Showing papers on "R-CAST published in 1996"

Distributed intelligent agents

Abstract: In Retsina, the authors have developed a distributed collection of software agents that cooperate asynchronously to perform goal-directed information retrieval and integration for supporting a variety of decision-making tasks. Examples for everyday organizational decision making and financial portfolio management demonstrate its effectiveness.

Coordination of Multiple Intelligent Software Agents

Abstract: We are investigating techniques for developing distributed and adaptive collections of information agents that coordinate to retrieve, filter and fuse information relevant to the user, task and situation, as well as anticipate user's information needs. In our system of agents, information gathering is seamlessly integrated with decision support. The task for which particular information is requested of the agents does not remain in the user's head but it is explicitly represented and supported through agent collaboration. In this paper we present the distributed system architecture, agent collaboration interactions, and a reusable set of software components for structuring agents. The system architecture has three types of agents: Interface agents interact with the user receiving user specifications and delivering results. They acquire, model, and utilize user preferences to guide system coordination in support of the user's tasks. Task agents help users perform tasks by formulating problem solving plans and carrying out these plans through querying and exchanging information with other software agents. Information agents provide intelligent access to a heterogeneous collection of information sources. We have implemented this system framework and are developing collaborating agents in diverse complex real world tasks, such as organizational decision making, investment counseling, health care and electronic commerce.

Evaluating ethical decision making and computer use

Abstract: The authors present the results of an exploratory survey identifying users' attitudes and behavior when computer privacy and resource-ownership situations are encountered. T HE Theory of Reasoned Action (TRA), frequently used to describe ethical decision-making behavior, relates attitudes and social norms to individual behavioral intentions [2, 17]. The components of this theory appear to be present in ethical decision situations [11], but they may be insufficient in their usual characterizations to completely describe an ethical decision process [17, 18]. Computer users face an increasing number of ethical dilemmas in their use of computers, and novel situations appear with each new technology. This article presents an exploratory evaluation of the TRA and extensions to attempt to better understand ethical decisions in computer use.

Facilitation, GSS, and Training as Sources of Process Restrictiveness and Guidance for Structured Group Decision Making: An Empirical Assessment

Abstract: Structured decision techniques have been a mainstay of prescriptive decision theory for decades. Group Support Systems GSSs automate many of the features found in decision techniques, yet groups often choose to ignore both the technique and the technology in favor of more familiar decision processes. This research empirically tests propositions and hypotheses for a specific instantiation of Adaptive Structuration Theory. A controlled laboratory experiment tests the ability of three appropriation mediators e.g., facilitation, GSS configuration, and training to directively affect group decision making through guidance and restrictiveness. The experiment used a hidden-profile task and structured decision technique which directed group members to reach a decision by identifying the problem, choosing criteria, and selecting a solution. The results supported the proposition that appropriation mediators can increase the faithful use of structured decision techniques and that faithful use can improve decision quality.

Decision Making in Complex Naval Command-and-Control Environments:

Abstract: This paper presents the findings of a study on how experienced naval officers make decisions in a complex, time-pressured command and control setting, the Combat Information Center of AEGIS cruisers. The decision processes invoked by the officers were consistent with the recognition-primed decision model. The majority of decisions concerned situation awareness and diagnosis in which the decision makers used feature-matching and story generation strategies to build situation awareness. Furthermore, awareness of the situation enabled the officers to recognize appropriate actions from published procedures or past experience. A recognitional strategy was used to identify 95% of the actions taken; decision makers compared multiple options in only 4% of the cases. These findings are discussed in terms of their implications for framing command-and-control problems, for emphasizing situation awareness, for a descriptive model of decision making, and for designing decision supports.

A process perspective on strategic decision making

Abstract: Posits that a process perspective on strategic decision making is more likely to yield a successful outcome. Conceives the strategic decision‐making process as a composite of the concept of strategic gap and the managerial decision‐making process. Presents six examples of real‐world strategic decision in support of a process approach to the making and implementing of such decisions. The evidence in support of a process perspective on strategic decision making suggests a need for further research and exposition of this critically important subject.

Establishing the Boundaries of a Paradigm for Decision-Making Research:

Abstract: In recent years a paradigm shift in decision-making research has occurred. Modern decision researchers are more interested in studying decision making as it occurs in the so-called real world under naturalistic conditions. As such, the naturalistic decision-making (NDM) movement has seen growing support in recent years. Our contention is that in order for a new paradigm of NDM to flourish, it is essential to define carefully and completely the nature of decisions that are of interest to investigate. Once this is done, it is then possible to determine the psychological processes and strategies that decision makers use to cope with this subset of naturalistic decisions and to establish acceptable methods of study. Given this need, the purpose of this article is to examine the concept of NDM as it has been treated in past work and to refine and expand this conceptualization.

