A Distributed Architecture for Cooperative Intelligent Decision Support ...

1 downloads 10368 Views 183KB Size Report
forming virtual teams of geographically distributed knowledge workers to ... decision support system is viewed as a set of computer based tools integrating expert ...
IEEE MULTIDISCIPLINARY ENGINEERING EDUCATION MAGAZINE, VOL. 2, NO. 2, JUNE 2007

22

A Distributed Architecture for Cooperative Intelligent Decision Support Systems Abdelkader Adla

Abstract—We consider, in this work, the paradigm of distributed decision-support systems, in which several decisionmakers who deal with partial, uncertain, and possibly exclusive information must reach a common decision. We propose, to this end, an integrated framework based on a distributed architecture where each decision-maker uses an individual cooperative intelligent support system allowing him to carry out his decision tasks. The individual cooperative intelligent DSS are viewed as a set of computer based tools integrating expert knowledge and using collaboration technologies that provide decision-maker with interactive capabilities to enhance his understanding and information base about options through use of models and data processing. The proposed framework is intended to be adapted to group learning using ComputerMediated Communication technology. Index Terms—Decision support systems, Cooperative DSS, Distributed group DSS, Collaborative learning.

D

I. INTRODUCTION

ecision making is considered one of the most critical activities done in organizations. To support this complex process for individuals, a variety of independent, standalone information systems called Decision Support Systems (DSS) have been developed in the two last decades. However, these traditional decision support systems give not enough possibility of intervention to the user. Indeed, these systems are reduced to an insular and very technical state in which the objective is not to support decision but to dump data on the screen in the hope that the user will know what to de with. Such regular decision support systems help decision-makers to manipulate data and models. They don’t play the role of intelligent assistants to the decision makers. Although several research communities have examined aspects of intelligent system behaviour, often these contributions are merely parts of a solution. These systems have failed in use due to their poor interaction and no cooperation with the people having to use them. In decision-making processes, contrary to structured problems, one of the most important aspects is that the man The author is with the Department of Computer Science, Oran University, Algeria, and IRIT – Paul Sabatier University, Toulouse, France ([email protected]). Publisher Identification Number 1558-7908-IMCL2007-13

takes advantage on the machine. To solve problems requires calling intuition and know-how of the decision-maker which becomes the preponderant element of the couple Man/machine. The system must be able to play collaborator's role with the decision-maker, that is, to know his intentions and the context of the decision problem, to be able to give an action coordinated with the one of the decision-maker. Despite their impressive functionalities, DSS of all of types must focus on supporting, not replacing, a human decision maker for important decision tasks, as many of the problem situations faced by managers are unstructured in nature and require the use of reasoning and human judgement. Therefore, the DSS and the decision maker must form a united problem solver. In the other hand, in many organizational or social settings, a decision does not appear as an outcome given by a “single” decision maker, but as a compromise between various divergent interests and points of view. In the early 1990s, a shift from a mainframe based DSS to a network DSS occurred. As decision making moves from being an individual activity toward a group activity, many organizations are forming virtual teams of geographically distributed knowledge workers to collaborate on variety of tasks. We consider, in this thesis, the paradigm of distributed decision-support systems, in which several decision-makers who deal with partial, uncertain, and possibly exclusive information must reach a common decision. We propose, to this end, an integrated framework, based on a distributed architecture where each decision-maker uses an individual cooperative intelligent support system allowing him to carry out his decision tasks. The individual cooperative intelligent decision support system is viewed as a set of computer based tools integrating expert knowledge and using collaboration technologies that provides decision-maker with interactive capabilities to enhance his understanding and information base about options through use of models and data processing, and collaborates with him. The proposed system is further facilitated, i.e. a facilitator has the responsibility to manage the group activity of decision making. In particular, we envision to further extend traditional DSS by embedding expert knowledge with the DSS to provide intelligent decision support, by integrating cooperation to put the decision maker effectively in the loop of such a decision support, and by

1558-7908 © 2007 IEEE Education Society Student Activities Committee (EdSocSAC) http://www.ieee.org/edsocsac

