The Secretary Problem Revisited - The Group Decision-Making Perspective

We consider a decision situation where there is an initial set of alternatives that may be augmented, a variant of the problem known in the literature as the "secretary problem". We focus the discussion on the special case of group decision-making, where a group or committee is charged with the responsibility of negotiating the decision. We investigate situations of explicitly stated (multiple) criteria and the simpler situation of no such stated criteria. The former case includes the congenial, the mixed, and the uncongenial cases, where the individuals agree on the direction of all, some, or none of the criteria, respectively. We offer a framework within which a group of individuals can be supported in such a decision process. In the case of explicit criteria, we provide the decision-makers with probability information of the likelihood of finding more preferred alternatives provided the initial set of alternatives is expanded. The framework is tested using a simulated real-world choice situation.     

A Calibrated Group Decision Process

In practice most organisational decisions are made by groups that bring into the problem multiple perspectives, both complementary and contradictory. When having a group of decision makers, usually individuals’ preferences are either led to consensus or are aggregated with the use of some function like the median, the arithmetic or geometric mean. We focus in the second case, where individual’s preferences need to be aggregated. Our approach is based on the fact that when two decision makers are asked to give their preference between a pair of criteria using a specific scale, it is possible that they will give slightly different answers, even when they actually have the same opinion. This difference will not affect the case of a single decision maker, as it will be consistent throughout the whole process. However, it can affect a group decision when the values will be used as an input for the aggregation function. A novel approach is presented that enhances group decision making through a group calibration process. The proposed process adjusts individuals’ preferences based on their answers on a set of standardized questions prior to the aggregation phase. The method focuses The whole concept is applied to the group analytical network process method and it is illustrated through a telecommunications project case. The decision under examination concerns the selection of the right place for deploying a new telecom service of a multinational-based telecommunications company where a group of geographically dispersed decision makers form an ad-hoc virtual team in order to select the location for a new technical support centre.     

Group Structured Reasoning for Coalescing Group Decisions

In this paper we present the notion of structured reasoning through a model, called the Generic/Actual Argument Model (GAAM). The model which has been used as a computational representation for machine modelling of reasoning and for hybrid combinations of human and machine reasoning can be used as a coalescent framework for decision making. Whilst the notion of structuring reasoning is not new, structured reasoning is advanced as a technique where group consensus on reasoning structures at various levels can be used to facilitate the comprehension of complex reasoning particularly where there are multiple perspectives. For an issue, the approach provides a scaffolding structure for cognitive co-operation and a normative reasoning structure against which group participants can identify points of difference and points in common as well as the nature of the differences and similarities. Intra-group transparency characterized by the ability to recognise points in common and understand the nature of differences is important to the process of coalescing group decisions that carry maximum group support.     

Aggregation of Individual Preference Structures in Ahp-Group Decision Making

This paper presents a new procedure, to which we have given the name Aggregation of Individual Preference Structures (AIPS), whose objective is to deal with multiactor decision making when using Analytic Hierarchy Process (AHP) as the methodological support. This procedure incorporates ideas similar to Borda count methods and transfers to the case of preference structures the principle of aggregation employed in the two approaches traditionally followed in AHP-group decision making (aggregation of individual judgments and aggregation of individual priorities). The new aggregation method allows us to capture: (i) the richness of uncertainty inherent to human beings; (ii) the vision of each decision maker within the context of the problem; (iii) the interdependencies between the alternatives being compared and (iv) the intensities of the preferences that each decision maker gives to these interdependencies. From the preference structure distribution associated to each decision maker, this new approach (AIPS) provides the holistic importance of each alternative and ranking, as well as the most representative preference structure distribution for the group. The knowledge derived from these could be employed as an initial step in the search for consensus, which characterises the negotiation processes followed by the actors involved in the resolution of decisional problems. An earlier version of this paper was presented at the GDN2005 Conference held at Vienna. The work has been partially funded under Research Projects “Electronic Government. Internet-based Complex Decision Making: e-democracy and e-cognocracy” (Ref. PM2004-052) and “Internet-based Complex Decision Making. Decisional Tools for e-cognocracy” (Ref. TSI2005-02511).     

