Extending Unbounded Systems Thinking with agent-oriented modeling: conceptualizing a multiple perspective decision-making support system
Introduction
Research on decision-making with its various components (information acquisition, problem formulation, information modeling, etc.) is prevalent in the literature. Much of the literature focuses on enabling decision-making by categorizing or focusing knowledge and information into uni-perspective packages ripe for choosing the “right” or “best” alternative. Very often, the perspective chosen is one that is widespread within the group or organization, or one that is associated with an influential individual within the group. According to Nonaka and Takeuchi [33], individuals are not passive in their reception of new information or knowledge and must interpret it to fit their own situation and perspective. This interpretation is tenuous at best; change in interpretation can occur simply through filtered diffusion of information or knowledge from a singular perspective. Only by employing multiple perspectives during information acquisition and diffusion can an organization provide “…employees with a conceptual framework that helps them make sense of their own experience” [33, p.15]. This necessitates a system design that is cognizant of the use and benefits of multiple perspectives. System design, irrespective of methodology, is the first critical step toward an efficiently working implementation.
This paper begins the process by examining agent-oriented system design implications for multiple perspective system support and proposing that Unbounded Systems Thinking (UST) can logically be extended by applying the Belief–Desire–Intention (BDI) model. A cooperative agent system is exemplified, indicating that agent-oriented modeling is an appropriate methodology for designing such a system, and that agent technology may be an appropriate implementation tool for multiple perspective decision-making systems.
Section snippets
Systems design for multiple perspectives
Many researchers have investigated the need for organizations to support multiple perspectives. Hine and Goul [22] stress the need for organizations to engage in interpretive learning, during which members develop their own perspective of the organizational environment and work toward forming an organization-wide perspective based on these multiple interpretations. Argyris and Schön [1] and DiBella and Nevis [10] suggest that information stagnation may be reduced and organizational learning
Facilitating multiple perspective systems with agent technology
Constructing complex information systems using software agents is not new. For instance, Elofson et al. [12] illustrate how using intelligent agents during information acquisition can facilitate knowledge sharing. Chang and Chen [5] discuss using autonomous agents for information management. Bradshaw et al. [2] indicate that agent technology provides potential for many applications and begin development of tools designed to allow agent technology to be used by non-experts in the field.
Fowler
A cooperative-agent framework for multiple perspective decision-making
The Belief–Desire–Intention (BDI) model of agents, developed by the Australian Artificial Intelligence Institute, has become one of the best known and most studied models for agents [34]. An agent's practical reasoning process involves repeatedly updating beliefs from information in the environment, deciding what options are available, filtering these options to determine new intentions, and acting on the basis of these intentions. In some cases, this reasoning may also include an effort to
Illustrating a cooperative agent implementation of the UST model
All universities have established admission guidelines, but the admissions procedure itself is conducted by individuals with different ideas of what information is important to a student's future success. Hence, although quantitative models can be and are applied during the admissions process, qualitative information is often important. The following example is patterned after the doctoral candidate admissions process for a Management Information Systems department at a large Southeastern
Conclusion
The admissions example discussed here illustrates the application of cooperative agents within the three-perspective framework of the UST model. Because specific plans and facts for each agent are domain specific, it is clear that this cooperative multi-agent system could be applied to other decision settings where consideration of multiple perspectives is important. Furthermore, the relationships among the agents may be determined by the actual context of the problem.
The need for organizations
Dianne J. Hall is an Assistant Professor of Management Information Systems at Auburn University. She received her doctorate at Texas A&M University. She has served as an instructor of MIS, computer science, and economics at Texas A&M University in College Station, Corpus Christi, and Kingsville and has served as a consultant. Her work has appeared in academic and practitioner journals and books. Her current research interests include applications of information technologies in support of
References (39)
- et al.
An agent-based framework for building decision support systems
Decision Support Systems
(1999) - et al.
Autonomous intelligent agent and its potential applications
Computers & Industrial Engineering
(1996) Decision making and knowledge management in inquiring organizations: a new decision-making paradigm for DSS
Decision Support Systems Special Issue on Knowledge Management
(2001)- et al.
An intelligent agent community approach to knowledge sharing
Decision Support Systems
(1997) The role of AI-based technology in support of the knowledge management value activity cycle
Journal of Strategic Information Systems
(2000)- et al.
Building an agent-mediated electronic commerce system with decision analysis features
Decision Support Systems
(2001) Agent-oriented programming
Artificial Intelligence
(1993)Work-oriented development of knowledge-based systems—a case for multiple perspectives
Knowledge-Based Systems
(1994)- et al.