IEEE MULTIDISCIPLINARY ENGINEERING EDUCATION MAGAZINE, VOL. 2, NO. 2, JUNE 2007

developing the cooperative group decision support system over the network to provide many advantages such as anytime-anywhere access, anonymity, ease of use and centralised corporate memory. The rest of the paper is as follows. Section II presents the paradigm of decision support systems. Intelligent and group decision support systems are presented respectively in sections III and IV, while Section V presents the collaborative learning concept. In Sections VI and VII the proposed architectures of the overall framework and the specific cooperative intelligent decision support system are detailed. Finally, Section VIII gives some concluding remarks and a direction of the future work. II. DECISION SUPPORT SYSTEMS The decision process is broadly defined as a bundle of correlated tasks that include: gathering, interpreting and exchanging information; creating and identifying scenarios, choosing among alternatives, and implementing and monitoring a choice [21]-[20]. Briefly, the decision process refers to some techniques or processing rules aiming at structuring the context, timing or content of communication. To support this complex process for individuals, a variety of independent, standalone information systems called Decision Support Systems (DSS) have been developed in the last decades to interactively support some or all phases of the decision making process. The field, which has evolved from the disciplines of management science and management information systems since the 1970s, has come to include personal decision support systems (DSS), group decision support systems, negotiation support systems, intelligent decision support systems, knowledge management based DSS, executive information systems (EIS) or business intelligence systems, and data warehouse based systems (DWS), expert and knowledge based systems (ES&KBS) [4]-[13]. Decision support systems (DSS) are computer-based systems designed to support and enhance managerial decision making. They are used to support complex decision-making and problem solving. They are designed to actively interact with an individual decision makers in order to solve ill or nonstructured decision problems and to assist them to make better decisions based on information obtained [19, 26]. Problems where priorities, judgements, intuitions and experience of the decision maker are essential, where the sequence of operations such as searching for a solution, formalization and structuring of problem is not beforehand known, when criteria for the decision making are numerous, in conflict or hard dependent on the perception of the user and where resolution must be acquired at restricted time. A complementary way of looking at DSS is associated with the role and functions that DSS have to fulfil [22], as seen from a user’s perspective: they assist managers in their decision processes in semi-structured tasks; they support and

23

enhance rather than replace managerial judgements; they improve the effectiveness of decision-making rather than its efficiency; they attempt to combine the use of models or analytical techniques with traditional data access and retrieval functions; they specifically focus on features that make them easy to use by non-computer people in an interactive mode; and they emphasize the flexibility and adaptability to accommodate changes in both the approach of the decision maker and the environment in which he acts. III. INTELLIGENT DECISION SUPPORT SYSTEMS Recently, many improvements have been witnessed in the DSS field, with the inclusion of artificial intelligence techniques and methods, as for example: knowledge bases, fuzzy logic, multi-agent systems, natural language, genetic algorithms, neural networks and so forth. The inclusion of AI technologies in DSS is an effort to develop computer based systems that mimic human qualities, such as approximate, reasoning, intuition, and just plain common sense. The new common denomination is: Intelligent Decision Support Systems – IDSS [24]. Intelligent decision support systems (DSS) are interactive computer-based systems that use data, expert knowledge and models for supporting decision makers in organizations to solve complex, imprecise and ill-structured problems by incorporating artificial intelligence techniques [24]. They draw on ideas from diverse disciplines such decision analysis, artificial intelligence, knowledge-based systems and systems engineering. The usage of IDSS is intended to improve the ability of operators and decision makers to better perform their duties and work together. There may be different ways to make a DSS more intelligent; the most frequently suggested method is to integrate a DSS with an ES. Turban and Watson [28] suggested two fundamental ES/DSS integration models: (1) ES are integrated into DSS components, the incorporation of ES aims to enhance the function of particular components in a DSS; for example, integrating an ES into the Data Base Management System (DBMS) of a DSS, which adds reasoning capability to data manipulation. This particular integration enables users to perform higher-level queries. According to Turban, the integration of ES in DSS components could be applied independently. (2) ES is integrated as a separate component in the DSS; an ES is an add-on to the original DSS. The ES complements the DSS in one or more steps of a decision making process. Such integration may be conceptualized as using the ES to play the role of a human expert who carries out strategy formulation, interpretation or alternative evaluation. In the second model, the DSS is responsible for both data and model manipulation, while the ES provides domain knowledge and recommends resolutions during the planning the process. During the process, data and models are manipulated through the Data

1558-7908 © 2007 IEEE Education Society Student Activities Committee (EdSocSAC) http://www.ieee.org/edsocsac