A Mixed-Data Evaluation in Group TOPSIS with Differentiated Decision Power

This main objective of this paper is to provide decision support for mixed data in group Technique for Order Preference by Similarity to Idea Solution (TOPSIS) with differentiated decision power. We use a signum function to compare the ordinal performance of alternatives on any qualitative criterion, or the partial information provided by decision makers. The proposed process for ordinal information is uniformly coherent with the traditional TOPSIS steps, preserving the characteristic of distance-based utilities. Ordinal weights are also considered herein, and the decision power of the group members is formulated by their weights under an agreement in the group. Two examples demonstrate that the proposed approach has some benefits and achieves robustness with two types of sensitivity analyses. Some discussions and their limitations to the approach are also provided.     

Coordination Structures and System Restrictiveness in Distributed Group Support Systems

Most GSS research has studied the impact of restricting group interaction to GSS-prescribed coordination structures with face-to-face groups, while Distributed GSS (DGSS) has been largely ignored. Due to the nature of mediated communication in asynchronous interaction, it is relatively difficult to coordinate distributed groups, and a special coordination structure must be arranged to overcome these difficulties. This study examines the effect of system restrictiveness of coordination structures in an asynchronous environment. A 2 × 2 factorial experiment was designed with two independent variables – sequential vs. parallel coordination mode, and with vs. without a leader – to construct coordination structures with varying degrees of restrictiveness. The study finds that less restrictive coordination structures are more appropriate to support asynchronously interacting distributed groups. Objective decision quality is equal for both parallel and sequential coordination mode, but is significantly better with a group leader. Groups with parallel coordination mode have a stronger belief that the decisions they made are of higher quality than those of groups with sequential coordination mode. In groups with a leader, communication effectiveness is better. Satisfaction with a decision process is higher in parallel coordination groups and in groups with a leader. There is also a significant interaction effect. Satisfaction with the decision process is higher in sequential coordination groups with a leader than sequential coordination groups without a leader.     

The Role of the DS/AHP in Identifying Inter-Group Alliances and Majority Rule Within Group Decision Making

DS/AHP is a nascent method of multi-criteria decision-making, based on the Dempster-Shafer theory of evidence and indirectly the Analytic Hierarchy Process. It is concerned with the identification of the levels of preference that decision makers have towards certain decision alternatives (DAs), through preference judgements made over a number of different criteria. The working result from a DS/AHP analysis is the body of evidence (BOE), which includes a series of mass values that represent the exact beliefs in the best DA(s) existing within certain subsets of DAs. This paper considers the role of DS/AHP as an aid to group decision-making, through the utilisation of a distance measure (between BOEs). Here, the distance measure enables the identification of the members of the decision-making group who are in most agreement, with respect to the judgements they have individually made. The utilisation of a single linkage dendrite approach to clustering elucidates an appropriate order to the aggregation of the judgements of the group members. This develops the DS/AHP method as a tool to identify inter-group alliances, as well as introduce a ‘majority rule’ approach to decision-making through consensus building.     

The Effect of a Quantitative Decision Aid – Analytic Hierarchy Process – on Group Polarization

The objective of this paper is to examine group polarization in subjects who have access to a quantitative decision aid, comparing to subjects who do not. Group polarization indicates the opinion of a group involved in a decision process will tend to be more extreme in the direction of the norm than the initial opinions of its members. Most of previous studies were focused on the theoretical explanation of this effect. Two theories are generally proposed to explain it: social comparison (SC) and persuasive arguments (PA). As we know, no study has been worked on how it can be dealt with. In this study, we use Analytic Hierarchy Process (AHP) as a quantitative decision aid in a group process to examine the changes of this effect. Two hundred and forty undergraduate students participated in this study. Results show that persuasive argument was significantly reduced by the use of AHP, while no significant effect was found in group polarization and social comparison. The implications of the findings are further discussed.     

