An Ontology-Based Expert System to Detect Service Level Agreement Violations

Abstract

In this paper, an expert system developed with an ontology-based approach to detect Service Level Agreement (SLA) violations is presented. The widespread use of SLAs in various areas complicates SLA management and in particular the detection of violations. Although it is necessary to automatically detect SLA violations, developing a different solution for each domain is quite costly. Several domains were investigated, and many common concepts have been identified in terms of SLAs. Nevertheless, it has been determined that each domain has its own distinct metrics and criteria. By combining familiar and changeable concepts, we have acquired the idea of creating a generic SLA ontology for different SLA domains. After generic SLA ontology was created, an expert system called SLAVIDES was developed using this ontology. The developed expert system is designed to detect SLA violations, check constraints, and make inferences. The developed system has been tested on the SLA data of the telecommunication domain. The results show that the proposed system can correctly detect SLA violations.