Explaining Novel Events in Process Control through Model Generative Reasoning
Michael J. Coombs, Roger T. Hartley · 1999
ABSTRACT: This article discusses the application of an abductive reasoning method termed Model Generative Reasoning (MGR) to the construction of explanations for novel events in physical systems. The MGR algorithm progressively develops intensional domain descriptions (models) to cover problem assumptions, which are then evaluated against domain facts as alternative explanations of queried phenomena. Illustrations of the workings of the MGR approach are given using concepts from process control.