IEEE MULTIDISCIPLINARY ENGINEERING EDUCATION MAGAZINE, VOL. 2, NO. 2, JUNE 2007

Base Management System (DBMS) and Model Base Management System (MBMS), respectively. Instructions for data modifications and model execution may come from the ES interface directly. The MBMS obtains the relevant input data for model executions from the DBMS and, in return, results generated from model executions are sent back to DBMS for storage. The database also provides facts for the ES as part of the knowledge base. Using these facts together with the predefined rules, the inference engine of the ES performs model validations and planning evaluations, according to what a domain expert is supposed to do. Conclusions and recommendations are then passed to the interface, where they are displayed to the decision maker and transferred to the MBMS and DBMS, in the form of procedures calls for various actions. On this way, the IDSS is able to capture the domain knowledge and provide intelligent guidance during the process. While the data and model manipulations are done through the DSS, decision makers can focus solely on the process issues. However, making a simple machine act intelligently may be much less useful or important than being able to cooperate in an environment with other intelligent agents, whether they are humans or machines. Beyond being able to communicate with others, detect and correct mistakes, and take advantage of others’ abilities, so that overall intelligence or effectiveness may be an emergent property of all the smart agents working on the problem in a relatively coordinated fashion. It is found that “lack of attention to the human and organizational aspects of IT is a major explanatory factor and is manifest in a failure to involve users appropriately [9]. As Keen [18] stated, decision support systems “support, rather than replace, judgement in that they do not automate the decision process nor impose a sequence of analysis on the user”. Therefore, judgement and decision making must occur throughout the entire problem solving process, that is, during the user’s physical interaction with the system, and as the final human decision is being made. Because of this, the user’s decision processes must be factored into the design process of successful cooperative DSS. In this context, Soubie [25] proposes an architecture for cooperative knowledge based systems in which the automated system and its user solve a problem jointly in a cooperation mode. Moreover, according to Keen [18], decisions may be individual, group, or organizational tasks. When it comes to group or organizational tasks, a distributed DSS should be considered as a convenient alternative when its users and/or components cannot be at the same place at the same time, not as a replacement. A review of intelligent DSS that combine mathematical modelling with knowledge-based systems can be found in [8].

24

to the complexity of business relationships, the greater number of decision makers and organizations that are involved in the decision process, online access to multiple external information sources, and the decreasing in the time allowed for decision making. Thus, complex economic, social and political structures require the decisions to be made in a framework of sophisticated processes involving many stakeholders, who more or less directly participate in the decision-making process. In performing group activities by teams while working together, the use of Group Decision Support Systems can provide various levels of support including idea generation, information sharing, decision analysis, and alternative evaluation. In addition, GDSS facilitates communication among team members, regardless of the geographical limitations and group decision traps (such as fear of expressing ideas). A number of frameworks or typologies have been proposed for organizing our knowledge about decision support systems [23]. The two most widely implemented approaches for delivering decision support are Data-Driven and ModelDriven DSS. Data-Driven DSS helps managers organize, retrieve, and synthesize large volumes of relevant data using database queries, OLAP techniques, and Data Mining tools. Model-Driven DSS uses formal representations of decision models and provided analytical support using the tools of decision analysis, optimization, stochastic modelling, simulation, statistics, and logic modelling. Three other approaches have become more wide spread and sophisticated because of collaboration and web technologies: KnowledgeDriven DSS can suggest or recommend actions to managers. Document-Driven DSS integrates a variety of storage and processing technologies to provide managers document retrieval and analysis. Finally, Communication-Driven DSS or Group Decision Support Systems rely on electronic communication technologies to link multiple decision makers who might be separated in space or time, or to link decision makers with relevant information and tools. Since solving complex problems requires that people collaborate in modern organizations, the GDSS has drawn much attention in the last two decades or so. DSS research has evolved to include several additional concepts and views. Executive information systems have extended the scope of DSS from personal or small group use to the corporate level. Model management systems and knowledge-based decision support systems provided smarter support for the decisionmaker. The latter began evolving into the concept of organizational knowledge management about a decade ago, and is now evolving into a broader notion of DSS – GDSS – serving as knowledge sources or connecting decision-makers with diverse sources [10].

