Artificial intelligence in simulation

The aims are to clarify some basic concepts on cognizant simulation as well as on the similarity and the relationships of AI and simulation; provide an inventory and a taxonomy of AI-based (cognizant) simulation and AI-assisted simulation (cognizant simulation environments); and to indicate some desirable research directions.     

Artificial intelligence and social forecasting*

Artificial intelligence (AI) is the study of how to write programs enabling computers to do things that would require intelligence if done by people, and it could engage with social forecasting in two ways. First, it is part of the overall social‐technological context within which forecasters work. Commercial Al‐programs will affect markets and life‐styles; and advice‐giving “expert” systems will raise novel legal, social, and psychological problems. Second, AI‐programs might be used for making the social forecasts. Unlike the (essentially quantitative) computer models used for this purpose today, they could reason (and explain themselves) in verbal form. Writing an expert system requires clarification of the theories, assumptions, and “rule‐of‐thumb” inferences concerned. It would be easier to identify the inherent moral‐political bias than it is in models comprising sets of differential equations.     

Intention to use business intelligence tools in decision making processes: applying a UTAUT 2 model

Central European Journal of Operations Research - The pressure on the speed of information processing ranks business intelligence technologies among the fastest growing decision support tools. The...     

Wisdom,decision support and paradigms of decision making

Many decision support tools have been developed over the last 20 years and, in general, they support what Simon termed substantive rationality. However, such tools are rarely suited to helping people tackle wicked problems, for which a form of procedural rationality is better suited. Procedurally rational approaches have appeared in both management science and computer science, examples being the soft OR approach of cognitive mapping and the design rationale based on IBIS. These approaches are reviewed and the development of Wisdom, a procedurally rational decision support process and accompanying tool, is discussed and evaluated.     

Graph-based multi-agent decision making

In lots of practical multi-criteria decision making (MCDM) problems, there exist various and changeable relations among the criteria which cannot be handled well by means of the existing methods. Considering that graphic or netlike structures can be used to describe the relationships among several individuals, we first introduce the graphic structure into MCDM and formalize the relations among criteria. Then, we develop a new tool, called graph-based multi-agent decision making (GMADM) model, to deal with a kind of MCDM problems with the interrelated criteria. In the model, the graphic structure is paid sufficient attention to in two main aspects: (1) how the graphic structure has influence on the benefits of agents (or the criteria values); and (2) the relation between the graphic structure and the importance weights of agents (criteria). In this case, we can select the best plan(s) (or alternative(s)) according to the overall benefits (the overall criteria values) resulting from the model. Moreover, a fuzzy graph-based multi-agent decision making (FGMADM) method is developed to solve a common kind of situations where the graphic structure of agents is uncertain (confidential or false). Three examples are used to illustrate the feasibility of these two developed methods.     

Hierarchical multi-objective decision making

This paper proposes a hierarchical procedure for solving decision problems with multiple objectives. The procedure consists of two levels, a top- and a base-level. The main idea is that the top-level only provides general preference information. Taking this information into account the base-level then determines a compromise solution. For a multi-objective linear program it will be shown how such a hierarchical procedure can be structured by deriving weight restrictions from the general preference information of the top-level and by using the interactive MODM procedure of Zionts and Wallenius on the base-level.     

Search during decision making

Nearly four hundred non-routine organizational decisions were investigated to discover search approaches––determining the frequency of use and success of each search approach uncovered. A “search approach” is made up of a direction and a means to uncover solution ideas. Direction indicates desired results and it can be either implicit or explicit, with an explicit direction offering either a problem or a goal-like target. Solutions can be uncovered by opportunity, bargaining, and chance as well as by rational approaches. Defining a search approach as a direction coupled with a means of search, search approaches were linked with indicators of success, measured by the decision's adoption, value and timeliness, noting frequency. A rational, goal-directed, search approach was more apt to produce successful outcomes. Bargaining with stakeholders to uncover solutions was always combined some of the search approaches in this study, and this merger improved the prospects of success. Searches with an opportunistic or chance (emergent opportunity) features and rational searches with a problem target were more apt to produce unsuccessful outcomes. The means used to come up with a solution had less bearing on success than did the type of direction, with goal-directed searches leading to the best outcomes. Each search approach is discussed to reveal best practices and to offer suggestions to improve practice.     

Possibilistic sequential decision making

When the information about uncertainty cannot be quantified in a simple, probabilistic way, the topic of possibilistic decision theory is often a natural one to consider. The development of possibilistic decision theory has lead to the proposition a series of possibilistic criteria, namely: optimistic and pessimistic possibilistic qualitative criteria [7], possibilistic likely dominance [2], [9], binary possibilistic utility [11] and possibilistic Choquet integrals [24]. This paper focuses on sequential decision making in possibilistic decision trees. It proposes a theoretical study on the complexity of the problem of finding an optimal strategy depending on the monotonicity property of the optimization criteria – when the criterion is transitive, this property indeed allows a polytime solving of the problem by Dynamic Programming. We show that most possibilistic decision criteria, but possibilistic Choquet integrals, satisfy monotonicity and that the corresponding optimization problems can be solved in polynomial time by Dynamic Programming. Concerning the possibilistic likely dominance criteria which is quasi-transitive but not fully transitive, we propose an extended version of Dynamic Programming which remains polynomial in the size of the decision tree. We also show that for the particular case of possibilistic Choquet integrals, the problem of finding an optimal strategy is NP-hard. It can be solved by a Branch and Bound algorithm. Experiments show that even not necessarily optimal, the strategies built by Dynamic Programming are generally very good.     

