Back to results list
Show full item record
Please use this identifier to cite or link to this item:
|Title:||Cognitive network process with fuzzy soft computing technique in collective decision aiding||Authors:||Yuen, Kevin Kam Fung||Degree:||Ph.D.||Issue Date:||2009||Abstract:||The multi-criteria and multi-expert decision aiding models investigate the problems of identifying candidates, analyzing the criteria, and selecting the best alternative(s) based on the aggregation of the perceptions and preferences of the group decision makers. Although many studies have investigated these problems, there are no conclusions as to a single decision model that can dominate others. Among the various well-known models, the Analytic Hierarchy Process (AHP) /Analytic Network Process (ANP) is popular, and is applied in various domains, although there are some limitations. The Cognitive Network Process (CNP) is developed on the improvement of AHP/ANP with the cognitive decision process.
The CNP model is one of the models of the multi-criteria and multi-experts decision aiding. It applies the interdisciplinary techniques of decision sciences, cognitive sciences and fuzzy soft computing, on the basis of the mathematical modeling development. The cognitive architecture of the CNP is mainly comprised of five processes: Problem Cognition Process (PGP), Cognitive Assessment Process (CAP), Cognitive Prioritization Process (CPP), Multiple Information Fusion Process (MIP), and Decisional Volition Process (DVP). In PGP, decision problems are formed as a Structural Assessment Network (SAN). In CAP, a Compound Linguistic Ordinal Scale (CLOS) model is proposed for the improvement of rating activities of the assessment. In CPP, a Cognitive Prioritization Operator (CPO) of a Pairwise Opposite Matrix (POM) is proposed to derive the utility set from the POM. In MIP, a Cognitive Style and Aggregation Operator (CSAO) model is proposed for selection of aggregation operators to aggregate the utility sets with respect to the attitudes or cognitive styles of the decision makers. In DVP, a valuation function of the utility sets is used to provide the decision solution. The framework of CNP includes primitive and extent types. The primitive type is a individual decision making model using linguistic variables represented by crisp numbers. The extent types include the notions of the collective judgments and fuzzy linguistic variables.
The main contribution of the CNP includes the mathematical developments of CLOS, POM, CPO, CSAO, fuzzy POM, and fuzzy CPO. The numerical analyses with the discussions of these concepts are performed respectively. Five cases selected from other publications illustrate the usability and validity of the CNP, with comparisons with the (fuzzy) AHP/ANP, and complementation with other decision models.
Like the impacts of AHP/ANP, the proposed CNP can be applied in many domains such as material management, transportation management, psychometrics, social sciences, business research, decision sciences, computer sciences, and engineering management. The CNP is the ideal alternative of the AHP/ANP.
|Subjects:||Hong Kong Polytechnic University -- Dissertations.
Decision making -- Methodology.
Decision making -- Mathematical models.
|Pages:||xix, 432, 51 p. : ill. ; 30 cm.|
|Appears in Collections:||Thesis|
View full-text via https://theses.lib.polyu.edu.hk/handle/200/4716
Citations as of May 22, 2022
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.