IV. GROUP DECISION SUPPORT SYSTEMS Currently, the need for Group Decision Making (GDM) techniques and support is greater than ever before. This is due

A group decision support system (GDSS) or group support systems (GSS) is developed to provide decision aid to groups or organizations by supporting collaborative and interactive

1558-7908 © 2007 IEEE Education Society Student Activities Committee (EdSocSAC) http://www.ieee.org/edsocsac

IEEE MULTIDISCIPLINARY ENGINEERING EDUCATION MAGAZINE, VOL. 2, NO. 2, JUNE 2007

works. One of the first formal definitions of a GDSS, evolved from the research of DeSanctis and Gallup: A GDSS combines communication, computing, and decision support technologies to facilitate formulation and solution of unstructured problems by a group of people [12]. DeSanctis and Gallup identify three environmental contingencies as critical to GDSS design: group size, member proximity, and group task. Group decision support systems augment the efficiency and effectiveness of group problem solving and decision making [12]-[28]. Such systems aim at the facilitation of communication between individuals and support of the expression of their opinions and beliefs in a way that is commonly accepted and understood, while providing the necessary technical infrastructure [11]-[16]. This is accomplished by providing communications facilities to support team problem solving and techniques for structuring the decision analysis via a systematic direction of the pattern, timing, or content of the related discussions [17]. In addition, based on differences in group size and dispersion of group members, four environmental settings are placed allowing the GDSS design and other technologies to be compatible. A space/time grid (including same-time and sameplace or different-time and different-place scenarios) is used to classify various collaboration technologies. Tools that support distributed teams which have been empirically tested are mainly synchronous computer conferencing systems (i.e. discussion forum software); these systems do not have explicit support for decision-making processes and often do not provide tools for alternative evaluation. GDSS products, such as GroupSystems [7], are LAN-based client-server applications often supporting same-time and same-place groups working in face-to-face settings. An audio/video conference system is a major example of a collaboration technology that supports groups working at the same time but at different places. This category of tools focuses on enhancing the virtual presence of meeting participants. The support for group processes and decision making are mostly missing from products in this category. Asynchronous technologies, such as e-mail and discussion forums, are commonly used as in the business world by distributed teams [7]. Asynchronous technologies tend to focus on supporting group information exchange and sharing [29]. However, these tools do not have strong support for group decision-making processes comparing to traditional GDSS. Holsapple and Whinston [15] made an exhaustive investigation into multiparticipant decision systems, and introduced three related areas: GDSS, organizational DSS, and negotiation support systems. These three areas are supported by technologies in the field of organizational computing, which include: coordination technology, computer-supported cooperative work, groupware, and computer-mediated communication. Furthermore, the technological infrastructure should fit within the organizational infrastructure in three respects: roles, relationships, and regulations. From the viewpoint of decision analysis, numerous techniques, in the areas of multi-criteria

25

/attribute decision making, multi-objective decision making, and group decision making can help decision makers to make a better choice. Namely, three techniques are naturally incorporated into GDSS to facilitate an efficient decision: brainstorming, Delphi and NGT [29]. Research that studied group decision support systems in the existing literature used mainly face-to-face facilitated GDSS. Some of its results may not apply to distributed teams [7] that, it is difficult for distributed teams to arrange face-to-face meetings or to meet at the same time virtually. Moreover, although most presented GDSS environments try to solve problems in the real world, the lack of an integrated procedure, from decision identification, basic information acquiring, to final decision proposed, makes the systems only partially supportive or even needful of outside assistance. Still, despite the existence as well as the extensive use of numerous general-purpose commercial systems, it is our belief that these systems do not readily fulfil the needs or operational usages of specialists or experts in different organizations to render their expertise in GDM processes. It is this belief that propels our study. V. COLLABORATIVE LEARNING Collaborative learning focuses primarily on knowledge management (the sharing and manipulation of knowledge that already exists somewhere within the workgroup), and knowledge building – the collaborative construction of knowledge that is new within the community. Individual learning, as a process of constructing personal knowledge, takes place within a learner’s personal perspective [6]. Collaborative learning involves an interaction among personal perspectives contributed by the participants and a merging of these into a group perspective definitive of the group discourse. Within individual perspectives, there is a strong tendency for ideas to diverge [14] – negotiation is required to bring ideas back into consensus and to promote individual ideas to the status of group knowledge. However, the negotiation required might be called “knowledge negotiation” because it is not just a matter of selecting among alternative existing states (propositions, proposals, activation functions), but of constructing new knowledge through collaborative interaction and discourse. The new knowledge is typically represented by or embodied in a shared “knowledge artefact”, such as a concept, theory, text or folder of structured information [27]. Collaborative knowledge building, itself, can be viewed as fundamentally a knowledge negotiation process. Proposed statements of knowledge by individuals are subjected to collaborative interactions, whereby meanings of terms are clarified, alternative related statements are compared, linguistic expressions are refined, etc. [27].

