Evaluating Knowledge Management Techniques to Augment Network Fault Diagnosis

This research in progress proposes to evaluate the effectiveness of using Knowledge Management (KM) approaches in conjunction with Decision Support Systems (DSS). The authors performed a preliminary field survey at a global optical telecommunications network that is characterized by dispersed groups of collaborating engineers and other stakeholders. The results indicate alarm correlation has been a major challenge for telecommunications systems. A DSS should theoretically be able to leverage expert engineering knowledge using KM techniques to alleviate these challenges. This research will evaluate two specific DSS/KM approaches for capturing and sharing this engineering knowledge across independent network nodes. A rule-based approach allows local field engineers to expand the system’s knowledge base. In addition, the organization’s Research & Development engineers will distribute a knowledge base model that includes a behavioral model of the equipment. The models will then be combined with an inference engine to perform root-cause analysis of network faults.