National Technical University of Athens (NTUA)
Πολυκριτηριακά συστήματα αποφάσεων σε ασαφές περιβάλλον και εφαρμογή τους στην αξιολόγηση της επιπροσθετικότητας έργων των μηχανισμών του προτοκόλλου του Κυότο
Abstract
dc:descriptionThe subject under study by this Ph.D. thesis is the research and analysis of the techniques for the modelling of imprecision in decision support systems and, especially in multicriteria decision problems. Using these techniques, multicriteria decision support methodologies are developed and applied to decision problems in which imprecision exists. The imprecision is not associated with likelihood and it cannot be modelled by probability theory because it has no stochastic features. The imprecision is originated from the absence of sharply defined criteria to determine the set membership rather than the presence of random variables. The imprecision in a decision problem means that the decision-maker(s) cannot exactly determine the numerical values of the problem’s parameters. The thesis analyses and investigates the following imprecision modelling techniques in a multicriteria decision problem: 1. Fuzzy numbers that can be used if the decision-maker(s) has the knowledge and the ability to determine membership functions suitable for the given context and if he can handle the complexity and computational effort of making arithmetic operations with fuzzy numbers. Moreover, a prerequisite for the use of fuzzy numbers is the existence of a mathematical model for the decision-making procedure or the ability to develop one. In this case, the fuzzy numbers can be incorporated in the mathematical procedure. 2. Ordinal scales which can be used if there is no knowledge and expertise to define membership functions. In this case too, it is necessary the decision-maker(s) to be able to formulate the mathematical model of their decision-making procedure. Knowing the mathematical model, it can be adapted to include values of the ordinal scale. 3. Fuzzy expert systems which have the ability to model the imprecision in the problem parameters with fuzzy sets and, additionally, they can represent and store human knowledge in a specific domain using fuzzy IF-THEN rules. The knowledge expressed through fuzzy rules is the decision-making procedure in a multicriteria problem. Furthermore, fuzzy expert systems emulate the human reasoning process and communicate with the experts in a “natural” language. They can be used if the mathematical model of the system or decision process is inexistent or it is too complex and severely non-linear. Considering the techniques for imprecision modelling, the research focuses on the development and study of: 1. A new fuzzy multicriteria method that uses fuzzy numbers for the ratings of the alternatives to the criteria and the criteria weights. The proposed method is based on the ideal and anti-ideal solutions and it also utilises distance and similarity measures for fuzzy numbers. Their aggregation is guided by logic-based decision rules in order to construct decision functions. 2. Another new multicriteria method that originates from the transformation of an arithmetic multicriteria method so as the new method to use criteria ratings and weights measured in an ordinal scale. The development of the proposed ordinal method is founded on the distance from the ideal and the anti-ideal solutions. It also necessitates the use of functions that have the ordinal scale as their domain and also their range so as to replace the usual arithmetic operations. These kinds of functions are investigated and applied in order to aggregate ratings and weights within the procedure of the proposed method. 3. A fuzzy expert system that utilises fuzzy IF-THEN rules. The fuzzy system models complex decision-making procedures which cannot be expressed in mathematical terms. It can be adapted and parameterized so as to model the decision-making procedure of the additionality evaluation for the projects under the Clean Development Mechanism of the Kyoto Protocol. The new fuzzy multicriteria method and the proposed ordinal method are applied to the additionality evaluation problem. Further, considering that a mathematical model for the decision-making procedure of this specific problem cannot be constructed, the fuzzy system is adapted to model this procedure. The application studies several candidate projects and their additionality degree is evaluated. The data source is the project labelled “Business Opportunities for CDM Project Development in the Mediterranean” No 4.104 l/D/02-003-507.21086 of the EU SYNERGY programme.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2007
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Αναγνωστόπουλος, Κωνσταντίνος
- Anagnostopoulos, Konstantinos
Subjects
dc:subject × 20- Πολυκριτηριακές μέθοδοι αποφάσεων
- Ασαφείς αριθμοί
- Μέτρα ομοιότητας
- Μέτρα απόστασης
- Διακριτή και διατεταγμένη κλίμακα
- Ασαφές έμπειρο σύστημα
- Ανακρίβεια
- Πρωτόκολλο του Κιότο
- Multicriteria decision methods
- Fuzzy numbers
- Similarity measures
- Distance measures
- Ordinal scale
- Fuzzy expert systems
- Imprecision
- Kyoto protocol
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/16892