Ieee Information Model Standards for Test and Diagnosis

In this paper we discuss the use of information modeling to develop information exchange standards and metrics for test and diagnostics. For example, fault information is being transferred from one diagnostic reasoner to another. How many attributes does a fault have? One, three, fifteen? How do these attributes relate to the fault, to diagnosis, and tests? If both reasoners do not understand the “model” of a fault, then there will not be an information exchange. Humans can, when they are aware of the differences, resolve the mismatch in information. Computer programs cannot “talk” until they resolve small differences in conceptual information. We present an overview of the current and future directions of the Artificial Intelligence Exchange and Service Tie to All Test Environments (AI-ESTATE) standards. We present a summary of the work completed so far on the diagnostic reasoner exchange standard and on the testability metrics standard. We address objectives, document organization, information modeling, service versus API specification, and other issues raised by the AI-ESTATE community. We also discuss the vision of the AI-ESTATE subcommittee in its work to integrate the AI-ESTATE information models and projects such as testability/diagnosability assessment and test/maintenance feedback.