FUNCTION-BASED CANDIDATE DISCRIMINATION DURING MODEL-BASED DIAGNOSIS

Abstract

We propose function for candidate discrimination, i.e., suspect ordering during model-based diagnosis. Function offers advantages over structure and fault probabilities currently being used for candidate discrimination. It is readily available from device design, unlike fault probabilities, which are hard to obtain. Function-based discrimination is not dependent on the topology of the device, unlike structure-based discrimination. We propose classes as a scheme for representation of function. As part of classes, we define a set of function primitives and provide a framework for identifying the functions of components and subsystems of a device. The representation scheme is domain independent. We propose a function-based technique for candidate discrimination called the default order technique, and outline a diagnosis algorithm that applies the technique to the class model of a device. Function-based diagnosis is in addition to and as a supplement for model-based diagnosis based on behavior and structure. We demonstrate by qualitative analysis that function-based discrimination is at least as effective as fault probabilities for candidate discrimination of simple devices. In complex devices, function facilitates explanation generation based on causality, which is a desirable feature of diagnosis systems. Our discrimination technique provides a functional basis for partitioning components in the practicable version of the minimum entropy technique proposed by deKleer.