1558-7908 © 2007 IEEE Education Society Student Activities Committee (EdSocSAC) http://www.ieee.org/edsocsac

IEEE MULTIDISCIPLINARY ENGINEERING EDUCATION MAGAZINE, VOL. 2, NO. 2, JUNE 2007

This process establishes a “common ground” of understanding concerning the meaning of the accepted expressions and its constituent terms. This does not necessarily means that every individual involved fully understands and accepts this common ground in his own mind, but rather that a group understanding has been established in the discourse of the community in which this knowledge is thereby accepted. The common ground provides a basis for the meaning that the artefact encapsulates to be understood in a shared way by the collaborative community [5]. VI. A DISTRIBUTED ARCHITECTURE FOR COOPERATIVE INTELLIGENT DECISION SUPPORT SYSTEMS We view that successful distributed cooperative intelligent decision support systems and their subsystems must act intelligently and cooperatively in a complex domain with potentially high data rates and make judgements that model the very best human decision makers. It is also crucial that human decision makers maintain control over the final judgments, either by focusing the system on particular reasoning goals, or by modifying the basic knowledge on which the systems judgements rely. Therefore, the benefits of the traditional DSS can be further leveraged by embedding expert knowledge and implementing the DSS using collaboration technologies. Embedding expert knowledge with the DSS provides intelligent decision support, and implementing the intelligent DSS using collaboration technologies puts the decision maker effectively in the loop of such DSS and provides many advantages, such as distributed decision, anytime-anywhere access, anonymity, ease of use, and centralized data storage. In addition, distributed decision means that several entities (humans and machines) cooperate to reach an acceptable decision, and that these entities are distributed and possibly mobile along networks. Distributed decision making must be possible at any moment. It might be necessary to interrupt a decision process and to provide another, more viable decision. The advantages of a distributed DSS running on a distributed architecture are twofold: one the one hand, it enhances communication and information exchange from various expertise domains. This increased knowledge reusability is essential in a DSS. A decision problem is rarely completely new and often reuse parts of the knowledge used to solve past problems. On the other hand, a DSS is only useful if it is able to quickly generate quality alternatives. The flexibility offered by a distributed architecture allows precisely a DSS to retrieve and combine data, models, and other types of knowledge under tight constraints, amongst the currently available knowledge sources, to generate the required decision alternatives. To this end, the guidelines of our research methodology are stated as follows: 1.

Literature review on group decision systems and cooperative systems;

2.

3. 4. 5.

6.

26

A proposal of an architecture model for cooperative decision support system based on distributed topology; Development of an individual cooperative intelligent decision support system; Development of a group facilitation tool; Implementation and validation of the overall architecture applied to the boilers management system of GLZ oil plant. Adapting the above architecture to a collaborative learning context by re-thinking the nature of the interactions within this differing context.

Decision making is mostly used as a multi-participant process, which is increasingly interactive. We consider here the paradigm of distributed decision-support systems, in which several decision-makers who deal with partial, uncertain, and possibly exclusive information must reach a common decision. To this end, the use of a cooperative system makes possible the collaboration of distant users dispersed over a network and important volumes of data. The cooperative work so initiated can be synchronous or asynchronous. A small group or a whole organization can be supported. The application can be carried in several sites over a common information base. The decision making participants may know each other and work together, their influence on the decision making process may vary according to their individual levels of responsibilities at work, or they may have different abilities, arguments and points of view. The consequently development of a GDSS was based on the distributed approach, building upon the following assumptions: (1) Prospective users of DSS are involved in the decision making process and should be supported locally, particularly in generation stage; (2) The user may be expert or non experienced in decision making; so embedding expert knowledge help both to generate alternatives; (3) The tools should allow for generate, submit and selection of alternatives and present consequences of selected decisions in a meaningful way; and (4) Time intervals between formulation of a query and receipt of response should be minimal to allow for an effective on-line experience while using the DSS. This perspective enables reaching decisions by combining personal judgements with information provided by these tools. Within this framework, decision-makers undertake environments assessment and strategic analysis and provide data, judgement, intuition and personal vision as inputs to the process. An intelligent reasoning process is performed by individual DSS to generate alternatives. Decision-makers review their overall viability and make a final decision. In the proposed architecture shown in Fig. 1, the group is constituted of two or several decision-makers (DM) (participants) and a facilitator. Each participant interacts with a specific Cooperative Intelligent DSS (CI-DSS) that integrates

