An integrated toolset for ontology-guided diagnostic knowledge discovery

Abstract

Today's diagnostic systems can generate a large amount data. Data from sources such as onboard reasoners and historical maintenance data are often stored in heterogeneous systems and cannot be collected immediately and aggregated for use. In our previous work we described a software visualization tool that allowed integration of different data sources and displayed the data with elements organized according to maintenance-oriented ontologies. This tool allows users to search quickly through available data to locate interesting relationships in the sequences of maintenance events. Additional previous work described a diagnostic maturation tool, called ModelMat, that updates causal relationships in a Timed Failure Propagation Graph based on historical diagnostic session data. In this paper, we present an update to both of these projects discussing enhancements to each as well as work in progress to create a single, integrated toolset, called Bobcat, to support ontologyguided diagnostic knowledge discovery.

Implementing AI-ESTATE with prognostic extensions in Java

“…This would allow for tighter, standards-based integration across the complete process for testing and maintenance. To that end, we could also look into how SAPPHIRE and Montana State University's Bobcat knowledge discovery toolset [33] could be integrated together to allow for greater coupling between the diagnostic and maintenance processes, such as is needed for model maturation.…”