International Journal of Information Technology & Decision Making

There are many parameters in challenging decision problems, including the alternatives, the criteria, resources, the perception of decision problems, decision makers and their psychological states, information inputs from the environment, and self-suggestion, etc. At any moment of time, some of these parameters can catch our attention, called alerted parameters; some cannot, called unalerted parameters. Some parameters are visible, some are invisible. In addition, the parameters themselves can vary over certain ranges or domains. All of these make challenging decision problems very complex. We call this kind of problems as decision problems with changeable spaces (parameters).

We may focus on certain parameters with certain assumed values to find an "optimal" solution, which may lead to solve wrong problem with bad solution. Quite often, our focus may be just a small part of what we know, or just a part of what we are most familiar with. We may often neglect what we are not familiar with, and pay no attention to what we do not know. As a consequence, we may see just a small part of the problem domain (including all parameters and their possible variations over time). The portion (of the problem domain) that we cannot see is our decision blind. Suppose our alerted domain (those parameters and their variations that are currently under our consideration) to be fixed in only a small part of the problem domains. Then very likely we could end up with a serious mistake. This situation is known as decision trap.

In this article, we will introduce a systematic scheme, based on habitual domain theory, to help us reduce decision blinds and avoid decision traps so that we could make decision with good quality. Then we will also introduce the concept of competence set analysis to help us cope with challenging decision problems. This including: (i) how to effectively expand our competence (resources, skill, know-how, information, ideas, effort, etc.) as to solve a given problem effectively; and (ii) given a set of competence, how to maximize its value by solving a set of value added problems. Furthermore, we will introduce innovation dynamics which describe the dynamics of how to solve a set of problems with our existent or acquired competence (to relieve the pains or frustrations of "certain customers or decision makers" at certain situations) as to create value, and how to distribute this created value so that we can continuously expand out competence set to solve more challenging problems and create more value.

Based on the notions of bipolar lattices and L-sets, YinYang bipolar fuzzy sets and fuzzy equilibrium relations are presented for bipolar clustering, optimization, and global regulation. While a bipolar L-set is defined as a bipolar equilibrium function L that maps a bipolar object set X over an arbitrary bipolar lattice B as L:X ⇒ B, this work focuses on the unit square lattice B_F = [-1, 0] × [0, 1]. A strong or weak bipolar fuzzy equilibrium relation in a bipolar set X is then defined as a reflexive, symmetric, and bipolar interactive (or transitive) fuzzy relation μ_R: X ⇒ B_F. Three types of bipolar α-level sets are presented for bipolar defuzzification and depolarization. It is shown that a fuzzy equilibrium relation is a non-linear bipolar generalization and/or fusion of multiple similarity relations, which induces disjoint or joint bipolar fuzzy subsets including quasi-coalition, conflict, and harmony sets. Equilibrium energy and stability analysis can then be utilized on different clusters for optimization and global regulation purposes. Thus, this work provides a unified approach to truth, fuzziness, and polarity and leads to a holistic theory for cognitive-map-based visualization, optimization, decision, global regulation, and coordination. Basic concepts are illustrated with a simulation in macroeconomics.

This paper proposes an enhanced interactive satisfying optimization method based on goal programming for the multiple objective optimization problem with preemptive priorities. Based on the previous method, the approach presented makes the higher priority achieve the higher satisfying degree. For three fuzzy relations of the objective functions, the corresponding optimization models are proposed. Not only can satisfying results for all the objectives be acquired, but the preemptive priority requirement can also be simultaneously actualized. The balance between optimization and priorities is realized. We demonstrate the power of this proposed method by illustrative examples.

Continuing attempts to align science and technology policies with industrial and societal needs have aroused interest in the determination of research priorities. In this paper, we report a case study where leading experts from industry and public administration were assisted by multicriteria decision analysis in the planning of a collaborative research program for Scandinavian forestry and forest industries. We also address processual and methodological challenges in the deployment of multicriteria methods, and argue that such methods can contribute to the quality of decision support processes in related contexts.