1558-7908 © 2007 IEEE Education Society Student Activities Committee (EdSocSAC) http://www.ieee.org/edsocsac

IEEE MULTIDISCIPLINARY ENGINEERING EDUCATION MAGAZINE, VOL. 2, NO. 2, JUNE 2007

DM 1

CI-DSS1

DM 2

CI-DSS2

DM n

CI-DSSn

Group Decision Toolkit

GF- DSS

Facilitator

Fig. 1. A distributed architecture for cooperative intelligent DSS

local expertise and allowing him to generate one or several alternatives of the problem submitted by the facilitator. The group (facilitator and participants) use the group toolkit for alternative generation, organization, and evaluation as well as for alternative choice which constitutes the collective decision. To this purpose, several techniques are available (vote, consensus, arbitration, multiple criteria analysis, etc.). The facilitator role is also aided via a Group Facilitation Support system (GF-SS). The Distributed Cooperative Intelligent Decision Support System (DCI-DSS) should be, in essence, decentralized in terms of databases, model bases and knowledge bases; and should make the best use of the available knowledge to offer optimally cooperation capabilities for the users. The main features of the intelligent cooperative DSS considered here are: they operate on knowledge-bases, they are based on decentralized distributed architecture; and they are designed to facilitate collaboration and communication among decision making groups. Due to their distributed nature, they urge for solutions that go beyond centralized client/server DSS designs. In this perspective, we envision to implement two f cooperation modes: (1) Man-machine cooperation allowing every decisionmaker to solve problem and to generate alternatives. Here, each man-machine association is considered as a whole and indivisible entity within the cooperation network. There must be at least one communication protocol between the two

27

components of the cooperative entity, as well as a machine, and formal data and information are directly available for computing. (2) Mediated man-man cooperation allowing the group of decision makers and the facilitator to make collective decision. This kind of cooperation uses a machine as an intermediate communication medium. The group decision system consists of a set of modules such: Interpersonal Communication Manager, Organizational Memory, Session Planning and a Group Toolkit. Each group tool has two versions: (a) participation version as private screen; it is used by a meeting participant engaging in a meeting activity; (b) Facilitation version as public screen; it is used by a meeting facilitator to set up parameters or data items associated with a meeting activity. In the proposed system each networked decision-maker is supported by a specific cooperative intelligent DSS. VII. SPECIFIC COOPERATIVE INTELLIGENT DSS By cooperative decision support systems, it is understood that the scenario implies a human decision maker and an automated system jointly working towards a solution. In addition, it couples the intellectual resources of individuals with the capabilities of computers to help make decisions. This system has as objective to offer resolutions or parties of resolutions to the decision makers. It achieves the task decomposition in sub-tasks as well as the assignment of roles to the both agents (system and decision maker) [1]-[2]-[3]. One way to enhance cooperation, within the decision process, is to produce a set of high quality candidate solutions, evaluated with respect to multiple objectives. This allows decision makers to gain important insights into the tradeoffs between multiple, possibly competing goals. We have to take into account: (a) A decision maker, facing a singular decision situation, is not familiar with the decision task, and (b) an expert decision maker, who is familiar with the decision problems, has to face and already has strategies compiled in his long term memory. To this end, embedded expert knowledge and software tools are used to learn the expert decision maker strategies in order to assist him (possibly continuously), help him to improve his performance and to capitalize his know-how. The proposed architecture, for the design of a cooperative intelligent decision support system extends the one of Soubie [25] developed for cooperative knowledge-based systems. To this end, we suggest several models: Conceptual Model of the Application (ACM), Conceptual Models of the Users (UCM) and Control Model of resolution (CM). The ACM is a set of three bases: Database (DB), Model base (MB), and Knowledge base (KB). Therefore, the model of the application integrates a representation of the domain knowledge (Domain

1558-7908 © 2007 IEEE Education Society Student Activities Committee (EdSocSAC) http://www.ieee.org/edsocsac