Q-analysis for modeling and decision making

This paper provides an overview of existing and potential applications of a system-theoretic approach called Q-analysis, using the examples of design and analysis of expert systems in medical image processing and analysis: namely the organization of a histopathologic knowledge base. Q-analysis is also applied to a multicriterion decision-making (MCDM) problem using a method called multicriterion Q-analysis (MCQA). A brief discussion of the advantages and limitations of Q-analysis is given, with suggestions for further applications.     

Cost-sensitive learning and decision making revisited

In many real-life decision making situations the default assumption of equal misclassification costs underlying pattern recognition techniques is most likely violated. Then, cost-sensitive learning and decision making bring help for making cost-benefit-wise optimal decisions. This paper brings an up-to-date overview of several methods that aim to make a broad variety of error-based learners cost-sensitive. More specifically, we revisit direct minimum expected cost classification, MetaCost, over- and undersampling, and cost-sensitive boosting.     

Stochastic multicriteria decision making and uncertainty

We define the notion of stochastic multicriteria decision problem to take into account uncertainty in the data. A general approach is proposed to analyse these problems. As a special case, project evaluation by experts is considered. Stochastic independence problems are discussed and the notion of expected preference function is defined to introduce a stochastic extension of the Promethee outranking method.     

Self-consistency of decision rules for group decision making

The author treats, in this paper, a group of decision makers, where each of them already has preference on a given set of alternatives but the group as a whole does not have a decision rule to make their group decision, yet. Then, the author examines which decision rules are appropriate. As a criterion of “appropriateness” the author proposes the concepts of self-consistency and universal self-consistency of decision rules. Examining the existence of universally self-consistent decision rules in two cases: (1) decision situations with three decision makers and two alternatives, and (2) those with three decision makers and three alternatives, the author has found that all decision rules are universally self-consistent in the case (1), whereas all universally self-consistent decision rules have one and just one vetoer in the essential cases in (2). The result in the case (2) implies incompatibility of universal self-consistency with symmetry. An example of applications of the concept of self-consistency to a bankruptcy problem is also provided in this paper, where compatibility of self-consistency with symmetry in a particular decision situation is shown.     

Local decision making and decision fusion in hierarchical levels

Hierarchical problem solving is preferred when the problem is overwhelmingly complicated. In such a case, the problem should better be analyzed in hierarchical levels. At each level, some temporary solutions are obtained; then a suitable decision fusion technique is used to merge the temporary solutions for the next level. The hierarchical framework proposed in this study depends on reutilization or elimination of previous level local agents that together perform the decisions due to a decision-fusion technique: a performance criterion is set for local agents. The criterion checks the success of agents in their local regions. An agent satisfying this criterion is reutilized in the next level, whereas an agent not successful enough is removed from the agent pool in the next level. In place of a removed agent, a number of new local agents are developed. This framework is applied on a fault detection problem.      

Energy models for decision making

Since the 1973 oil crisis there has been an explosion in energy modelling activities throughout the world. A bewildering array of models have been or are being developed. The models differ in their geographical scope (local, national, regional or global), their technical scope (a process, an industry, an energy source or all energy industries), their timescale (one to one hundred years) and their systems boundaries (energy, economy, society). In addition there are crucial differences in the level of detail modelled, in the way time is structured and in the way in which decisions are included or perhaps excluded from the model.The natural questions to ask are what sort of models are now available; which models are suitable for which purposes; what has been learnt so far; which are the promising directions for future developments? This paper draws on modelling experience both within the National Coal Board and elsewhere to discuss these questions. The paper contains the following sections: energy models and decision making processes; choosing model boundaries; logic and facts, the basis of the model; choice of energy model; links to the economy; future directions. The subject will be of interest to specialist energy modellers and those interested more generally in strategic modelling for government and industry.     

Confidence structures in decision making

Decision making is defined in terms of four elements: the set of decisions, the set of outcomes for each decision, a set-valued criterion function, and the decision maker's value judgment for each outcome. Various confidence structures are defined, which give the decision maker's confidence of a given decision leading to a particular outcome. The relation of certain confidence structures to Bayesian decision making and to membership functions in fuzzy set theory is established. A number of schemes are discussed for arriving atbest decisions, and some new types of domination structures are introduced.This research was partly supported by Project No. NR-047-021, ONR Contract No. N-00014-75-C-0569 with the Center for Cybernetic Studies, The University of Texas, Austin, Texas, and by ONR Contract No. N-00014-69-A-0200-1012 with the University of California, Berkeley, California.     

Improving long-run industrial decision making

The paper describes the present status of new product evaluation and decision making in large and small companies. It discusses the need to draw from a wide range of sources for societal as well as financial and technical information. While deficiencies and the need for improved analytical capabilities are identified, the emphasis is on what can be done, and near term problems using as an example the decision to produce a new type of vehicle engine.