A Comparison of Full Profile and Hierarchical Information Integration Conjoint Methods to Modeling Group Preferences

This paper presents a comparative test of full profile (FP), original hierarchical information integration (HII-O), and integrated hierarchical information integration (HII-I) conjoint methods for modeling group preferences. It is hypothesized that in settings where groups need to decide about complex multi-attribute alternatives, HII-O will predict holdout profiles better than FP, and that HII-I will perform better than HII-O. The predictive ability of the three methods is tested for the case of housing preferences of housing co-ops, which are groups consisting of three to five people who jointly own a house. The results confirm that HII-I outperforms the other two methods and further suggest that FP and HII-O perform equally well. In addition, two variations of HII-I are developed. One of these provides independent estimates of the relative influence of group members on the decision outcomes, however at the expense of requiring larger designs. The two HII-I variations are also tested and found to be equivalent in predictive ability.     

An Integrated Group Decision-Making Method Dealing with Fuzzy Preferences for Alternatives and Individual Judgments for Selection Criteria

Organizations often require decisions to be made by a group, and decision makers often have fuzzy preferences for alternatives and individual judgments when attempting to reach an optimal solution. In order to deal with the fuzziness of preference of decision makers, this paper proposes an integrated fuzzy group decision-making method. This method allows group members to express fuzzy preferences for alternatives and individual judgments for solution selection criteria. It also allowed for the weighting of group members. The method then aggregates these elements into a compromise group decision which is the most acceptable for the group as a whole. This method has been implemented and tested. An example is presented to illustrate the method.     

The Effects of Integrating Cognitive Feedback and Multi-attribute Utility-Based Multicriteria Decision-Making Methods in GDSS

Cognitive conflicts arise within groups because the members of a group view a problem from different perspectives, even when they have similar interests in achieving a goal. Disagreement within a group may occur due to: (a) differing judgment policies among the members, (b) inconsistency by any member in using a judgment policy, (c) group process losses that prevent group members from understanding each other better, or (d) limited processing capability which may prevent group members from processing all information effectively. Disagreement is especially likely when policies, processes, or information are conflicting in nature.A level 2 GDSS to aid judging in cognitive conflict tasks is presented that combines cognitive feedback and Multi-attribute utility (MAU) theory based multicriteria decision-making techniques with the communication structure and activity-structuring capabilities of a level 1 GDSS. Though cognitive feedback and MAU methods have been used separately to help groups resolve cognitive conflicts, never before have the two decision aids been used together in a computer-based collaborative system.The contributory effects of the components of this GDSS design were empirically tested in a laboratory setting. Three treatments: an unaided face-to-face meeting, a level 1 GDSS supported meeting, and a level 2 GDSS supported meeting were compared in a repeated measures experimental design.Results largely supported the proposed research hypotheses. Some specific findings include: (1) the level 2 GDSS reduced disagreement between group members and improved consistency of judgments better than the other meeting environments did; (2) there was no significant difference in the reduction of disagreement between the level 1 GDSS and face-to-face meetings; and (3) while there was no difference in improvement of consistency of individual judgments between the face-to-face and level 1 GDSS supported meetings, group judgments made in face-to-face meetings were more consistent.     

Aggregation and Disaggregation of Preferences for Collective Decision-Making

Group decisions are of longstanding interest to researchers from a wide spectrum of disciplines. Group Decision Support Systems (GDSS) can play a vital role in situations where multiple persons are involved, each having their own private perceptions of the context and the decision problem to be tackled. In such an environment the conflict between the members of the planning group is not an unusual situation. Multiple criteria decision aid (MCDA) methods may be a useful tool in coping with such interpersonal conflicts where the aim is to achieve consensus between the group members. This paper combines two well-known multicriteria methods, based on the notion of aggregation of preferences, in order to construct a consensus seeking methodology for collective decision-making.     

