aust, 2003). [6] J. M. Corchado, R. Laza, Constructing Deliberative Agents with Case-based Reasoning Technology, International Journal of Intelligent Systems , 18(12), 2003, 1227-1241. Figure 3: CBR-BDI Agent base SCM [7] Rosalia Laza, A Case-Based Reasoning The intelligent agents handle the flow of Approach to the Implementation of BDImaterials, product & information flow with mobility Agents, Proc. European Workshop onfeature. The case-based reasoning provides the Case-Based Reasoning - EWCBR, 2002,mechanism to take the decisions in sourcing, 27-30.inventory, transport and demand forecasting. Hence [8] Juan M. Corchado-Rodríguez, An Agent-CBR-BDI agents are more suitable for Based Architecture for Developingimplementing the supply chain system. Internet-Based Applications, The European Journal for the Informatics Professional,VI. CONCLUSION 5(4), 2004, 51-55. [9] Juan M. Corchado and M. A. Pellicer, SCM has become the key strategic area that Development of CBR-BDI Agents,has direct impact over the success of any enterprise International Journal of Computer Sciencein today’s highly competitive business environment. & Applications, 2(1), (2005), 25 -32.We have represented the supply chain and related [10] Javier Bajo, Constructing Deliberativeproblems through a unified, flexible, and scalable Agents Using a Hybrid System, Proc.framework. The CBRR-BDI agent based SCM is International Joint Conference, Brazil,capable of taking decision on basis of past 2006.experience. The system maintains the flow of [11] Mart´Navarro and Stella Heras, Guidelinesinformation between the entities i.e. suppliers, to apply CBR in Real-Time Multi-Agentmanufacturers, distributors, retailers and the Systems, Journal of Physical Agents,customers. (2009) 3(3), 39-43. In future, the distributed database may be [12] Costin Bădică, Software Agents:applied for storing the case-base for every agent. Languages, Tools, Platforms, ComSIS 8(2),This factor will provide the access to case-base Special Issue, (2011), 256-295stored at different locations. 1292 | P a g e