Decision making in emergencies

Abstract: Proposes that decision making is part of all management tasks and that it is particularly important for emergency managers as they often need to take decisions quickly on very inadequate information. Briefly reviews some of the particular problems of emergency decision. Looks at the usefulness of Vroom and Yetton’s decision process model for emergencies, before proposing a simplified problem classification based on three problem characteristics. Concludes by reviewing a collection of “emergency” decisions and analysing some of the common factors to suggest a number of simple action rules to be used in conjunction with the simplified decision process model proposed, the “emergency manager’s decision cube”.

A Theory of Decision Support System Design for User Calibration

Abstract: A theory is proposed for designing decision support systems (DSS) so that the confidence a decision maker has in a decision made using the aid equals the quality of that decision. The DSS design theory for user calibration prescribes properties of a DSS needed for users to achieve perfect calibration. Relevant calibration, decision making, and DSS literatures are synthesized; and related behavioral theories are borrowed to identify the properties of expressiveness, visibility, and inquirability as requisite components of the DSS design theory for user calibration.

Decision Support Systems: An Applied Managerial Approach

Abstract: INTRODUCTION TO DECISION SUPPORT SYSTEMS. Decision Making. COMPONENTS OF A DSS. Data Components. Model Components. Intelligence and Decision Support Systems. User-Interface Components. Mail Components. ISSUES OF DESIGN. International Decision Support Systems. Designing a Decision Support System. Object-Oriented Technologies and Decision Support Systems Design. Implementation and Evaluation of Decision Support Systems. RELATED SYSTEMS. Group Decision Support Systems. Executive Information Systems. Photo Credits. Index.

Arguing for decisions: a qualitative model of decision making

Abstract: We develop a qualitative model of decision making with two aims: to describe how people make simple decisions and to enable computer programs to do the same. Current approaches based on Planning or Decision Theory either ignore uncertainty and tradeoffs, or provide languages and algorithms that are too complex for this task. The proposed model provides a language based on rules, a semantics based on high probabilities and lexicographical preferences, and a transparent decision procedure where reasons for and against decisions interact. The model is no substitute for Decision Theory, yet for decisions that people find easy to explain it may provide an appealing alternative.

The effects of decision support and task contingencies on model formulation: a cognitive perspective

Abstract: This paper takes a cognitive cost-benefit approach to understanding model formulation. Work in the behavioral decision literature on the role of effort and accuracy in choice tasks indicates that effort, or cognitive cost, is a key factor in understanding decision behavior. However, the model formulation literature does not discuss how effort interacts with other factors, such as task complexity and decision aids, to influence model formulation. In this paper, based on the work on the cost-benefit theories of cognition, we posit that two types of effort, namely that associated with building or formulating a model and that associated with utilizing that model in the solution of a problem, will influence model formulation. We then examine how the methods used in the behavioral decision making literature and the reported findings concerning the interaction of effort with task and decision aids can be utilized to understand model formulation.

Assessment and decision making in animals: A mechanistic model underlying behavioral flexibility can prevent ambiguity

Abstract: Understanding how animals make decisions is a fundamental question in behavioral ecology which has cascading effects on how animals respond to environmental variation. An explicit model of the mechanisms of information processing and decision making can help prevent conflated definitions and ambiguous interpretations. Unambiguous definitions are crucial for clear communication between theoreticians and empiricists and for the rapid advancement of studies of decision making. Moreover, employing a clear model of underlying proximal processes will help bridge the gap between cognitive psychology and behavioral ecology and should aid scientific advancement. We present a simple model to guide studies of assessment and decision making. According to the model, individuals assess perceived stimuli and evaluate them for useful information. The association between perceived stimuli and evaluated information involves assessment rules. Based on evaluated information, individuals can employ trade-offs and make decisions. The association between the result of assessment and observed behavior involves decision rules. The model clearly emphasizes that the study of decision rules requires knowledge of the results of assessment, and we acknowledge the difficulty of studying assessment. However, without this knowledge, we can only study decision rules when we assume assessment rules between subjects are identical (i.e., with a uniform group of subjects). The simple model can be used to structure the design and interpretation of studies of assessment and decision making and help theoreticians and empiricists work together to understand behavioral flexibility.