Electronic Meetings and Intense Group Conflict: Effects of a Policy-Modeling Performance Support System and an Audio Communication Support System on Satisfaction and Agreement

Ad-hoc decision teams were used to examine the effects of an electronic meeting system (EMS) on group satisfaction and agreement. The decision task provoked intense conflict of values. The EMS had two core features - a policy-modeling group performance support system (incorporating structured decision methods and computer-supported cognitive feedback using Multi-Attribute Utility Analysis and Social Judgment Analysis), and an audio-based group communication support system (allowed dispersed members to communicate by voice). Policy groups reached higher agreement than conventional decision-making groups, apparently due primarily to the structure for cognitive-conflict tasks that was imposed on group discussion rather than computer-supported cognitive feedback displays. Audio groups were more satisfied with the conflict process than face-to-face groups. Decision agreement was equivalent across the two media. These audio effects for a highly equivocal task represent a further challenge to media richness theory.     

Internet Multiattribute Group Decision Support in Electronic Commerce

The rapid growth of the Internet has provided the means for distributed organizational decision making for electronic commerce. Members of organizations can jointly investigate products, exchange information, and make decisions on-line from remote sites. Internet-based multiattribute group decision making is characterized by three aspects, (i) individual interactive decision making, (ii) communication means, and (iii) group consensus reaching. The purpose of this research was to study the role of communication and individual decision strategies and their influence on multiattribute group decision making and consensus reaching in organizational electronic commerce settings. The results of this study indicate that analytic decision support is indispensable in collaborative Internet-based decision making, that a perfect match of analytic decision support and communication channels must be achieved, and that efficiency of individual decision support should be compromised for higher confidence in the group's decisions. The results of this study also confirm findings by Häubl and Trifts (2000) that interactive decision analytic support has positive effects on the quality and efficiency of individual decision making, and findings by Limayem and DeSanctis (2000) and Todd and Benbesat (2000), that decision makers will use normative decision models if they require little effort and if decisional guidance is provided. The conclusion drawn from this study is that the continuously evolving Internet technology for collaborative decision making is only one aspect for better organizational decision making – the crucial aspect, however, will be the development and optimal integration of analytic decision models, communication channels, and consensus reaching mechanisms.     

PROMETHEE Group Decision Support System and the House of Quality

Quality function deployment (QFD) is a multi-step method that monitors customer needs throughout a product development process. The House of Quality (HOQ) exercise undertaken in the first phase of QFD is considered as the most important, since customer needs must be accurately translated into a set of technical requirements for the final product. This paper provides a PROMETHEE group decision support system (GDSS) approach that integrates the design preferences of the QFD team. We highlight the selection and ranking of the technical requirements in the HOQ exercise, where a group of multidisciplinary decision makers (DMs) in a globally dispersed QFD team is required to input their individual preferences. Our approach advances the HOQ group decision making context in three important areas. First, it treats each criterion and DM as unique in terms of the preference function and threshold levels. Second, it seeks a multi-criteria approach for the HOQ process, where some DMs may play a more important role than others on a certain criterion. Third, sensitivity analysis through the Geometrical Analysis for Interactive Assistance (GAIA) plane provides valuable information about the conflicts, similarities, or independencies between the criterion and the DMs, respectively. A case on an automotive part illustrates the performance of the PROMOTHEE approach with GAIA.     

Technology and Group Decision Process in Going-Concern Judgements

Accounting and auditing practices are continually being affected by advances in technology. This study empirically examined the effect of group decision processes and technological advances on group going-concern decision making. Groups with access to group decision support systems (GDSS) were compared to groups without access to GDSS for their going-concern judgments. The results show group discussion induced auditors to be more conservative and to consider factors which may have overlooked at the individual level, though neither structure significantly reduced the considerable variance in the individual going-concern judgments. Further, as compared to their counterparts in the face-to-face discussion groups, GDSS groups indicated much higher confidence in their group's final assessment of the client's going-concern status and a higher level of satisfaction and agreement with the group decision processes. The findings suggest that while group discussions did not significantly reduce auditors' considerable variance in going-concern judgments, future research should investigate which explicit models would improve the consensus on going-concern evaluations.     