Organizational Learning II
(1996) - et al.
Application of Intelligent agent technology for managerial data analysis and mining
The Data Base for Advances in Information Systems
(1999)
A framework for organizational memory and organizational memory information system: insights from Churchman's five inquirers
The Design of Inquiring Systems: Basic Concepts of Systems and Organizations
Information seeking in social context: structural influences and receipt of information benefits
IEEE Transactions on Systems, Man and Cybernetics. Part C, Applications and Reviews
Knowledge management and the framing of information: a contribution to OR/MS practice and pedagogy
Journal of the Operational Research Society
How Organizations Learn: An Integrated Strategy for Building Learning Capability
Modelling social agents: communication as action
Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence
A framework for integrating multiple perspectives in system-development—viewpoints
International Journal of Software Engineering and Knowledge Engineering
Is it an agent, or just a program? A taxonomy for autonomous agents
Cited by (15)
Piper diagram – A novel visualisation tool for process design
2016, Chemical Engineering Research and DesignCitation Excerpt :However, some of these tools do not provide good insights for process engineers, while some are lack of visualisation capability which is useful during decision making phase. The lack of visualisation ability greatly reduces user's imaginary and idea delivery capability, hence affecting decision making capability and confusion might arise when optimisation idea are to be delivered to different individual (Dong and Zhang, 2014; Hall et al., 2005). This effect is even more pronounced when optimisation work is being presented to personnel with non-technical background, e.g. management level, inexperience labours as they might be short of relevant mathematical optimisation knowledge and experience (Hall et al., 2005).
Introducing a novel model of belief-desire-intention agent for urban Land use planning
2013, Engineering Applications of Artificial IntelligenceEngaging multiple perspectives: A value-based decision-making model
2007, Decision Support SystemsA Multi-Attribute Decision Support System for Allocation of Humanitarian Cluster Resources Based on Decision Makers’ Perspective
2022, Sustainability (Switzerland)In-situ simulation in design science research: Evaluation of complex design artifacts
2018, International Conference on Information Systems 2018, ICIS 2018
Dianne J. Hall is an Assistant Professor of Management Information Systems at Auburn University. She received her doctorate at Texas A&M University. She has served as an instructor of MIS, computer science, and economics at Texas A&M University in College Station, Corpus Christi, and Kingsville and has served as a consultant. Her work has appeared in academic and practitioner journals and books. Her current research interests include applications of information technologies in support of multiple-perspective and value-based decision-making.
Yi “Maggie” Guo (B.E. in Information Engineering; M.S. in Management Information Systems, University of Nebraska at Omaha; PhD in Management Information Systems, Texas A&M University) is currently an Assistant Professor at the University of Michigan—Dearborn. Yi's research interests include e-commerce, online shopping experience, flow theory, and agent-based systems in knowledge management. Her work has appeared in academic journals and conferences, and in Advanced Topics of Information Resources Management (M. Khosrowpour, ed.).
Robert Davis is an Associate Professor in the Computer Information Systems and Quantitative Methods Department of the McCoy College of Business Administration at Texas State University-San Marcos. He earned his MBA and PhD degrees from the University of South Carolina. He is the co-author of Operations Management: Concepts in Manufacturing and Services. He has published many articles dealing with the effective and efficient utilization of resources for operations and supply chain management. His articles have appeared in academic journals such as Management Science, Engineering Costs and Production Economics, Journal of Marketing, National Productivity Review, International Journal of Production Research, Competitiveness Review, Journal of Business Research, and Production and Inventory Management Journal. He has served as Co-Principal Investigator on several research projects with industrial clients dealing operations and supply chain issues.
Casey G. Cegielski, PhD, is an Assistant Professor of Management Information Systems in the College of Business on the faculty of Auburn University in Auburn, Alabama. He earned a doctorate in Business Administration with a concentration in Management Information System from the University of Mississippi. Additionally, he earned a Master of Accountancy and a Bachelor's degree from the University of Alabama. His current research interests are in the areas of innovation diffusion, emerging information technology, computer-facilitated speech recognition, and the strategic use of information technology. His research has appeared in several international information systems journals including Communications of the ACM, Information and Management, and the Information Systems Journal. Dr. Cegielski has more than 10 years of professional experience within the domain of information technology.
- 1
Present Address. Department of Management Studies, School of Management, University of Michigan—Dearborn, Dearborn, MI 48126-2638. Tel.: +313 593 5336.
- 2
Tel.: +1 512 245 8676.
- 3
Tel.: +1 334 844 6542.