Multi‐attribute decision support in the event of a nuclear accident

Abstract: Retrospective studies of nuclear accidents such as those at Three Mile Island and Chernobyl have emphasized the need for preparing structured decision support methodologies for use in any future emergency. This paper discusses the multi-attribute aspects of such decision support, i.e. methods for supporting decisions when there are several conflicting objectives. Considerable progress has been made in application of these methods over the past 4 or 5 years. In particular, the decision-conferencing format has proved successful in helping senior decision makers understand and deal with the issues that arise in considering long-term countermeasures made. Within the RODOS project, a European initiative to build a decision support system for emergency response, multi-attribute value and utility techniques for short-term and medium-term countermeasures are being developed. However, much remains to be done, particularly in relation to the incorporation of uncertainty when there is a risk of an imminent accident.

Supporting facilitation in group support systems: techniques for analyzing consensus relevant data

Abstract: In this paper, we present a set of techniques and an approach to support the facilitator in building consensus during group decision making in computer supported group work. The approach utilizes data about each participant's expressed preferences (scores and ranks) for a set of decision alternatives under consideration. The data are analyzed to provide the facilitator with information about the level of group consensus, coalescing of subgroups, and areas of strong disagreement. An illustration from a real-world case situation demonstrates the approach.

Technical Briefing: Making Multiple-Objective Decisions

Abstract: This text presents methods for solving multiple criteria decision problems. It shows how multiple criteria methods differ according to: the problem; the timing of decision maker information; the information required from the decision maker; and the availability of supporting software.

HMSS: a management support system for concurrent hospital decision making

Abstract: Mounting health care costs have escalated the pressure on hospitals and other health care providers to control expenses. Conventional hospital information systems help meet the challenge by providing data necessary for policy formation and outcome measurement. Additional decision support systems deliver models that can be used to systematically evaluate the policies. When deployed successfully, each stand-alone system can effectively support a segment of the hospital decision making process. Integrating the stand-alone functions can enhance the quality and efficiency of the segmented support, create synergistic effects, and augment decision making performance and value. A high-level integration framework, known as the management support system (MSS), can be adapted to provide the desired synthesis. This paper demonstrates how management support systems can improve hospital decision making. It overviews the hospital decision making process, presents an MSS for supporting this process, and measures the impact of the MSS on the process and outcomes of decision making. The paper also examines the implications of the analyses for information systems research and health care practice.

Information technology support for collaborative decision making in auditing: an experimental investigation

Abstract: Several studies in behavioral decision making have found that decisions made by interacting groups tend to “shift” from the mean judgment for the same task when performed individually. While this “choice-shift” phenomenon has been documented in experiments where group participants interact face-to-face, the impact of Group Decision Support Systems (GDSS) on this phenomenon has not been investigated. GDSS permit simultaneous and anonymous interaction and provide a structured environment for group discussion. This paper reports the results of an experiment investigating whether decisions reached by groups using GDSS technology differ from those made by traditional “face-to-face” groups for the same judgment task. The results revealed that GDSS-mediated communication, whether anonymous or non-anonymous, resulted in no significant choice shift while the face-to-face communication mode resulted in a significant cautious shift. The results also revealed that GDSS-mediated communication was more efficient, and that there was no significant difference in subjects' perceived satisfaction with the group process between GDSS-mediated and face-to-face groups.

The Multiple Criteria Location Problem: 2. Preference-Based Techniques and Interactive Decision Support:

Abstract: This is the second of two papers in which multiple criteria location problems (MCLPs) are discussed. In this paper two major approaches to locational decisionmaking are overviewed: optimizing decision rules (utility-function-based methods) and satisficing decision rules (goal-programming methods). Their advantages and disadvantages are discussed. From these two concepts a quasi-satisficing decision rule is developed and operationalized through a reference point method. A framework for an interactive decision support system (DSS) for tackling MCLPs is proposed. The system integrates a network model with the quasi-satisficing approach. It is argued that the DSS data and analytical components can be effectively integrated by means of the interactive decision support concept that involves a feedback exchange of information between a decisionmaker and a computer-based support system. This concept allows for the exploration of the locational decision problem and the alternative solutions both in decision space ...