Influence of Computer-Based Meeting Support on Process and Outcomes for a Divisional Coordinating Group

This research examines the introduction of computer-based group decision support systems (GDSS) to members of a division level coordinating group. Participants performed authentic problem formulation tasks which varied naturally in degree of structuredness, in two non-GDSS meetings then were provided GDSS for four additional meetings. It was proposed that the introduction of the GDSS and task structuredness would influence (1) group process in terms of the amount of divergent and convergent thinking communicated during meetings and (2) perceived outcomes regarding quality, satisfaction, understanding, confidence, and commitment to group positions. Results suggest that GDSS use affected both the total amount and pattern of group communication but not perceived outcomes. Structuredness of the task affected perceived outcomes but neither amount nor patterns of communication. Additional qualitative data regarding perceived of costs and benefits in using GDSS provide richer explanation for study findings and suggest further lines of inquiry.     

Reaching informed agreement in multispecialist cooperation

Very often, complex decisions must be made by a group of specialists rather than a single decision maker. To make an effective decision, the combination of the group's expertise must be brought to bear on the situation. Fusing expertise where individuals have very detailed knowledge in their own areas and much weaker understanding of others is characterized by many difficulties: (1) agents cannot communicate their expertise in an intelligible way to nonexperts because of differences in vocabulary and conceptual content; (2) the process allows for incorrect inferences; and (3) no one knows what anyone else needs to know. This impasse cannot be broken until shared mental models are developed to provide a level of agreement in evaluating alternatives needed to focus the activity of the group. This article presents a model of decision making by teams of specialists in which agents' evaluations confound expert and naive inferences in judging alternatives. A partitioning of agent knowledge into expert and naive models is proposed. The naive portion of agents' models provides both a common language and the inferential skeleton needed for the development of shared models. Communications are categorized into types of evaluation or justification based on their form and the entities they involve within the agent models. A process of model refinement is outlined, linking communications among agents to modifications of the naive/ shared portions of their models. The process of cooperative problem solving by a team of specialists is characterized as a search among alternatives in which model refinement continually alters the agents' evaluations, leading to progressively greater accuracy and more precisely directed search. The model is intended as a research tool for investigating multiagent problem solving among people and machines.     

The Role of Information Feedback in the Management Group Decision-Making Process Applying System Dynamics Models

The research project addresses the influence of feedback information on the decision process supported by the application of system dynamics models. A user-friendly application was developed and used in the experiment with decision groups. The participants were 174 undergraduate management science students. They had the task of determining the optimum business strategy by maximizing the multiple criteria function under three experimental conditions: a ₁) an indivIDual decision process without the support of a system dynamics model, a ₂) an indivIDual decision process supported by a system dynamics model, and a ₃) a decision process supported by a system dynamics model and subject interaction via computer mediation. The hypotheses that the indivIDual decision process supported by a system dynamics model yields higher Criteria Function values than one without a system dynamics model, as well as the decision process supported by both a system dynamics model and subject interaction yields higher Criteria Function values than one supported by a system dynamics model alone were confirmed.     

Monte Carlo Simulation Techniques in a Decision Support System for Group Decision Making

This paper describes a group decision support system based on an additive multi-attribute utility model for identifying a consensus strategy in group decision-making problems where several decision-makers or groups of decision-makers elicit their own preferences separately. On the one hand, the system provides procedures to quantify the DMs or group of DMs preferences separately. This involves assessing the DMs or group of DMs component utilities that represent their preferences regarding the respective possible attribute values and objective weights that represent the relative importance of the criteria. On the other hand, we propose Monte Carlo simulation techniques for identifying a consensus strategy. An iterative process will be carried out, where, after the simulations have been performed, the imprecise component utilities and weights corresponding to the different DMs or groups of DMs are tightened to output more meaningful information in the next simulations to achieve a consensus strategy. Finally, an application to the evaluation of remedial strategies for restoring contaminated aquatic ecosystems illustrates the usefulness and flexibility of this decision support tool.