This paper presents a fuzzy set-based approach to the evaluation of information technology (IT) projects. We assume a multi-criteria decision-making framework, where sets of general and domain-specific criteria are used to judge the relative performance of alternative technologies. The methodology was originally developed for DIAS.net, an EU project aiming at the development of the Information Society in insular and isolated regions of Europe. In this paper, we present many aspects of our evaluation framework, including the synthesis of evaluation teams, the assessment of the importance of criteria, the evaluation of the performance of the alternatives and the final ranking and selection of projects. The methodology presented has the innovative feature of embodying techniques of fuzzy sets theory into the classical multi-criteria decision analysis. This combination enables us to handle efficiently the subjectiveness that often characterizes expert judgements on a decision problem. Fuzzy linguistic terms, such as "poor," "fair," "very important," etc. are proposed for assessing the relative merit of alternatives and criteria. The paper concludes by exploring the potentiality of the above methodology in providing a flexible and robust IT evaluation framework.

This paper presents N-Site, a distributed consensus building and negotiation support system, which is used to provide geographically dispersed teams with agile access to a Web-based group decision support system. Four teams located in France, Mexico, the Ukraine, and the United States participated in the N-Site project. Each team was required to research the problem using the World Wide Web (WWW). With this background, each team identified opportunities, threats and alternatives as a basis for developing a response to the Cuban Missile Crisis that confronted President Kennedy in October 1962. The strategic assessment model (SAM) (M. Tavana, J. Multi-Criteria Decision Anal.11 (2002) 75–96; M. Tavana and S. Banerjee, Decision Sci.26 (1995) 119–143.) was used by each team to choose a strategy that best fit the team's perspective. SAM and WWW enabled the teams to evaluate strategic alternatives and build consensus based on a series of intuitive and analytical methods including environmental scanning, the analytic hierarchy process (AHP) and subjective probabilities. The WWW was used to achieve interaction among the international teams as they attempted to negotiate a decision framework and select a diplomatic response. The project was assessed with a Web-distributed survey instrument. This use of the WWW has implications for international diplomacy as well as global business.

Quantitative models explaining and forecasting the growth of new technology like the Internet in global business operation appear infrequently in the literature. This paper introduces two artificial intelligence (AI) models such as the neural network and fuzzy regression along with an augmented diffusion model to study and predict the Internet growth in several OECD nations. First, a linear version of an augmented diffusion model is designed. An augmented diffusion model is constructed by including an economic indicator, gross domestic product per capita, into the model. In the next step, two soft AI models are calibrated from the augmented diffusion model. Performance measures of predictions from these models on new samples show that these soft models provide improved forecast accuracy over the augmented diffusion model. The results confirm the major contribution of this research in predicting global Internet growth.

Safety-critical wide area networks are comprised of human and technical elements cooperatively performing tasks in a safety-critical setting. In such settings, both human and technical dimensions are critical in performance evaluation. It is the relationships between changes in network performance and their impacts on human operator performance with safety-critical wide area networks that we investigate in this research. The paper begins by examining the theoretical background for this research, and then describes the research model. It was found that decreases in network reliability were associated with expected degradations in operator satisfaction, operator confidence, and increases in operator workload. Likewise, it was found that an increased number of network tasks processed was associated with a decrease in operator accuracy and decreased operator communication. The results of this study suggest that understanding the impact of the network performance on human performance is important in safety-critical settings employing wide-area networks.

Discounted cash flow techniques are the generally accepted methods for valuing firms. Such methods do not provide explicit acknowledgment of the value determinants and overlook their interrelations. This paper proposes a different method of firm valuation based on fuzzy logic and expert systems. It does represent a conceptual transposition of discounted cash flow techniques but, unlike the latter, it takes explicit account of quantitative and qualitative variables and their mutual integration. Financial, strategic and business aspects are considered by focusing on 29 value drivers that are combined together via "if–then" rules. The output of the system is a real number in the interval [0, 1], which represents the value-creation power of the firm. To corroborate the model a sensitivity analysis is conducted. The system may be used for rating and ranking firms as well as for assessing the impact of managers' decisions on value creation and as a tool of corporate governance.

Stock selection rules are extensively utilized as the guideline to construct high performance stock portfolios. However, the predictive performance of the rules developed by some economic experts in the past has decreased dramatically for the current stock market. In this paper, C4.5 decision tree classification method was adopted to construct a model for stock prediction based on the fundamental stock data, from which a set of stock selection rules was derived. The experimental results showed that the generated rules have exceptional predictive performance. Moreover, it also demonstrated that the C4.5 decision tree classification model can work efficiently on the high noise stock data domain.