Multi-Agent Integration of Information Gathering and Decision Support

Abstract: We are investigating techniques for developing distributed and adaptive collections of information agents that coordinate to retrieve, filter and fuse information relevant to the user, task and situation. In our system of agents, information gathering is seamlessly integrated with decision support. In this paper we present the distributed system architecture, agent collaboration interactions, and a reusable set of software components for structuring agents. The system has three types of agents: Interface agents interact with the user receiving user specifications and delivering results. They acquire, model, and utilize user preferences to guide system coordination in support of the user’s tasks. Task agents help users perform tasks by formulating problem solving plans and carrying out these plans through querying and exchanging information with other software agents. Information agents provide intelligent access to a heterogeneous collection of information sources. We have implemented this system framework and are developing collaborating agents in diverse complex real world tasks, such as organizational decision making, and financial portfolio management.

On-Line Group Decision Support by Preference Programming in Traffic Planning

Abstract: Preference programming is a decision support technique which allows decision makers to give preference statements of weight ratios in terms of intervals instead of single numbers in a value tree. Individual preferences, based on single number statements, can be combined into an interval model, and the negotiation proceeds by focusing on decreasing the width of the intervals. The preference programming approach was evaluated with a realistic traffic planning problem by using the HIPRE 3 + Group Link software. The results from nine test groups indicate that preference programming is an operational group decision support technique which initiates negotiations and efficiently directs the discussion towards issues which are relevant in reaching a consensus.

The contributions of organizational science to the development of decision support systems research subspecialties

Abstract: This study is conducted to infer the intellectual structure of the decision support systems (DSS) field by means of an empirical assessment of the DSS literature over the period 1971 through 1990. Three multivariate data analysis tools (factor analysis, multidimensional scaling, and cluster analysis) are applied to an author cocitation frequency matrix derived from a large database file of comprehensive DSS literature over the same period. We conclude that decision support systems are grounded in several contributing disciplines, such as organizational science, multiple criteria decision making, group decision making, and strategic planning. We further conclude that the DSS area is in the process of building its own articulated theories in the subareas of foundations, group decision support systems, model management, user interface/individual differences, and decision support systems implementation, and that organizational scientists have made important contributions to the development of DSS research subspecialties.

Perspectives on Representation and Analysis of Negotiation: Towards Cognitive Support Systems

Abstract: The rapid expansion of Decision and Negotiation Support Systems has been built mainly on decision-theoretic approaches. This has resulted in the decision maker being viewed through the lens of the problem. In this article, the focus is on the decision maker’s view of the problem. Three levels of problem articulation are described. Special emphasis is placed on the needs level and the implications it carries for the cognitive and instrumental levels. The three levels of articulation, the organizational model of making decision in social settings, and the three basic approaches to decision making form the basis for computer support focused on understanding and change rather than preferences and outcomes. We argue that in the dynamic, interactive context characteristic of negotiations, a cognitive support system based on restructurable modeling provides a richer basis for support.

The impact of unreliable information on the management of software projects: a dynamic decision perspective

Abstract: The task of managing software projects is universally plagued by cost and schedule overruns. A fundamental problem in software projects is the presence of unreliable information. In initial information as well as in subsequent status reports. The authors report an experiment that investigates decision making in software projects as exemplars of complex, dynamic environments reactive to the actions of the decision maker. The experiment shows that in coping with unreliable information in such environments, decision makers are susceptible to self-fulfilling prophesies created by the environment, and are prone to demonstrate conservatism. A process tracing extension of the experiment shows that subjects demonstrate a low capacity for handling complexity. The implications of the results for managing software projects and for research in dynamic decision making are discussed.

Integrating AI and optimization for decision support: a survey

Abstract: Optimization models have been the workhorses of computer based decision support. Their emphasis is on model structure, quantification and solution efficiency. However, there also are important meta-modeling, analysis and interpretation activities associated with practical decision making. AI methods, because of their aim of emulating human reasoning and thinking activities, have the potential of providing computer based support for these other decision making activities. There has been a growing literature on the integration of AI and optimization techniques for decision support. Much of this body of work describes techniques that are application or problem specific. Others describe more general methods addressing different specific aspects of decision making. In this paper we use a conceptual framework to survey and analyze these efforts. Our survey shows that efforts to integrate AI and Optimization have been focused mainly on model formulation and selection. Other activities such as post solution analysis or solver selection, have received considerably less attention for automated support. The dramatic difference in paradigms between AI and Optimization result in vastly different data structures and control primitives in their respective software implementations. We conjecture that this disparity will continue to thwart the development of general tools for seamless AI/Optimization integration.