IEEE MULTIDISCIPLINARY ENGINEERING EDUCATION MAGAZINE, VOL. 2, NO. 2, JUNE 2007

Conceptual Model) and a representation of the task expertise (Task Conceptual Model) based on the task-method paradigm. Thus we can express the Application Conceptual Model in terms of tasks, methods and domain knowledge. The Architecture of a cooperative intelligent decision support system consists of the following modules: The Data Base Management System (DBMS) mainly contains a relational database which is managed by a the database management system, and which provides speed data retrieval, updating, and appending; The Model Base Management System (MBMS) includes many statistical, management scientific models, or other quantitative models that offer the system’s analytical or forecasting capability to solve future outcomes. Optimization models, such linear programming and dynamic programming, are often adopted to determine the optimal resource allocation to maximize or minimize an objective function; The Knowledge Base Management System (KBMS) can support any of the other subsystems or play an independent role. It suggests alternatives or actions to decision makers. Additionally, it can be inter-connected with the knowledge base. The Dialogue Manager allows the decision maker to interact with the system to achieve the task. The Control and cooperation Management: Sharing tasks is a condition to implement cooperation between the two agents. The task that is the subject of cooperation must be decomposed in consistent subsets. The task distribution among the two agents is dynamically made, according to the performances of the couple man/machine and of the workload of the user. Competences of the user and the system are sometimes complementary, sometimes “redundant”. In the latter case, user and system are often able to play the same role. The choice question of the appropriate agent which will have to play one role settles therefore. According to the context, different indications could be made to direct this choice. The set of indications on the manner to allocate different roles to the agents defines the cooperation modes. VIII. CONCLUSIONS AND FUTURE WORK We considered here the paradigm of distributed decisionsupport systems. , in which several decision-makers who deal with partial, uncertain, and possibly exclusive information must reach a common decision. The development of the Distributed Cooperative Intelligent DSS is based on two cooperation modes: (1) Man-machine cooperation allowing every decision maker to solve problem and to generate an alternative, (2) Mediated man-man cooperation allowing the

28

group of decision makers and the facilitator to make collective decision. This kind of cooperation uses a machine as an intermediate communication medium. To support each decision maker, we defined a cooperative architecture for intelligent decision system that takes into account user abilities and integrate him in the loop of decision problem solving process particularly while generating alternatives. We assume that putting the human operator effectively in the loop of such decision support system represents the major guarantee of mastering efficiently the inherent complexity of the problems. An initial application of the boilers management system was also developed Currently, the aspects of the work are undertaken: (1) integrating an expert system approach capable to develop facilitation skills, (2) developing a planning tool allowing the assignment of the tasks beforehand modelled to software or human agents, and (3) adapting the proposed framework to collaborative learning. The facilitation expert system considers the support to expert and inexperienced facilitators by incorporating a decision making process model in the support system (GFSS). Therefore, the facilitator is supported in managing the group activity along the three phases of the decision process, i.e. in pre-meeting, during meeting, and post meeting phases. He or she uses the GF-SS in a cooperation mode. The model of the decision-making process, we are developing, is based on task-method paradigm: (a) Task, the unique task that must be performed; (b) Method, the technique (s) or the tool(s) that must be used to perform the task. It is currently under construction. As for the planning tool, it is particularly useful to manage the task allocation in a dynamic way and, according to the control of the different tasks, and to change planning initially implemented. The interest of such tool is to be able to provide an interactive and dynamic support during the problem collective resolution; this type of support – interactive and dynamic – being a precondition for the decision support. Other types of task planning tools can be envisaged in a global decision process. Indeed, more classical tools issued from operational research, namely in optimization, can be also used. This tool constitutes essential provision in cooperation management. The man-management cooperation is possible only if this task management tool allows a quick re-planning as well as a re-counting of allocations if there is a context modification or an evolution of the problem. This approach insures the dynamic character of the tool. Finally, one perspective of this work is to adapt our cooperative framework to group learning using CMC technology.

1558-7908 © 2007 IEEE Education Society Student Activities Committee (EdSocSAC) http://www.ieee.org/edsocsac

IEEE MULTIDISCIPLINARY ENGINEERING EDUCATION MAGAZINE, VOL. 2, NO. 2, JUNE 2007

ACKNOWLEDGMENT The author would like give thanks to the anonymous reviewers for their helpful and insightful comments and suggestions about the contents of this paper. REFERENCES [1]

