Intelligent Support of Decision Making in Management of Large-Scale Systems Using Case-Based, Rule-Based and Qualitative Reasoning over Ontologies

Abstract

The current trend in intelligent support of decision making is an integration of different knowledge representation models and reasoning mechanisms, it allows improving quality and efficiency of obtained decisions. In this paper, we present an ontology-based approach to intelligent support of decision making in the management of large-scale systems using case-based, rule-based and qualitative reasoning. A concept of the reasoning mechanisms integration implies that case-based reasoning (CBR) takes on the role of leading reasoning mechanism, while rule-based (RBR) and qualitative reasoning (QR) support the different phases of CBR-cycle - adaptation and revision phases respectively. The paper describes a modified CBR-cycle and ontological knowledge representation model which supports the proposed concept of reasoning integration. A formal qualitative model of decision making was developed for revision of case solution, it includes the following components: system state model, action model, and assessment model. An ontological representation of the qualitative model was proposed for integration with structural case model in an ontological knowledge base. Implementation of the proposed approach is illustrated by a number of examples of decision making support in various subject domains.