National Technical University of Athens (NTUA)
Τεχνικές ανάλυσης και παρακολούθησης ορθής λειτουργίας πληροφοριακών συστημάτων πολλαπλών επιπέδων
Abstract
dc:descriptionThis thesis deals with the design and development of goal model reasoning frameworks. The goal models aim to denote system requirements, system policies, and stakeholder objectives. The reasoning frameworks aim to a) assist on run-time system verification, and b) assist on the formation of remedial plans when one or more requirements are failed to be delivered. For the run-time system verification, the thesis proposes a bottom-up reasoning framework that utilizes fuzzy goal models proposed as an extension of standard goal models. More specifically, the stakeholders of a software system define sets of goals that the software system under review should comply with. This can be achieved by designing instances of a meta-model proposed as an extension of AND/OR goal trees with well defined formal semantics. These instances (i.e. models) can then be transformed to a set of weighted fuzzy logic programs that constitute the knowledge base of a fuzzy controller. Additionally, values for all observable characteristics of the system are collected by actively monitoring the system, and are transformed to fuzzy facts through a proper fuzzification process. Finally, a fuzzy inference engine is utilized in order to evaluate the degree of satisfaction for the various goals posed by the stakeholders, using run-time system information.Similarly, for the formation of remedial plans, the thesis proposes a top-down reasoning framework that utilizes SAT solvers and genetic algorithms. More specifically, the thesis proposes temporal dependencies between nodes as extensions to standard goal models, in order to represent sequences of possible remedial actions. The reasoning strategy evaluates these models, in order to devise optimal remedial plans when the system fails to comply with, or deliver its requirements. The objective is for these remedial plans to have a positive impact towards reconfiguring the system to a state that can either satisfy its requirements, or satisfice them within an acceptable degree. Finally, the thesis reports on evaluation experiments on the performance, scalability, and stability of the proposed reasoning frameworks, and identifies areas of future research.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2015
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Chatzikonstantinou, Georgios
- Χατζηκωνσταντίνου, Γεώργιος
Subjects
dc:subject × 10- Μοντέλα δένδρων στόχων
- Ασαφής μηχανισμός συμπερασμού
- Επαλήθευση ορθής λειτουργίας χρόνου εκτέλεσης
- Goal models
- Fuzzy reasoning
- Run-time verification
- 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/37516