A. Adla, “Système coopératif d’aide à la décision”, in 2006 Proc. IHM Conf., Montréal, Canada, 2006. [2] A. Adla, J.L. Soubie, and P. Zarate, “A distributed architecture for cooperative decision support systems” in 2006 Proc. EuroWorgroup Workshop on decision support systems, London, England. [3] A. Adla and P. Zarate, “A cooperative intelligent decision support system”, in 2006 Int. Conf. IEEE. Service Systems and Service Management, Troyes, France, October 25-27 2006. [4] D. Arnott and G. Pervan, “A critical analysis of decision support systems research, Journal of Information Technology, vol. 20, no. 2, pp. 67-87, 2005. [5] M. Baker, “The role of grounding in collaborative learning tasks”, in Collaborative learning: Cognitive and computational Approaches, Dillenbourg, P., Editor, 1999, Pergamon: Oxford, UK. [6] R.J. Borland and R.V. Tenkasi, “Perspective making and perspective taking in communities of knowing”, Organization Science, vol. 6, no. 4, pp. 350-372, 1995. [7] M. Chen, “TeamSpirit: Design, implementation, and evaluation of Webbased group decision support system”, Elsevier B;V., Decision Support System, 2005. [8] W. Cheung, “An Intelligent decision support system for service network planning”, Decision Support Systems, Vol. 39, pp. 415-428, 2003. [9] C. Clegg, “Information technology: a study of performance and the role of human and organizational factors”, Ergonomics, vol. 40, no. 9, pp851-871, 2002. [10] J.F. Courtney, “Decision making and knowledge management in inquiring organizations: toward a new decision making paradigm for DSS”, Decision Support Systems, vol. 31, no. 1, pp. 17-38, 2000 [11] A. Dennis, “Group Support Systems for Strategic Planning”, Journal of Management Information Systems, vol. 14, no. 1, pp 155-184, 1997. [12] G. DeSanctis and B. Gallup, “A foundation for the study of group decision support systems”, Management Science, vol. 13 no 12, pp.1589-1609, 1987.

29

[13] G. Forgionne “I-DMSS: A Conceptual Architecture for the Next Generation of decision Making Support Systems in the Internet Age”, in Proc. Conf. IFIP TC8/WG8.3, Cork, Island, pp 154-165, 2002. [14] J. Hewitt and C. Teplovs, “An analysis of growth patterns in computer conferencing threads”, in Proc. Conf. Computer Supported Collaborative Learning (CSCL’99), 1999, Palo Alto, CA. [15] C.W. Holsapple and A.B. Whinston, Decision Support Systems: A Knowledge-Based Approach, West Pub., St. Paul, MN, 1996. [16] L. Jessup and J. Valacich, “Future directions and challenges in the evolution of group support systems, in Group support systems, new perspectives, L. Jessup and J. Valacich (eds), 1992. [17] N.I. Karacapidilis and C. Pappis, “A framework for group decision support systems: combining AI tools and OR techniques”, European Journal of Operational Research, vol. 103, pp. 373-388, 1997. [18] P.G.W. Keen, “Value analysis: Justifying Decision Support Systems, MIS Quarterly, vol. 5, no. 1, pp. 15-25, 1981. [19] P. Keen and M. Scott-Morton, Decision Support Systems: an organizational perspective, Addison-Wesley Publishing, 1978. [20] J.E. McGrath, “Groups – Interaction and Performance”, Prentice Hall, Englewood Cliffs, NJ, 1984. [21] H. Mintzberg, The structuring of Organization, Prentice Hall, 1979. [22] B.J. Parker and G.A. Al-Utaibi, “Decision support systems: the reality that seems hard to accept”, OMEGA International Journal of Management Sciences, vol. 14, no. 2, pp. 135-143, 1986. [23] D.J. Power “Supporting Decision-Makers: An Expanded Framework”, Available: http://dssresources.com, 2000. [24] R. Ribeiro, “Intelligent Decision Support Tool for Prioritizing Equipment Repairs in Critical/Disaster Situations”, in Proc. Euro Working On Decision Support Systems Workshop, London, England, 2006. [25] J.L. Soubie, “Modelling in cooperative based systems”, In Proc. Conf. COOP’98, Cannes France, may 1998. [26] R. Sprague and D. Carlson, Building Effective Decision Support Systems, Prentice-Hall, Inc, Englewood Cliffs, 1982. [27] G. Stahl, “Knowledge Negotiation in Asynchronous learning networks”, in Proc. 36th Hawaii Int. Conf. System Sciences, IEEE, 2003. [28] E. Turban and J. Aronson, “Decision support systems and intelligent systems”, Prentice-Hall International, Upper Saddle River, New Jersey, 2001. [29] P. Zarate, « Des systèmes interactifs d’aide à la décision aux systèmes coopératifs d’aide à la décision: Contributions conceptuelles et fonctionnelles », Mémoire HDR, INP de Toulouse, 2005.

1558-7908 © 2007 IEEE Education Society Student Activities Committee (EdSocSAC) http://www.ieee.org/edsocsac